Beck, David E. - Kann, Brian R. - Margolin, David A. - Vargas, H. David - Whitlow, Charles B - Improving Outcomes in Colon and Rectal Surgery (2019, CRC Press - Taylor & Francis Group)

Improving Outcomes in
Colon and Rectal Surgery

Improving Outcomes in
Colon and Rectal Surgery
Edited by
Brian R. Kann, MD, FACS, FASCRS
David E. Beck, MD, FACS, FASCRS
David A. Margolin, MD, FACS, FASCRS
H. David Vargas, MD, FACS, FASCRS
Charles B. Whitlow, MD, FACS, FASCRS
Department of Colon and Rectal Surgery
Ochsner Clinic Foundation
New Orleans, Louisiana
CRC Press
Taylor & Francis Group
6000 Broken Sound Parkway NW, Suite 300
Boca Raton, FL 33487-2742
© 2018 by Taylor & Francis Group, LLC

CRC Press is an imprint of Taylor & Francis Group, an Informa business
No claim to original U.S. Government works
Printed on acid-free paper
International Standard Book Number-13: 978-1-138-62683-6 (Pack- Hardback and eBook)
This book contains information obtained from authentic and highly regarded sources. While all reasonable efforts have been made to publish reliable
data and information, neither the author[s] nor the publisher can accept any legal responsibility or liability for any errors or omissions that may be
made. The publishers wish to make clear that any views or opinions expressed in this book by individual editors, authors or contributors are personal
to them and do not necessarily reflect the views/opinions of the publishers. The information or guidance contained in this book is intended for use
by medical, scientific or health-care professionals and is provided strictly as a supplement to the medical or other professional’s own judgement, their
knowledge of the patient’s medical history, relevant manufacturer’s instructions and the appropriate best practice guidelines. Because of the rapid
advances in medical science, any information or advice on dosages, procedures or diagnoses should be independently verified. The reader is strongly
urged to consult the relevant national drug formulary and the drug companies’ and device or material manufacturers’ printed instructions, and
their websites, before administering or utilizing any of the drugs, devices or materials mentioned in this book. This book does not indicate whether
a particular treatment is appropriate or suitable for a particular individual. Ultimately it is the sole responsibility of the medical professional to make
his or her own professional judgements, so as to advise and treat patients appropriately. The authors and publishers have also attempted to trace the
copyright holders of all material reproduced in this publication and apologize to copyright holders if permission to publish in this form has not been
obtained. If any copyright material has not been acknowledged please write and let us know so we may rectify in any future reprint.
Except as permitted under U.S. Copyright Law, no part of this book may be reprinted, reproduced, transmitted, or utilized in any form by any elec-
tronic, mechanical, or other means, now known or hereafter invented, including photocopying, microfilming, and recording, or in any information
storage or retrieval system, without written permission from the publishers.
For permission to photocopy or use material electronically from this work, please access www.copyright.com (http://www.copyright.com/) or con-
tact the Copyright Clearance Center, Inc. (CCC), 222 Rosewood Drive, Danvers, MA 01923, 978-750-8400. CCC is a not-for-profit organization that
provides licenses and registration for a variety of users. For organizations that have been granted a photocopy license by the CCC, a separate system
of payment has been arranged.
Trademark Notice: Product or corporate names may be trademarks or registered trademarks, and are used only for identification and explanation
without intent to infringe.
Library of Congress Cataloging-in-Publication Data
Names: Kann, Brian R., editor. | Beck, David E., editor. | Margolin, David A., editor. | Vargas, H. David., editor. | Whitlow, Charles
B., editor.
Title: Improving outcomes in colon and rectal surgery / edited by Brian R. Kann, David E. Beck, David A. Margolin, H. David Vargas,
Charles B. Whitlow.
Description: Boca Raton : CRC Press, [2018] | Includes bibliographical references and index.
Identifiers: LCCN 2018011083| ISBN 9781138626836 (pack - book and e-book : alk. paper) | ISBN 9781351816786 (e-book)
Subjects: | MESH: Colonic Diseases--surgery | Rectal Diseases--surgery | Treatment Outcome | Colon--surgery | Rectum--surgery
| Colorectal Surgery--methods
Classification: LCC RD543.C57 | NLM WI 650 | DDC 617.5/547--dc23
LC record available at https://lccn.loc.gov/2018011083
Visit the Taylor & Francis Web site at

http://www.taylorandfrancis.com
and the CRC Press Web site at
http://www.crcpress.com
Contents
Preface ix
Editors xi
Contributors xiii
1 Preexisting conditions 1
Arida Siripong and Farouq Manji
2 Preoperative bowel preparation 14

Aaron L. Klinger and David A. Margolin
3 Anesthesia and intraoperative positioning 18

W. David Sumrall, III and David E. Beck
4 Sepsis 29
Jennifer S. Beaty and Moriah E. Wright
5 Intraoperative anastomotic challenges 38

David E. Beck
6 Other intraoperative challenges 49

Jesus Flores and Sharon G. Gregorcyk
7 Optimizing use of electronic health records 61

Allison B. McCoy and Genevieve B. Melton-Meaux
8 Postoperative anastomotic complications 66

Traci L. Hedrick
9 General postoperative complications 74

J. Marcus Downs and Kristina K. Booth
10 Care paths and optimal postoperative management 87

Benjamin D. Shogan and Heidi K. Chua
11 Limitations of anorectal physiology testing 93

Thomas E. Cataldo and Syed G. Husain
12 Limitations of colorectal imaging studies 104

Andrew C. Matthews and Charles C. Matthews
13 Transanal endoscopy 121

Aaron L. Klinger and Brian R. Kann
v
vi Contents
14 Laparoscopy for colorectal disease 129

Bradley R. Davis and Kevin R. Kasten
15 Medical legal issues 141

Terry C. Hicks and David E. Beck
16 Miscellaneous conditions 148

Alexander T. Hawkins and M. Benjamin Hopkins
17 Outcomes and quality 155

Shaun R. Brown and Kurt G. Davis
18 Hemorrhoidal surgery 161

Jeffery Mino and Massarat Zutshi
19 Nonoperative therapy for hemorrhoidal disease 171

Joseph C. Adongay and Scott A. Brill
20 Surgery and nonoperative therapy of perirectal abscess and anal fistulas 177
Mohammed Iyoob Mohammed Ilyas and Craig A. Reickert
21 Surgery and nonoperative therapy of anal fissure 188

Mary T. O’Donnell and Cary B. Aarons
22 Surgery for pilonidal disease and hidradenitis suppurativa 195

John D. Hunter and Leander M. Grimm Jr.
23 Surgical treatment of fecal incontinence 207

Nicole M. Saur and Joshua I.S. Bleier
24 Surgery for rectal prolapse 224

Steven R. Hunt
25 Management of diverticulitis 235

David E. Beck, H. David Vargas, and Molly M. Ford
26 Abdominal surgery for colorectal cancer 244

Jason F. Hall and Rocco Ricciardi
27 Abdominal restorative surgery for rectal cancer 253

Charles C. Vining and Nijjia N. Mahmoud
28 Transanal approaches to rectal cancer 266

Charles B. Whitlow and Lara McKean Basté
29 Abdominoperineal resection 271

W. Brian Perry and Huisar Dao Campi
30 Management of rectal cancer after complete clinical response to neoadjuvant chemoradiotherapy 279
Rodrigo O. Perez and Laura Melina Fernandez
31 Indications and outcomes for treatment of recurrent rectal cancer and colorectal liver/lung metastases 288
Luanne M. Force and David J. Maron
32 Evaluation and management of peritoneal metastatic disease 296

James Fleshman and Katerina O. Wells
Contents vii
33 Chemotherapy for colon and rectal cancer 306

Jonathan Lu and Marc R. Matrana
34 Adjunctive treatment of rectal cancer with radiation and the adverse effects of radiation exposure
of the rectum 310
Roland Hawkins
35 Surgical management of ulcerative colitis 323

Shannon McChesney and Brian R. Kann
36 Surgery for Crohn disease 333

Emily Steinhagen and Sharon L. Stein
37 Evaluation and management of lower gastrointestinal bleeding 347

Arjun N. Jeganathan and Skandan Shanmugan
38 Ostomies 352
Danielle Pickham and Supriya S. Patel
39 Operative and nonoperative therapy for chronic constipation 365

Robert J. Sinnott, Michelle C. Julien, and Daniel E. Sarmiento
40 Colorectal trauma 380

Alison Althans and Scott R. Steele
41 Prevention and management of urologic complications after colorectal surgery 395

Jacob A. McCoy, J. Christian Winters, and Scott E. Delacroix Jr.
Index 407
Preface
Quality measures and outcomes are receiving greater atten- future of their specialty. In addition to reviewing the avail-
tion by the lay and medical communities. The occurrence or able literature, they have described their personal approach
mismanagement of complications often results in poor out- to complications in colorectal surgery. Numerous technical
comes, increased cost, and significant morbidity. Answering descriptions and highlights from multiple discussions held
the call for transparency and improvement requires action in surgical locker rooms, morbidity and mortality confer-
by all involved in the care of patients. Collection of objective ences, and the hallways of conferences and symposiums
data and quality measures allows documentation of opti- have been included. Using this approach, we hope this book
mal care and desired outcomes while identifying areas for will provide initial guidance to the less-experienced pro-
improvement. vider and stimulate additional thought and research to the
The goal of this book is to present the current knowledge more-experienced provider.
of outcomes, as well as the techniques for minimizing and The editors gratefully acknowledge the efforts of the
managing complications from the common diseases and many individuals who made this book possible. This book
procedures of this specialty. This information will aid pro- carries on the vision of previous editors and contributors
viders in optimizing care and encourage research in out- to the first two editions of Complications in Colon and
come and quality measurement. Rectal Surgery and Improved Outcomes in Colon and Rectal
Improving Outcomes of Colon and Rectal Surgery rep- Surgery.
resents the collaborative efforts of many individuals. The
contributing authors were selected for their knowledge of Brian R. Kann, MD, FACS, FASCRS
colorectal surgery and ability to present their surgical judg- David E. Beck, MD, FACS, FASCRS
ment and experience in written form. They represent a spec- David A. Margolin, MD, FACS, FASCRS
trum of experienced providers who have made significant H. David Vargas, MD, FACS, FASCRS
contributions to younger individuals who will shape the Charles B. Whitlow, MD, FACS, FASCRS
ix
Editors
Brian R. Kann, MD, FACS, David E. Beck, MD, FACS,

FASCRS earned his under- FASCRS was a distin-
graduate degree from Old guished graduate of the
Dominion University in United States Air Force
Norfolk, VA in 1993, fol- (USAF) Academy and
lowed by his medical degree attended medical school at
from Hahnemann University the University of Miami.
in Philadelphia, PA in 1997. He completed his resi-
He completed his internship dency in general surgery at
and residency in general sur- Wilford Hall USAF
gery, including a one-year Medical School, Lackland
research fellowship focusing AFB, Texas and completed
on transplant immunology, a fellowship in colorectal
at Cooper University Hospital, University of Medicine and surgery at the Cleveland Clinic Foundation, Cleveland,
Dentistry of New Jersey (UMDNJ)/Robert Wood Johnson Ohio. He is board certified in general and colon and rectal
Medical School in Camden, New Jersey in 2003, then went on surgery and is a fellow of the American College of Surgeons
to complete a residency in colon and rectal surgery at the and the American Society of Colon and Rectal Surgeons.
Ochsner Clinic in New Orleans, Louisiana in 2004. He has After retiring from the Air Force, where he was chairman
previously held academic appointments at Cooper University and residency program director of the Department of
Hospital, UMDNJ/Robert Wood Johnson Medical School, General Surgery at Wilford Hall USAF Medical Center and
where he was assistant professor of surgery and program the military consultant to the air force surgeon general in
director of the General Surgery Residency from 2004 to 2010, colon and rectal surgery, Dr. Beck joined Ochsner in 1993,
and at the University of Pennsylvania Perelman School of served as chairman of the Colon and Rectal Surgery
Medicine, where he was assistant professor of clinical surgery Department from 1995 to 2014, and is now chairman
and program director of the Colon and Rectal Surgery emeritus.
Residency from 2010 to 2014. In 2015, he returned to the Dr. Beck conducts research on colorectal diseases, has
Ochsner Clinic in New Orleans, where he currently serves as a authored and edited nine medical textbooks, and written
senior staff surgeon and associate program director of the over 300 scientific publications. He was the president of the
Colon and Rectal Surgery Residency. American Society of Colon and Rectal Surgeons (ASCRS)
Dr. Kann is certified by both the American Board of from 2010 to 2011 and is a member of many other medi-
Surgery and the American Board of Colon and Rectal cal associations. He served as a member of the Board of
Surgery. He is an active member of the American College Governors of the Ochsner Clinic Foundation from 2005
of Surgeons, the American Society of Colon and Rectal to 2012 and currently serves on the American Board of
Surgeons (ASCRS), the Association of Program Directors Colon and Rectal Surgery. Dr. Beck is on the editorial
in Colon and Rectal Surgery, and the Society for Surgery board of several prestigious medical journals. In addition
of the Alimentary Tract. He has served on numerous com- to his clinical and administrative duties, he is a professor
mittees within these organizations and has been selected of surgery at the University of Queensland, Australia and
as program chair for 2019 ASCRS national meeting. He is on the clinical faculty at the Louisiana State University
has published numerous original scientific manuscripts, and Uniformed Services University Schools of Medicine.
reviews, and book chapters in the field of colon and rectal Dr. Beck is listed in Who’s Who, Best Doctors in America,
surgery. While his practice spans the field of colon and rec- Good Housekeeping’s Top Cancer Doctors for Women, and
tal surgery, he has particular clinical interest in the surgical Best Surgeons in New Orleans. Dr. Beck is a nationally rec-
management of inflammatory bowel disease. ognized expert in inflammatory bowel disease, anal, rectal
xi
xii Editors
and colon cancer, stomas, adhesions, bowel preparation, highest score on the written examination administered by
sphincter saving surgery for cancer, laparoscopic surgery, the American Board of Colon and Rectal Surgery.
and postoperative pain management. After leaving fellowship training in New Orleans, Dr.
Vargas gained extensive clinical experience in both pri-
David A. Margolin, MD, vate practice and academic surgery all the while remain-
FACS, FASCRS earned his ing involved in medical education and postgraduate surgery
medical degree from the resident training. He has been active in local, regional, and
Medical College of Ohio in national professional organizations and has held positions of
Toledo and completed his leadership. He has authored multiple book chapters as well as
internship and residency scientific articles published in peer-reviewed surgical journals.
at Case Western Reserve While Dr. Vargas’ clinical practice encompasses the
University in Cleveland, breadth of Colon Rectal Surgery, in particular he has
Ohio. He completed his gained recognition for minimally invasive or laparoscopic
fellowship in colon and colorectal surgery which led to his recruitment in 2007 to
rectal surgery at Ochsner. the University of Kentucky College of Medicine as Chief of
Dr. Margolin is board cer- Colon Rectal Surgery.
tified in general surgery and colon and rectal surgery and After five years in Kentucky, he returned to New Orleans
has been on staff at Ochsner since the beginning of 2003. in 2012 and joined the Department of Colon and Rectal
Dr. Margolin is te current President of the American Surgery of the Ochsner Clinic where he now serves as Staff
Society of Colon and Rectal Surgeons and is listed in Best Surgeon and Program Director of the Colon Rectal Surgery
Doctors in America. He serves as Director of Colon and Fellowship. In addition, he is a Assistant Medical Director
Rectal Surgery Research. His professional interests of Endoscopy Services at Ochsner Medical Center.
include laparoscopic colon and rectal surgery, inflamma-
tory bowel disease, complex anorectal conditions and Charles B. Whitlow, MD,
incontinence. FACS, FASCRS earned his
medical degree from the
H. David Vargas, MD, University of Arkansas and
FACS, FASCRS graduated completed his internship
with distinction with a and residency at William
major in Religious Studies Beaumont Army Medical
from the University of Center in El Paso, Texas. He
Virginia in Charlottesville, completed his colon and
Virginia. He stayed at the rectal surgery fellowship at
University of Virginia and Ochsner. Dr. Whitlow is
graduated from the School board certified in general
of Medicine. He completed surgery and colon and rec-
a residency in General tal surgery and has been on staff at Ochsner since the Summer
Surgery and served as of 2002. Dr. Whitlow is the chairman of the Department of
Chief Surgical Resident at Colon and Rectal Surgeins. He is listed in Best Doctors in
the Lehigh Valley Hospital in Allentown Pennsylvania. Dr. America and is a Past President of the American Board of
Vargas then pursued additional training here in New Colon and Rectal Surgery. His particular areas of professional
Orleans performing a fellowship at the Ochsner Clinic in interest include transanal excision of large benign rectal
the specialty of Colon Rectal Surgery. He became board cer- tumors, sphincter-preserving surgery for rectal cancer, surgi-
tified (and has been re-certified) by the American Board of cal treatment of inflammatory bowel disease, laparoscopic
Surgery and the American Board of Colon Rectal Surgery. and robotic colon and rectal surgery, surgery for rectal pro-
Of note, Dr. Vargas distinguished himself by attaining the lapse, and treatment of anorectal fistulas.
Contributors
Cary B. Aarons MD, FACS, FASCRS Shaun R. Brown MD, FACS, FASCRS
Department of Surgery Department of Colon and Rectal Surgery
University of Pennsylvania Womack Army Medical Center
Philadelphia, Pennsylvania Fort Bragg, North Carolina
Joseph C. Adongay MD Huisar Dao Campi MD

Section of Colon and Rectal Surgery Department of Surgery
Grant Medical Center University of Texas Health Science Center
Columbus, Ohio San Antonio, Texas
Alison Althans BA Thomas E. Cataldo MD, FACS, FASCRS

Case Western Reserve University School of Medicine Department of Colon and Rectal Surgery
Cleveland, Ohio Brown University
Providence, Rhode Island
Lara McKean Basté MD
Heidi K. Chua MD, FACS, FASCRS
Ochsner Clinic Foundation Department of Colon and Rectal Surgery
New Orleans, Louisiana Mayo Clinic
Rochester, Minnesota
Jennifer S. Beaty MD, FACS, FASCRS
Bradley R. Davis MD, FACS, FASCRS
Department of Colon and Rectal Surgery, Inc.
Omaha, Nebraska
Carolinas Medical Center
Charlotte, North Carolina
David E. Beck MD, FACS, FASCRS
Department of Colon and Rectal Surgery Kurt G. Davis MD, FACS, FASCRS
Louisiana State University
Joshua I. S. Bleier MD, FACS, FASCRS
Department of Surgery Scott E. Delacroix, Jr. MD
University of Pennsylvania Department of Urology
Philadelphia, Pennsylvania Louisiana State University School of Medicine
Kristina K. Booth MD, FACS, FASCRS
Division of Colon and Rectal Surgery J. Marcus Downs MD, FACS, FASCRS
University of Oklahoma Health Science Center Department of Colon and Rectal Surgery
Oklahoma City, Oklahoma UT Southwestern Medical School
Dallas, Texas
Scott A. Brill MD
Department of Colon and Rectal Surgery Laura Melina Fernandez MD
OhioHealth Colon and Rectal Surgeons Angelita & Joaquim Gama Institute
Columbus, Ohio São Paulo, Brazil
xiii
xiv Contributors
James Fleshman, Jr. MD, FACS, FASCRS M. Benjamin Hopkins MD, FACS, FASCRS
Department of Surgery Division of General Surgery
Baylor University Department of Colon and Rectal Surgery
Dallas, Texas Vanderbilt University Medical Center
Nashville, Tennessee
Jesus Flores MD
Steven R. Hunt MD, FACS, FASCRS
Texas Colon and Rectal Surgeons
Dallas, Texas
Washington University
St. Louis, Missouri
Luanne M. Force MD
Department of colorectal Surgery John D. Hunter MD
University of Miami Division of Colon and Rectal Surgery
Miami, Florida Department of Surgery
University of South Alabama
Molly M. Ford md, facs, fascrs Mobile, Alabama
Vanderbilt University Syed G. Husain MD, FACS, FASCRS
Nashville, Louisiana Department of Colon and Rectal Surgery
Brown University
Providence, Rhode Island
Sharon G. Gregorcyk MD, FACS, FASCRS
Texas Colon and Rectal Surgeons
Mohammed Iyoob Mohammed Ilyas MBBS, MS, MRCS
Dallas, Texas
Division of Colon and Rectal Surgery
Department of Surgery
Leander M. Grimm, Jr. MD, FACS, FASCRS Henry Ford Hospital
Division of Colon and Rectal Surgery Detroit, Michigan
Department of Surgery
University of South Alabama Arjun N. Jeganathan MD
Mobile, Alabama Department of Surgery
University of Pennsylvania
Jason F. Hall MD, FACS, FASCRS Philadelphia, Pennsylvania
Boston Medical Center Michelle C. Julien MD, FACS, FASCRS
Boston, Massachusetts Division of Colon and Rectal Surgery
Lehigh Valley Hospital
Alexander T. Hawkins MD, MPH Allentown, Pennsylvania
Division of General Surgery
Department of Colon and Rectal Surgery Brian R. Kann MD, FACS, FASCRS
Vanderbilt University Medical Center Department of Colon and Rectal Surgery
Nashville, Tennessee Ochsner Clinic Foundation
Roland Hawkins MD
Kevin R. Kasten MD, FACS, FASCRS
Radiation Oncology
Ochsner Cancer Institute
Carolinas Medical Center
Charlotte, North Carolina
Traci L. Hedrick MD, MS, FACS, FASCRS Aaron L. Klinger MD

Department of Surgery Department of Colon and Rectal Surgery
University of Virginia Health System Ochsner Clinic Foundation
Charlottesville, Virginia New Orleans, Louisiana
Terry C. Hicks MD, FACS, FASCRS Jonathan Lu MD

Department of Colon and Rectal Surgery Department of Hematology and Oncology
Ochsner Clinic Foundation Ochsner Clinic Foundation
New Orleans, Louisiana New Orleans, Louisiana
Contributors xv
Farouq Manji MD Jeffery Mino MD

Department of Colon and Rectal Surgery Department of Colon and Rectal Surgery
Ferguson Clinic Spectrum Health Medical Group Cleveland Clinic
Michigan State University Cleveland, Ohio
Grand Rapids, Michigan
Mary T. O’Donnell MD
David A. Margolin MD, FACS, FASCRS Department of Surgery
Department of Colon and Rectal Surgery University of Pennsylvania
Ochsner Clinic Foundation Philadelphia, Pennsylvania
Supriya S. Patel MD, FACS, FASCRS
David J. Maron MD, MBA, FACS, FASCRS Kaiser Permanente Department of General Surgery
Department of Colorectal Surgery San Jose, California
Cleveland Clinic Florida
Weston, Florida Rodrigo O. Perez MD
University of São Paulo School of Medicine
Marc R. Matrana MD, MS São Paulo, Brazil
Department of Hematology and Oncology
Ochsner Clinic Foundation W. Brian Perry MD, FACS, FASCRS
New Orleans, Louisiana Department of Surgery
University of Texas Health Science Center
Nijjia N. Mahmoud MD, FACS, FASCRS San Antonio, Texas
University of Pennsylvania Danielle Pickham MD, FACS, FASCRS
Philadelphia, Pennsylvania Kaiser Permanente Department of General Surgery
San Jose, California
Andrew C. Matthews MD
Craig A. Reickert MD, FACS, FASCRS
Department of Radiology
Division of Colon and Rectal Surgery
Henry Ford Hospital
Detroit, Michigan
Charles C. Matthews MD, FACR Rocco Ricciardi MD, MPH, FACS, FASCRS
Department of Radiology Department of Colon and Rectal Surgery
Ochsner Clinic Foundation Massachusetts General Hospital
New Orleans, Louisiana Boston, Massachusetts
Shannon McChesney md Daniel E. Sarmiento MD

Department of General Surgery Division of Colon and Rectal Surgery
Tulane University School of Medicine Lehigh Valley Hospital
New Orleans, Louisiana Allentown, Pennsylvania
Allison B. McCoy PhD Nicole M. Saur MD, FACS, FASCRS

Tulane University School of Public Health Department of Surgery
and Tropical Medicine University of Pennsylvania
New Orleans, Louisiana Philadelphia, Pennsylvania
Jacob A. McCoy MD Skandan Shanmugan MD

Department of Urology Department of Surgery
Ochsner Clinic Foundation University of Pennsylvania
New Orleans, Louisiana Philadelphia, Pennsylvania
Genevieve B. Melton-Meaux MD, PhD, FACS, FASCRS Benjamin D. Shogan MD

Division of Colon and Rectal Surgery Department of Colon and Rectal Surgery
University of Minnesota University of Chicago
Minneapolis, Minnesota Chicago, Illinois
xvi Contributors
Robert J. Sinnott DO, FACS, FASCRS Charles C. Vining MD

Division of Colon and Rectal Surgery Department of Colon and Rectal Surgery
Lehigh Valley Hospital University of Pennsylvania
Allentown, Pennsylvania Philadelphia, Pennsylvania
Arida Siripong MD, FACS, FASCRS Katerina O. Wells MD

Department of Colon and Rectal Surgery Department of Surgery
Ferguson Clinic Spectrum Health Medical Group Baylor University
Michigan State University Dallas, Texas
Grand Rapids, Michigan
Scott R. Steele MD, FACS, FASCRS Charles B. Whitlow MD, FACS, FASCRS
Colorectal Surgery Department of Colon and Rectal Surgery
Cleveland Clinic Ochsner Clinic Foundation
Cleveland, Ohio New Orleans, Louisiana
Sharon L. Stein MD, FACS, FACRS J. Christian Winters MD

University Hospitals Cleveland Medical Center Department of Urology
Case Western Reserve University Louisiana State University School of Medicine
Cleveland, Ohio New Orleans, Louisiana
Emily Steinhagen MD Moriah E. Wright MD

University Hospitals Cleveland Medical Center Department of Colon and Rectal Surgery, Inc.
Case Western Reserve University Omaha, Nebraska
Cleveland, Ohio
W. David Sumrall, III MD Massarat Zutshi MD, FACS, FASCRS

Department of Anesthesia Department of Colon and Rectal Surgery
Ochsner Clinic Foundation Cleveland Clinic
New Orleans, Louisiana Cleveland, Ohio
H. David Vargas MD, FACS, FASCRS

1
Preexisting conditions
ARIDA SIRIPONG AND FAROUQ MANJI
CHALLENGING CASE PREOPERATIVE TESTING
A 63-year-old woman is referred to your office with a

For healthy patients undergoing surgery, routine laboratory
right colon cancer found on screening colonoscopy.
tests may be unnecessary. In a large study of 2,000 patients
Her past medical history is significant for stable coro-
undergoing elective surgery, only 0.22% of routine preoper-
nary artery disease.
ative laboratory results found abnormalities that prompted
intervention (1). This testing can unnecessarily delay sur-
gery and lead to an increase in health-care costs without
CASE MANAGEMENT clinical benefit. Therefore, the decision to order preopera-
tive tests in the healthy individual should be guided by the
You order a complete blood count (CBC) and patient’s clinical history, physical exam findings, disease
basic metabolic profile. A chest x-ray and a com- pathology, and risk of planned surgery.
puted tomography (CT) scan of the abdomen and Ambulatory anorectal procedures are considered low risk
pelvis are obtained for staging purposes. No other and do not require routine laboratory testing in the asymp-
workup is needed prior to scheduling the patient tomatic patient (2). The remaining colorectal procedures,
for surgery. however, generally involve intraabdominal dissection and
are classified as elevated risk. Baseline blood count and type
and screen are indicated in these patients undergoing major
surgery where significant blood loss is a potentiality and/or
INTRODUCTION
is anticipated. Coagulation studies are reserved for patients
with a history of bleeding or coagulopathy, on chronic antico-
A thorough preoperative assessment is essential to identify, agulation medications, or with comorbidities that may affect
treat, and optimize preexisting comorbidities and minimize normal coagulation (renal disease or liver failure). Creatinine
morbidity and mortality after colorectal surgery. Paying level is warranted in older patients (older than 60 years old),
early attention to patient risk factors and recognizing their as well as in those with a history of diabetes or baseline renal
potential impact on outcomes is the surgeon’s responsibil- insufficiency. Patients with recent weight loss, infection, or
ity. This will often require a multidisciplinary approach and hospital admission may benefit from albumin assessment to
coordination with other physicians; however, these extra guide recommendations for preoperative nutritional support
steps and dedicated attention are as critical as the technical and discussion regarding the role of stoma creation.
aspects of the procedure to ensure maximal benefit for each
patient.
This chapter highlights preoperative optimization of
the patient undergoing elective colorectal surgery. In the CARDIAC EVALUATION
setting of emergent surgery, additional testing or modi-
fication of preexisting conditions is a luxury, and the Cardiovascular disease (CVD) is the leading cause of death
risk of delaying surgery will seldom justify the benefit in the industrialized world (3), and 25%–30% of all patients
of additional workup. In these cases, intensive intraop- undergoing noncardiac surgery have significant coronary
erative and perioperative care is necessary to minimize artery disease at the time of operation (4). The goal of pre-
complications. operative cardiac evaluation is to identify those who will
1
2 Preexisting conditions
benefit from additional testing and intervention prior to to <4 METs. The role of cardiac stress testing is closely
surgery. In general, prophylactic cardiac interventions are related to METs and functional capacity of the patient.
not advised, and additional workup should only be recom- Patients with elevated surgical risk and poor (<4 METs)
mended if also warranted outside of the surgical setting. or unknown functional capacity should undergo exercise
Perioperative risk stratification for noncardiac surgical or pharmacological stress testing if it will change manage-
patients is calculated based on procedure-related risk, car- ment. In the setting of elective surgery, findings of severe
diac risk indices, and assessment of exercise tolerance based cardiac ischemia on stress testing should prompt interven-
on metabolic equivalents (METs). Currently the American tion with medical therapy and/or preoperative revascular-
College of Cardiology/American Heart Association (ACC/ ization. Of note, in those with <4 METs, additional cardiac
AHA) categorizes noncardiac surgery into low risk, which testing or intervention should not be pursued if it will not
conveys a risk of myocardial infarction or major adverse impact surgical decision-making (decision to proceed with
cardiac event (MACE) <1%, and elevated risk, which con- surgery or palliative measures).
veys risk ≥1% (5). Historically, this risk was stratified into
three groups; however, recommendations for those of inter- TESTING AND INTERVENTIONS
mediate and high risk were similar. Therefore, in current
guidelines, these groups have been combined and most As mentioned previously, prophylactic cardiac interven-
major colorectal procedures are classified as elevated risk. tions have no proven benefit in outcomes and should only
Various cardiac risk indices have been described. The be considered in patients who would also require it in the
Goldman Cardiac Risk Index is a multivariate risk index nonsurgical setting (5). In those who present with indica-
and precursor to the widely used Lee Revised Cardiac Risk tions for urgent cardiac intervention before noncardiac sur-
Index (RCRI) (6,7). The RCRI is a simple and validated eval- gery, the type of cardiac intervention should be guided by
uation tool, based on six predictors of perioperative cardiac the urgency of the noncardiac surgery.
risk (high-risk surgery, history of ischemic heart disease, Coronary artery bypass graft (CABG) is indicated in
congestive heart failure, cerebrovascular disease, diabetes patients with triple vessel disease or myocardial ischemia
mellitus requiring insulin treatment, and renal dysfunc- with concomitant decreased left ventricular function, who
tion with creatinine >2). A patient with none, one, two, or require elective noncardiac surgery with a high bleeding
more than three risk factors has a MACE rate of 0.4%, 1.0%, risk (9). There is a paucity of data regarding optimal timing
2.4%, and 5.4%, respectively (6). The American College of of elective noncardiac surgery after CABG, although one
Surgeons National Surgical Quality Improvement Program compelling study suggests a significant increase in mortality
(ACS NSQIP) Myocardial Infarction and Cardiac Arrest for patients undergoing high-risk vascular surgery within
(MICA) Risk Calculator is a tool based on multivariate 30 days of CABG (10). Therefore, when possible, noncardiac
analysis, derived from prospectively collected data from surgery should be postponed for 30 days after recent CABG
over 500 hospitals and one million operations (8). The and may not be a feasible option for the patient with symp-
advantage of the NSQIP calculator is the greater number of tomatic colorectal pathology.
input variables required to generate risk estimations, there- Percutaneous coronary intervention (PCI) with drug-
fore deriving more accurate results. MICA has been shown eluting or bare metal stents is indicated in (1) patients with
to outperform the RCRI in discriminative power among left main disease whose comorbidities preclude bypass sur-
the same group of patients (8). Regardless of which model gery without undue risk and (2) patients with unstable cor-
is chosen, practitioners should be comfortable using one of onary artery disease who would be appropriate candidates
these risk indices in the preoperative assessment. for emergency or urgent revascularization (11,12). While
The recommended 2014 ACA/AHA algorithm for pre- bare metal stents require uninterrupted antiplatelet therapy
operative cardiac evaluation is shown in Figure 1.1 (5). for 30 days, this recommendation is extended to 365 days
As previously noted, patients needing emergent surgery after placement of a drug-eluting stent. These recommen-
require close perioperative monitoring and management dations are based on several studies that show convincing
but often cannot delay surgery for additional testing. Those evidence that disruption of dual-antiplatelet therapy within
with acute coronary symptoms undergoing nonemergent a short time period results in higher adverse cardiac out-
surgery should be treated based on guideline-directed comes, and is the leading predictor of coronary thrombosis
medical therapy (GDMT). Asymptomatic patients with and restenosis (13,14). In the setting of a recent stent and
cardiac risk factors are stratified based on surgical and urgent indications for major abdominal surgery, discussion
clinical risk. Low-risk procedures do not require additional with the patient’s cardiologist regarding the use of bridging
testing, whereas those undergoing elevated-risk procedures antiplatelet agents, such as Integrillin or Tirofiban, may be
are further categorized based on METs. METs, a measure beneficial (15). When used as a bridging agent for Plavix,
of exercise tolerance, can be evaluated based on a few sim- these short-acting agents should be started as an infusion
ple questions during the initial encounter. Patients unable therapy 24 hours after the last dose of Plavix (5 days prior
to walk two blocks on level ground or carry two bags of to surgery) and continued up to 4 hours prior to surgery.
groceries up one flight of stairs without symptoms of The infusion is then resumed 2 hours postoperatively until
angina or dyspnea have poor exercise tolerance, equivalent Plavix is restarted.
Cardiac evaluation / Perioperative β-blockade 3
*See Sections 2.2, 2.4, and 2.5 in

Patient scheduled for surgery with the full-text CPG for
known or risk factors for CAD* recommendations for patients
(Step 1) with symptomatic HF, VHD, or
arrhythmias.
†See UA/NSTEMI and STEMI
CPGs (Table 2).
Emergency Clinical risk stratification

Yes
and proceed to surgery
No
ACS† Evaluate and treat

Yes
(Step 2) according to GDMT†
No
Estimated perioperative risk of MACE

based on combined clinical/surgical risk No further
(Step 3) testing
(Class IIa)
Excellent
(>10 METs)
Low risk (<1%) Elevated risk Moderate or greater Proceed to

(Step 4) (Step 5) (≥4 METs) functional surgery
capacity
Moderate/Good
(≥4–10 METs)
No further
No further No or testing
testing unknown (Class IIb)
(Class III:NB)
Proceed to Poor or unknown

surgery functional capacity
(<4 METs): Pharmacologic
Will further testing impact Yes stress testing
decision making or (Class IIa)
perioperative care?
(Step 6)
If If
normal abnormal
No Coronary
revascularization
according to
Proceed to surgery existing CPGs
according to GDMT or (Class I)
alternate strategies
(noninvasive treatment,
palliation)
(Step 7)
Figure 1.1 ACC/AHA cardiac risk assessment algorithm. (Data from Fleisher LA et al. Circulation. 2014;130(24):2215–45.)
Balloon angioplasty should be considered for those who PERIOPERATIVE β-BLOCKADE

do not meet criteria for CABG, require time-sensitive non-
cardiac surgery, and/or are at high risk for bleeding (16). A β-Blockers should be continued in patients who were using
patient with a new diagnosis of colon cancer with findings them chronically, and postoperative management should be
of severe cardiac ischemia may fall into this category. The guided by clinical circumstances, such as hypotension, bra-
main benefit of balloon angioplasty is the ability to carry out dycardia, or bleeding (5). In patients at intermediate or high
noncardiac surgery immediately without necessitating dual- risk of myocardial ischemia, or with three or more RCRI
antiplatelet therapy, although ideally such surgery should be factors, it may be reasonable to start β-blockers in the pre-
postponed a minimum of 14 days (5). Ultimately, decisions operative setting. Importantly, though, β-blockers should
regarding type of PCI and management of perioperative be initiated in advance of surgery to assess safety and titrate
antiplatelet therapy should be coordinated between the sur- dosage appropriately. β-Blockers should not be started on
geon, cardiologist, anesthesiologist, and patient, weighing the day of surgery, and this may actually be harmful, as
the relative risk of bleeding versus stent thrombosis. illustrated by results of the POISE (Perioperative Ischemic
Evaluation Study) trial (17). In this study of 9,000 partici- was based on a small study published in 1989, which sug-
pants, although β-blockade diminished the incidence of gested PPCs could be higher in patients who cease smok-
myocardial infarction, patients also experienced higher ing less than 8 weeks before surgery versus those who
rates of death, stroke, hypotension, and bradycardia. continue smoking (30). More recent analysis, however,
challenges these results. Two recent meta-analyses demon-
strate no evidence to suggest an increased risk of PPC when
smoking cessation occurs within a few weeks of surgery.
PULMONARY ASSESSMENT Furthermore, there is a time-related decrease in postopera-
tive complications the longer smoking is stopped before sur-
Postoperative pulmonary complications (PPCs) are com- gery (31,32). The current data demonstrate that it is safe to
mon after noncardiac surgery and play an important role encourage patients to stop smoking any time in the preop-
in patient outcomes. Definitions of PPCs vary across stud- erative period, and ideally 6–8 weeks before the procedure.
ies, and therefore, the true incidence is difficult to describe, Obstructive sleep apnea (OSA) is defined by a state of
with reported rates ranging from 6% to 80% (18). Patient- upper airway obstruction leading to apneic episodes. The
related factors that increase risk of PPCs include smoking, incidence of OSA has increased with the rise in obesity and
age older than 60 years, congestive heart failure, chronic is associated with higher risk of postoperative hypoxemia,
obstructive pulmonary disease (COPD), functional depen- cardiopulmonary events, intensive care unit admission,
dency, and American Society of Anesthesiologists (ASA) and increased hospital length of stay (33). Unfortunately,
Physical Status Classification of III or above (18–20). OSA may be undetected in the preoperative setting, as
Surgery-specific factors include general anesthesia, longer symptoms may deviate from the traditional description
operating room times (more than 2–3 hours), emergency of daytime sleepiness and snoring, and instead manifest
surgery, and site-specific surgery, with the greatest risk as headaches, difficulty concentrating, altered mood, and
among upper abdominal and thoracic procedures, which nocturia. Given the challenges in diagnosing OSA based on
contribute to splinting and a restrictive pulmonary physi- symptoms alone, screening tools including the STOP-Bang
ology (21,22). In a recent multicenter prospective study of questionnaire have helped identify patients who may benefit
ASA III patients undergoing prolonged general anesthesia from pulmonary evaluation prior to major abdominal sur-
(more than 2 hours), 33.4% of patients experienced at least gery (34). This questionnaire includes four objective patient
one PPC. In this study, even mild PPCs, including atelec- measures and four additional questions regarding sleeping
tasis or prolonged oxygen requirement, were predictors of habits. Preoperative recognition of OSA can minimize anes-
increased mortality, intensive care unit admission, and pro- thetic complications as well as PPC with the anticipated use
longed length of stay. Furthermore, modifiable factors from of continuous positive airway pressure postoperatively.
this review included colloid administration, higher intraop- Guidelines regarding preoperative chest radiography
erative blood loss, prolonged surgery and anesthesia time, and spirometry emphasize clinical assessment, relying on
and higher intraoperative tidal volumes (18). the history and physical exam (21). Guidelines from the
COPD is a significant predictor for pulmonary compli- American College of Physicians do not recommend rou-
cations, with an observational study based on the NSQIP tine preoperative chest radiography for predicting risk
database describing risk of pneumonia, prolonged ven- of PPC, as it does not alter outcomes (35). Patients who
tilation, and reintubation at 6.5%, 8.8%, and 5.5% among should have chest radiography include those with new or
COPD patients (23). However, despite the increased risk unstable cardiopulmonary signs or symptoms, and patients
seen in COPD patients, there is no prohibitive level of at increased risk of postoperative pulmonary complication
pulmonary function that serves as a contraindication to if the results would alter perioperative management (i.e.,
noncardiac surgery. Prior studies demonstrate that COPD informed decision-making, timing, and type/technique
severity does not incrementally correlate with risk of PPCs; of surgery). For example, a COPD patient diagnosed with
therefore, routine spirometry is also not recommended pneumonia on chest x-ray may benefit from delaying elec-
in COPD patients without clinical changes in pulmonary tive surgery until the infectious process is treated and pul-
function (21). monary status is optimized to baseline.
Smoking is widely accepted as a risk factor for PPC. Rates
of respiratory failure, pneumonia, and other related compli-
cations are demonstrably higher in active smokers (24–26).
These patients are more likely to have prolonged hospi-
RENAL DISEASE
tal stay, obtain wound infections, and experience venous
emboli and cardiac complications (27–29). Thus, patients Chronic renal failure is present in over 20% of patients
should be screened for smoking status, previous smoking over the age of 60, and is reported in 15% of the population
history, and in specific cases, occupational or secondhand overall (36). Renal failure encompasses a wide range of kid-
exposure. Prior debate centered on the duration of smoking ney dysfunction, ranging from glomerular filtration rate
cessation before intervention and the potential increase in <60 mL/min to dialysis-dependent renal failure. Regardless
PPC if patients stop smoking shortly before surgery. This of disease severity, it is crucial to prevent additional kidney
Liver disease 5
injury in these patients who are highly susceptible to post- (MELD) scores, are used to assess surgical risk in liver fail-
operative acute renal failure. Intraoperatively, significant ure patients. The CTP classification was originally described
blood loss and hypovolemia are poorly tolerated and should to assess operative risk in patients undergoing shunt surgery
be minimized. Avoidance of nephrotoxic agents, including for portal hypertension but has also been utilized in risk
nonsteroidal anti-inflammatory agents and IV contrast, assessment for other abdominal surgeries (39). Designed
and recognizing the impact of decreased renal function on to quantify liver dysfunction, the CTP score uses albumin,
medication clearance, such as nondepolarizing neuromus- bilirubin, prothrombin time (INR), presence of ascites, and
cular blocking agents, are critical components to periop- encephalopathy to assign points and subsequently classify
erative care. patients into three categories, A–C (maximal dysfunction).
Chronic kidney disease is associated with a host of Mortality associated with Child’s class A, B, and C has been
comorbidities, but most significantly it increases risk of reported at 10%, 17%, and 63% (40), respectively, in a review
CVD and is an independent predictor of adverse cardiac of nonhepatic abdominal procedures. In a study of cirrhotic
events. CVD and kidney disease are closely related, and patients undergoing colectomy, in-hospital mortality was
in the nonsurgical setting, CVD is the leading cause of 24%, with the highest mortality rates in those with enceph-
morbidity and mortality in chronic renal failure patients. alopathy, ascites, hypoalbuminemia, and anemia (38). The
Postoperatively, these patients experience a higher rate of MELD score is derived from a complex formula based on
cardiovascular complications and noncancer mortality INR, bilirubin, and creatinine and is calculated using Web-
after colorectal cancer surgery; the rate in this population based tools. In general, MELD scores classified as less than
has been reported at 5%–10% after elective and up to 40% 10, 10–15, and greater than 15 correlate to Child’s class A, B,
after emergency procedures (37). Given this relationship and C, respectively. MELD scores greater than 15 have been
between CVD and chronic kidney disease, there should a associated with a higher risk of complications, mortality
high index of suspicion for underlying cardiac disease in all due to complications, and overall mortality after colorectal
renal failure patients. surgery (41).
In end-stage renal failure, surgery should be timed Postoperative morbidity in the cirrhotic patient is largely
soon after dialysis to minimize electrolyte and fluid shift related to anastomotic, bleeding, and stoma complica
changes. Prior to surgery, it is also important to recognize tions (42). Damaged hepatocytes decrease production of
the physiologic changes that accompany underlying chronic clotting factors, with subsequent coagulopathy, and pre-
kidney disease. Progressive renal disease can lead to hypo- operative anticipation of bleeding risk is critical. Although
albuminemia, anemia, hyperkalemia, decreased leukocyte minimizing the severity of a significant anastomotic leak,
and immunologic function, and increased bleeding time stoma creation in the setting of ascites has inherent risks of
due to uremic platelet dysfunction. These changes contrib- peristomal leakage and varices as well. Furthermore, ascites
ute to higher rates of infectious and wound complications can increase infectious complications, wound dehiscence,
in this population and should be carefully considered and or evisceration. Meunier evaluated 41 cirrhotic patients
discussed with the patient when consenting for surgery. As who underwent colorectal surgery and identified postopera-
coagulopathy is secondary to platelet dysfunction in the tive infection as the most significant risk factor for mortal-
uremic patient, in the emergent setting, DDAVP (desmo- ity, increasing it from 11% to 53% (43).
pressin acetate) or dialysis may be used to mitigate bleeding Preoperative findings, such as portal hypertension, vari-
complications. ces, and a large amount of ascites, represent decompensated
liver failure and may be an indication to consider preopera-
tive transjugular intrahepatic portosystemic shunt (TIPS),
if colectomy is deemed necessary. One study of severely
LIVER DISEASE cirrhotic patients with abdominal malignancies reported
outcomes of abdominal surgery 1 month after TIPS was
Cirrhosis and underlying liver disease represent the most performed, noting decreased portal hypertension, ascites,
significant predictors of mortality after colorectal surgery, and venous congestion; less intraoperative blood loss; and
noting a 6.5-fold increased risk (38). Fortunately, it is rare decreased need for blood transfusion (44). Nonetheless,
for patients to present with colorectal disease in the set- TIPS increases the rate and severity of hepatic encepha-
ting of cirrhosis. For these unique situations, it is critical lopathy, and 1-year mortality rate after TIPS is estimated
to consider the natural history of the colorectal pathology, at 50%, related to the overwhelming severity of liver fail-
the severity of liver dysfunction, and potential candidacy ure. Regardless of whether or not TIPS is pursued, medical
for liver transplantation. Thorough preoperative counseling optimization of ascites with diuretic agents should also be
facilitates informed decision-making, allowing the surgeon employed throughout the perioperative period to minimize
to review goals of care and outline realistic expectations fluid overload.
regarding risks of any intervention. A rare but significant dilemma arises when a cirrhotic
Liver failure is a well-known predictor of mortality patient presents with colorectal cancer. It is important
after abdominal surgery. Two available metrics, the Child- to remember that a cirrhotic patient will not be a trans-
Turcotte-Pugh (CTP) and Model for End-Stage Liver Disease plant candidate until deemed cancer free for 5 years. These
cases should be discussed in a multidisciplinary setting,

to review overall goals of care, both short and long term. OBESITY
Nonetheless, metastatic colorectal cancer to the liver is
low (<10%), with theorized low rates due to poor tissue Based on the most recent data from the Centers for Disease
environment for tumor growth (45). Given the host of Control and Prevention, 38% of the adult U.S. population is
physiologic changes that accompany this disease process, obese, defined as body mass index (BMI) >30 kg/m2 (56).
patients with liver failure present several perioperative Furthermore, more than 5% of men and 10% of women
challenges. are classified as morbidly obese, with BMI >40 kg/m2. The
obesity epidemic places a significant socioeconomic burden
on our health-care system and predisposes patients to addi-
tional comorbidities, including insulin resistance and diabe-
DIABETES
tes, CVD, hypertension, and obstructive sleep apnea (57,58).
Physically, these patients pose specific technical chal-
In 2012, it was estimated that 29.1 million people in lenges for the colorectal surgeon, and the obese patient
the United States had diabetes, with approximately 20% significantly benefits from a minimally invasive approach
(8.1 million) of patients undiagnosed (46). Diabetic patients when feasible (59). Despite these technical challenges, obese
harbor microvascular and macrovascular pathology that patients have a decreased mortality rate compared to their
contribute to long-term complications, including nephrop- nonobese counterparts, referred to as the “obesity para-
athy, CVD, cerebrovascular disease, neuropathy, and reti- dox” (60). This favorable outcome is thought to be a result
nopathy and have been associated with increased morbidity of increased nutritional stores and a chronically inflamed
and mortality after colorectal surgery (47). This increase in state of obesity that may better prepare these patients for the
complications is likely a manifestation of both hyperglyce- physiological stress of surgery.
mia and the associated comorbid conditions, with higher Despite this lower mortality rate, obese patients harbor a
rates of postoperative renal failure and myocardial infarc- prothrombotic and pro-inflammatory state and experience
tion identified in diabetic patients (48). increased morbidity after colorectal surgery, with higher
Surgery induces a physiologic disruption in glucose rates of venous thromboembolism (VTE) and surgical site
homeostasis due to release of stress hormones and insulin infection (SSI). In a study of 7,020 colectomy patients, obese
resistance. For diabetic patients, who have marginal insu- patients had an increased rate of SSI compared with non-
lin secretion at baseline, the above factors contribute to a obese patients (14.5% versus 9.5%, p < 0.001) and overall
significant catabolic state. The subsequent hyperglycemia increased risk of SSI by 60% (61). In addition, obese patients
is a well-described risk factor for delayed wound heal- demonstrate a significantly increased risk of pulmonary
ing and infectious complications after colorectal surgery. embolism (2.18; 95% confidence interval [CI], 2.16–2.19)
Hyperglycemia impairs monocyte and neutrophil function, and deep vein thrombosis (relative risk [RR] 2.5; 95%
with a resultant increase in surgical site infections (49–51). CI, 2.49–2.51), with these results magnified in those under
In a review of 11,633 general surgery patients undergoing the age of 40 (62). Measures to decrease VTE and infectious
colorectal and bariatric surgeries, hyperglycemia (>180 mg/ complications in the obese population may include patient
dL) was associated with adverse infectious complications, selection for secondary or delayed primary closure and con-
with effects mitigated in those who received insulin (52). sideration for extended thromboprophylaxis, in the section
Close glycemic monitoring, however, is critical to prevent “Considerations for Extended Thromboprophylaxis”.
and treat not only hyperglycemia but also hypoglycemia.
Prolonged hypoglycemia can lead to neurologic sequelae,
including somnolence, unconsciousness, seizures, and irre-
versible neurologic d amage (53), and play an equally detri- MALNUTRITION
mental role in the recovery period.
Hemoglobin A1c serves as an indicator of global long-term One of the most important yet often underrecognized fac-
glucose control, and studies suggest a relationship between tors to assess at the initial encounter is nutritional status.
preoperative levels and postoperative morbidity (54). A Malnutrition is common among colorectal cancer patients,
prospective study of 438 patients undergoing abdominal and it is a well-known predictor of postoperative compli-
surgery reported higher rates of major postoperative com- cations (63). Furthermore, obstruction, fistulization, and
plications in those with preoperative HbA1c levels >6.5% infection related to inflammatory bowel disease (IBD) or
and perioperative hyperglycemia (54). Similarly, Gustaffson malignancy can prevent adequate nutrient and fluid absorp-
and colleagues reported increased morbidity after colorec- tion. Therefore, strategies to optimize nutrition are a critical
tal s urgery in patients with HbA1c >6 mg/dL (55). Although component of preoperative planning.
firm guidelines do not exist regarding optimal preopera- Malnutrition is a well-known predictor of adverse post-
tive levels, a recent HbA1c level should be obtained prior to operative outcomes after colorectal surgery and is tradition-
surgery to optimize glycemic control leading up to elective ally defined by an albumin level 3.5 g/dL, BMI <18.5 kg/m2,
procedures. or weight loss of >10% of total body weight over a 6-month
Immunosuppression 7
period. A recent review of the ACS NSQIP database showed

a high rate of malnutrition in colorectal cancer patients IMMUNOSUPPRESSION
(27.8%), much higher than all other cancer types (64). In this
study, the three listed criteria were used to define malnutri- Several medications alter the immune response of colorec-
tion, but only albumin <3.5 g/dL independently predicted tal surgery patients. In elective surgery, the surgeon may
30-day mortality and postoperative outcomes including minimize the adverse effects of these medications by careful
sepsis, renal failure, and cardiovascular events, return to decision-making regarding surgery timing, technique, and
operating room, and need for reintubation. In reality, how- selection for stoma creation.
ever, several factors can affect fluctuations in albumin level, Steroids are often used to treat IBD in the acute setting
and it should not be taken as a sole indicator of nutritional and achieve symptomatic control prior to surgical inter-
status and the implied risks. vention. However, steroids negatively impact all phases of
When identified, malnutrition should be addressed and wound healing and are associated with increased rates of
treated in the nonemergent setting. In the last two decades, venous thromboembolism (68), decreased bone density,
the role of immune-enhancing nutritional supplementa- and adrenal suppression. Available literature investigating
tion, or immunonutrition (IMN), has been well studied the relationship between steroids and surgical outcomes
with promising results from several randomized controlled is largely based on varying definitions of “recent steroid
trials. These formulas are composed of specific immune- administration” and dosages; however, results are consis-
modulating substances, including arginine, nucleotides, tent, describing the negative impact of steroids on healing.
glutamine, and omega-3 fatty acids/fish oil, which have been In a review of NSQIP data, perioperative steroid use was
shown to modify postsurgical stress and immune response, associated with increased rates of superficial SSI (5% ver-
resulting in lower infectious complications and shorter sus 2.9%), deep surgical site infections (1.8% versus 0.8%), a
length of stay (65). Arginine is an essential amino acid two- to threefold higher risk of dehiscence and organ space
found at low resting levels in the normal state, but serves SSI, and a fourfold increase in mortality in those under-
as the primary fuel source for T cells. During trauma or going abdominal surgery (69). In a study of 250 colorectal
surgery, arginine production cannot meet the demands of patients undergoing left-sided resections with anastomoses,
the body; therefore, IMN can significantly support immune Slieker et al. identified a 7.5% leak rate, with corticosteroid
function by supplementing this deficiency. use inferring a sevenfold increased risk of anastomotic leak
Even a short course of high arginine-rich protein and (70). In particular, the authors emphasized a significantly
nutrient supplements prior to colorectal surgery signifi- higher leak rate in those receiving “long-term steroids”
cantly reduces postoperative morbidity. In a randomized (50% leak rate), although duration and dosage of steroids
controlled trial of patients undergoing surgery for gas- were not described and numbers were small for this sub-
trointestinal malignancies (colorectal, stomach, and pan- group. Similarly, a meta-analysis of 12 studies including
creas), Braga et al. found significant clinical benefit in those 9,565 patients identified an anastomotic leak rate of 6.77%
randomized to receive immunonutrition versus a control compared to 3.27% in those receiving steroids versus no ste-
enteral formula (66). In this study, patients were separated roids (71). Despite the variability in defining perioperative
into four groups: (1) preoperative IMN, (2) preoperative and steroids regimens, these studies emphasize the increased
postoperative IMN, (3) preoperative control isoenergetic/ surgical risk inherent to patients unable to wean or stop
isonitrogenous formula, and (4) no supplementation, dem- long-term or perioperative steroids prior to surgery.
onstrating a significant reduction in infection rates in those In the last two decades, the treatment of IBD has trans-
who received IMN (12% versus 32%, p < 0.05). These results formed with the introduction of biologic agents. Biologics
have been replicated with subsequent randomized con- are monoclonal antibodies or fusion proteins that bind to
trolled trials, and most recently, Thornblade et al. published strictly defined molecules that play a crucial role in the
results from Surgical Care Outcomes Assessment Program inflammatory process. Antitumor necrosis factor-alpha
(SCOAP) in a community setting (67). In this prospective (anti-TNF-α) agents, including infliximab and adalimumab,
cohort study of 3,357 colorectal surgery patients, the authors and more recently vedolizumab, are commonly utilized in
reported the applicability of these results outside of a clinical the treatment of IBD with promising results. Despite these
trial. In general, patients receiving IMN had a higher ASA advancements in medical management, up to one-third of
class (III–IV) and were more likely to require an ostomy patients with Crohn disease will still require surgical resec-
(18% versus 14%, p = 0.02). Although results were not station within 5 years of diagnosis and overall two-thirds of
tistically significant, those receiving IMN had lower rates Crohn disease patients will require major abdominal sur-
of prolonged length of stay (13.8% versus 17.3%, p = 0.04) gery at some point in their lifetime (72,73). Therefore, the
and decreased rates of serious adverse events (6.8% versus impact of biologics on perioperative outcomes remains an
8.3%, p = 0.25). In combination with other components of area of active interest.
enhanced recovery pathways, including avoidance of naso- TNF-α increases angiogenesis and collagen production,
gastric tubes, early enteral feeding, carbohydrate loading, and therefore it is hypothesized that inhibition of TNF-α
and close glucose control, IMN has proven to be a critical delays wound healing, increasing postoperative complica-
component of optimizing surgical outcomes. tions. While this topic has been extensively studied, results
are conflicting. In 2008, Appau et al. demonstrated the strength was measured out to 7 days only, while in reality
negative impact of recent infliximab administration (within anastomotic complications may present up to 2–3 weeks
3 months) on ileocolic resection for Crohn disease, report- after the index procedure. Although similar studies of
ing a significantly higher rate of postoperative sepsis (20% human cohorts are unavailable, Dean and colleagues iden-
versus 5.8%, p = 0.021), anastomotic leak (10% versus 1.4%, tified a higher incidence of SSIs and incisional hernias in
p = 0.045), and hospital readmission (20% versus 2.9%, patients randomized to either sirolimus versus tacrolimus
p = 0.007) among those receiving infliximab. Syed and after renal transplantation (47% versus 8%, p < 0.0001) (78).
colleagues similarly published a single-center study of anti- Given the lack of evidence to support cessation of these
TNF agents in 325 patients undergoing surgery for Crohn drugs in the perioperative period and the critical role they
disease, highlighting the negative impact of biologics. In play in preventing transplant rejection, transitioning from
this cohort, 150 patients were exposed to anti-TNF therapy sirolimus to tacrolimus 6 weeks leading up to surgery may
within 8 weeks of abdominal surgery, noting no differ- be reasonable.
ence in preoperative nutritional status or corticosteroid or Among colorectal cancer patients, 28,000 patients (20%)
immunomodulator use in the two groups. On multivariate will present with metastatic disease at the time of diagno-
analysis, recent anti-TNF therapy was a predictor for overall sis. Metastatic colorectal cancer requires a patient-specific,
infectious (odds ratio [OR] 2.43; 95% CI, 1.18–5.03) and sur- multidisciplinary approach due to the variation in disease
gical site (OR 1.96; 95% CI, 1.02–3.77) complications (74). burden and distribution. Patients who are asymptomatic at
More recently, however, emerging studies have chal- initial presentation often benefit from initial chemotherapy,
lenged these findings and repeatedly demonstrate the safety and those with a favorable response may be appropriate sur-
of continuing biologics in the perioperative period. In a gical candidates in the future. For those patients who pres-
Danish study of 2,293 patients with Crohn disease who ent for elective surgery after neoadjuvant chemotherapy,
underwent intestinal resection, biologic therapy within surgery timing is left to the discretion of the surgeon. The
12 weeks of surgery did not predict a higher rate of morbid- cytotoxic effect of chemotherapy leads to induction of cell
ity and mortality. Furthermore, a subanalysis of this data death in the setting of colorectal cancer, and theoretically
showed no increased risk of postoperative complications also delays wound and anastomotic healing. Bevacizumab,
when given within 14 days of surgery (75). Waterman et al. a humanized monoclonal antibody targeting vascular endo-
similarly looked at a cohort of 195 IBD patients who were thelial growth factor (VEGF) receptor, is considered first-
exposed to biologic therapy before surgery and found no dif- line therapy with FOLFOX in the treatment of metastatic
ference in postoperative infectious rates when exposure was colorectal cancer. Bevacizumab prevents tumor growth by
within 14 days, 15–30 days, or 31–180 days before surgery inhibiting neoangiogenesis but can also lead to deleteri-
compared with controls (76). Review of the available data ous effects on healthy tissue in the postoperative setting,
highlights the controversial nature of this topic but increas- delaying wound healing and increasing the risk of infec-
ingly supports the practice of continuing biologic therapy. tious, ischemic, and bleeding complications. Bevacizumab
Perhaps more relevant, however, is the overall impact of has been associated with increased rates of early and late
combined immunosuppressive agents on wound healing. In anastomotic complications, including fistula formation up
the study by Waterman et al., while shorter interval between to 5 months after surgery (79–81). With a half-life of 20
last dose of biologic therapy and surgery did not increase (11–50) days, this drug should be held at least 28–40 days
surgical complications, combination therapy with thiopu- prior to elective surgery and postoperatively resumed no
rine and biologics was associated with higher rates of peri- earlier than 28 days, and ideally 6 weeks after surgery (82).
operative morbidity. This point underscores the cumulative For those who are unable to afford a drug holiday due to
effect of immunomodulating agents and cautions one to clinical deterioration (obstruction and perforation), stoma
consider temporary stoma creation in the setting of multi- cre
ation should be strongly considered, and increased
modal immunosuppression. bleeding risk should be anticipated prior to arriving to the
With continuing advancements in medical immuno- operating room.
suppression regimens, transplant recipients are living lon-
ger, and it is not uncommon for the colorectal surgeon to
encounter these patients in practice. In the emergent set-
CHRONIC ANTICOAGULATION AND
ting, particularly after initial transplantation, these patients
are often receiving high-dose immunosuppression, with
PERIOPERATIVE MANAGEMENT
minimal physiologic reserve, and intestinal anastomoses
should be avoided when possible. In the elective setting, With advancements in CVD management, a host of new
however, limited data exist to guide the surgeon in preop- anticoagulant agents are available. Familiarity with these
erative counseling and decision-making. A study of rodent various medications, including an understanding of their
models undergoing intestinal anastomoses and abdominal mechanism of action, reversal agent, bioavailability, and
wall closure showed that tacrolimus was associated with no half-life, is important to minimize perioperative mor-
difference in wound healing or tensile strength in the early bidity. The most common agents will be discussed in this
postoperative period (77). In this study, however, wound section, acknowledging that we are unable to provide a
Considerations for extended thromboprophylaxis 9
comprehensive review of all anticoagulation options in this Dabigatran etexilate is a direct thrombin inhibitor. Ease of
chapter. administration and the lack of routine drug-level surveil-
Aspirin irreversibly inhibits cyclooxygenase-1 (COX-1) lance are attractive features of these new medications. These
and subsequently impairs platelet function. The effects of medications should be stopped at least 2 days prior to elec-
aspirin last the duration of the platelet life span, from 8 to tive surgery, and resumed as soon as able. In situations that
10 days. Continuation of low-dose aspirin (81 mg) through require a more specific description of the drug’s impact on
the perioperative period is safe in patients undergoing major clotting, anti-Factor 10a levels can be measured. No reversal
abdominal surgery, without an increased risk of major post- agent is currently available for Factor Xa inhibitors.
operative bleeding complications (83). Dedicated studies Although hemostasis is a key concern for any surgeon,
investigating bleeding complications in the setting of high- timing of resuming anticoagulation should be coordinated
dose aspirin (325 mg), however, are limited. If chosen to dis- with the patient’s cardiologist or primary care physician to
continue prior to surgery, high-dose aspirin should be held minimize the risk of thromboembolic events as well.
for 7 days.
Clopidogrel is a platelet receptor inhibitor and most
commonly used in patients with prior cardiac or vascular
stents, or history of stroke. Due to its irreversible effects HYPERCOAGULABLE CONDITIONS
on platelet aggregation, when possible it should be discon-
tinued 5–7 days prior to elective surgery. In specific set- Patients with thrombophilia conditions deserve specific
tings where a major abdominal procedure with high risk attention regarding perioperative medical management.
of bleeding is required in a patient with a recent cardiac Hypercoagulable conditions, such as Factor V Leiden defi-
stent, bridging agents such as tirofiban may be used, as pre- ciency, antithrombin deficiency, and protein C and S defi-
viously mentioned in the section “Cardiac Evaluation.” ciency, can lead to spontaneous thrombosis in the form
Warfarin inhibits vitamin K-dependent clotting factor of venous thromboembolism or arterial thrombosis. For
synthesis, with a half-life of 36–42 hours. Most indications patients on chronic anticoagulation, such as warfarin, ther-
for anticoagulation have a therapeutic range between 2 and apeutic low molecular weight heparin or a heparin infusion
3 or 2.5 and 3.5, and warfarin should be held 5 days before can be used to bridge therapy in the perioperative setting,
surgery, with the level rechecked the day prior to or morn- as noted in the section “Chronic Anticoagulation and
ing of surgery. Ideally, the INR level should be 1.4 or less Perioperative Management.” It is important to remember
before major abdominal surgery to minimize bleeding com- that timing to resume anticoagulation in this population
plications. In the event of an emergent surgery, FFP can be with inherent hypercoagulability (84) is often more critical
administered as a reversal agent, acknowledging that FFP compared to those on chronic anticoagulation for preven-
has an INR of 1.6. Postoperative resumption of anticoagulation of thrombotic complications related to atrial fibrilla-
tion should be at the discretion of the surgeon. For patients tion. When therapeutic anticoagulation is suspended for
at high risk of clotting, bridging options with Lovenox injec- an extended period of time, patients can experience signifi-
tions or a continuous heparin infusion are effective options. cant complications, ranging from mesenteric thrombosis to
Heparin inactivates antithrombin III and can be admin- limb-threatening ischemia.
istered as an IV infusion for therapeutic indications or
subcutaneous injection for thromboembolic prophylaxis.
With a half-life of 45 minutes, heparin infusions provide
CONSIDERATIONS FOR EXTENDED
the advantage of implementing therapeutic anticoagula-
tion with easier and faster reversibility if bleeding compli-
THROMBOPROPHYLAXIS
cations arise. Low molecular weight heparin (enoxaparin)
has a half-life of 3–5 hours and is attractive for bridging Indications for extended thromboprophylaxis after surgery
of anticoagulation in the outpatient setting. Compared are well established in colorectal cancer patients but poorly
to enoxaparin, unfractionated heparin is cheaper, has a defined otherwise. Extended thromboprophylaxis after
shorter half-life, and is safer to use in renal failure patients. abdominal surgery for colorectal cancer is recommended
However, enoxaparin has a lower risk of heparin-induced for 3–4 weeks, using enoxaparin (85). In a randomized
thrombocytopenia (HIT) and provides the flexibility of use study of 1-week versus 4-week prophylaxis for VTE events
in the ambulatory setting. For patients receiving enoxa- after laparoscopic colorectal cancer surgery, Vedovati and
parin twice daily, the evening dose the night prior to sur- colleagues demonstrated a significant reduction of VTE
gery should be held. Heparin infusions should be stopped events (9.7% versus 0%, p = 0.001) without increased bleed-
6 hours prior to surgery. If necessary, protamine sulfate can ing risk in those receiving extended therapy (86).
be used to reverse the effects of heparin products. Similar guidelines in other high-risk populations, includ-
Rivaroxaban, Apixaban, and Dabigatran etexilate are ing IBD and bariatric patients, are lacking but emerg-
oral agents that have been recently introduced as alterna- ing as areas of interest. Aminian and colleagues recently
tives to traditional options for therapeutic anticoagula- introduced a risk assessment tool to guide indications for
tion. Rivaroxaban and apixaban inhibit Factor Xa, while extended postoperative thromboprophylaxis after bariatric
surgery. Based on ACS NSQIP data, the authors found that latter group, no VTE events occurred within 30 days of sur-
more than 80% of VTE events occurred after discharge, with gery compared to 4.5% in those not receiving postdischarge
the most significant predictors including congestive heart anticoagulation (p = 0.006). No significant difference in
failure, paraplegia, dyspnea at rest, and reoperation (87). In bleeding events or mortality was noted. These results are
addition, prior history of DVT and procedure-specific fac- encouraging and may be relevant in the prevention of VTE
tors, including open approach, operating time more than 3 events in morbidly obese colorectal surgery patients as well.
hours, and revision surgeries have been associated with more A recent review of NSQIP data similarly reported
VTE events after bariatric surgery (84,88,89). A promising higher rates of VTE in IBD compared to colorectal can-
study by Raftopoulos compared administration of twice- cer patients (2.7% versus 2.1%) (91). Additional studies
daily enoxaparin 30 mg until hospital discharge versus 176 have identified a 1.5- to 1.8-fold higher risk of VTE in IBD
patients who also received enoxaparin 40 mg daily for 10 patients, with highest rates among ulcerative colitis ver-
days following discharge after bariatric surgery (90). In the sus Crohn patients, with an OR of 2.1 versus 0.96 (92,93).
Figure 1.2 ACS NSQIP surgical risk calculator. (From the American College of Surgeons National Quality Improvement
Program, http://riskcalculator.facs.org/RiskCalculator.)(Continued)
Risk assessment tools 11
A recent study by Scarpa et al. evaluated VTE rates in

patients undergoing major colorectal surgery and identi- RISK ASSESSMENT TOOLS
fied a significantly increased risk of VTE in ulcerative coli-
tis patients (OR 7.4; 95% CI, 1.4–44.4; p = 0.017). Again, During the initial encounter, a limited amount of time is
in this study DVT rates were higher in ulcerative colitis available to evaluate and address each patient’s pathology,
compared to colorectal cancer patients (p = 0.009) (94). review comorbidities, and have an informed discussion
Given the multitude of variables that contribute to VTE regarding surgical indications, alternatives, risks, and ben-
risk, development of a VTE risk calculator for colorectal efits of a proposed procedure. In the last decade, surgical
surgery may help stratify and better select patients who risk calculators have been adopted as helpful adjuncts in
will benefit from extended therapy. this discussion and the informed consent process.
Figure 1.2 (Continued) ACS NSQIP surgical risk calculator. (From the American College of Surgeons National Quality
Improvement Program, http://riskcalculator.facs.org/RiskCalculator.)
Traditionally, risk evaluation was based on physi- accurate assessment of these preexisting conditions helps
cian experience and his/her interpretation of a patient’s optimize clinical outcomes, and along with risk calculators,
preexisting conditions. For example, the ASA Classification allows the surgeon to have an educated, informed discus-
was developed by anesthesiologists to define a patient’s sion with patients who present for surgery.
operative risk by assessing comorbidities, and ranges from
ASA I to VI. ASA class has been shown to be an accurate REFERENCES
predictor of surgical outcomes (95), but predictive capacity
of this classification depends heavily on clinical judgment 1. Kaplan EB et al. JAMA. 1985;253(24):3576–81.
and physician experience rather than objective measures. 2. Benarroch-Gampel J et al. Ann Surg. 2012;256(3):
Given these limitations, risk calculators such as 518–28.
POSSUM (Physiological and Operative Severity Score for 3. Davenport DL et al. J Am Coll Surg. 2007;204(6):
enumeration of Mortality and Morbidity) were introduced 1199–210.
to include objective measures into this estimation (96). 4. Maddox TM. Mt Sinai J Med. 2005;72(3):185–92.
POSSUM, first introduced in 1991, calculates complication 5. Fleisher LA et al. Circulation. 2014;130(24):2215–45.
and mortality rates using 12 physiologic and 6 operative 6. Lee TH et al. Circulation. 1999;100(10):1043–9.
variables. However, POSSUM and modifications of the tool 7. Goldman L. N Engl J Med. 1994;330(10):707–9.
including the Colorectal-POSSUM (CR-POSSUM) score 8. Gupta PK et al. Circulation. 2011;124(4):381–7.
have repeatedly overestimated mortality after colorectal 9. Kirklin JW et al. J Am Coll Cardiol. 1991;17(3):543–89.
cancer surgery (97). 10. Breen P et al. Anaesthesia. 2004;59(5):422–7.
In 2013, the ACS NSQIP Surgical Risk Calculator was 11. Hillis LD et al. Circulation. 2011;124(23):2610–42.
introduced as a Web-based decision aid tool to facilitate 12. Levine GN et al. Circulation. 2011;124(23):e574–e651.
the informed consent process (98). This tool calculates 13. Kaluza GL et al. J Am Coll Cardiol. 2000;35(5):
patient-specific, empirically derived risk estimates based on 1288–94.
21 patient-specific risk factors and the type of procedure. 14. Hawn MT et al. JAMA. 2013;310(14):1462–72.
This tool was created from an updated, high-quality data- 15. Kristensen SD et al. Eur Heart J. 2014;35(35):
base recording demographics, comorbidities, and 30-day 2383–431.
outcomes after surgery from more than 500 institutions in 16. Fleisher LA et al. Circulation. 2014;130(24):e278–333.
the United States and Canada. Measured outcomes include 17. Group PS et al. Lancet. 2008;371(9627):1839–47.
readmission, ileus, anastomotic leak, and the most com- 18. Fernandez-Bustamante A et al. JAMA Surg. 2017;
mon complications associated with colorectal surgery, as 152(2):157–66.
seen in Figure 1.2. Estimates of potential for discharge to 19. Garibaldi RA et al. Am J Med. 1981;70(3):677–80.
skilled nursing or rehabilitation facilities are also calculated 20. Fujita T, Sakurai K. Am J Surg. 1995;169(3):304–7.
and can be an important component of the preoperative 21. Smetana GW et al. Ann Intern Med. 2006;144(8):
discussion. Furthermore, the Surgical Adjustment Score 581–95.
(SAS) allows the surgeon to subjectively adjust estimated 22. Canet J et al. J Am Soc Anesthesiol. 2010;113(6):
risk based on clinical assessment of variables not yet mea- 1338–50.
sured with this tool. For example, the ACS NSQIP Surgical 23. Gupta H et al. Chest. 2013;143(6):1599–606.
Risk Calculator currently does not assess hypoalbumin- 24. Wetterslev J et al. Acta Anaesthesiol Scand. 2000;
emia, a known predictor of postoperative complications, 44(1):9–16.
into calculated risk estimates. However, a recent study by 25. Morton H. The Lancet. 1944;243(6290):368–70.
Hu and colleagues demonstrated that addition of hypoal- 26. Schwilk B et al. Acta Anaesthesiol Scand. 1997;41(3):
buminemia as the 22nd factor in the calculator increased 348–55.
accuracy of morbidity and mortality estimates (99). Initial 27. Frick W et al. Tex Dent J. 1994;111(6):21–3.
evaluation of the risk calculator’s utility has been promising 28. Theadom A, Cropley M. Tob Control. 2006;15(5):
and highlights the opportunity for continued development 352–8.
with additional modifiers. 29. Mills E et al. Am J Med. 2011;124(2):144–54. e8.
30. Warner MA et al. (eds). Mayo Clinic Proceedings
1989 Jun 1;64(6):609–616. Elsevier.
31. Myers K et al. Arch Intern Med. 2011;171(11):983–9.
32. Wong J et al. Can J Anesth/J Canadien d’anesthésie.
CONCLUSION 2012;59(3):268–79.
33. Kaw R et al. Chest J. 2012;141(2):436–41.
Close attention to patient comorbidities in the preoperative 34. Chung F et al. Br J Anaesth. 2012;108(5):768–75.
setting is an essential component of surgical planning. A 35. Feely MA et al. Am Fam Physician. 2013;87(6):414–8.
thorough history and physical exam are the foundation of 36. https://www.niddk.nih.gov/health-information/kid-
this evaluation and will help guide additional testing, inter- ney-disease/chronic-kidney-disease-ckd. Accessed
ventions, and consultations, when indicated. Ultimately, July 23, 2018.
References 13
37. Currie A et al. Colorectal Dis. 2014;16(11):879–85. 68. Higgins PD et al. Clin Gastroenterol Hepatol. 2015;
38. Metcalf AM et al. Dis Colon Rectum. 1987;30(7): 13(2):316–21.
529–31. 69. Ismael H et al. Am J Surg. 2011;201(3):305–8; discus-
39. Peng Y et al. Medicine (Baltim). 2016;95(8):e2877. sion 8–9.
40. Neeff H et al. J Gastrointest Surg. 2011;15(1):1–11. 70. Slieker JC et al. Arch Surg. 2012;147(5):447–52.
41. Ghaferi AA et al. Ann Surg. 2010;252(2):345–50. 71. Eriksen TF et al. Colorectal Dis. 2014;16(5):O154–60.
42. Mansour A et al. Surgery. 1997;122(4):730–5; 72. Bouguen G, Peyrin-Biroulet L. Gut. 2011;60(9):
discussion 5–6. 1178–81.
43. Meunier K et al. Dis Colon Rectum. 2008;51(8): 73. Lazarev M et al. Inflamm Bowel Dis. 2010;16(5):
1225–31. 830–5.
44. Azoulay D et al. J Am Coll Surg. 2001;193(1):46–51. 74. Syed A et al. Am J Gastroenterol. 2013;108(4):
45. Gervaz P et al. J Am Coll Surg. 2003;196(6):874–9. 583–93.
46. Centers for Disease Control and Prevention. National 75. Norgard BM et al. Aliment Pharmacol Ther. 2013;
Diabetes Statistics Report, 2014: Estimates of Diabetes 37(2):214–24.
and its Burden in the United States. Atlanta, GA: US 76. Waterman M et al. Gut. 2013;62(3):387–94.
Department of Health and Human Services. 2014. 77. Willems MC et al. PLOS ONE. 2013;8(9):e76348.
47. Fransgaard T et al. Colorectal Dis. 2016;18(1):O22–9. 78. Dean PG et al. Transplantation. 2004;77(10):1555–61.
48. Yeh CC et al. Diabetes Care. 2013;36(10):3216–21. 79. Yoshioka Y et al. Surg Today. 2014;44(7):1300–6.
49. Delamaire M et al. Diabet Med. 1997;14(1):29–34. 80. Eveno C et al. Clin Res Hepatol Gastroenterol. 2011;
50. McConnell YJ et al. J Gastrointest Surg. 2009;13(3): 35(2):135–9.
508–15. 81. Deshaies I et al. J Surg Oncol. 2010;101(2):180–3.
51. Ata A et al. Am Surg. 2010;76(7):697–702. 82. Gordon CR et al. Ann Plast Surg. 2009;62(6):707–9.
52. Kwon S et al. Ann Surg. 2013;257(1):8–14. 83. Mantz J et al. Br J Anaesth. 2011;107(6):899–910.
53. Investigators N-SS et al. N Engl J Med. 2012;367(12): 84. Jamal MH et al. Surg Endosc. 2015;29(2):376–80.
1108–18. 85. Bergqvist D et al. N Engl J Med. 2002;346(13):
54. Goodenough CJ et al. J Am Coll Surg. 2015;221(4): 975–80.
854–61 e1. 86. Vedovati MC et al. Ann Surg. 2014;259(4):665–9.
55. Gustafsson UO et al. Br J Surg. 2009;96(11):1358–64. 87. Aminian A et al. Ann Surg. 2017;265(1):143–50.
56. National Center for Health Statistics (US. Health, 88. Finks JF et al. Ann Surg. 2012;255(6):1100–4.
United States, 2016: with Chartbook on Long-term 89. Chan MM et al. Surg Obes Relat Dis. 2013;9(1):
Trends in Health. 88–93.
57. Jung UJ, Choi MS. Int J Mol Sci. 2014;15(4):6184–223. 90. Raftopoulos I et al. Surg Endosc. 2008;22(11):
58. Must A et al. Jama. 1999;282(16):1523–9. 2384–91.
59. Martin ST, Stocchi L. Clin Colon Rectal Surg. 2011; 91. Gross ME et al. Dis Colon Rectum. 2014;57(4):482–9.
24(4):263–73. 92. Wilson MZ et al. Ann Surg. 2015;261(6):1160–6.
60. Mullen JT et al. Ann Surg. 2009;250(1):166–72. 93. Nguyen GC, Sam J. Am J Gastroenterol. 2008;103(9):
61. Wick EC et al. Arch Surg. 2011;146(9):1068–72. 2272–80.
62. Stein PD et al. Am J Med. 2005;118(9):978–80. 94. Scarpa M et al. Int J Colorectal Dis. 2009;24(9):
63. Huhmann MB, Cunningham RS. Lancet Oncol. 2005; 1049–57.
6(5):334–43. 95. Menke H et al. Int Surg. 1993;78(3):266–70.
64. Hu WH et al. Nutr J. 2015;14:91. 96. Copeland GP et al. Br J Surg. 1991;78(3):355–60.
65. Bharadwaj S et al. Gastroenterol Rep (Oxf). 2016;4(2): 97. Senagore AJ et al. Dis Colon Rectum. 2004;47(9):
87–95. 1435–41.
66. Braga M et al. Surgery. 2002;132(5):805–14. 98. Bilimoria KY et al. J Am Coll Surg. 2013;217(5):
67. Thornblade LW et al. Dis Colon Rectum. 2017;60(1): 833–42.
68–75. 99. Hu WH et al. Medicine (Baltim). 2016;95(10):e2999.
2
Preoperative bowel preparation
AARON L. KLINGER AND DAVID A. MARGOLIN
(2–4). This is due largely to the high concentration of aerobic

CHALLENGING CASE and anaerobic bacteria that reside within the gut. Surgical
wound infection has been shown to increase length of stay
A 57-year-old man with a sigmoid colon cancer found by 1 week and hospital costs by over $17,000 (4). Currently,
via colonoscopy for a history of anemia and weight to help mitigate these infectious complications, IV antibi-
loss is awaiting surgery. The patient had a mechanical otics with the addition of a preoperative bowel preparation
and oral antibiotic bowel preparation ordered. While consisting of mechanical bowel preparation (MBP), oral
you are seeing the patient in the preop holding area, antibiotics (OAP), or a combination of the two have become
he tells you that he took the oral antibiotics, but “I common practice. The purpose of bowel preparation is to
tried drinking that prep last night and after only half decrease the bacterial load in the gut in order to reduce rates
a glass I got sick and threw up. I just couldn’t take it.” of infectious complications including wound infection,
Do you perform elective surgery? abscess, and anastomotic leaks. Suggested secondary ben-
efits of bowel preparation include ease of bowel handling
(especially in minimally invasive surgery) and facilitation of
intraoperative endoscopy. Those who question the benefits
CASE MANAGEMENT of preoperative bowel preparation note suboptimal patient
compliance, associated patient discomfort, potential elec-
Yes. Multiple studies have demonstrated the safety trolyte imbalances, as well as the aspiration risk associated
of elective bowel surgery in the absence of a with consuming large volumes of liquid.
mechanical bowel preparation.
SYSTEMIC ANTIBIOTIC PROPHYLAXIS

INTRODUCTION
There is no debate regarding the importance of systemic
In 1887, Halsted published his landmark study, “Circular antibiotic prophylaxis in colorectal surgery. In the 1860s
Suture of the Intestine” (1). Here he further supported the Joseph Lister, having been made aware of the work of
novel realization that microorganisms, not air exposure, Louis Pasteur, began work on antiseptic techniques for
lead to peritonitis and noted the high rate of death at the wounds (5). He famously sprayed incisions, dressings, and
time following bowel surgery. He concluded that “the chief instruments with carbolic acid in hopes of preventing infec-
danger of infection of the peritoneal cavity is manifestly tion. Experimentation with preoperative systemic antibi-
from the contents of the intestine, in case they find their otic prophylaxis began in the 1940s following the release of
way through the wound in the intestine or along the lines of penicillin, a time when SSI following colon surgery was as
suture.” Since then the idea of cleaning the bowel to reduce high as 90% (6). Use of perioperative penicillin was com-
bacterial burden and prevent infectious complications has mon practice by the mid-1950s, and its use became expected
been adapted into surgical dogma, but practice continues to by the general public. This prompted Altemeier to state in
be challenged. 1955 that “antibiotics are most effective prophylactically
Rates of surgical site infection (SSI) following elective when used as adjuvants to careful and adequate surgery
colon and rectal resection remain among the highest of the but not as substitutes for it.” He noted that the evidence at
general surgical procedures with estimates at 20% or higher the time supported the use of antibacterial agents in select
14
Mechanical bowel preparation 15
conditions, including surgical procedures through contam-

inated areas such as the gastrointestinal tract, and warned MECHANICAL BOWEL PREPARATION
of rising antibiotic resistance (7).
Song and Glenny, in their landmark review, evaluated Physical clearance of the colonic lumen before surgery has
147 trials of antibiotic prophylaxis in colorectal surgery been common practice for over a century. Historic prepara-
performed between 1984 and 1995. They showed that the tions included extended period dietary restriction, cathar-
infection rate of those receiving prophylaxis had improved tics, whole bowel irrigation, and enemas (6). Today several
significantly from 22% (mean rate 1965–1980) to 11% oral preparations are available for bowel cleansing. While
(1984–1995). Pooled data from four trials showed a much these have U.S. Food and Drug Administration (FDA) indi-
higher infection rate in patients receiving no prophylaxis cations for cleansing for colonoscopy and radiologic studies,
than in those receiving any systemic antibiotics (40% none are approved for preoperative cleansing for surgery.
versus 13%). They also concluded from their review of 17 Available preparations vary from large volume isomotic
trials that there is no benefit to multiple antibiotic doses preparations to smaller hyperosmotic solutions. Osmotic
compared with a single preoperative dose. Oral antibiotics agents include polyethylene glycol (PEG), sodium phos-
alone were shown to be significantly worse at preventing phate (NaP), sodium sulfate, sodium ascorbate, and sodium
SSI than oral antibiotics with additional systemic antibiot- picosulfate.
ics (odds ratio [OR] 3.34). They concluded that oral antibi- PEG is an inert osmotically active polymer that is mixed
otics alone were inadequate prophylaxis, that the regimen with an electrolyte solution, typically to a volume of 4 L,
chosen must provide adequate coverage of aerobic and which a patient is instructed to ingest over 2–3 hours. The
anaerobic bacteria, and that parenteral antibiotic admin- added electrolyte content in concert with the osmotic activ-
istration must be timed to assure that tissue concentration ity of PEG prevents net absorption or excretion of water and
of antibiotics is sufficient at the time of bacterial contami- electrolytes acting to clear bowel contents without caus-
nation (8). ing significant fluid shifts or electrolyte abnormalities in
In addition to recommending combined OAP and MBP most patients. Due to the potential for these imbalances,
prior to colorectal surgery, the 2013 guidelines from the however, special care needs to be taken in elderly patients,
Surgical Infection Society strongly recommend the use of patients with congestive heart failure, and patients with
preoperative IV antibiotics (3). They note infectious com- renal insufficiency. A modified, reduced volume prep con-
plication rates as high as 30%–60% without antibiotics sists of 2 L PEG solution taken with four bisacodyl tablets.
compared to <10% with prophylaxis. Pooled data analysis While not FDA approved, some surgeons have begun using
by this group also shows a significantly decreased mortal- a “MiraLAX prep,” which consists of over-the-counter elec-
ity rate with the use of preoperative IV antibiotics (11.2% trolyte-free PEG solution dissolved in a sports drink (i.e.,
versus 4.5%). The authors echo that systemic antibiotics Gatorade). This prep with the addition of bisacodyl utilizes
should offer coverage for both aerobic and anaerobic bac- less volume than traditional PEG and is more palatable for
teria. Recommended regimens include a single dose of a patients, allowing for better compliance (10).
second-generation cephalosporin with appropriate cover- NaP is a sodium-based hyperosmolar oral laxative. Its
age (cefoxitin or cefotetan) or cefazolin plus metronida- use requires smaller volumes than PEG, which like the
zole. Alterative regimens are provided for those in resistant “MiraLAX prep” may result in better patient compliance.
communities or with allergies to recommended agents. It This regimen consists of 45 mL NaP in 240 mL clear liquids
is noted that ertapenem has been shown to be superior to taken twice, 10 hours apart. Patients should be instructed
cefotetan at preventing SSI (18.1% versus 31.1%), but its use to take their prep early so that they are not kept awake the
remains controversial due to concern of creating resistant night prior to their procedure. Alternatively, patients can
organisms and potential for Clostridium difficile infection take NaP in tablet form—4 tablets and 240 mL clear liq-
(increased, but not statistically significant rates have been uids are taken every 15 minutes until 28 tablets have been
reported). Systemic prophylaxis should be provided within consumed. The FDA issued a black box warning in May
1 hour to the start of surgery, drugs with shorter half-lives 2006 regarding the use of oral NaP for bowel preparation in
may require re-dosing for longer operations. This can be elderly patients, those with underlying kidney disease, those
avoided by using an antibiotic with a longer half-life. with dehydration, or those taking medications that alter
To better determine the ideal prophylactic IV antibiotic renal perfusion (such as angiotensin-converting enzyme
regimen, Poeran et al. analyzed 90,725 patients receiving inhibitors, angiotensin receptor blockers, diuretics, and
open colectomies at 445 hospitals between 2006 and 2013 (9). nonsteroidal anti-inflammatory drugs). This population is
The overall SSI prevalence for the study was 5.2%. Although at an increased risk for development of acute renal failure,
cefoxitin was the most commonly used preoperative anti- or nephrocalcinosis, secondary to fluid shifts, phosphate
biotic (42% of cases), they found ampicillin- sulbactam load, and decreased intravascular volume when taking NaP
(OR 0.71), ertapenem (OR 0.65), and metronidazole with prep. This complication can be avoided by consuming a
cefazolin (OR 0.56) all to be superior in preventing SSI. The larger (2–3 L) volume of clear liquids while taking this prep.
authors confirmed lack of benefit to providing antibiotics However, oral NaP solutions are not currently commercially
beyond the day of surgery. available.
16 Preoperative bowel preparation
The efficacy of MBP in reducing infectious complications at 19, 18, and 9 hours prior to surgery). With this prepara-
has recently been questioned. Multiple randomized con- tion, they showed a SSI reduction from 43% to 9%. This is
trolled trials and meta-analyses have failed to show a benefit still the most common protocol used today (17). A modified
to MBP alone. In a 2011 Cochrane Review, Güenaga et al. version with equally efficacious outcomes substitutes eryth-
compared 5,805 patients from 18 trials who underwent elec- romycin, which can have significant gastrointestinal side
tive colorectal surgery (11). Of these, 2,906 received MBP, effects, with 500 mg of metronidazole (18).
and 2,899 received no preparation. All patients received IV These outcomes have been challenged, and several small
antibiotics. They found no significant difference between trials have failed to show a benefit to OAP when broad-
these groups in rates of anastomotic leakage or wound spectrum IV antibiotic prophylaxis is also administered.
infection in low anterior resection or colonic resection. Furthermore, in a study in which all patients received IV
Despite these findings, MBP remains common practice antibiotics and MBP, Wren et al. found a higher rate of
in colorectal surgery. A survey of members of the American C. difficile infection in patients taking oral antibiotics (19).
Society of Colon and Rectal Surgeons (ASCRS) performed As skepticism of the utility of MBP grew, surgeons ques-
by Beck and Fazio in 1990 showed 100% of respondents tioned the utility of OAP alone. Like Poth and his contem-
using some form of mechanical (most commonly enemas poraries 70 years prior, Atkinson et al. hypothesized that
and cathartics) and antibiotic preparation, most using com- oral antibiotics without MBP would not alter incidence of
bined oral and parenteral agents (12). A repeat survey in SSI (20). They reviewed 6,399 patients from the American
1997 by Nichols again found 100% compliance with some College of Surgeons National Surgical Quality Improvement
form of MBP prior to elective colorectal surgery with the Program (ACS-NSQIP) who underwent elective segmental
majority using PEG solutions. Eighty-seven percent of colon resection. They found a significantly lower rate of SSI
respondents used both oral and parenteral antibiotics (13). (9.7%) in those taking oral antibiotics than in those who
A 2005 survey by Lassen et al. of colorectal surgeons in five did not (13.7%); however, this study did not compare this to
Northern European countries found over 90% using some patients receiving combined oral/mechanical prep.
form of mechanical prep prior to elective left colectomy In 2010, Englesbe reviewed 1,553 elective colectomies
(14). Finally, a survey of ASCRS surgeons in 2016 found that performed over 16 months in 23 hospitals in Michigan (2).
94.3% of respondents used a mechanical bowel preparation He compared infectious outcomes and ileus in patients
always or selectively (15). receiving a combined (MBP + OAP) prep to MBP alone.
It is difficult to establish the true value of MBP in isolation Patients who received a combined prep were significantly
of other factors. A common endpoint in most studies is SSI, less likely to develop surgical site infection (4.5% versus
which as discussed previously and in the following section 11.8%), organ space infection (1.8% versus 4.2%), or pro-
is influenced by the use of oral and/or systemic antibiotics. longed ileus (3.9% versus 8.6%). There was no significant
difference in C. difficile rate. Importantly, these groups were
matched for various other factors, including use of paren-
teral prophylactic antibiotics. Scarborough and colleagues
ORAL ANTIBIOTIC BOWEL collected data from the 2012 Colectomy-Targeted ACS-
PREPARATION NSQIP database (21). They compared 4,999 patients divided
into four groups: no preparation, MBP only, OAP only, and
Attempts to reduce the colon’s bacterial load with poorly combined prep. Unlike Atkinson, they found no signifi-
absorbed oral antibiotics have been made since the cant benefit to either modality alone but noted significantly
1930s–1940s. At this time the mortality of colon surgery decreased rates of SSI (3.2% versus 9%), anastomotic leak
was 10%–12% with, as mentioned previously, 80%–90% of (2.8% versus 5.7%), and length of stay (4 days versus 5 days)
survivors experiencing a wound infection (6). It was noted in combined mechanical/OAP prep patients compared
that MBP reduced overall bacterial counts but did not alter to those receiving no bowel prep. A Cochrane Review of
the concentration of remaining colonies. Oral antibiotic studies published between 1980 and 2007 including 43,451
preparation is aimed at clearing the colonic lumen of aero- patients was performed to determine the ideal antimicro-
bic (Escherichia coli) and anaerobic (Bacteroides fragilis) bial prophylaxis for CRS. Their meta-analysis, in which all
species. In 1943 Firor and Poth showed preoperative oral patients received MBP, showed a statistically significant
sulfanilamide reduced mortality from peritonitis from 10% decrease in SSI with receipt of combined oral and paren-
to 4%. Their protocol, like most of that era, consisted of a teral antibiotics compared to parenteral alone (relative risk
mechanical bowel preparation followed by a clear liquid or [RR] 0.56) or oral alone (RR 0.56). However, the authors did
low residue diet and a week of oral antibiotics prior to sur- state that while their review shows that antibiotics delivered
gery. It was agreed that oral antibiotics were not adequate within this framework can reduce the risk of postoperative
alone without mechanically cleared bowel (16). surgical wound infection by as much as 75%, it is not known
In 1972, Nichols and Condon developed a protocol con- whether oral antibiotics would still have these effects when
sisting of a mechanical bowel preparation followed by 1 g the colon is not empty (22).
doses of neomycin and erythromycin taken three times The authors of this chapter performed a retrospec-
starting 19 hours prior to the time of surgery (administered tive case-control study of the prospectively collected
References 17
Colectomy-Targeted ACS-NSQIP database from 2012 to

2015. We evaluated over 27,000 patients using doubly robust 2. In addition, combined mechanical bowel prepa-
propensity score–adjusted multivariable regression and ration and oral antibiotic bowel preparation
found that patients receiving IV antibiotics combined with significantly reduce infectious complications,
MBP/OAP have reduced odds of SSI, organ space infection, including C. difficile colitis, in elective colon and
wound dehiscence, and anastomotic leak than no prepara- rectum resections.
tion and lower odds of SSI than OAP alone. Importantly,
unlike Wren, the authors found a reduction in C. difficile
infections with the use of combined bowel preparation (23).
REFERENCES
1. Halsted WSMD. Am J Med Sci. 1887;188:436–460.
SPECIAL CIRCUMSTANCES 2. Englesbe MJ et al. Ann Surg. 2010;252(3):514–9,
discussion 519–20.
Patients with partial large bowel obstruction create a 3. Bratzler DW et al. Surg Infect (Larchmt). 2013;14(1):
unique and challenging population. These patients are at 73–156.
an increased risk for infectious complication, and the use 4. Hedrick TL et al. Dis Colon Rectum. 2013;56(5):
of bowel prep would be ideal. Since large volumes will not 627–37.
be tolerated by mouth, PEG cannot be used. Endoscopically 5. Lister J. Br Med J. 1867;2(351):246–8.
placed stents may allow for a bowel prep to be completed 6. Poth EJ. World J Surg. 1982;6(2):153–9.
prior to operative intervention; however, the use of stenting 7. Altemeier WA et al. AMA Arch Surg. 1955;71(1):
is beyond the scope of this chapter. A “slow prep” with NaP 2–6.
or small volumes of PEG may be attempted in select patients; 8. Song F, Glenny AM. Br J Surg. 1998;85(9):1232–41.
they must be well hydrated without uncorrected electrolyte 9. Poeran J et al. Dis Colon Rectum. 2016;59(8):
abnormalities. A low threshold must be held to abort the 733–42.
bowel preparation or to operate. An initial dose of 15 mL of 10. Siddique S et al. Am J Gastroenterol. 2014;109(10):
NaP or 17 g of PEG in 120 mL of water is given to a moni- 1566–74.
tored and serially examined patient. Two additional doses 11. Guenaga KF et al. Cochrane Database Syst Rev. 2011;
are given at 4-hour intervals, at which point the doses of NaP (9):CD001544.
can be increased to 30 mL if tolerated or PEG continued. 12. Beck DE, Fazio VW. Dis Colon Rectum. 1990;33(1):
Abdominal pain or emesis should be considered indications 12–5.
to cancel the prep and consider early operative intervention. 13. Nichols RL et al. Clin Infect Dis. 1997;24(4):609–19.
14. Lassen K et al. BMJ. 2005;330(7505):1420–1.
15. Beck DE, McCoy AB. Perioperative management of
SUMMARY the colorectal; surgery patient: Current status and an
ASCRS survey. Submitted.
While the risk of infectious complications in colorec- 16. Firor WM, Poth EJ. Ann Surg. 1941;114(4):663–71.
tal surgery can never be completely eliminated, there 17. Nichols RL et al. Ann Surg. 1972;176(2):227–32.
are a variety of things that should be done to mitigate 18. Pollock AV, Evans M. N Engl J Med. 1980;303(18):
the risk: 1066.
19. Wren SM et al. Arch Surg. 2005;140(8):752–6.
1. Intravenous antibiotics are mandatory and should 20. Atkinson SJ et al. Surg Infect (Larchmt). 2015;16(6):
be given within 1 hour of skin incision. It is 728–32.
incumbent that surgeons be familiar with their 21. Scarborough JE et al. Ann Surg. 2015;262(2):331–7.
hospital’s antibiotograms so that they can choose 22. Nelson RL et al. Cochrane Database Syst Rev. 2014;(5):
the most efficacious drugs. CD001181.
23. Klinger AL et al. Ann Surg. 2017.
3
Anesthesia and intraoperative positioning
W. DAVID SUMRALL, III AND DAVID E. BECK
(SCIP), including prophylactic antibiotic administration,

CHALLENGING CASE temperature control, and the proper positioning of patients
for colorectal surgery, will be reviewed.
A 47-year-old male is undergoing a transanal excision
of a rectal villous adenoma under IV sedation and
local infiltration of xylocaine. During the procedure,
the patient complains of lightheadedness and numb- ANESTHETIC TECHNIQUE
ness of the tongue. The anesthesiologist notices
bradycardia and hypotension.
Most anorectal operations involve only a relatively small
external anatomic area. This allows for a number of anes-
thetic options to be the selection of which should be carefully
CASE MANAGEMENT undertaken. Techniques of regional and local anesthesia can
provide excellent relaxation and comfort to the patient in
Local anesthetic systemic toxicity is suspected. The most instances. Unfortunately, not all patients may agree to a
patient should be moved to the supine position and technique other than general anesthesia anorectal operations.
supported with supplemental oxygen via mask. The For these patients, general anesthesia is therefore the anes-
patient’s blood pressure is supported with IV fluid thetic of choice. Medically, however, the benefits of regional
and epinephrine. Further treatment may require con- and local anesthesia may outweigh their disadvantages and
tinued resuscitation and intralipid. may be safer than a general anesthetic. The selection of anes-
thetic technique must also take into consideration the com-
fort level of the perioperative team with the type of anesthetic
selected. To make the choice that is in the best interest of the
INTRODUCTION patient, one must be familiar with the options as well as their
advantages, disadvantages, and complications.
The American Society of Anesthesiologists (ASA) defines
anesthesiology as a discipline within the practice of medi- GENERAL ANESTHESIA
cine that specializes in the (1) medical management of
patients who are rendered unconscious and/or insensible A patient’s preference for general anesthesia may be based
to pain and emotional stress during surgical, obstetric, on a fear of the operation itself or a fear of the operating
and certain other medical procedures; (2) protection of life room. Some people’s modesty makes them uncomfortable
functions and vital organs under the stress of anesthetic, to be awake while they are exposed for an operation. Others
surgical, and other medical procedures; and (3) manage- worry that a local or regional anesthetic will wear off in the
ment of problems in pain relief (1). This chapter discusses middle of the procedure, leaving them vulnerable to severe
the various kinds of anesthesia used in colorectal surgery, pain. They may also worry about repeated needle punctures.
including their relative benefits and risks. Anesthetic risks Still others simply do not want to know what occurs in an
and enhanced recovery as managed by anesthesiologists operating room and believe that anything less than a gen-
(such as intraoperative opioid sparing, goal-directed fluid eral anesthetic will make them too aware of the course of
management, and new treatments for postoperative pain events. If these fears can be identified, they can frequently
relief) will be reviewed. Finally, anesthetic considerations be addressed and overcome by careful consultation and
as they apply to the Surgical Care Improvement Project agreement among all parties involved.
18
Anesthetic technique / Local anesthesia 19
Potential disadvantages to a general anesthetic include response to hypoxia and hypercarbia, and while a restless
the need for endotracheal intubation if the procedure patient may simply be a restless patient, there is always the
is to be performed in the prone-jackknife position. possibility that the patient is agitated due to relative hypoxia
Furthermore, there is a significant incidence of nausea and or hypercarbia.
vomiting following general anesthesia, and patients with One must always keep in mind the possibility of local
significant cardiac or pulmonary insufficiency may be at anesthetic systemic toxicity (LAST) when using these drugs.
increased risk. The typical doses used for local infiltration in colorectal
A distinct advantage of general anesthesia is total relax- procedures are far below the threshold needed for systemic
ation of the pelvic floor, which can be important when oper- toxicity (Table 3.1). However, unintended IV or intraarte-
ating on complicated and high fistulas or when approaching rial injection could result in systemic toxicity. As such, it is
the rectum for transanal excisions. Preoperative consulta- important to recognize the signs and symptoms of systemic
tion with the surgeon is advisable to ascertain the nature toxicity when they first appear, as toxicity progresses in a
and complexity of the surgical procedure. Local anesthesia dose-dependent fashion.
is frequently inadequate in these instances. If regional anes- At lower plasma concentrations, the patient begins to
thesia is contraindicated by a neurologic condition, a back experience central nervous system (CNS) toxicity charac-
injury, anticoagulation, infection at the site of needle entry, terized by lightheadedness, tinnitus, and numbness of the
or for other reasons, general anesthesia may be the anes- tongue. As plasma concentrations increase, the patient
thetic of choice. begins to experience CNS excitation, resulting in seizures,
followed by unconsciousness, coma, and respiratory arrest.
LOCAL ANESTHESIA At higher plasma concentrations, cardiovascular (CV) tox-
icity occurs, as the local anesthetic blocks sodium channels
The earliest local anesthetic used was cocaine (prepared of the myocardium.
in weak solutions and injected in high volumes) for field Relative potency of the local anesthetic plays a role here.
block at the turn of the nineteenth century (2). However, Lidocaine toxicity will result in bradycardia and hypoten-
the toxicity of cocaine, its irritant properties, and its strong sion prior to cardiac arrest, while the longer-acting, more
potential for physical and psychological dependence led to potent bupivacaine often results in sudden cardiovascular
the development of alternative local anesthetics. Many of collapse due to ventricular dysrhythmias. Maintenance of
these—such as lidocaine—are still used today, as much as perfusion and ventilation through prolonged cardiopul-
half a century after their introduction (3). monary resuscitation (CPR) is the key, as the patient will
While there are relatively few instances in colorectal sur- not convert into a life-sustaining cardiac rhythm until the
gery where it is used as the sole anesthetic, local anesthesia local anesthetic has had a chance to completely dissociate
still has a place. It requires, however, a cooperative patient from the sodium channels of the conducting system of the
who can remain immobile for both the infiltration of the heart. Cardiopulmonary bypass may even be considered.
local anesthetic, as well as for the actual procedure. Dissociation of local anesthetic from sodium channels has
It is important to be cognizant of the patient’s underlying been shown to take a considerable length of time, and pro-
health status and the position that the patient will be in for longed, intensive, and continuous support is warranted.
the procedure. A healthy patient in his or her mid-twenties Treatment of CNS toxicity, including the cessation of
can tolerate the prone-jackknife position much better than seizure activity, is with the use of benzodiazepines, propo-
an obese geriatric patient with a pulmonary history who fol, or thiopental. Treatment of CV toxicity is supportive
uses supplemental oxygen. Restlessness and inability to tol- in nature and may require electric cardioversion, epineph-
erate local anesthesia may be due to patient factors but also rine, and magnesium (4). Systemic toxicity following local
could be a result of sedation, including from self-admin- anesthetic administration is thankfully rare. More com-
istered medications when the procedure is performed in mon, however, is inadequate analgesia following local
an office setting. Any degree of sedation blunts the body’s anesthetic infiltration. This can be multifactorial in nature.
Table 3.1 Local anesthetic drugs
Maximum dose
Agent Onset Duration Maximum dose with epinephrine
Tetracaine
Lidocaine 2–5 minutes 30–45 minutes 5 mg/kg 7 mg/kg
Mepivacaine
Prilocaine
Bupivacaine 30 minutes 2 hours 2 mg/kg 4 mg/kg
Procaine 5–10 minutes 15–30 minutes 10 mg/kg
Liposomal bupivacaine 30 minutes 48–72 hours 266 mg (with 133 mg bupivacaine) 266 mg
20 Anesthesia and intraoperative positioning
Inadequate analgesia resulting from insufficient quantities analgesia. The two pivotal studies leading to approval were
placed in the correct location is easily resolved with the in hemorrhoidectomy and bunionectomy patients (5,6). The
addition of further local anesthetic at the site. Inadequate drug is provided in a 20 cc vial that contains 266 mg of lipo-
analgesia can also result from tachyphylaxis to local anes- somal bupivacaine. It can be diluted up to 14 times if desired.
thetics, which is defined as repeated injection of the same Since its release, this drug has seen increasing adoption, but
dose of local anesthetic leading to diminishing efficacy. the reported experience has been limited to date (7). A series
Additionally, inadequate analgesia can be a consequence of of four consecutive patients undergoing loop ileostomy clo-
the tissue pH into which the local anesthetic is injected. sure were successfully managed with multimodality postop-
Local anesthetics exist in both ionized and nonionized erative pain management (including liposomal bupivacaine,
states; it is only in the nonionized state that local anesthet- IV paracetamol, and ibuprofen) as 23-hour procedures.
ics can penetrate the nerve sheath, thus producing anal- Utilization of local infiltration as part of a multimodality
gesia. In an acidic environment (i.e., an infected pilonidal approach appears to have great potential.
cyst), more of the anesthetic is converted into the ionized
state, leading to far less of the nonionized form available to MONITORED ANESTHETIC CARE
produce analgesia. It is not uncommon for infected tissues
to prove nearly impossible to be rendered totally insensitive Monitored anesthetic care (MAC) is defined by the ASA
despite more than adequate amounts of local anesthetic as “a procedure in which an anesthesiologist is requested
infiltration. or required to provide anesthetic services,” and includes
A perianal block (Figure 3.1) can be performed with the (1) the diagnosis and treatment of clinical problems during
patient in either the prone or lithotomy position and pro- and immediately following the procedure; (2) the support
vides relaxation of the sphincter as well as anesthesia. The of vital functions; (3) the administration of sedatives, anal-
anesthetic solution of choice is infiltrated in a fan fashion gesics, hypnotics, anesthetic drugs, or other medications
from the lateral positions to superficially encompass the necessary for patient safety; (4) physical and psychological
anal margin. Emphasis should be placed in the posterolat- comfort; and (5) the provision of other services as needed
eral positions where the greatest concentration of nerves is to complete the procedure safely (8). When it comes to the
found. A finger or retractor is placed within the canal. At care of a patient undergoing MAC, all of the precautions
the anterior, posterior, and lateral positions, anesthetic is and equipment needed to perform a safe general anesthetic
injected submucosally or intramuscularly through the pre- must be present, as it is always possible that an escalation of
viously infiltrated tissue. The needle is held parallel to the care will be needed. While uncommon, it is possible that a
finger, with care to avoid entering the canal. patient cannot safely undergo a MAC for a specific proce-
A new formulation of liposomal bupivacaine (Exparel, dure. Most commonly, this is due to the inability to safely
Pacira Pharmaceuticals, Parsippany, New Jersey) has received prevent a patient from moving in response to painful stim-
approval from the U.S. Food and Drug Administration and uli without producing oversedation and/or apnea. Some
can provide analgesia for up to 72 hours. It was approved patients, when sedated as part of MAC, appear either to
for injection into the surgical site to produce postsurgical move excessively in response to stimuli or to develop airway
(a) (b)
External
1st sphincter
injection
Internal
sphincter
2nd
injection
Submucosa
3rd
injection
Figure 3.1 Technique for anal block. (a) Perianal view of submucosal injection. (b) Sagittal view of injection of anal canal.
Anesthetic technique / Regional anesthesia 21
compromise, requiring further intervention and escalation of the medications used: hyperbaric, isobaric, and hypo-
of care by the anesthesiologist. baric. Hypobaric local anesthetics are less dense than normal
The same limits of positioning and patient tolerance that cerebrospinal fluid (CSF), which allows these medications to
were discussed with local anesthetics apply to procedures rise in the CSF following injection. This is commonly used
under MAC as well. While most patients will be able to for perineal procedures that will be performed in the prone
tolerate a lithotomy or prone position with mild sedation, jackknife position. The local anesthetic is injected into the
others are unable to tolerate these positions without endo- intrathecal space, and the patient is immediately placed in
tracheal intubation, positive pressure ventilation, and high the jackknife position to allow the hypobaric solution to drift
inspired oxygen concentration. Additionally, there are those upward, or caudad. After approximately 5 minutes, the spinal
patients who are unable to understand or comply with the anesthetic will have “set up,” meaning the uptake and distri-
requirement to remain immobile. Young children, mentally bution of the local anesthetic across nerve membranes have
challenged patients, or extremely ill patients are examples of occurred. No further migration of the drug should occur at
poor candidates for MAC. this point.
There is an erroneous perception on the part of patients— By adding a small amount of glucose to the local anes-
and even physicians—that a patient undergoing MAC is at thetic used, the solution will become hyperbaric. The den-
decreased risk for serious anesthesia-related complications sity of the solution will cause it to sink in relation to the
when compared to general anesthesia, that MAC is safer. CSF (10). An alternative approach to perineal analgesia per-
This can best be appreciated by examining the ASA Closed formed in the prone jackknife position is performing the
Claims Project database. The ASA Closed Claims Project is intrathecal block using a hyperbaric solution, then keeping
a structured evaluation of all adverse anesthetic outcomes patients in the sitting position for 5 minutes to allow the spi-
obtained from the closed claim files of 35 professional liabil- nal anesthetic to sink caudad, thus blocking the lumbosa-
ity insurance companies in the United States. A 2006 review cral nerves. Once the block has “set up,” the patient is placed
showed more than 40% of claims associated with MAC in the prone jackknife position. These two techniques have
involved death or permanent brain damage, which was simi- allowed the use of significantly less local anesthetic for the
lar to the percentage seen in claims associated with general spinal anesthesia, compared to isobaric solutions, which
anesthesia (9). Respiratory depression was the most common have the same density as CSF. Isobaric solutions require a
(21%) damaging mechanism; nearly half of such occurrences higher dose of local anesthetic to evenly distribute through-
were judged to be preventable through better monitoring. out the CSF, resulting in a larger volume needed to achieve
Cardiovascular events comprised another 14% of the the same blockade of the lumbosacral nerves. The benefits
claims made in patients undergoing MAC, which was simi- related to reducing the total amount of local anesthetic
lar in frequency to that seen following general anesthesia. injected are a decreased risk of toxicity, along with pro-
The average payment made to a plaintiff in these cases was viding adequate analgesia and allowing faster recovery of
$159,000. These data demonstrated that the significant risks motor function.
of injury and death are similar among MAC and general A caudal anesthetic is the placement of a local anesthetic
anesthesia. and/or narcotic into the epidural space from an approach
through the sacral hiatus. This is typically performed in either
REGIONAL ANESTHESIA the prone or lateral position. While uncommon in adults, this
procedure is used frequently in children, where the caudal
Central neuraxial blockade space is more easily accessible and a relatively safe and easy
approach to infuse local anesthetic and/or narcotic for post-
Regional anesthesia encompasses a wide variety of periph- operative analgesia while still under general anesthesia.
eral and central neuraxial blocks, many of which do not The third and final type of central neuraxial block is the
pertain to colorectal surgery. The most common regional epidural anesthetic. While epidural anesthesia can be used
anesthesia technique applied in colorectal surgery is the as a single injection for colorectal procedures, it is more
spinal, or intrathecal, blockade. The spinal block is rela- common to place a catheter within the epidural space to
tively easy to place, has a fast onset of sensory and motor provide analgesia during and after the procedure. The loca-
blockade, and has a predictable length of efficacy. This is a tion of the block is determined by the anesthesiologist based
very old technique, dating back to the late 1800s, when it on several anatomic factors; however, a thoracic approach
was performed using cocaine as the anesthetic agent, to the has been shown to be more effective in reducing postopera-
great amazement of surgeons of the day (2). tive ileus and early return of bowel and bladder function
With the advent of newer local anesthetics, we can now than a lumbar approach (11).
tailor the duration of the spinal blockade to the projected Most commonly, patients will receive a postoperative
length of the surgery by varying the type and amount of continuous infusion of a local anesthetic and narcotic mix-
local anesthetic used. The goal is to provide adequate anal- ture through the epidural catheter. In addition, they may be
gesia for the duration of the procedure, yet allow safe ambu- given the opportunity to provide themselves small amounts
lation and encourage urination within a short time frame of analgesia through their epidural catheter on demand.
after cessation of surgery. There are three different densities This is termed patient-controlled epidural analgesia (PCEA),
and it provides excellent pain control while minimizing the The final relative contraindication is abnormal coagula-
undesirable side effects typically seen with IV narcotics. tion status. Patients with abnormal coagulation—either due
Provided the patient does not manifest signs of systemic to endogenous factors such as liver disease or thrombocy-
infection, the epidural catheter can remain in place for sev- topenia, or due to the administration of anticoagulants—
eral days following surgery if needed to control pain. This must be considered carefully. Additionally, patients who
benefit must be weighed against the risk of withholding are receiving or will be receiving anticoagulants postop-
anticoagulant prophylaxis and a possible resultant throm- eratively have different needs than patients who receive a
boembolic event. general anesthetic alone. For spinal and caudal anesthesia,
While initial studies examining PCEA were performed the greatest risk of spinal hematoma (a neurosurgical emer-
using lumbar epidural, more recent studies have exam- gency) occurs at the time the block is placed. For epidural
ined the impact of thoracic epidural analgesia on patients anesthesia, the risk of hematoma formation is just as great
undergoing elective colorectal surgery. In 2001 Carli et al. at the time of epidural catheter removal as during place-
reported 42 patients undergoing open large bowel resection, ment. As a result, certain guidelines should be instituted in
randomized to receive either an IV patient-controlled anal- order to reduce the risk of spinal hematoma formation upon
gesia (IVPCA) morphine or a thoracic (T7-8) epidural with removal of the epidural catheter.
bupivacaine and fentanyl (11). Patients who received thoracic Heparin is often administered perioperatively as pro-
epidural had distinctly superior analgesia as compared to the phylaxis against deep vein thrombosis (DVT) formation.
IVPCA morphine group; time to first flatus and first bowel While the effect of IV heparin administration is immediate,
movement occurred, on average, 36 hours sooner in the epi- subcutaneous administration requires 1–2 hours to effect a
dural group, and time to readiness to discharge was the same change on coagulation. Small doses of heparin administered
in both groups. In 2007, Taqi et al. examined thoracic epi- subcutaneously prior to surgery for DVT prophylaxis are
dural analgesia compared to postoperative IV morphine for not a concern in terms of risk of spinal hematoma formation
laparoscopic colectomy (12). Recovery from postoperative (13). Postoperatively, subcutaneous DVT prophylaxis dos-
ileus occurred sooner in the epidural group by 1 or 2 days, ing twice daily of heparin while an epidural catheter is in
and a full diet was resumed earlier. The epidural group place is acceptable. The catheter is removed 2 hours prior to
experienced significantly less pain at rest, with coughing, the next heparin dosing to maximize safety.
and with ambulation. These studies demonstrate the effec- Therapeutic heparin, however, is a different matter. Ruff
tiveness of thoracic epidural analgesia and its superiority in et al. demonstrated that neuraxial procedures performed
allowing early return of bowel function, ability to resume a less than 1 hour after heparin therapy is discontinued
full diet, and early ambulation, as compared to IV narcotics. resulted in a 25-fold increase in spinal hematoma (14). The
All three of these techniques—spinal, caudal, and epidural— effect is even more pronounced if the patient also received
have one thing in common: contraindications. Specifically, aspirin.
absolute contraindications to neuraxial techniques include Low molecular weight heparin (LMWH) was intro-
patient refusal, infection at the planned site of needle punc- duced in 1993 as an alternative to heparin prophylaxis for
ture, elevated intracranial pressure, and bleeding diathesis. prevention of DVT. There have been numerous reports of
There are also several relative contraindications. Bacteremia spinal hematoma in patients receiving LMWH with a neur-
raises the concern that the needle puncture site of the neuraxial blockade. For patients receiving low-dose LMWH for
axial block might allow an epidural abscess or meningitis to thromboprophylaxis preoperatively, it is recommended
develop; however, a clinical scenario may exist where the need that neuraxial anesthesia occur at least 12 hours after the
to avoid a general anesthetic might outweigh the small risk of last dose. In patients who are receiving high-dose LMWH,
such bacterial tracking into the neuraxial space. neuraxial anesthesia should be delayed for 24 hours after
While chronic back pain is not a contraindication to the last dose. Postoperatively, the typical prophylactic
neuraxial techniques, patients with underlying neurological twice-daily dosing of LMWH should only begin 24 hours
disease should be considered carefully, as neuraxial block- after the neuraxial block, and any epidural catheter should
ade might exacerbate their condition, such as in multiple be removed prior to initiation of twice-daily dosing. Once-
sclerosis. The presence of cardiac disease also indicates that daily thromboprophylactic dosing, however, can safely
caution should be applied, as patients who receive a neur- occur with an epidural catheter in place, provided that the
axial block typically experience a sudden decrease in lower first dose occurs at least 8 hours following the initial block-
extremity vascular tone, leading to rapid vasodilation and ade and that any epidural catheter is removed 12 hours after
a significant decrease in systemic vascular resistance. The the last dose prior to its removal (12).
resultant precipitous drop in systolic and diastolic blood Warfarin therapy is another concern. Warfarin antico-
pressure can be extremely dangerous in patients with severe agulation must be stopped 4–5 days prior to surgery, and
coronary artery disease, aortic stenosis, and idiopathic the prothrombin time/international normalized ratio (PT/
hypertrophic subaortic stenosis (IHSS). It is still arguable INR) assessed prior to surgery. Anticoagulation with warfa-
whether the presence of IHSS or aortic stenosis is an absolute rin can be used for thromboprophylaxis in patients with an
contraindication to neuraxial blockade, and many centers indwelling epidural catheter, though the catheter should be
avoid them in the presence of these coexisting morbidities. removed while the INR is still less than 1.5. Typically, this
Anesthetic technique / Awareness under anesthesia 23
is approximately 36 hours following the initial administra- 3 cm medial and 3 cm inferior to the anterior superior iliac
tion of warfarin. Neurologic and motor testing should be spine, in a cephalolateral direction through the abdominal
routinely performed on these patients (15). muscles until contact is made with the iliac bone. As the
All three of the neuraxial techniques have possible side needle is removed, local anesthetic solution is injected. This
effects. Patients can become hypotensive, as their systemic is repeated one to two more times to cover a fan-shaped
vascular resistance decreases. This is due to the sympathec- area, for a total of approximately 10–20 mL of local anes-
tomy caused by blockade of sympathetic fibers along the thetic (19).
thoracic sympathetic chain. Rarely, patients can develop an
unintentionally high spinal anesthetic, leading to bradycar- AWARENESS UNDER ANESTHESIA
dia, apnea, and even loss of consciousness. This “high spi-
nal” must be treated as a general anesthetic, with immediate Awareness under anesthesia is a rare complication of anes-
securing of the airway with endotracheal intubation and thesia, but one which has risen to prominence in the public
supportive therapy until the local anesthetic is metabolized. eye recently. Studies of large numbers of patients in Sweden
Some patients can experience mild back pain at the site demonstrated an overall incidence of 0.16% (20). One can
of needle placement, especially when multiple attempts imagine that this would be a distressing event; the frequency
are needed to place the block. Postdural puncture head- of posttraumatic stress disorder in the 2 years following an
ache (PDPH) can occur, typically following unintended incident of awareness under anesthesia approached 50%,
dural puncture with an epidural needle—a “wet tap.” These even if the patient was not initially distressed by the inci-
headaches are characterized by a slow leak of CSF from dent. A similarly large study in the United States found an
the puncture, leading to a headache that is strongest when overall incidence rate for confirmed intraoperative aware-
standing and lessened when lying. They are often treated ness of 0.13%, and a rate of 0.24% of possible awareness (21).
conservatively with oral fluid therapy, oral caffeine, and the It has long been known that awareness occurs with greater
recumbent position. If relief is not obtained after conserva- frequency in emergent trauma surgery cases, cases involv-
tive treatment, an epidural blood patch can be performed. ing cardiopulmonary bypass, and emergency cesarean sec-
For this procedure, 20 mL of sterile, autologous blood is tions. These are situations where patients may experience
injected into the epidural space, resulting in thrombus for- significant hypotension, requiring a reduction in volatile
mation, sealing of the dura, and cessation of CSF leak. If anesthetic agents below the level that ensures amnesia. If
the diagnosis of PDPH is correct, there is typically immedi- there is a question whether a patient has had an episode of
ate relief of symptoms. Epidural abscess and meningitis are awareness under anesthesia, it is imperative the anesthesi-
possible if proper sterile technique is not used, or if systemic ologist be contacted, and the patient reassured. Psychiatric
infection is present (10). evaluation is usually necessary to help the patient deal with
the potentially distressing nature of this complication.
Transversus abdominis plane block A device available that attempts to determine the depth
of consciousness is the bispectral index (BIS), a monitor
The transversus abdominis plane (TAP) block is a relatively of anesthetic depth approved by the U.S. Food and Drug
new procedure for blocking the abdominal wall afferent Administration. The frontal electroencephalograph is mea-
nerves by way of the lumber triangle of Petit. It can be per- sured, processed using proven algorithms, and reported on
formed using a landmark technique or under ultrasound an arbitrary scale of 0–100. On this scale, 100 equates to
guidance; 20 mL of 0.375% of bupivacaine or levobupi- completely awake and responsive, and zero represents com-
vacaine is then injected into the transversus abdominis plete electrical silence of the brain. A BIS of less than 60 is
neurofascial plane (16–18). In a prospective, randomized generally considered a safe level to ensure adequate depth
controlled trial, McDonnell et al. reported patients under- of anesthesia and lack of awareness under anesthesia. In
going large bowel resection who received the TAP block the B-Aware trial, patients at high risk for awareness under
required 75% less morphine in the first 24 hours, and had anesthesia were randomized to two groups, either routine
significantly lower pain scores at all time points over the care or a BIS-guided anesthetic. While the incidence of
first 24 hours. Additionally, these patients experienced sig- awareness among even high-risk patients was very low, the
nificantly less postoperative nausea and vomiting (16). This risk of awareness was 82% lower in patients with BIS-guided
is an excellent block for patients having smaller abdominal anesthetics (22).
procedures (e.g., ileostomy takedown/revision), on an out- However, there is controversy surrounding the reliabil-
patient basis. ity of the BIS monitor. Use of the BIS monitor and mainte-
nance within the proper depth of anesthesia (as indicated by
Ilioinguinal and iliohypogastric nerve block the BIS algorithm) is still no assurance that the patient will
not have an episode of awareness, as there are numerous
These are field blocks of the terminal branches of the lum- reports to the contrary (23). Additionally, there are numer-
bar plexus, primarily from the L1 root. These blocks are ous conditions that can influence the BIS, causing BIS levels
relatively simple to perform and provide anesthesia in the that are paradoxically high, such as ketamine administra-
inguinal and genital region. A 22-gauge needle is inserted tion or the use of halothane, or paradoxically low, such as
following nitrous oxide termination (24). An analysis of inhibiting the enzyme cyclooxygenase (COX), thereby
the ASA Close Claims Project database demonstrates that blocking the production of prostaglandins resulting in an
between the years of 1961 and 1995 there were 79 claims for anti-inflammatory response. NSAIDs are classified by their
awareness made in the United States; 18 claims for awake selectivity of the COX isoenzymes. There is a risk of bleed-
paralysis, i.e., the inadvertent administration of a muscle ing with these agents; use of NSAIDs is dependent on the
relaxant to an awake patient, and 61 claims for recall under individual patient’s risk factors. Nonselective agents such as
general anesthesia, i.e., recall of events while receiving gen- ibuprofen have an increased side effect profile (bleeding and
eral anesthesia. Most of the claims for awake paralysis rep- antiplatelet effect); however, COX-1 inhibitors are preferred
resented substandard care; less than half of the claims for over selective COX-2 inhibitors such as celecoxib given the
recall were the result of substandard care. The majority of recent evidence of cardiovascular risks associated with
patients experienced temporary emotional distress; 10% of COX-2 agents (28–30).
patients were later diagnosed with posttraumatic stress dis- IV ibuprofen (Caldolor, Cumberland Pharmaceuticals,
order. The awareness of sound without pain was the most Nashville, Tennessee) is administered as 800 mg every
common intraoperative event; 21% of patients experienced 6 hours (31). In a prospective randomized study, Cataldo
pain while aware under anesthesia (25). et al. compared the effect of intramuscular ketorolac in
combination with PCA (morphine) to that of PCA alone.
Narcotic requirements were decreased by 45% (32). They
INTRAOPERATIVE GOAL-DIRECTED suggested that this combination may be particularly
THERAPY beneficial in patients especially prone to narcotic-related
complications.
With the introduction of enhanced surgical recovery pro-
Acetaminophen is a centrally acting analgesic but lacks
grams, there has been renewed interest in optimizing sur-
peripheral anti-inflammatory effects. Oral acetaminophen
gical fluid regimens. The historical debate between liberal
is widely used for acute pain relief. Acetaminophen is a
versus restrictive fluid regimens has been reevaluated, and
common ingredient in many combination oral pain medi-
the idea of individualized goal-directed therapy has been
cations; it is vital to counsel the patient not to exceed the
introduced and subjected to a number of randomized con-
4,000 mg daily maximum dose due to the risk of hepato-
trolled trials. While untreated hypovolemia can be detri-
toxicity. Systematic reviews of randomized controlled tri-
mental to patients, fluid overload can be just as (if not more)
als (RCTs) confirm the efficacy of oral acetaminophen for
hazardous. By tailoring fluid administration to an indi-
acute pain (33). However, acetaminophen has a slow onset
vidual patient’s needs using a treatment algorithm based
of analgesia and until recently the nonavailability of the oral
on closely monitored flow variables, postoperative recovery
can be improved with reduced morbidity, less gastrointesti- route immediately after surgery limited its value in treating
nal dysfunction, and reduced hospital stay (26). immediate postoperative pain. An IV form of acetamino-
phen is now commercially available (Ofirmev, Mallinkrodt
Pharmaceuticals, St. Louis, Missouri). Acetaminophen’s
INTRAOPERATIVE OPIOID SPARING major advantage over NSAIDs is its lack of interference
with platelet function and safe administration in patients
A recent development in general anesthesia techniques is with a history of peptic ulcers or asthma. Opioid-sparing
opioid-sparing anesthetics. Opioids are known to be associ- effects have been associated with acetaminophen adminis-
ated with some adverse side effects, including postoperative tered intravenously (34). A mixed trial comparison found
delirium, ileus, allergic reactions, and nausea. Opioid- a decrease in 24-hour morphine consumption when acet-
sparing protocols attempt to use multimodal strategies to aminophen, NSAIDs, or COX-2 inhibitors are given in
limit these side effects while still providing safe and com- addition to PCA morphine after surgery with a reduction
fortable care for the patient. Most guidelines use a combi- in morphine-related adverse effects. However, the study did
nation of acetaminophen, ibuprofen, ketamine, lidocaine not find any clear differences between the three nonopioid
infusion, Decadron, dexmedetomidine, and local anesthet- agents (35). A systematic review identified 21 studies com-
ics. There are several studies that have shown great success paring acetaminophen alone or in combination with other
adopting this technique (27). NSAIDs and reported increased efficacy with the combina-
tion of two agents than with either alone (35). Current dos-
POSTOPERATIVE PAIN MANAGEMENT ing is 1000 g IV every 6 hours.
In patients undergoing major colorectal operations, Beck
A number of the pain management options have been and colleagues compared 66 patients who received multi-
grouped into multimodality pain management pro- modality pain management to 167 patients managed with
grams. Many of these include opioid-sparing techniques. opioid PCA and found that the multimodality patients had
Nonsteroidal anti-inflammatory drugs (NSAIDs) reduce lower pain scores, used less opioids, had less opioid-related
the amount of opiates requested and administered to the adverse events, and had decreased lengths of postoperative
patient and therefore opioid side effects (28). They are hospital stay (average 1.8 days) (36). A sample multimodal-
useful in treating mild to moderate pain. NSAIDs act by ity pain management regimen is presented in Table 3.2.
Positioning / Prone 25
Table 3.2 Sample multimodality pain management
Preoperative
Acetaminophen (paracetamol) 1000 mg IV in preop Cervical
Ibuprofen 800 mg IV in preop plexus
Intraoperative
Brachial
Liposomal bupivacaine 266 mg wound inﬁltration plexus
Postoperative
Acetaminophen (paracetamol) 1000 mg IV every 6 hours Radial
until patient taking oral meds nerve
Ibuprofen 800 mg IV every 8 hours until patient taking Ulnar
oral meds nerve
PCA (morphine or Dilaudid) for severe pain (scale 6–10)
until patient taking oral meds
Oxycodone 10 mg PO every 4 hours for moderate pain
when taking oral medication
Abbreviations: IV, intravenously; PCA, patient-controlled
anesthesia; PO, by mouth.
Peroneal
nerve
POSITIONING
SUPINE
This is the most common surgical position; it results in the Figure 3.2 Nerves at risk for injury during positioning for
least hemodynamic and ventilatory changes and is frequently a surgical procedure.
the best position for surgical exposure. The supine position
is not perfect, of course, as it creates certain pressure points and head in line with the rest of the body. There are sev-
that, given time, result in ischemia over certain bony promi- eral different pillow types that allow for proper positioning
nences, such as the heels, sacrum, and back of the head. The of the head in a neutral position with the remainder of the
head should rest on a soft support to spread the pressure, body, while keeping the eyes, nose, and chin free from pres-
decreasing the incidence of pressure points, thus preventing sure. There is a low, but significant, risk that pressure on the
alopecia. Particular care must be given to the arms, including eye or surrounding orbit will lead to increased intraocular
careful padding of the elbows and wrists. Abduction of the pressure, decreased retinal artery blood flow, and resultant
arms must not exceed 90° from the body to prevent compro- blindness, if the intraocular pressure exceeds systemic pres-
mising blood flow to the distal arm (37). Trendelenburg posi- sure. Although this is a rare complication associated with
tioning while supine has several anesthetic implications, as it the prone position, it is, nevertheless, devastating. Extreme
causes the diaphragm to move cephalad, causing increased care must be taken to avoid this life-changing occurrence.
airway pressures and possibly advancing the endotracheal A soft but firm roll is placed under the hips of the prone
tube into an endobronchial position. patient. Placement is critical since positioning of the roll too
Shoulder braces are sometimes used to prevent the patient far caudally will lead to poor exposure by not elevating the
from sliding off the table during extreme Trendelenburg buttocks high enough to separate them. A low roll will also
positioning, though this can cause injury by compressing cause discomfort in the thighs during a local anesthetic and
the brachial plexus (38). The most common upper extrem- postoperatively from a general or regional anesthetic. A roll
ity injury is to the ulnar nerve, which is three times more placed too far cephalad may restrict venous return through
likely in men who undergo general anesthesia. This seems to the inferior vena cava resulting in transient hypotension
occur despite padding of the extremity (39). Other nerves at and the possibility of increased intraoperative bleeding.
risk due to positioning are illustrated in Figure 3.2. Excessive compression of the abdomen can also impair
ventilation. Bean bags or chest rolls can improve ventila-
PRONE tion. An additional roll is placed beneath each ankle to keep
the toes off the surface of the operating table. All pressure
Even when a procedure is planned in the prone position, points including knees, elbows, breasts, and male genitalia
induction of general anesthesia and intubation of the tra- need to be carefully monitored and adequately padded.
chea should occur in the supine position. The patient is The thorax should be supported with chest rolls that
then turned prone, taking care to keep the cervical spine extend from the clavicle to the iliac crest. The arms can be
placed at the side of the body in a neutral position, with care- in the United States, this examination can also be performed
ful padding of the elbows to prevent injury. Alternatively, with the patient in the left lateral position. The position is
the arms can be positioned alongside the head, taking care particularly suited to the pregnant patient, patients with
that the arms are not abducted greater than 90° to prevent severe chronic obstructive pulmonary disease, some ampu-
injury to the brachial plexus (37,38). Great care must be tees, and patients with spinal cord or other neurologic inju-
taken to not inadvertently dislodge the endotracheal tube ries. For local anesthetics, this position appears to eliminate
while prone, as it is exceedingly difficult to reintubate or complaints of pubic pain and low backache. It may limit visu-
mask ventilate a patient in the prone position. alization of the right posterior quadrant (42) but is a good
The prone position has much to offer for most anorec- alternative for sphincterotomies, fistulotomies, and pilonidal
tal surgery. The entire operating teams including surgeon, cysts being treated under local anesthetic.
assistant surgeon, and scrub nurse have a full and unob- Just as with prone positioning, it is imperative that the
structed view of the operative field. Thus, both the assistant head be kept in a neutral position while turning the patient.
and the nurse can always see the maneuvers of the surgeon, Additionally, extra cushioning is needed under the head to
offering the best chance for improved efficiency; lighting of keep the cervical and thoracic spines in line. An axillary
any type is enhanced. Anorectal anatomy is clearly delin- roll needs to be placed just caudad to the dependent axilla in
eated, and any bleeding automatically drains away from the order to prevent compression injuries to the brachial plexus.
operative field instead of onto the floor (39). Disadvantages It should not be placed in the axilla, as the purpose is for
include concern about the airway as mentioned. the weight of the thorax to be borne by the chest wall. The
dependent arm is extended perpendicular to the body on
LATERAL DECUBITUS a padded armboard, while the nondependent arm is simi-
larly extended on an armrest suspended in such a way that
The modified left lateral or Sims position has been champi- the arm is not abducted greater than 90° from the body.
oned by the Ferguson Clinic in their descriptions of hem- Additionally, the arm should not be raised superior to the
orrhoidectomy technique (Figure 3.3) (40,41). This position level of the deltoid. A pillow or cushion should be placed
allows an assistant easy access, but unless the patient is cor- between the knees (37,38).
rectly positioned, the upper buttock, in a natural response
to gravity, may coapt with the lower buttock, compromis- LITHOTOMY
ing visibility. Furthermore, the assistant usually has no
opportunity to see any of the operation as the upper buttock The lithotomy position is very common in colorectal surgery.
retraction is most effectively accomplished from the contra- The hips are flexed 80°–100° from the trunk, and the legs are
lateral side. abducted 30°–45° from midline. It is important that the legs
The patient is placed in a left lateral position with the always be moved simultaneously to prevent lumbar spine tor-
buttocks extended beyond the side of the table. As shown sion, and that the legs be carefully padded to reduce the risk of
in Figure 3.3, the patient’s back is slightly flexed and angled injury. In a retrospective review of patients undergoing sur-
across the table, which positions the head at the opposite side gery in the lithotomy position, Warner et al. found that the
of the table to prevent the patient from falling off the side of most common lower extremity nerve injury was to the com-
the table. The legs can be positioned in a variety of ways: the mon peroneal nerve, accounting for 78% of nerve injuries. It
left thigh and leg are extended straight to the bottom of the was postulated that the cause was compression of the nerve
table and the right thigh and leg flexed on a pillow into a near between the leg support and the lateral head of the fibula (43).
knee-chest position or both legs can be flexed (with a small While rare (1 in 8,720), the incidence of compartment
pillow between the legs for padding). The patient’s arms can syndrome of the lower extremities is markedly higher in
be crossed either in front of the head or under the forehead. the lithotomy position than in all other surgical positions.
The left lateral position can be particularly useful for office Compartment syndrome occurs when high tissue pressure
examinations. Although the prone-jackknife position is prob- builds within the closed space of the anterior compartment.
ably the most popular position for rigid proctosigmoidoscopy Ischemia of the tissue in the compartment results in edema
of the interstitium, thereby raising compartment pressure.
Since perfusion is dependent on compartment pressure
being lower than mean arterial pressure to allow tissue per-
fusion, any situation where increased compartment pres-
sure and/or decreased arterial flow into the tissue can result
in ischemia. The result is capillary endothelial damage and
even greater interstitial edema. Unfortunately, it is not com-
pletely understood why some patients develop a compart-
ment syndrome, while others do not. As a result, no safe
maximum time limit can be defined (44). Early diagnosis
and treatment with fasciotomy is imperative. Analysis of
Figure 3.3 Modified left lateral or Sims position. closed claims in cases of compartment syndrome due to the
Conclusion 27
lithotomy position during colorectal surgery demonstrated allowed 24 hours, the colorectal surgeon must document,
an average indemnity payment of $426,000. Great care must in the medical record, the reason for continuation of the
be taken in the positioning and padding, as patients them- antibiotic.
selves cannot express any pain or discomfort they may be Venous thromboembolism was discussed earlier in this
experiencing while under general or regional anesthesia. chapter, and involvement of the anesthesiologist and associ-
ated regional anesthesia play a significant role here. As pre-
viously mentioned, an epidural catheter must be removed
SURGICAL CARE IMPROVEMENT at an appropriate time surrounding the initiation and dis-
PROJECT continuation of heparin, LMWH, or warfarin. The risks of
inadequate venous thromboembolism prophylaxis must
be weighed against the benefits of regional anesthesia for
PROCESS AND OUTCOME MEASURES colorectal surgical patients.
Postoperative pneumonia is a complication where the
The SCIP of the United States is a national quality initia- cause is multifactorial. Ventilator management and weaning
tive involving the American Society of Colorectal Surgeons, protocols for patients requiring postoperative mechanical
the American College of Surgeons, the ASA, the American ventilation may fall under the purview of the anesthesiologist.
Hospital Association, the Association of Perioperative
Registered Nurses, and a host of governmental agencies ded-
icated to improvement in health care (45). The goals of the
SCIP partnership were to reduce the incidence of surgical CONCLUSION
complications by 25% by the year 2010 and to promote the
use of evidence-based care processes known to reduce surgi-
Although the sum total of anesthesia practice can hardly
cal complications.
be related in a chapter, we have attempted in the preceding
Out of approximately 40 million major operations each
pages to highlight areas in anesthesia practice of which the
year, postoperative complications account for up to 22%
colorectal surgeon should be aware. Improved patient sat-
of preventable deaths among patients, depending on the
isfaction through reduction of postoperative pain, earlier
complication. These complications accounted for 2.4 mil-
ambulation, and quicker return of bowel function and diet
lion additional hospital days and $9.3 billion in additional
will have a marked impact on surgical outcomes. Thoracic
charges each year (46).
epidural anesthesia/analgesia is becoming a standard for
SCIP focuses on areas where the incidence and cost of
many colorectal surgical procedures, whether as the sole
the most common and preventable complications are high:
anesthetic, or in conjunction with general anesthesia.
1. Surgical site infections Awareness under anesthesia is a rare, but serious concern,
2. Adverse cardiac events highlighted more recently in the media and receiving much
3. Venous thromboembolism greater appreciation among surgical patients. Timely support-
4. Postoperative pneumonia ive care, including psychological counseling, may improve
outcome and reduce the incidence of posttraumatic stress dis-
Although not limited to anesthesia care, the anesthesi- order. Oversedation resulting in hypoventilation, hypoxemia,
ologist and colorectal surgeon must partner in attempts to and hypercarbia can produce devastating results. Extreme
meet the expectations set by the national SCIP initiative. caution must be given to the patient who is restless, but
One such initiative is the administration of prophylac- sedated. Loss of airway is the ultimate disaster under general
tic antibiotics within 1 hour of surgical incision. Although anesthesia and is a surgical, as well as anesthetic, emergency.
not typically considered “anesthetic agents,” antibiotics may Proper positioning requires the vigilance of the anesthe-
best be given within 1 hour before incision if administered sia provider, the colorectal surgeon, and the operating room
by the anesthesia provider. Frequent operating room and nurses. Severe nerve injuries can generally be avoided with
turnover delays may result in an antibiotic administration the use of padding. Even with appropriate padding, there is
well before the 1 hour limit if given in the preoperative hold- an increased incidence of neurologic injury with the use of
ing area. Late patient arrivals for same-day admit surgery or stirrups in the lithotomy position. Extra care must be taken
administrative paperwork delays may result in inadequate with the patient in the prone position, as neck injuries from
or insufficient time to infuse the antibiotic prior to going to improper turning, endotracheal tube dislodgement, or peri-
the operating room, with the result of no antibiotic being operative blindness from periorbital pressure can all result
given or being given only if the “missed dose” is noticed by in devastating outcomes.
someone in the operating room. Partnership of the surgeon and the anesthesiologist may
Although no longer reportable as public information, help improve outcomes, reduce surgical site infections,
prophylactic antibiotic selection for surgical patients is improve perioperative cardiac morbidity and mortality, and
monitored, as is discontinuation of the antibiotic within 24 reduce the incidence of venous thromboembolism. Whether
hours after the surgery end time (48 hours for cardiac sur- in the office setting, outpatient center, or surgical hospital,
gery patients). If an antibiotic is felt to be needed beyond the safe anesthesia practice is paramount.
REFERENCES 22. Myles PS et al. Lancet 2004;363:1757–63.

23. Rampersad SE, Mulroy MF. Anesth Analg 2005;100:
1. Stoelting RK, Miller RD. Scope of anesthesia practice. 1363–4.
In Basics of Anesthesia, 5th Edition. Philadelphia, PA: 24. Dahaba AA. Anesth Analg 2005;101:765–73.
Churchill Livingstone Elsevier, 2007, p. 11. 25. Domino K et al. Anesthesiology 1999;90(4):1053–61.
2. Hutson LR, Vachon CA. Anesthesiology 2005;103(4): 26. Noblett SE, Horgan AF. Sen Colon Rectal Surgery
885–9. 2010;21:160–4.
3. Stoelting RK, Miller RD. Local anesthetics. In Basics 27. Mulier JP. Practical Implications 4th ESPCOP Meeting,
of Anesthesia, 5th Edition. Philadelphia, PA: Churchill Bruges, December 2013.
Livingstone Elsevier, 2007, pp. 123–34. 28. Lowder JL et al. Am J Obstet Gynecol 2003;189:
4. Barash PG et al. Local Anesthetics in Clinical 1559–62; discussion 62.
Anesthesia, 5th Edition. Philadelphia, PA: Lippincott 29. Dajani EZ, Islam K. J Physiol Pharmacol 2008;
Williams and Wilkins, 2006, pp. 449–67. 59(Suppl 2):117–33.
5. Haas E et al. Am Surg 2012;78(5):574–81. 30. De Oliveira GS, Jr. et al. Anesth Analg 2012;114:
6. Golf M et al. Adv Ther 2011;28(9):776–88. 424–33.
7. Cohen SM. J Pain Res 2012;5:567–72. 31. Southworth S et al. Clin Ther 2009; 31(9): 1922–35.
8. Stoelting RK, Miller RD. Choice of anesthetic 32. Cataldo P et al. Surg Gynecol Obstet 1993;176:
technique. In Basics of Anesthesia, 5th Edition. 435–8.
Philadelphia, PA: Churchill Livingstone Elsevier, 2007, 33. Toms L et al. Cochrane Database Syst Rev 2008:
pp. 178–84. CD004602.
9. Bhananker SM et al. Anesthesiology 2006;104: 34. Maund E et al. Br J Anaesth 2011;106:292–7.
228–34. 35. Ong CK et al. Anesth Analg 2010;110:1170–9.
10. Stoelting RK, Miller RD. Spinal and epidural anesthe- 36. Beck DE et al. Ochsner J 2015;15(4): 408–12.
sia. In Basics of Anesthesia, 5th Edition. Philadelphia, 37. Barash PG et al. Patient positioning. In Clinical
PA: Churchill Livingstone Elsevier, 2007, pp. 241–71. Anesthesia, 5th Edition. Philadelphia, PA: Lippincott
11. Carli F et al. Dis Colon Rectum 2001;44(8):1083–9. Williams and Wilkins, 2006, pp. 643–65.
12. Taqi A et al. Surg Endoscopy 2007;21:247–52. 38. Stoelting RK, Miller RD. Positioning and associ-
13. Liu SS, Mulroy MF. Reg Anesth Pain Med ated risks. In Basics of Anesthesia, 5th Edition.
1998;23:157–63. Philadelphia, PA: Churchill Livingstone Elsevier, 2007,
14. Ruff RL, Dougherty JH. Stroke 1981;12:879–81. pp. 291–303.
15. Horlocker TT et al. The Second ASRA Consensus 39. Goldberg SM et al. Essentials of Anorectal Surgery.
Conference on Neuraxial Anesthesia and Philadelphia, PA: JB Lippincott; 1980:80.
Anticoagulation. http://www.asra.com/consensus- 40. Ferguson JA et al. Surgery 1971;70:480–4.
statements/RAPM-Anticoagulation.pdf 41. Nivatvongs S. Dis Colon Rectum 1980;23:308–9.
16. McDonnell JG et al. Anesth Analg 2007;104(1):193–7. 42. Cheney FW et al. Anesthesiology 1999;90(4):
17. Rafi AN. Anaesthesia 2001;56(10):1021–6. 1062–9.
18. El-dawlatly AA et al. Internet J Anesthesiol 43. Warner MA et al. Anesthesiology 1994;81:6–12.
2006;16(2):1–3. 44. Beraldo S, Dodds SR. Dis Colon Rectum 2006;49:
19. Stoelting RK, Miller RD. Peripheral nerve blocks. In 1772–80.
Basics of Anesthesia, 5th Edition. Philadelphia, PA: 45. “Medicare Quality Improvement Community”
Churchill Livingstone Elsevier, 2007, pp. 273–90. website. http://www.medqic.org accessed March 31,
20. Sandin RH et al. Lancet 2000;355:707–11. 2008.
21. Sebel PS et al. Anesth Analg 2004;99:833–9. 46. Zhan C, Miller MR. JAMA 2003;290:1868–74.
4
Sepsis
JENNIFER S. BEATY AND MORIAH E. WRIGHT
together “to provide a conceptual and a practical frame-

CHALLENGING CASE work to define the systemic inflammatory response to
infection” (1). This is the origin of the concept of systemic
Five days after a subtotal colectomy with end inflammatory response syndrome (SIRS). This term came
ileostomy for medically refractory ulcerative colitis, about because of recognition that a large number of clini-
your patient is febrile with tachycardia. Your patient cal conditions not involving infection could still initiate
also has leukocytosis with left shift and decreased an inflammatory response (1). Sepsis was defined as SIRS
urine output. Your patient rules in for sepsis criteria. with a known source of infection (Table 4.1) (1). This was
also the advent of multiple organ dysfunction syndrome
(MODS), which describes a continuum of organ dysfunc-
tion (1). Changes in organ function over time can be viewed
CASE MANAGEMENT
as an important prognostic indicator (1). These definitions
were revisited in 2001 at the International Sepsis Definitions
Your patient is transferred to the intensive care unit
Conference (2). This included ACCP, SCCM, European
(ICU), and the Surviving Sepsis Campaign Bundles
Society of Intensive Care Medicine (ESICM), American
are initiated. The patient receives a bolus of IV fluids,
Thoracic Society, and Surgical Infection Society (SIS) rep-
serum lactate is drawn, blood cultures are obtained,
resentatives. At this conference, the description of variables
empiric antibiotics are initiated, and computed
that may contribute to the diagnosis of sepsis was expanded
tomography (CT) scan of the abdomen and pelvis is
without changing the definition of SIRS or MODS. In short,
performed. The study reveals intraabdominal abscess
the committee felt that the signs and symptoms of sepsis
surrounding the rectal stump staple line, which
were more varied than the initial criteria, but the exist-
appears amenable to drainage.
ing concepts of sepsis, severe sepsis, and septic shock were
robust and should remain as previously described (2).
INTRODUCTION SEPSIS-3
There are times in the postoperative period when patients In 2015, the SCCM and the ESICM met and determined
do not follow a normal postoperative course. The patient new definitions for sepsis and associated conditions (3).
may become septic, require transfer to the ICU, or require This group sought to differentiate between uncomplicated
an unplanned operation. Since 1991, there have been infection and sepsis as well as to update the definitions of
increasing efforts to define the disease process of sepsis and sepsis and septic shock to better reflect current understand-
improve upon its high mortality rate. ing of pathobiology. The key concepts of sepsis include the
following: it is the primary cause of death from infection;
it is a syndrome shaped by pathogen factors and host fac-
SYSTEMIC INFLAMMATORY RESPONSE tors with characteristics that evolve over time; organ dys-
SYNDROME, SEPSIS, AND MULTIPLE function may be occult, and therefore, its presence should
ORGAN DYSFUNCTION SYNDROME be considered in any patient presenting with infection and
vice versa; the clinical and biological phenotype of sepsis
In 1991, the American College of Chest Physicians (ACCP) can be modified by preexisting acute illness, long-standing
and the Society of Critical Care Medicine (SCCM) came comorbidities, medications, and interventions; and specific
29
30 Sepsis
Table 4.1 Sepsis definitions
2001—SCCM/ESICM/ACCP/
1991—ACCP/SCCMa ATS/SISb 2015—Sepsis-3c
SIRS Two or more of the Unchanged
following:
• Temperature >38°C or
<36°C
• Heart rate >90 bpm
• Respiratory rate >20
breaths/min or
PaCO2 <33 mm Hg
• White blood cell count
>12,000/mm3, <4,000/
mm3, or >10% bands
Sepsis SIRS with known source of Infection with some of the following: Life-threatening organ dysfunction
infection General variables caused by a dysregulated host
• Fever response to infection.
• Hypothermia Organ dysfunction can be described
• Heart rate >90/min by an increase in SOFA score of
• Tachypnea two points or more.
• Altered mental status
• Significant edema or positive
fluid balance
• Hyperglycemia
Inflammatory variables
• Leukocytosis
• Leukopenia
• Normal WBC count with >10%
bands
• Plasma C-reactive protein >2 SD
above normal
• Plasma procalcitonin >2 SD
above normal
Hemodynamic variables
• Arterial hypotension
• SVO2 >70%
• Cardiac index >3.5 L/min/M23
Organ dysfunction variables
• Arterial hypoxemia
• Acute oliguria
• Creatinine increase >0.5
• Coagulation abnormalities
• Ileus
• Thrombocytopenia
• Hyperbilirubinemia
Tissue perfusion variables
• Hyperlactatemia
• Decreased capillary refill
or mottling
Severe Sepsis associated with Unchanged Removed, felt to be redundant
sepsis organ dysfunction,
hypoperfusion, or
hypotension
(Continued)
Evaluation and diagnosis / Sepsis bundles 31
Table 4.1 (Continued) Sepsis definitions
2001—SCCM/ESICM/ACCP/
1991—ACCP/SCCMa ATS/SISb 2015—Sepsis-3c
Septic Sepsis with hypotension Unchanged A subset of sepsis with profound
shock despite adequate fluid circulatory, cellular, and metabolic
resuscitation with abnormalities; associated with
perfusion abnormalities greater mortality than sepsis alone.
(lactic acidosis, oliguria, Persisting hypotension requiring
or acute alteration in vasopressors to maintain MAP
mental status) ≥65 mm Hg, serum lactate
>2 mmol/L despite adequate
volume resuscitation
MODS Presence of altered organ Unchanged Unchanged
function in acutely ill
patient such that
homeostasis cannot be
maintained without
intervention
Abbreviations: ACCP, American College of Chest Physicians; ATS, American Thoracic Society; ESICM, European Society of Intensive Care
Medicine; MAP, mean arterial pressure; MODS, multiple organ dysfunction syndrome; SCCM, Society of Critical Care Medicine;
SD, standard deviation; SIRS, systemic inflammatory response syndrome; SIS, Surgical Infection Society; SOFA, Sequential Organ
Failure Assessment; WBC, white blood cell.
a Bone RC et al. Chest 1992;101(6):1644–55.
b Levy MM et al. Crit Care Med 2003;31(4):1250–6.
c Singer M et al. JAMA 2016;315(8):801–10.
infections may result in local organ dysfunction without sepsis (5). Four measures should be performed within the
generating a dysregulated systemic host response (3). Sepsis first 3 hours (Table 4.3), including measuring serum lactate
is now recognized to involve early activation of pro- and level, drawing blood cultures, initiating empiric antibiotics,
anti-inflammatory responses, as well as major modifica- and administering crystalloid bolus for hypotension of lac-
tions in nonimmunologic pathways (cardiovascular, neu- tic acidosis (5). Three additional measures should be imple-
ronal, autonomic, hormonal, bioenergetic, metabolic, and mented within the first 6 hours (Table 4.3).
coagulation), all of which have prognostic significance (3). The International Multicentre Prevalence Study on Sepsis
Organ dysfunction severity can be described using a vari- (IMPreSS) evaluated compliance with the SSC bundles and
ety of scoring systems, but the most common one in current correlated compliance with mortality in 1794 patients (6).
use, and the one recommended by Sepsis-3 is the Sequential This is the first report of compliance with the 2012 SSC bun-
Organ Failure Assessment (SOFA) (Table 4.2) (3). The baseline dles. IMPreSS found the overall hospital mortality rate in
SOFA score is assumed to be zero in patients without preexist- sepsis to be 28.4% (6). Compliance with all of the evidence-
ing organ dysfunction. A SOFA score greater than or equal based bundle metrics for the treatment of sepsis was low:
to two reflects an overall mortality risk of approximately 10% only 19% compliance for 3-hour bundle and 35.5% for the
in a general hospital population with suspected infection. It 6-hour bundle. Patients whose care included compliance
is important to note that SOFA is not intended to be a stand- with all of the metrics in the 3-hour bundle had a 40% reduc-
alone definition of sepsis, and failure to meet two or more tion in the odds of dying in the hospital (20% versus 31%,
criteria should not lead to a deferral of investigation or treat- p < 0.001). Patients whose care included compliance with all
ment of an infection (3). It is also important to note that these metrics in the 6-hour bundle had a 36% reduction in odds of
updated definitions were not endorsed by the ACCP (4). dying in the hospital (22% versus 32%, p < 0.001) (6).
In another recent study in 2014, Levy et al. evaluated
29,470 patients with sepsis and measured compliance with
the SSC bundles and outcomes (7). If compliance with resus-
EVALUATION AND DIAGNOSIS citation bundle was high, mortality was 29%, as opposed to
low compliance with resuscitation bundle, the mortality
SEPSIS BUNDLES was 38.6% (p < 0.001) (7). Hospital and ICU length of stay
decreased 4% for every 10% increase in compliance with
The Surviving Sepsis Campaign (SSC) was developed to resuscitation bundle. Increased compliance with SSC bun-
reduce mortality in severe sepsis. Specific measures should dles corresponded to a 25% relative risk reduction in mor-
be completed in the workup and treatment of patients with tality rate. This study demonstrates performance metrics
32 Sepsis
Table 4.2 Sequential Organ Failure Assessment (SOFA) score
Score
System 0 1 2 3 4
Respiration
PaO2/FiO2, ≥400 <400 <300 <200 with respiratory <100 with respiratory
mm Hg support support
Coagulation
Platelets, ×103/µL ≥150 <150 <100 <50 <20
Liver
Bilirubin, mg/dL <1.2 1.2–1.9 2–5.9 6–11.9 >12
Cardiovascular MAP MAP Dopamine <5 or Dopamine 5.1–15 or Dopamine >15 or
≥70 mm Hg <70 mm Hg dobutamine epinephrine ≤0.1 epinephrine >0.1
(any dose) or norepinephrine or norepinephrine
≤0.1 >0.1
Central nervous
system
Glascow Coma 15 13–14 10–12 6–9 <6
Score
Renal
Creatinine, mg/dL <1.2 1.2–1.9 2.0–3.4 3.5–4.9 >5
Urine output, <500 <200
mL/day
Source: Adapted from Singer M et al. JAMA 2016;315(8):801–10.
can drive change in clinical behavior, improve quality of not demonstrate a survival benefit for EGDT. These trials
care, and may decrease mortality in patients with severe include the 2014 ProCESS, the 2014 ARISE, and the 2015
sepsis and septic shock (7). ProMISe trials (8–10). In the ProCESS study, 1,341 patients
In contrast to the IMPreSS trial and Levy’s study, three diagnosed with sepsis in the emergency room were assigned
recently completed, large multicentered randomized tri- to EGDT versus usual care (8). There were no differences
als comparing early goal-directed therapy (EGDT) to in 90-day mortality, 1-year mortality, or need for organ
usual therapy in academic and community hospitals did support (8). The ARISE study randomized 1,600 patients
Table 4.3 Surviving sepsis campaign bundles
To be completed within 3 hours:

1. Measure lactate level
2. Obtain blood cultures prior to administration of antibiotics
3. Administer broad-spectrum antibiotics
4. Administer 30 mL/kg crystalloid for hypotension or lactate ≥4 mmol/L
To be completed within 6 hours:
5. Administer vasopressors for hypotension not responding to initial fluid resuscitation
• Maintain MAP ≥65 mm Hg
6. If persistent hypotension after initial volume resuscitation (MAP ≤65 mm Hg) or if lactate >4 mmol/L,
• Either
• Repeat focused exam (after initial fluid resuscitation) by licensed independent practitioner including vital signs,
cardiopulmonary, capillary refill, pulse, and skin findings
• Or two of the following:
• Measure central venous pressure (CVP), goal CVP ≥ 8 mm Hg
• Measure central venous oxygen saturation (Scvo2), goal Scvo2 ≥ 70%
• Perform bedside cardiac ultrasound
• Conduct dynamic assessment of fluid responsiveness with either fluid challenge or passive leg raise
7. Remeasure lactate if initial lactate was elevated
Source: Adapted from Dellinger RP et al. Crit Care Med 2013;41:580–637.
Infection / Surgical site infection 33
diagnosed with sepsis in the emergency room to EGDT ver- tobacco use, altered immune response, corticosteroid use,
sus usual care. There were no significant differences in sur- recent hospitalization, length of preoperative hospitaliza-
vival time, in-hospital mortality, duration of organ support, tion, and colonization with microorganisms (11).
or length of hospital stay (9). The ProMISE study enrolled Superficial incisional SSI occurs within 30 days postop-
1,260 with sepsis and randomized to EGDT or usual care. eratively and involves skin or subcutaneous tissue of the
There was no difference in mortality between the groups, incision and at least one of the following: purulent drain-
EGDT (29.5%) versus control (29.2%) (10). Interestingly, the age from the superficial incision; organisms isolated from
EGDT group had higher costs, and the probability it was an aseptically obtained culture of fluid or tissue from the
cost effective was only 20% (10). The true benefit of mod- superficial incision; at least one of the following signs or
ern therapies for sepsis may lie in the early identification of symptoms of infection—pain or tenderness, localized
those with the condition, which would ultimately lead to an swelling, redness, or heat; superficial incision is deliberately
early treatment, rather than a specific algorithm. opened by a surgeon and is culture-positive or not cultured
(a culture-negative finding does not meet this criterion);
and diagnosis of superficial incisional SSI by the surgeon or
attending physician (11). When stratifying by wound clas-
INFECTION sification, the clean, clean-contaminated, contaminated,
and dirty wound classifications had superficial SSI rates of
PREVENTION 1.76%, 3.94%, 4.75%, and 5.16%, respectively (12).
Deep incisional SSI occurs within 30 days after the opera-
Surgical prophylaxis has been a topic of discussion for many tive procedure if no implant is left in place or within 1 year if
years. Guidelines from the Infectious Diseases Society of implant is in place and the infection appears to be related to
America (IDSA) were originally published in 1999, but they the operative procedure, involves deep soft tissues (e.g., fas-
received an update in 2013 (11). This update included timing cial and muscle layers) of the incision, and the patient has
of preoperative antibiotic dosing. IDSA guidelines recom- at least one of the following: purulent drainage from the
mend dosing within 60 minutes prior to incision, and anti- deep incision but not from the organ/space component of
biotic agents with a prolonged administration time should the surgical site, a deep incision spontaneously dehisces or
be administered such that the dose is completed within 60 is deliberately opened by a surgeon and is culture-positive
minutes of incision (11). For antibiotic selection and dosing, or not cultured and the patient has at least one of the fol-
the ISDA provided guidance on agents for specific proce- lowing signs or symptoms—fever (>38°C) or localized pain
dures as well as dosing for obese patients, the specifics of or tenderness (a culture-negative finding does not meet this
which are outside the scope of this chapter. Duration of pro- criterion); an abscess or other evidence of infection involv-
phylactic antibiotics should be no more than 24 hours after ing the deep incision is found on direct examination, during
surgery, and this includes a lack of need for prophylaxis for reoperation, or by histopathologic or radiologic examina-
the presence of indwelling drains or catheters (11). tion; and diagnosis of a deep incisional SSI by a surgeon or
attending physician (11). When stratifying by wound classi-
SURGICAL SITE INFECTION fication, the clean, clean-contaminated, contaminated, and
dirty wound classifications had deep incisional infection
Surgical site infection (SSI) is a common complication after rates of 0.54%, 0.86%, 1.31%, and 2.1%, respectively (12).
any surgical procedure, the risk of which is delineated by Organ/space SSI involves any part of the body, exclud-
patient factors as well as the type of procedure. One way the ing the skin incision, fascia, or muscle layers, that is opened
risk of SSI has been stratified is with the National Healthcare or manipulated during the operative procedure. Specific
Safety Network wound classification system. This system sites are assigned to organ/space SSI to further identify
divides wounds into four classes: clean, clean-contaminated, the location of the infection (e.g., endocarditis, endome-
contaminated, and dirty or infected. Clean wounds include tritis, mediastinitis, and osteomyelitis). Organ/space SSI
wounds without infection and without transection of the must meet the following criteria: infection occurs within
respiratory, biliary, gastrointestinal, genital, or uninfected 30 days after the operative procedure if no implant is in
urinary tracts. Clean-contaminated wounds involve tran- place or within 1 year if implant is in place and the infection
section of one of the above tracts. Contaminated wounds appears to be related to the operative procedure, infection
involve fresh open accidental wounds or operations with involves any part of the body, excluding the skin incision,
major break in sterile technique or gross spillage from the fascia, or muscle layers, that is opened or manipulated dur-
gastrointestinal tract. Dirty wounds include old traumatic ing the operative procedure, and the patient has at least one
wounds with devitalized tissue, existing clinical infection, of the following: purulent drainage from a drain that is
or wounds associated with perforated viscera. With increas- placed through a stab wound into the organ/space, organ-
ing contamination of the wound comes increased risk for isms isolated from an aseptically obtained culture of fluid
SSI. Patient factors that contribute to an increased risk or tissue in the organ/space; an abscess or other evidence of
for SSI include extremes of age, nutritional status, obesity, infection involving the organ/space that is found on direct
diabetes mellitus, concurrent remote body-site infection, examination, during reoperation, or by histopathologic or
34 Sepsis
radiologic examination; and diagnosis of an organ/space SSC specifically recommends against the use of hydroxy-
SSI by a surgeon or attending physician (10). When stratify- ethyl starches for fluid resuscitation of severe sepsis and
ing by wound classification, the clean, clean-contaminated, septic shock (5).
contaminated, and dirty wound classifications had organ/
space SSI rates of 0.28%, 1.87%, 2.55%, and 4.54%, respec- Vasopressors
tively (12).
Management of SSI depends largely on the level of infec- The goal when initiating vasopressor therapy is to maintain
tion and the severity of illness. Superficial SSI is generally mean arterial pressure (MAP) of 65 mm Hg. All patients
managed by opening the wound. Organ/space SSI is gener- requiring vasopressors should have an arterial catheter
ally managed with drainage; further detail on management placed as soon as practical if resources are available (5).
of this is provided in Chapter 9. Norepinephrine is the first choice vasopressor, and
epinephrine can be added to and potentially substituted
for norepinephrine when an additional agent is needed to
maintain adequate blood pressure (5,18–20).
TREATMENT Vasopressin 0.03 units/min can be added to norepi-
nephrine with intent of either raising MAP or decreasing
Despite the diversity of specific processes in these infec- norepinephrine dosage. Low-dose vasopressin is not rec-
tions, the basic tenets of management of sepsis are simi- ommended as the single initial vasopressor for treatment of
lar: resuscitate patients who have SIRS, control the source sepsis-induced hypotension, and vasopressin doses higher
of contamination, remove most of the infected or necrotic than 0.03–0.04 units/min should be reserved for salvage
material, and administer antimicrobial agents to eradicate therapy (i.e., failure to achieve adequate MAP with other
residual pathogens (13–15). vasopressor agents) (5,20).
Phenylephrine is not recommended in the treatment of
SURVIVING SEPSIS CAMPAIGN 2012 septic shock except in circumstances where norepinephrine
INITIAL RESUSCITATION is associated with serious arrhythmias, cardiac output is
known to be high and blood pressure persistently low, or
Fluid therapy in severe sepsis as salvage therapy when combined inotrope/vasopressor
drugs and low-dose vasopressin have failed to achieve the
Crystalloids are the initial fluid of choice in the resuscitation MAP target (5).
of severe sepsis and septic shock. SSC of 2012 recommends Dopamine should be reserved as an alternative vasopres-
an initial fluid challenge in patients with sepsis-induced tis- sor agent to norepinephrine only in selected patients (e.g.,
sue hypoperfusion with suspicion of hypovolemia to achieve patients with low risk of tachyarrhythmias and absolute or
a minimum of 30 mL/kg of crystalloids (a portion of this relative bradycardia). Low-dose dopamine should not be
may be albumin equivalent) (5). More rapid administration used for renal protection (5).
and greater amounts of fluid may be needed in some patients.
The SSC recommends a fluid challenge technique be applied, Inotropic therapy
wherein fluid administration is continued as long as there is
hemodynamic improvement based on either dynamic (e.g., A trial of dobutamine infusion up to 20 micrograms/
change in pulse pressure or stroke volume variation) or static kg/min can be administered or added to vasopressor (if in
(e.g., arterial pressure and heart rate) variables (5). use) in the presence of myocardial dysfunction as suggested
by elevated cardiac filling pressures and low cardiac output,
Albumin use or ongoing signs of hypoperfusion, despite achieving ade-
quate intravascular volume and adequate MAP (5).
The Saline versus Albumin Fluid Evaluation (SAFE) ran-
domly evaluated 6,997 patients, comparing albumin and STEROID USE
crystalloids. Albumin demonstrated a nonsignificant trend
toward lower all-cause mortality at 28 days (p = 0.09) (16). The Annane Trial published in 2002 randomized 300 sep-
The 2014 Albumin Italian Outcomes Sepsis (ALBIOS) trial tic shock patients within 8 hours of diagnosis to receive
randomized 1,818 patients and further investigated the use 7 days of corticosteroids or placebo (21). The dose of hydro-
of albumin administration to maintain patients with severe cortisone was 50 mg IV every 6 hours and fludrocortisone
sepsis or septic shock to a serum albumin level ≥3 g/dL 50 microgram enterally per day. All patients had a short
(17). There was no difference in survival at 28 and 90 days, adrenocorticotropic hormone stimulation test at time of
although those treated with albumin had more favorable enrollment and were classified as responders (change in
SOFA subscores and received fewer vasopressors or inotro- cortisol >9 microgram/dL) or nonresponders (change
pes (17). The SSC of 2012 recommends albumin in the fluid <9 micrograms/dL) (21). Among nonresponders, cortico-
resuscitation of severe sepsis and septic shock only when steroid use was associated with a 10% absolute reduction
patients require substantial amounts of crystalloids. The in 28-day mortality (53% versus 63%), with no difference
Treatment / Recommendations for other supportive therapy of severe sepsis 35
at 1 year (69% versus 77%). Corticosteroids were associated medical or surgical ICU patients, and there were no dif-
with more rapid reversal of shock. ferences in length of stay, duration of ventilator therapy, or
However, the CORTICUS Trial (Corticosteroid Therapy need for renal replacement therapy (24). A 2012 follow-up
of Septic Shock) by Sprung in 2008 evaluated 499 patients publication by the authors demonstrated that excess mortal-
with septic shock in the past 72 hours. This was a multi- ity was due to moderate to severe hypoglycemia, especially
center, randomized, prospective, placebo-controlled in patients with distributive shock (25). This SSC approach
trial (21). The hydrocortisone regimen was 50 mg IV every should target an upper blood glucose ≤180 mg/dL rather
6 hours for 5 days, then 50 mg IV every 12 hours (days than an upper target blood glucose ≤110 mg/dL. Blood
6–8), then 50 mg IV every 24 hours (days 9–11), then stop. glucose should be monitored every 1–2 hours until glucose
No difference in mortality (39.2% versus 36.1%) was found values and insulin infusion rates are stable and then every 4
(22). This study was not powered to assess for mortality, so hours thereafter (5).
type II error exists. However, time to reversal of shock (SBP
>90 mm Hg without vasopressors for at least 24 hours) was Deep vein thrombosis prophylaxis
quicker with hydrocortisone (3.3 versus 5.8 days) for both
responders and nonresponders. Patients had high incidence Patients with severe sepsis should receive daily pharma-
of new infections occurring 48 or more hours after study coprophylaxis against venous thromboembolism with daily
drug (odds ratio [OR] 1.37, 95% confidence interval [CI] subcutaneous low molecular weight heparin. If creatinine
1.05–1.79), hyperglycemia and hypernatremia (22). clearance is <30 mL/min, use dalteparin or another form
The 2012 Surviving Sepsis Campaign guidelines for of low molecular weight heparin that has a low degree of
severe sepsis and septic shock only recommend using renal metabolism (5). Intermittent pneumatic compres-
hydrocortisone 200 mg IV divided daily in septic shock sion devices should also be used whenever possible. Septic
if fluid resuscitation and pressors are not able to reverse patients who have a contraindication for heparin (e.g.,
hemodynamic instability (5). thrombocytopenia, coagulopathy, or active bleeding)
should not receive p
harmacoprophylaxis but should receive
RECOMMENDATIONS FOR OTHER mechanical prophylaxis (5).
SUPPORTIVE THERAPY OF SEVERE SEPSIS
Nutrition
Blood product administration
Administer oral or enteral (if necessary) feedings, as tol-
Once tissue hypoperfusion has resolved and in the absence erated, rather than either complete fasting or provision of
of extenuating circumstances (myocardial ischemia, severe only IV glucose within the first 48 hours after a diagno-
hypoxemia, acute hemorrhage, or ischemic heart disease), sis of severe sepsis/septic shock (5). Avoid mandatory full
the SSC recommends the red blood cell transfusion occur caloric feeding in the first week, start low dose feeding (e.g.,
only when hemoglobin decreases to <7 g/dL (5). In the set- up to 500 calories per day), and advance as tolerated. Use
ting of thrombocytopenia, prophylactic platelet adminis- IV glucose and enteral nutrition rather than total parenteral
tration is recommended when counts are <10,000 mm3 in nutrition (TPN) alone or parenteral nutrition in conjunc-
the absence of bleeding. If the patient has significant risk tion with enteral feeding in the first 7 days after diagnosis
factors for bleeding, transfuse platelets prophylactically of severe sepsis/septic shock. Use nutrition with no specific
if <20,000 mm3. If active bleeding is present or if surgery immunomodulating supplementation rather than nutrition
or an invasive procedure is planned, transfuse platelets to providing specific immunomodulating supplementation in
achieve counts >50,000 mm3 (5). There is no current role for patients with severe sepsis (5).
erythropoietin or antithrombin in the treatment of severe
sepsis (5,23). Setting goals of care
Glucose control It is critical to discuss goals of care and prognosis with

patients and their families. Incorporate goals of care into
A protocol should be used to manage blood glucose in ICU treatment and end-of-life care planning, utilizing pallia-
patients with severe sepsis, commencing insulin dosing tive care principles where appropriate. Address goals of care
when two consecutive blood glucose levels are >180 mg/dL as early as feasible, but no later than 72 hours into an ICU
(5). The Normoglycemia in Intensive Care Evaluation— admission for sepsis (5,26,27).
Survival Using Glucose Algorithm Regulation (NICE-
SUGAR) trial randomized 6,104 medical and surgical Other recommendations
ICU patients to intensive glycemic control (target glucose
81–108 mg/dL) versus conventional glycemic control (target There is no role for IV immunoglobulins or IV selenium for
glucose ≤180 mg/dL) (24). Unexpectedly, the intensive gly- the treatment of severe sepsis (28,29). Bicarbonate therapy
cemic control arm had a higher 90-day mortality and more should not be used for the purpose of improving hemody-
hypoglycemic events. There was no difference between namics or reducing vasopressor requirements in patients
36 Sepsis
with hypoperfusion-induced lactic acidemia with pH >7.15 In the initial, abbreviated laparotomy to control intraab-
(5). Stress ulcer prophylaxis with proton pump inhibitors dominal sepsis, surgeons need to assess the degree of
should be used in patients with bleeding risk factors in physiologic derangement early in the operation. If severe
severe sepsis. There are specific guidelines for mechanical derangements exist, then the operative interventions need
ventilation of sepsis-induced acute respiratory distress syn- to be truncated. The primary aim in the operating room is to
drome (ARDS). These are outside the scope of this chapter. control the source of infection, resect nonviable bowel, close
or divert bowel perforations, and wash out the abdomen.
Initial surgical treatment focuses on source control using a
ANTIBIOTIC THERAPY
combination of resection and/or wide drainage. Failure to
Current guidelines from the SIS and the IDSA recommend achieve source control will lead to death. No definitive pro-
an antibiotic treatment course of 4–7 days, depending on cedures should occur during this operation, including anas-
clinical response (13–15). Observational studies show anti- tomosis and abdominal closure. An anastomosis is likely to
microbial therapy is typically administered for 10–14 days fail in the setting of extreme physiologic derangement. The
(30–32). A newly published randomized Study to Optimize abdomen is quickly and temporarily closed, with the goal to
Peritoneal Infection Therapy (STOP-IT) trial compared two contain the viscera, avoid potential injury and contamina-
strategies guiding the duration of antimicrobial therapy for tion, and control peritoneal effluent (34).
management of complicated intraabdominal infection (15).
In this study, 518 patients with complicated intraabdomi-
nal infection were randomly assigned adequate source con-
trol to receive antibiotics until 2 days after the resolution of
KEY POINTS
fever, leukocytosis, and ileus, with a maximum of 10 days of
●● Early recognition of sepsis is critical.
therapy (control group), or to receive a fixed course of anti-
●● Compliance with SSC bundles may improve sur-
biotics (experimental group) for 4 +/− calendar days (15).
vival and should be implemented as routine order
The primary outcome was a composite of surgical-site infec-
sets if not already in place.
tion, recurrent intraabdominal infection, or death within
●● Infection source should be removed in the safest,
30 days after the index source-control procedure, accord-
least invasive method possible.
ing to treatment group. There was no significant differ-
●● If surgery is required to control sepsis, principles
ence between groups after fixed duration antibiotic therapy
of “damage control surgery” should be followed.
(approximately 4 days) compared to after a longer course of
antibiotics (approximately 8 days) that extended until after
the resolution of physiological abnormalities (15).
REFERENCES
Source control 1. Bone RC et al. Chest 1992;101(6):1644–55.
2. Levy MM et al. Crit Care Med 2003;31(4):1250–6.
A specific anatomical diagnosis of infection requiring con-
3. Singer M et al. JAMA. 2016;315(8):801–10.
sideration for emergent source control should be sought and
4. Simpson SQ. Chest 2016;149(5):1117–8.
diagnosed or excluded as rapidly as possible. An interven-
5. Dellinger RP et al. Crit Care Med 2013;41:580–637.
tion should be undertaken for source control within the first
6. Rhodes A et al. Intensive Care Med 2015;41:1620–8.
12 hours after the diagnosis is made. When source control
7. Levy MM et al. Intensive Care Med 2014;40:1623–33.
in a severely septic patient is required, the effective inter-
8. Yealy DM et al. N Engl J Med 2014;370(18):1683–93.
vention associated with the least physiologic insult should
9. Peake SL et al. N Engl J Med 2014;371:1496–1506.
be employed (e.g., percutaneous rather than surgical drain-
10. Mouncey PR et al. N Engl J Med 2015: doi: 10.1056/
age of an abscess, if possible). If intravascular access devices
NEJMoa1500896
are a possible source of severe sepsis or septic shock, they
11. Bratzler DW et al. Am J Health-Syst Pharm 2013;70:
should be removed promptly after other vascular access has
195–283.
been established.
12. Ortega G et al. J Surg Res 2012;174(1):33–8.
13. Solomkin JS et al. Surg Infec (Larchmt) 2010;11:
SURGERY IN THE SEPTIC PATIENT 79–109.
14. Solomkin JS et al. Clin Infec Dis 2010;50:133–64.
With the success of damage-control surgery for the treat- 15. Sawyer RG et al. N Engl J Med 2015;372:1996–2005.
ment of hemorrhagic trauma, it has been adapted to shock 16. Finfer S et al. N Engl J Med 2004;350(22):2247–56.
secondary to peritonitis. The structured approach of dam- 17. Caironi P et al. N Engl J Med 2014;370(15):1412–21.
age control is now a five-part process: ground zero is the 18. De Backer D et al. N Engl J Med 2010;362:779–89.
initial evaluation or prehospital, part 1 is initial abbreviated 19. De Backer D et al. Crit Care Med 2012;40:725–30.
laparotomy, part 2 is ICU resuscitation, part 3 is later defini- 20. Russell JA et al. N Engl J Med 2008;358:877–88.
tive repair, and part 4 is definitive abdominal closure (33). 21. Annane D et al. J Am Med Assoc 2002;288(7):862–71.
References 37
22. Sprung CL et al. N Engl J Med 2008;358(2):111–24. 29. Angstwurm MW et al. Crit Care Med 2007;35:118–26.
23. Liumbruno G et al. Blood Transfus 2009;7:132–50. 30. Riccio LM et al. Surg Infec (Larchmt) 2014;15:417–24.
24. The NICE-SUGAR Study Investigators. N Engl J Med 31. Guirao X et al. J Antimicrob Chemother 2013;
2009;360:1283–97. 68(Suppl 2):ii37–44.
25. Finfer S et al. N Engl J Med 2012;367(12):1108. 32. Samuelsson A et al. Scand J Infect Dis 2012;44:820–7.
26. Nelson JE et al. Crit Care Med 2010;38:1765–72. 33. Waibel BH, Rotondo MF. Surg Clin North Am 2012;
27. Lee Char SJ et al. Am J Respir Crit Care Med 2010; 92:243–57.
182:905–9. 34. Waibel BH, Rotondo MF. Rev Col Bras Cir 2012;39(4):
28. Werdan K et al. Crit Care Med 2007;35:2693–2701. 314–21.
5
Intraoperative anastomotic challenges
DAVID E. BECK
divided into patient factors and surgeon factors (1–3). Patient-

CHALLENGING CASE related factors, such as the patient’s nutritional status and
associated medical conditions or medications, are not under
A 28-year-old man is undergoing a restorative procto- the operating surgeon’s control and are discussed in other
colectomy using a double-stapled technique for the chapters. This chapter focuses on anastomotic principles and
ileoanal anastomosis. During insertion of the stapler problems (e.g., leakage, ischemia, stenosis, and hemorrhage)
into the anus, the anal canal distal linear staple line is that can be identified and managed during the procedure.
disrupted. What are your options?
CASE MANAGEMENT PREANASTOMOTIC CONSIDERATIONS
The initial action is to visualize the distal staple line PREOPERATIVE DISCUSSION AND
using retractors. If the ends of the partially closed dis-
PLANNING
tal bowel can be visualized and grasped with clamps
or traction sutures, the amount of residual bowel Prior to surgery, the surgeon should have a plan that includes
can be assessed. If adequate length is present, one the expected operative findings or pathology and restora-
option is to reclose the bowel with a linear stapler tion of intestinal continuity, if possible. If the preoperative
placed below the disrupted staple line. After the sta- findings are confirmed, the operation should proceed along
pler is fired, the residual bowel end can be resected an organized pathway. Unexpected findings will obviously
with scissors or a scalpel. A second option is to require modifications. Prior to the procedure, the surgeon
reclose the disrupted staple line with sutures placed should also have a discussion with the patient, which includes
from the abdominal side or placed intraluminally via a these considerations with special emphasis on aspects of the
retractor (lighted Chelsea-Eaton, Hill-Ferguson, etc.) anastomosis and the possible need for a temporary or per-
placed into the anal canal. If the defect and bowel are manent stoma should restoration of intestinal continuity be
successfully closed, the anastomosis can proceed. impossible or ill advised (4). Proximal diversion will reduce
If the distal segment of bowel is impossible to the clinical sequelae from an anastomotic dehiscence. This is
visualize or close, a mucosectomy can be performed more likely in patients receiving preoperative chemotherapy
via the anus and a hand-sewn ileo-anal anastomosis and/or radiation, with poor nutrition, associated infection,
can be performed, as described in Chapter 35. Most or comorbid conditions (steroid use, hypotension, etc.). The
surgeons will create a diverting loop ileostomy when appropriate site for a potential stoma should be chosen pre-
the anastomosis has been this challenging. operatively, with the assistance of an enterostomal therapist.
The selection and marking of a stoma site provide another
opportunity for dialogue between the surgeon and patient.
INTRODUCTION
OPERATIVE PRINCIPLES
Colon and rectal surgery is a technique-oriented specialty
with many procedures requiring an anastomosis to reestab- The key to uncomplicated healing of an intestinal anasto-
lish bowel continuity. Achievement of a successful anastomo- mosis depends on adherence to well-established principles
sis is related to a number of surgical principles, which can be as well as the specifics of the technique. The principles of
38
Preanastomotic considerations / Bowel status 39
intestinal anastomosis include the following: (1) appropriate (7,8). A Cochrane Review of five randomized trials showed
access and exposure to the two ends of the bowel, (2) healthy equal or better morbidity or mortality in 576 patients with
bowel to be joined, (3) good blood supply, (4) gentle han- a mechanical bowel preparation and 583 patients without a
dling of the bowel, and (5) good apposition of ends with no mechanical preparation (9). An additional meta-analysis of
tension on the anastomosis (5). Any compromise of these seven randomized trials containing 1,454 patients showed
principles places the anastomosis at risk for complications. no significant differences for wound infection and septic
Exposure and access to bowel ends is maximized by tak- and nonseptic conditions (10). Certain situations, such as
ing the time to set up and position retractors and intraab- laparoscopic procedures, potential need for intraoperative
dominal packs. Headlights and lighted retractors minimize colonoscopy, or avoidance of spillage from proximal stool
the frustration of inadequate overhead lights. Deep pelvic loading after a low colorectal anastomosis, still require ade-
retractors such as the St. Mark’s retractor or Fazio pelvic quate mechanical bowel preparation. For other situations,
retractors assist the visualization of the distal rectum prior many surgeons are minimizing or eliminating a mechani-
to anastomosis. Extending the midline incision to the sym- cal bowel preparation in elective situations. The current evi-
physis pubis likewise allows maximum exposure of the dis- dence suggests that intraluminal contents should not be the
tal rectum. Operating with poor lighting and inadequate primary factor in deciding if an anastomosis should be per-
exposure not only jeopardizes the anastomosis but also formed. Other options to consider with unprepared bowel
increases operating room time. are to perform a subtotal colectomy with ileocolonic or ileo-
Techniques of intestinal anastomosis should also be per- rectal anastomosis. This option has been shown to be a safe
formed following the principle of gentle bowel handling. option for avoiding a stoma in left colon obstruction (11).
Clamps should be used only when absolutely necessary Alternatively, intraoperative colonic lavage in some hands
with the least amount of closure required to occlude the offers the ability to construct an anastomosis in patients
lumen. Care should be taken to exclude mesenteric blood with this condition (11–13).
vessels within the intestinal clamp. Gingerly inserting Whatever the bowel preparation used, it is critical that
appropriately sized intraluminal staplers prevents inadver- spillage of intraluminal contents be avoided to minimize
tent splitting and tearing of bowel ends to be anastomosed. complications and neoplastic dissemination. Most sur-
Excessive use of electrocautery at the anastomosis can cause geons agree that a clean, empty colon has less potential for
unappreciated tissue necrosis with potential for disrup- spillage, but that cannot compensate for poor technique.
tion. Mobilization of intestinal ends is required for expo- Additional protection against spillage of residual intestinal
sure, access, and freedom of tension on the anastomosis. contents is provided by controlling the ends of bowel used
However, during mobilization, it is important to preserve in the anastomosis. This can be accomplished by elevating
those blood vessels required for adequate anastomotic heal- the ends (with traction sutures) or by occluding the bowel
ing. For example, excessive skeletonization of the cut intes- proximal and distal to the anastomosis with tapes or non-
tinal ends may compromise their blood supply. crushing clamps (14).
Having two healthy ends of bowel to anastomose is ideal.
In some cases (bowel obstruction, diverticular disease, BOWEL STATUS
radiation enteritis, and Crohn disease), this situation may
not be possible and the plan to anastomose may be ques- When intestinal surgery is being performed for urgent situ-
tioned. Optimizing patient nutrition, treating infection, ations or even during certain elective operations, the first
and minimizing inflammation in the preoperative period decision is whether or not an anastomosis is appropriate.
may improve the bowel status. At operation, all diseased Healing of an anastomosis is at risk in certain clinical situa-
bowel is resected whenever possible to provide soft, pliable tions. Traditionally, intestine that is unprepared, obstructed,
bowel ends for anastomosis. irradiated, inflamed, or ischemic may not be suitable for
anastomosis (15–17). However, other than ischemia, current
BOWEL PREPARATION evidence suggests that constructing an anastomosis is safe
in selected cases of obstruction, irradiation, inflammation,
As described in Chapter 2, bowel preparation has under- and without bowel preparation (5,11,18,19). In addition to
gone major changes over the last 70 years. Until the last these local factors, patient factors such as malnutrition, dia-
decade, mechanical bowel preparation was a standard betes, renal failure, chronic hepatic disease, anemia, shock,
feature of elective bowel surgery, and the lack of a bowel steroid use, and other immunocompromised states may
preparation or poor results with a mechanical preparation place an anastomosis at risk for failure (4–6).
was in many surgeons’ views a contraindication to a pri- The safety of intestinal anastomosis in any particular clin-
mary anastomosis. Recent studies have failed to support ical scenario thus depends on patient and intestinal factors
the accepted view that bowel cleansing, in the presence of that must be carefully weighed by the operating surgeon (5).
appropriate antibiotics, reduced the risk of anastomotic leak Optimally, the bowel will have a good blood supply (docu-
or wound infection (6). Case series and reports from the mented by pink color, peristalsis, and pulsatile bleeding from
trauma literature suggested that good or better outcomes the cut edge), lack edema, be free of tension (see the section
could be achieved in unprepared bowel with an anastomosis “Obtaining Adequate Length”), and have adequate lumen for
40 Intraoperative anastomotic challenges
the type of anastomosis. The decision to perform an intes- legs often improves visualization during splenic flexure
tinal anastomosis ultimately depends on surgical judgment mobilization. Adequate visualization is imperative for safe
derived from an understanding of documented risks as well splenic flexure mobilization, and additional retraction or
as knowledge of one’s own ability and experience. incision extension should be one of the first considerations
if mobilization is difficult.
EXPOSURE Different retractors are available to improve exposure.
Those that are fixed to the bed, such as the Bookwalter, “Upper
The importance of adequate exposure cannot be overem- Hand,” Omnitract, or Polytract (Teleflex. Inc, Morrisvile, NC)
phasized. Exposure is facilitated by patient position, ade- make life easier for surgical assistants and provide a consis-
quate length of incision, appropriate choice of retractors, tent view throughout the operation (4). When placing retrac-
and lighting. If the possibility of a left-sided anastomosis tor attachments, the surgeon must be aware of the relation of
exists, the patient should be placed in lithotomy position, the retractor to the femoral vessels, nerves, and iliac crests.
using stirrups, after anesthesia is induced (Figure 5.1). Great Prolonged constant traction on the bowel may also be a prob-
care is taken to avoid pressure on the peroneal nerves and lem, and consideration of relieving the pressure intermittently
hips (20). The perineum should extend slightly over the during long cases may be appropriate.
end of the operating table to allow easy access for transanal Finally, adequate lighting is extremely important during
stapled anastomosis, upward pressure on the perineum for pelvic dissection as well as to provide a view of the anasto-
exposure of the distal rectum, or a two-surgeon combined mosis, deep in the pelvis. Equipment available to enhance
approach to hand-sewn coloanal anastomosis or abdomi- vision includes headlights, lighted retractors, cautery
noperineal resection. Once the patient is correctly posi- instruments, and suction devices.
tioned, irrigation of the rectum should be performed to
ensure the quality of the bowel preparation and to evacuate OBTAINING ADEQUATE LENGTH
any remaining fecal residue. Leaving a large mushroom-
shaped or Foley catheter in the rectum alerts the surgeon After the appropriate resection is completed, sufficient prox-
to the level of dissection in the low pelvis, prevents rectal imal and distal mobilization provides tension-free bowel
distension and possible enterotomy during mobilization, ends for a secure anastomosis. Tension is rarely a problem
and allows drainage of rectal contents, which minimizes for small bowel or ileocolic anastomosis. The small bowel
luminal spillage. mesentery has an avascular plane anterior to the aorta to the
The trend toward smaller incisions should be critically takeoff of the superior mesenteric artery. The right and left
evaluated when planning a colorectal anastomosis (4). colon have a posteriolateral fusion plane anterior to Gerota
Pelvic exposure is greatly facilitated by incisions that extend fascia. This avascular plane can be opened using a lateral or
to the pubic bone. The incision may require proximal exten- medial approach.
sion if mobilization of the splenic flexure is required. The Difficulty in obtaining tension-free bowel occurs more
extent of this extension will depend on factors such as the commonly with a left-sided (e.g., colorectal) anastomosis.
patient’s body habitus, disease process, and surgical tech- Additional left colon length is obtained using the follow-
nique. Adequate exposure with less generous incisions is ing maneuvers in this order: division of the lateral colonic
often possible in thin patients or those with low splenic flex- attachments, division of the splenic flexure attachments,
ures, mobile colons, or left-sided Crohn disease (the splenic division of the inferior mesenteric artery at its aortic takeoff,
flexure that is contracted down into the abdominal cavity). and division of the inferior mesenteric vein at the inferior
Placement of the operating surgeon between the patient’s border of the pancreas (Figure 5.2) (4). If these maneu-
vers do not provide adequate bowel length, branches of
the distal middle colic artery and veins may need division.
Unfortunately, this last action may compromise the blood
supply to the remaining colonic end. If this occurs, the isch-
emic bowel must be resected, and additional vessels will
need to be divided to provide the required bowel length. In
some cases the middle colic vessels will have to be divided
proximally, and the blood supply of the residual colon will
need to be based on the right and/or ileocolic artery. In most
patients, these vessels will provide adequate blood supply to
the proximal transverse colon or hepatic flexure, which can
be made to reach the rectum with one of two techniques.
One method is to open a window in the ileal mesentery
medial to the ileocolic artery and vein. The proximal trans-
verse colon is brought through this window to reach the pel-
vis (Figure 5.3) (21). Another option is to completely mobilize
Figure 5.1 Stirrups for modified lithotomy position. the right colon and then rotate it counterclockwise. This
Anastomotic technique 41
2 rotates the cecal tip to the right middle abdomen (pointed

toward the liver), reverses the direction of the colon, and
provides enough length for the hepatic flexure to reach the
pelvis (Figure 5.4). As this maneuver moves the cecum to an
abnormal position, it is important to remove the appendix,
since development of appendicitis would produce confusing
signs and symptoms.
In addition to a lack of tension, it is also critically impor-
4 5
tant that the anastomotic site have a good blood supply.
Before the anastomosis is constructed, the bowel should
be routinely checked for viability (normal color and peri-
3 stalsis and a pulsatile blood supply). Bleeding from divided
appendices epiploicae (or marginal vessels) at the cut end
1
of the proximal bowel intended for the anastomosis is fre-
quently used to objectively confirm an adequate blood sup-
ply. In addition to the factors described previously, bowel
used for an anastomosis must not be edematous, radiated,
or ischemic. An appropriate resection should remove isch-
emic or radiated bowel, while edema of the residual bowel
(e.g., associated with peritonitis) may mandate forgoing an
anastomosis in favor of a diverting stoma.
ANASTOMOTIC TECHNIQUE
In performing an anastomosis, surgeons have multiple

options. Each technique has associated advantages and dis-
Figure 5.2 Operative techniques to obtain left colon
advantages, and the specific one favored by an individual
length. (1) Division of lateral colonic attachments;
(2) division of the splenic flexure; (3) division of the inferior
surgeon depends more on training, personal experience,
mesenteric artery at its aortic takeoff and the inferior and perhaps blind faith than on the results of randomized,
mesenteric vein; (4) second division of the inferior mes- prospective studies. There has been no consistent scientific
enteric vein at the inferior border of the pancreas; and proof that one intestinal anastomotic technique is supe-
(5) incision of splenic flexure mesentary. (From Rafferty JF. rior. Options range from the physical configuration of the
Clin Colon Rectal Surg 2001;14:25–31. With permission.) anastomosis (end-to-end, side to side, end to side, side to
(a) (b)
Figure 5.3 (a) Window in mesentary is created medial to the ileocolic artery and vein. (b) Transverse is brought through
ileal mesenteric window to reach the pelvis.
(a) (b)
Figure 5.4 (a) Right colon is mobilized, right colic vessels are divided, and appendix is removed. (b) The right colon is
derotated to allow the hepatic flexure to reach the pelvis.
end, etc.) (Figure 5.5) and the method used to construct it: correct tension of the suture, and secure knots appear to be
sutures, staples, a combination of these, and experimental more important than the experimental findings discussed
methods such as compression devices or adhesives. Several previously. Experience, training, clinical judgment, and
of these merit discussion. ability are major factors in a surgeon’s choice of anastomotic
technique; however, some of the reported experience with
STAPLES VERSUS SUTURES suturing merits additional comment.
An extensive experience using a running monofilament
Suturing has been used since the beginning of intestinal technique has been described by Max and colleagues (28).
surgery. Different suture materials have shown some exper- In a retrospective report of 1,000 single-layer continuous
imental differences, but the clinical difference is arguable. polypropylene intestinal anastomoses, the authors believed
In general a stapled anastomosis usually takes less time but that this technique was quick, simple, economical, and safe.
is more expensive (22,23). Blood flow may be higher with Although an intraoperative leak rate was not reported, their
a stapled anastomosis, and in certain situations, such as a postoperative leak rate of 1% with the technique compares
low colorectal anastomosis, the use of staples is technically very favorably with others in reports using alternate tech-
easier (24). A final consideration is that any device can mal- niques (28).
function and lead to the need for use of additional staplers Six trials with 955 ileocolic participants were reviewed in
or conversion to a sutured anastomosis (5). a Cochrane Database Systemic review (29). The three larg-
A meta-analysis of 13 trials comparing hand-sewn with est prospective randomized trials comparing stapled versus
stapled anastomoses showed similar mortality, leak rates, hand-sewn methods for ileocolic anastomoses conducted
local cancer recurrences, and wound infections (25). This between 1970 and 2005 showed fewer leaks with stapled
review revealed a higher rate of postoperative strictures anastomosis. All other outcomes—stricture, anastomotic
with the stapled anastomosis, most of which were asymp- hemorrhage, anastomotic time, reoperation, mortality,
tomatic and easily managed with dilation. intraabdominal abscess, wound infection, and length of
Suture techniques, such as the number of layers or use of stay—showed no significant difference.
interrupted versus running sutures, have shown some clini-
cal differences. An inverting anastomosis is superior to an END-TO-END
everting technique. A number of investigators advocate a
single-layer anastomosis because they believe it causes less The use of surgical staplers has advantages in certain situa-
narrowing of the lumen since a smaller amount of tissue tions (e.g., the very low colorectal anastomosis), and they have
is strangulated (26,27). A single-layer anastomosis is also enjoyed widespread clinical usage. These mechanical devices,
felt to cause less devascularization, infection, and necrosis, however, do not compensate for improper or poor technique.
while the continuous suture distributes tension more evenly In an early survey of stapler complications by the
around the lumen (27,28). In clinical practice, however, American Society of Colon and Rectal Surgeons (ASCRS)
technical factors such as the correct placement of sutures, published in 1981, 243 surgeons responded that they had
Anastomotic technique / End-to-end 43
(a) Early experience with 73 consecutive stapled end-to-end

colorectal anastomoses by Gordon and Vasilevsky identified
intraoperative complications in 19 patients (26%) (31). These
included instrument failure (4), incomplete or inadequate
“donuts” (5), bleeding (3), bowel injury associated with use
of sizers (1), anvil extraction (1), anvil insertion (3), diffi-
culty with stapler extraction (1), and anvil not extractable
(1). The relative high incidence of these problems reflects the
early learning curve with stapling instruments and the early
developmental nature of the instruments used. Increased
experience and advances in instruments have minimized
the occurrence of these problems.
(b)
A prospective randomized multicenter study by
Dochetry and colleagues described 652 patients who were
randomized to a sutured (n = 321) or stapled large bowel
anastomosis (n = 331) between 1985 and 1989 (32). During
the study, 5 of the 331 patients (1.5%) randomized to a sta-
pled anastomosis had an instrument or technical failure.
Intraoperative anastomotic testing was not routinely per-
formed, but postoperative radiologic leaks were identified
in 14.4% of the sutured and 5.2% of the stapled colorectal
anastomoses. Clinical anastomotic leakage was evident in
4.4% of the sutured patients and 4.5% of the stapled patients.
Proper technique is a critical component to obtaining a
good anastomosis with a circular intraluminal stapler. To
minimize problems, the largest diameter stapler that can be
(c) accommodated by both bowel ends should be used (33). As
originally described, an intraluminal stapler entails usage
of purse-string sutures to hold the bowel over the stapler
cartridge and anvil during stapler closure. This purse-
string suture can be placed by hand (with a baseball or in-
and-out technique), with a fenestrated purse-string clamp
(Purse String Device, Davis & Geck, Wayne, New Jersey),
or with a stapling device (Purse String Instrument-65,
U.S. Surgical Corp., Norwalk, Connecticut). To work prop-
erly, the sutures must be placed correctly (approximately
1–2 mm back from the bowel ends and 2–3 mm apart). If
the sutures are placed too close, the bowel will not close
tightly around the stapler shaft. This nonconstricting
purse-string may be corrected by carefully cutting the
Figure 5.5 Types of anastomoses. (a) End-to-end anasto- bowel overlying the suture in two or more places to release
mosis. (b) Side-to-side functional end-to-end anastomo- additional suture to bunch up more of the bowel end. If the
sis. (c) End-to-side anastomosis. (From Beck DE. In Beck sutures are placed too far apart or some tear through, gaps
DE, ed. Handbook of Colorectal Surgery. St. Louis, MO: in the bowel ends will appear when the suture is tightened.
Quality Medical, 1997, pp. 400–30. With permission.)
This can be repaired by use of a “Pulley Stitch” (Figure 5.6)
(1,34). These interrupted 4-0 or 3-0 braided sutures (e.g.,
performed 3,594 end-to-end anastomoses (EEA) (30). silk or braided polyester) hold the purse-string suture to
Intraoperative complications were reported in 15.1% of the bowel ends and assist in pulling it tightly around the
patients. These complications included anastomotic leak shaft. Finally, placement of sutures too near the bowel end
(9.8%), tear during extraction (1.9%), anvil not extractable results in their tearing through the bowel, while placing the
(1.2%), complete anastomotic failure that required conver- sutures too far back from the bowel ends will produce an
sion to another technique (0.9%), instrument failure (0.8%), excessive bulk of tissue around the shaft.
and bleeding (0.5%). This report represents surgeons’ early If a purse-string clamp is used, it is important that the
experience with the use of staplers, and therefore, the results bowel be divided close to the clamp before the clamp is
must be evaluated in the proper context. Improvements in released. Leaving excess tissue adjacent to the clamp may
the instruments, anastomotic technique, and surgeon expe- result in too much tissue at the purse-string, which may pre-
rience have resulted in fewer complications. vent the stapler from closing and firing properly. Releasing
(a) a secure anastomosis. The use of open-ended clamps (e.g.,

Babcock clamps) allows for the possibility of placing the
purse-string through the end of the clamp. If this occurs,
the purse-string suture will need to be cut. Use of solid-
ended clamps eliminates the chance of this happening. Large
clamps increase the difficulty in inserting an anvil in bowel
diameter close to the diameter of the anvil.
Double staple
Another end-to-end stapling option involves a double-
staple technique (35,36). With this method a linear staple
line is placed across the distal bowel, and a circular stapler
is inserted into this bowel (via the anus for a left-sided anas-
tomosis). To avoid creating an ischemic area, the trocar of
the circular stapler should exit adjacent or as close as pos-
sible to the linear staples. The anvil is placed in the proximal
bowel and secured with a purse-string as described previ-
ously. When closed and fired, the circular stapler removes a
(b) portion of the crossed linear staple line to create the anas-
tomosis. Concern was initially expressed about these cross-
ing staple lines. However, subsequent experimental and
clinical evidence has confirmed the relative safety of this
method (37,38). The double-staple technique is helpful in
anastomosing bowel ends of dissimilar size and in ultralow
colorectal or coloanal anastomoses. Outside of these situa-
tions, the extra cost of using a stapler rather than a sutured
purse-string argues more for the use of a purse-string.
With low distal staple lines, it can be challenging to insert
the stapler into the anus and not disrupt the staple line.
Distal staple line disruption can occur if the distal bowel is
tenuous or under too much traction. If this occurs, several
options are available. The initial action is to visualize the
distal staple line. If the ends of the partially closed bowel
can be grasped with clamps or traction sutures, the amount
of residual bowel can be assessed. If adequate length is pres-
ent, the bowel can be closed with a linear stapler placed
below the disrupted staple line. After the stapler is fired, the
Figure 5.6 Repair of purse-string stitch. (a) Gap is identi- residual bowel end can be resected with scissors or a scalpel.
fied in purse-string suture. (b) Gap is closed with “pulley” A second option is to recluse the disrupted staple line with
sutures. sutures placed from the abdominal side or placed intralu-
minally via a retractor placed into the anal canal (39). If the
the clamp before dividing the bowel may result in inade- bowel is successfully closed, the anastomosis can proceed.
quate tissue to hold the purse-string. Difficulties in using If the distal segment of bowel is impossible to close, a mus-
the purse-string clamp low in the pelvis are minimized ectomy can be performed via the anus, and a hand-sewn
by the use of a double-armed suture (e.g., 2-0 monofila- coloanal or ileoanal anastomosis can be performed.
ment polypropylene, double-armed TS-9, David and Geck, A serious problem associated with double stapling of the
Wayne, New Jersey). Both needles are placed through the low rectum is the inadvertent creation of a rectovaginal fis-
clamp, and the needles can be bent several times while the tula. This unfortunate complication results from incorpo-
needle is withdrawn to allow the needles to be removed in rating the posterior wall of the vagina into the staple lines.
the confined pelvis. Maneuvers to reduce this occurrence include an adequate
Many surgeons use clamps to hold the bowel ends while dissection of the rectum off the posterior vagina, careful
placing the purse-string or to hold the bowel open to assist visualization of the bowel ends during closure of the sta-
placement of the anvil or stapler. Several problems can occur pler, and intravaginal palpation of the posterior vaginal wall
with use of these clamps. If the clamps are placed too far prior to firing the stapler (1).
back from the bowel end and placed too tightly, an injury to A variation of double stapling is triple stapling. In this
the bowel wall can occur, which can produce a leak despite anastomotic method, an extralinear stapler is used to close
Anastomotic testing 45
the bowel end after placement of the anvil into the proximal A proctoscopic (or flexible endoscopic) examination of the
bowel. The anvil trocar is then advanced through the closed rectum ensures an adequate lumen, confirms an adequate
bowel. This technique has been suggested for intracorporeal preparation and mobilization, and assists in identifying the
laparoscopic techniques; however, it is costly and produces apex of a Hartmann pouch.
another linear suture line that must be incorporated into the Once the stapler is closed and fired, it must be removed.
final anastomotic staple line. The technique has not gained Stapler extraction from the anastomotic area may be aided
widespread acceptance due to the relative ease in placing the with a traction stitch. Bowel spasm or a stapler misfire may
proximal purse-string. cause extraction difficulty. Gentle traction and careful sta-
Difficulty with anvil insertion in the proximal bowel pler manipulation usually allow it to be removed. If a mis-
lumen usually occurs when the stapler is too large for the fire results in an inability to remove the stapler, it may be
diameter of the bowel. Experience or the use of scissors necessary to excise and reaccomplish the anastomosis.
allows accurate selection of the correct size of circular sta-
pler. Additional helpful techniques include the use of dila- END TO SIDE AND SIDE TO SIDE
tors or glucagon to overcome bowel spasm, lubrication of (FUNCTIONAL END-TO-END)
the anvil head (with betadine, saline, or blood), and distrac-
tion of the bowel ends with three small-ended forceps or An end-to-side or side-to-end (the proximal bowel is usu-
clamps. Use of a recently developed low-profile anvil (CDH ally listed first) anastomosis is useful for joining bowel of
Ethicon-Endosurgery, Inc., Cincinnati, Ohio) has dimin- different diameter. The size of the anastomosis is not limited
ished this occurrence. by the bowel diameter. This configuration is often used for
ileocolic or ileorectal anastomoses. A side-to-side anasto-
Detachable staplers mosis is frequently used to join bowel with a linear cutting
stapler. Use of the bowel ends for a side-to-side anastomosis
For colorectal anastomosis, the circular stapler is usually serves as a functional end-to-end anastomosis. A surgical
placed through the anus. With currently available detach- atlas should be consulted for additional technical details.
able head staplers, the flat stapler shaft may be difficult to A meta-analysis of studies published between 1992 and
pass atraumatically through the anal sphincter muscles. 2005 of end-to-end versus other anastomotic configurations
Khoury and Opelka, in 1995, described a technique to in Crohn disease used eight studies including 661 patients
facilitate this maneuver (40). A Faensler or Chelsey-Eaton (41). The authors conclude that a side-to-side anastomosis
anoscope allows a gradual controlled dilation of the sphinc- led to fewer anastomotic leaks and overall complications, a
ters. After removal of the obturator, the stapler shaft can shorter hospital stay, and a perianastomotic recurrence rate
easily be passed through the anoscope (Figure 5.7). Once comparable to end-to-end anastomoses.
through the sphincter, the stapler must be inserted up to the
resected end of the rectum. Knowledge of rectal anatomy,
adequate mobilization of the posterior rectum, and selec-
tion of an appropriate size of stapler assist in accomplishing ANASTOMOTIC TESTING
this advancement. Incorrect insertion can tear or split the
rectum. Such an injury to the rectum mandates a very low All surgeons should test their anastomoses in some way. At
or coloanal anastomosis to reestablish intestinal continuity. a minimum, the anastomotic site is inspected and in some
(a) (b) (c)
Figure 5.7 Anoscopic-assisted stapler insertion. (a) Faensler anoscope is inserted after gentle anal dilation. (b) The ano-
scope obturator is removed, and the circular stapler is inserted through the anoscope. (c) The anoscope is withdrawn and
taken off the shaft of the stapler.
cases palpated. A visual inspection of a side-to-side anas- anastomotic lumen can be sized by palpation or visually
tomosis may be performed prior to closing the ends of the inspected. The ability to remove the anvil of a circular stapler
bowel. Gentle constriction of the bowel proximal or distal confirms a lumen corresponding to the size of the stapler,
to the anastomosis will confirm a patent lumen and the while distal rectal anastomosis can be evaluated by a procto-
absence of a gross leak. A more sensitive test can easily be scope. An alternative technique for colorectal anastomosis is
performed in the colorectal anastomosis (which is at higher an isoperistaltic side-to-side anastomosis (Figure 5.8).
risk for a leak) (42–45).
The author prefers to test low colorectal anastomosis LEAKAGE
with intraluminal instillation of a dilute solution of pro-
vidine-iodine (Betadine, Purdue Frederick Co, Norwalk, An accurate incidence of anastomotic leakage is difficult to
Connecticut). After the bowel is occluded above the anas- determine. Few studies have reported the incidence of intra-
tomosis with finger pressure, the testing solution is instilled operatively identified anastomotic problems. The incidence
gently with a bulb syringe inserted into the anus. Any leak is of leaks identified in the postoperative period is described
readily apparent. Irrigation with this dilute providine-iodine in Chapter 6.
solution also provides antimicrobial and tumoricidal activ- If a defective anastomosis is identified, it may be repaired
ity. Others have suggested testing with a dilute solution of in several ways. Additional sutures can approximate a small
methylene blue (46). Larger volumes are infused via a rectal gap, or the anastomosis can be resected and completely
tube, and with care even ileocolic anastomosis can be tested redone using a stapler or hand-sewn technique. Another
for leaks with this technique. The optimal pressure recom- option is to replace purse-string sutures around the defective
mended for detecting intraoperative leaks with air/water test- anastomosis and reinsert a new stapler through the lumen.
ing is 25–30 cm H2O (47,48). If an infusion s ystem is used, the The purse-string sutures are tightened, which should close
pressure can be controlled by the height of the infusion bag. the defect and hold the previously placed staples toward the
Some surgeons prefer to test their anastomosis with air stapler shaft. After closure and firing of the new stapler, the
(45). The pelvis is first filled with saline and the distal bowel new donuts (which should also contain the old staples) are
(containing the anastomosis) is distended with air (instilled removed with the stapler (49). If the anastomosis is very low,
transanally). Any anastomotic defect will produce air bub- the defect may also be repaired transanally.
bles. Unfortunately, with this method it is often difficult to
accurately identify the location of the leak if any blood has ANASTOMOTIC HEMORRHAGE
mixed with the saline. The saline must also be removed before
any identified leak can be repaired. Testing with air may be Hemorrhage can occur at both a staple and a suture line.
preferable for higher colorectal anastomosis as infused intra- Proper size staple height and correct tension of sutures
lumenal fluid may not reach a higher anastomosis.
A proctoscope (or colonoscope) can also be used to
inspect the colorectal anastomosis. Sufficient lumen size is
usually confirmed by the lack of stenosis, hemostasis is con-
firmed, and the bowel can easily be distended with air.
Finally, some surgeons inspect the intraluminal stapler
“donuts.” The author has not found this to be helpful as
complete donuts do not ensure the absence of a leak at the
anastomotic site (e.g., due to a tear of the bowel or staple
lines during stapler removal). Also, an incomplete donut
may be produced with an intact anastomosis. Intraoperative
testing as described above is more sensitive and specific.
Whatever method is used to inspect or test an anasto-
mosis, it is important to act on any defect or leak identified.
Options include suture reinforcement, reconstruction, or
proximal diversion.
CHALLENGES
INADEQUATE ANASTOMOTIC LUMEN

Adequate luminal patency is important for several rea-
sons. Bowel edema occurs in the perioperative period, and Figure 5.8 Isoparastaltic side-to-side functional end-to-
a marginal lumen may lead to a partial obstruction. The end anastomotic technique.
References 47
minimize the occurrence of this problem. For side-to-side removes contaminated fluid and blood, and should a leak
anastomoses performed with linear staples, care should be occur, it would be controlled. Opponents argue that the
taken to avoid incorporating any portion of the mesentery drain is dangerous as it allows bacteria a portal of entry
in the staple line. Techniques to stop hemorrhage include and it may erode the anastomosis. Trials have clearly shown
cautery of the bleeding vessels or placement of a suture at no benefit from drainage of intestinal anastomoses (58,59).
the site of bleeding. Excessive cautery is to be avoided as Despite evidence to the contrary, the practice of closed suc-
the staple line has the potential to transfer the electrical tion drainage for low pelvic anastomoses the first few days
energy to adjacent portions of the bowel. Reduction or stop- postoperatively continues due to individual surgeon’s beliefs
page of the bleeding may also be helped by digital compres- (5). It should be emphasized that lack of enteric contents in
sion or intraluminal instillation of an epinephrine solution a drain placed near an anastomosis does not preclude an
(1–100,000 or 1–200,000 u/mL). Another option is submu- anastomotic leak.
cosal injection of an epinephrine solution (50).
PROXIMAL PROTECTION (STOMAS) SUMMARY

For high-risk anastomosis, a proximal diverting stoma is
often used. A diverting stoma will not prevent an anasto- Adherence to established surgical principles and
motic leak but will reduce the septic morbidity and mortal- techniques should minimize anastomotic problems.
ity associated with the leakage. A properly constructed loop Mechanical devices cannot overcome limitations in
stoma is almost totally diverting (51). However, if absolute experience, skill, or judgment. Intraoperative iden-
total diversion is desired, a Prasad type of end-loop stoma tification of problems that occur permits correction
may be constructed (52). with minimal morbidity.
If diversion is needed, the author and editors prefer an
ileostomy over a colostomy. A diverting colostomy follow-
ing a colonic resection has several problems. A colostomy REFERENCES
includes a larger stoma, and due to its proximal location, the
ostomy output is loose or liquid and very odorous. If a sig- 1. Beck DE. Intraoperative anastomotic complica-
nificant colonic resection has been performed, the remain- tions. In: Hicks TC, Beck DE, Opelka FG, Timmcke
ing colon length is often insufficient to easily reach the AE. (eds.) Complications of Colon and Rectal
abdominal wall at a preferred stomal location. Closure of Surgery. Baltimore, MD: Williams and Wilkins, 1996,
a loop colostomy puts the marginal blood supply to a more pp. 70–81.
distal anastomosis at risk. 2. Goligher JC. Surgery of the Anus, Rectum, and
A loop ileostomy has several advantages (53). (1) It is Colon, 5th Edition. London, UK: Balliere Tindall,
easy to construct and close. As it is usually created in bowel 1984.
removed from the anastomotic site, (2) tension and blood 3. Steichen FM, Ravitch MM. Surg Clin North Am 1984;
supply are rarely a problem. (3) Ileostomy output is liquid, 64:425–40.
has little odor, and (4) unless the mesentery is abnormally 4. Rafferty JF. Clin Colon Rectal Surg 2001;14:25–31.
shortened, an ileostomy will reach almost any site on the 5. Sweeney WB. Clin Colon Rectal Surg 2001;14:15–23.
abdominal wall. 6. Delaney CP, Mackeiggan JM. Preoperative man-
agement—Risk assessment, medical evaluation,
ADJUVANTS AND DRAINS and bowel preparation. In Wolff BG, Fleshman JW,
Beck DE, Pemberton JH, Wexner SD. (eds.) ASCRS
Due to the morbidity associated with leaks, several adju- Textbook of Colorectal Surgery. New York, NY:
vants have been used in cases where there is high risk of Springer-Verlag, 2007, pp. 116–29.
potential compromised anastomoses. Wrapping the anas- 7. Demetriades D et al. J Traum 2001;50:765–75.
tomosis with omentum is a popular adjunct that is felt by 8. van Geldere D et al. J Am Coll Surg 2002;194:40–7.
many surgeons to prevent disruption. Unfortunately, there 9. Guenga KF et al. Cochrane Database Syst Rev 2003;
is no evidence to support this practice in humans (54,55). 2:CD001544.
The use of foreign materials around the anastomosis has 10. Slim K et al. Br J Surg 2004;91:1125–30.
been shown to be harmful (56,57). Reinforcing sutures posi- 11. Torralba JA et al. Dis Colon Rectum 1998;41:18–22.
tioned around a stapled anastomosis, while not routinely 12. Kressner U et al. Eur J Surg 1994;160:287–92.
necessary, may provide security, especially for low rectal 13. Forloni B et al. Dis Colon Rectum 1998;41:23–7.
anastomoses. 14. Wexner SD, Beck DE. Sepsis prevention in colorectal
Controversy continues regarding the use of drains as surgery. In Fielding LP, Goldnerg SM (eds.) Operative
an adjunct to intestinal anastomosis. The abdominal cav- Surgery, 5th Edition. London, UK: Butterworth-
ity cannot be adequately drained, but in cavities like the Heineman, Ltd., 1993, pp. 41–6.
low pelvis it is possible. Proponents believe that the drain 15. Irvin TT, Goliger JC. Br J Surg 1973;60:461–4.
16. Schrock TR et al. Ann Surg 1973;177:513–8. 38. Ravitch MM. Surg Clin North Am 1984;64:543–54.
17. Khoury GA, Waxman BP. Br J Surg 1983;70:61–3. 39. Tan WS et al. Tech Coloproctol 2007;11:266–7.
18. Hsu T-C. Dis Colon Rectum 1998;41:28–32. 40. Khoury DA, Opelka FG. Dis Colon Rectum 1995;38:
19. Weiber S et al. Eur J Surg 1994;160:47–51. 553–4.
20. Karulf R. Anesthesia and intraoperative positioning. 41. Simillis C et al. Dis Colon Rectum 2007;50:1674–87.
In Hicks TC, Beck DE, Opelka FG, Timmcke AE. (eds.) 42. Beard JD et al. Br J Surg 1990;77:1095–7.
Complications of Colorectal Surgery. Baltimore, MD: 43. Griffith JM, Trapnell JE. J R Coll Surg Edinb 1990;
Williams and Wilkins, 1996, pp. 34–49. 35:35–6.
21. Le TH et al. Dis Colon Rectum 1993;36:197–8. 44. Yalin R et al. Eur J Surg 1993;159:49–51.
22. Graffner H et al. Dis Colon Rectum 1984;27:767–71. 45. Davies AH et al. Ann R Coll Surg Engl 1988;70:345–7.
23. Fingerhut A et al. Surgery 1995;118:479–85. 46. Smith S et al. BMC Surg 2007;7:15.
24. Wheeless CR, Jr, Smith JJ. Obstet Gynecol 1983;62: 47. Gilbert JM, Trapnell JE. Ann R Coll Surg Engl 1988;
513–8. 70:158–60.
25. MacRae HM, McLeod RS. Dis Colon Rectum 1998;41: 48. Wheeler JM, Gilbert JM. Ann R Coll Surg Engl 1999;
180–9. 81:105–8.
26. Gambee LP et al. Am J Surg 1956;92:222–7. 49. Makabeli G, Williams LG. Dis Colon Rectum 1984;27:
27. Templeton JL, McKelvey ST. Dis Colon Rectum 1985; 490–1.
28:38–41. 50. Perez RO et al. Tech Coloproctol 2007;11:64–6.
28. Max E et al. Am J Surg 1991;162:461–7. 51. Pearl RK, Abcarian H. Diverting stomas. In
29. Choy PY et al. Cochrane Database Syst Rev 2007, MacKeigan JM, Cataldo PA. (eds.) Intestinal Stomas:
Issue 3. Art. No.: CD004320. DOI: 10.1002/14651858. Principles, Techniques, and Management. St. Louis,
CD004320.pub2. MO: Quality Medical, 1993, pp. 107–26.
30. Smith LE. Dis Colon Rectum 1981;24:236–42. 52. Prasad ML et al. Arch Surg 1984;119:975–6.
31. Gordon PH, Vasilevsky CA. Surg Clin North Am 1984; 53. Williams NS et al. Br J Surg 1986;73:566–70.
64:555–66. 54. Carter DC et al. Br J Surg 1972;129–33.
32. Dochetry JG et al. Ann Surg 1995;221:176–84. 55. McLachlin AD, Denton DW. Am J Surg 1973;125:
33. Fazio VW. Surg Clin North Am 1988;68:1367–82. 134–40.
34. Last MD, Fazio VW. Dis Colon Rectum 1985;28:979–80. 56. Trowbridge PR, Howes EL. Am J Surg 1967;113:
35. Cohen Z et al. Dis Colon Rectum 1983;26:231–5. 236–40.
36. Griffen FD, Knight CD. Surg Clin North Am 1984;64: 57. Laufman H, Method H. Surg Gynecol Obstet 1948;86:
579–90. 669–73.
37. Julian TB, Ravitch MM. Surg Clin North Am 1984;64: 58. Hoffmann J et al. Dis Colon Rectum 1987;30:449–52.
567–78. 59. Sagar PM et al. Br J Surg 1993;80:769–71.
6
Other intraoperative challenges
JESUS FLORES AND SHARON G. GREGORCYK
arouse suspicion for bleeding tendencies. Medication lists

CHALLENGING CASE are particularly important as many prescription and non-
prescription medications increase the risk for intraoperative
You are performing a laparoscopic sigmoid colectomy bleeding.
on a 45-year-old man with recurrent diverticulitis. The Medications with antiplatelet properties such as aspi-
sigmoid colon is inflamed, the mesentery bleeds eas- rin, nonsteroidal anti-inflammatory drugs, and clopidogrel
ily, and the dissection is difficult as the sigmoid colon should be stopped 5–7 days prior to surgery. Warfarin, a
is stuck down to the left pelvic sidewall. As you are well-known, commonly used anticoagulant is best stopped
dissecting the sigmoid colon free, you note sudden 3–5 days before surgery (1). The surgeon should be familiar
dark blood quickly pooling in the operative field. You with the newer direct oral anticoagulants including rivar-
suspect a major venous injury to the left external iliac oxaban and apixaban (Factor Xa inhibitors) and dabigatran
vein. Despite aggressive suctioning, you struggle to (direct thrombin inhibitor). Rivaroxaban should be held at
find the venous injury. least 24–48 hours prior to surgery, while the timing of when
to stop apixaban and dabigatran ranges from 24 to 96 hours,
depending on the patient’s baseline renal function (2,3). For
CASE MANAGEMENT patients at a higher risk, such as those with a recent coronary
stent, a recent (<3 month) history of venous thromboem-
You alert the anesthesiologist of bleeding from the bolism, or mechanical cardiac valve in the mitral position,
suspected venous injury. You push the bowel you a discussion with the patient’s treating physician is advised.
were just mobilizing into the side wall to help hold The surgeon must remember that common supplements,
pressure as you quickly convert your laparoscopic such as garlic, gingko, ginseng, fish oil, and vitamin E, can
procedure to a hand-assisted or open procedure. also prolong bleeding time and should be stopped 1 week
The colorectal surgeon can face various intraopera- prior to surgery (4). One must be prepared to transfuse
tive challenges in any given case. One must remain platelets or fresh frozen plasma in an antiplatelet therapy or
calm and recall sound clinical judgment and surgical anticoagulation patient requiring an emergency operation.
principles in dealing with complications. In obtaining the surgical history, the indication, type of
operation, postoperative hospital course, and any complica-
tions are important. Reviewing relevant operative reports
can provide information about a patient’s anatomy and can
PREOPERATIVE EVALUATION help anticipate intraoperative challenges, such as adhesions
or altered surgical planes. Abdominal abscesses, fistulae,
and use of mesh are red flags for adhesions. On physical
The preoperative evaluation is the first step in planning any
examination, a stiff, scarred abdominal wall is concerning
operation. It should start with a thorough history and physical
for challenging adhesions.
exam, which may provide clues to intraoperative challenges.
One may discover increased risks for bleeding, extensive
adhesive disease, or unusual anatomical considerations.
Indicators for potential bleeding risk include easy bruis- INTRAOPERATIVE HEMORRHAGE
ing, heavy menses, bleeding gums with brushing teeth, or
family history of coagulation disorders. The patient with Intraoperative hemorrhage can vary, from minimal inter-
hepatic disease, portal hypertension, or renal disease should mittent oozing to major life-threatening exsanguination.
49
50 Other intraoperative challenges
Even minimal intermittent oozing can be problematic, as bank, should transfusion be necessary. Communication
it stains the tissues, absorbs light, and can quickly obscure with the rest of the operating room personnel is also essen-
the operative field laparoscopically. Use of electrocautery tial. They should have topical hemostatic agents, additional
and energy devices, such as the Harmonic scalpel or Enseal suction devices, laparotomy sponges, a long laparoscopic
(Johnson and Johnson), Ligasure (Covidien), or Thunderbeat instrument set, and an open laparotomy tray available in the
(Olympus) can help reduce this problem. These instruments room. Once the anesthesiologist is prepared, the gauze can
can be used for careful blunt or sharp dissection, and divid- be removed and the splenic injury assessed. Topical hemo-
ing the mesentery and major vessels, minimizing the need static agents, such as Avitene (Davol), Surgicel Nu-Knit
to repeatedly exchange instruments. (Ethicon), or Evicel fibrin sealant (Ethicon) can be used to
The surgeon must always be prepared for the possibility of stop bleeding from small splenic capsular tears. An argon
bleeding during division of a major blood vessel, regardless beam coagulator can also be useful and is available for both
of the method of transection and operative approach. During laparoscopic and open procedures (5).
an open procedure, one can often grasp the vessel remnant Despite best efforts, sometimes splenic injuries incurred
and suture ligate it. During a laparoscopic or robotic proce- laparoscopically cannot be controlled and require conver-
dure, the bleeding can stain the camera lens, fill the opera- sion to an open procedure to stop the bleeding. During an
tive field quickly, and prevent viewing of the bleeding vessel. open procedure, the avascular splenic ligaments can be
Preparing for this possibility before dividing a major vessel is divided with electrocautery, allowing medialization of the
essential. One strategy is to hold on to the proximal portion spleen. Manual pressure can be held on the spleen or the
of the vessel, as it is being divided, so that it can be quickly splenic hilum to slow the bleeding. The surgeon must now
occluded should bleeding start. This allows the deploy- decide whether to attempt repair or perform a splenectomy.
ment of hemoclips or an Endoloop (Johnson and Johnson) Different techniques for splenorrhaphy have been
around the base of the bleeding vessel during minimally described. This includes partial splenectomy, mattress
invasive procedures. It is important to have these supplies suture repair, use of argon beam coagulation, mesh wrap, or
in the operating room in case they are immediately needed. a combination. Mattress sutures can be placed directly over
Just as important is deciding what method to use to divide a the splenic surface or over Teflon pledgets (6). The mesh
major blood vessel. In older patients who may have calcified wrap uses a polyglycolic mesh. A keyhole defect is made
vessels, it may be safer to divide them with a linear cutting in the mesh, and the spleen is passed through the defect
stapler with a vascular load than using an energy device. In until the keyhole encircles the splenic hilum. The mesh is
Crohn patients with inflamed, thickened mesentery, it may wrapped around the spleen and sutured to itself, creating a
be smarter to clamp, sharply divide, and tie the mesentery tamponade effect (7).
with sutures rather than use an energy device. The surgeon should make an aggressive attempt to pre-
The spleen can be a major source of intraoperative bleed- serve the spleen. However, splenic preservation should not
ing. Its proximity to the colon and omentum, along with the be undertaken in the face of ongoing uncontrollable hemor-
frequent need to mobilize the splenic flexure during colorec- rhage or hemodynamic instability. In these two situations,
tal operations makes it vulnerable to injury. Techniques a splenectomy is warranted to stop the bleeding and get the
to minimize the risk of splenic injury include incising the patient off the operating room table as soon as possible. The
correct plane, avoiding excessive traction, using an energy surgeon and the patient should be aware that splenectomy
device, and strategically approaching the splenic flexure carries a 5% lifetime risk of overwhelming postsplenectomy
from different angles. If the splenic flexure cannot be fully infection (OPSI), primarily from encapsulated bacteria such
mobilized using a straight laparoscopic approach, conver- as Streptococcus pneumoniae, Haemophilus influenzae, and
sion to hand-assisted laparoscopic procedure can allow bet- Neisseria meningitides (8–10). Vaccinations against these
ter traction, tactile feedback, and safe mobilization. These bacteria are recommended in the 2-week postoperative
extra steps to avoid injury to this delicate organ can save the period. Antibiotic prophylaxis and early aggressive treat-
surgeon headaches from troublesome splenic bleeding. ment of infections should be considered in asplenic patients
The most common splenic injury is a capsular tear, which as OPSI carries a mortality rate as high as 50% (10). Early
is usually controlled with electrocautery. This is true even postoperative complications from splenectomy can include
with minimally invasive approaches. If brisk bleeding is pneumonia, pancreatitis, and subphrenic abscess.
encountered, the spleen should be packed off with gauze. Another challenging situation the colorectal surgeon may
Ray-Tec sponges offer the advantage of being x-ray detect- face is hemorrhage in the pelvis. There are several potential
able and helping estimate blood loss. The surgeon and oper- sources of hemorrhage, including the internal iliac arteries
ating room personnel need to keep count of the number of and veins in the pelvic sidewalls, enlarged pelvic collateral
Ray-Tec sponges placed intraabdominally to ensure that no vessels from portal hypertension, and the presacral venous
sponge is left inside the abdomen. Next, one must address plexus. Surgery in this area can be especially challenging
the bleeding. The anesthesiologist should be notified of the due to the limited work space, presence of pelvic organs
splenic injury and potential for significant blood loss. This such as the uterus, a bulky rectal tumor, inflammation from
allows adequate resuscitation of the patient, placement of abscesses, or absence of anatomical planes from previous
additional IV access, and early notification of the blood pelvic surgery.
Intraoperative hemorrhage 51
Should pelvic hemorrhage occur, immediate pressure 1.5–2 cm2 fragment of rectus abdominis muscle, holding it
should be placed over the area. The surgeon needs to iden- in place with forceps to occlude the bleeding site, and apply-
tify the source of bleeding and make preparations to control ing electrocautery at 100 Hz to the forcep, which transmits
it. This may require conversion to an open procedure. The to the muscle fragment to weld close the bleeding point (15).
pelvis can be packed with laparotomy sponges. The surgeon The muscle fragment may fall off, but the bleeding source
should ask for long instruments, lighted pelvic retractors, is controlled as the underlying vessel has been coagulated
and possibly a second working suction device. As pelvic (Figure 6.1).
hemorrhage can be rapid and severe, the anesthesia team Another described technique involves harvesting a
should be made aware of the situation early on. Cell saver 4 × 2 × 1 cm piece of rectus abdominis muscle as a free
use can allow for autotransfusion of blood. Once the anes- flap and sewing it over the bleeding area to tamponade the
thesia team has adequately resuscitated the patient, the presacral bleeding (16) (Figures 6.2 and 6.3).
packs can be removed and the pelvis assessed for a bleeding In cases of severe pelvic hemorrhage that does not
source. respond to any attempts, the pelvis should be tightly packed
The pelvic floor and sidewalls are lined by the endopel- with laparotomy pads and the patient taken to the surgical
vic fascia. Dissection in the correct plane without violation intensive care unit (ICU) for resuscitation and correction
of this fascia can prevent injury to and bleeding from the of any coagulopathy, acidosis, hypothermia, and anemia.
underlying vascular structures. Violation of the endopel- Successful balloon tamponade with a Sengstaken-Blakemore
vic fascia along the pelvic sidewall can lead to injury to
one of the internal iliac veins, leading to profuse bleed-
ing. Direct pressure over the area is applied. If a vascu-
lar surgeon is readily available, their expertise can prove
very helpful (11). They can attempt suture repair of these
thinned-walled vessels. If they are not immediately avail-
able, suture ligation may be the best option. Other useful
instruments for this situation include clip appliers and
laparoscopic instruments, which in an open case can allow
extra length needed deep in the pelvis. Internal iliac artery
injuries also warrant intraoperative vascular surgery con-
sultation for repair. While an assistant holds pressure
over the injured artery, the surgeon can help prepare the
operative field and ask for a vascular instrument set while
awaiting the vascular surgeon. In the rare case where an
arterial injury is not immediately found in the operating
room, postoperative angiography with embolization can Figure 6.1 An overview of the operative technique of
be useful. rectus abdominis muscle welding. A forceps is used
Violation of the endopelvic fascia covering the sacrum to hold the muscle fragment over the bleeding point,
can lead to injury to the underlying presacral venous plexus. occluding the vessel. Electrocautery is then applied to the
The anatomy and fragility of this plexus makes control of forceps to weld closed the bleeding site. (From Harrison
bleeding difficult (12). This plexus contains avalvular veins JL et al. Dis Colon Rectum 2003;46:1115–7.)
that can attain high pressures. The presacral veins commu-
nicate with the internal vertebral venous system through
the basivertebral vein. They are thin walled and often retract
into the sacral foramen after injury. Use of electrocoagula-
tion or suture ligation to control this bleeding often results
in increased bleeding (13). Direct pressure using laparotomy
sponges to pack the pelvis should be used to gain control
of the bleeding, allowing the anesthesia team to “catch up”
with resuscitation. When possible, the specimen should be
resected to optimize access to the pelvis. Sterile thumb-
tacks or hemorrhage occluder pins can be placed directly
into the sacrum (midline) to occlude the bleeding site. If the
bleeding does not stop but is sufficiently minimized, use of
topical agents or repacking the pelvis may achieve complete
hemostasis (14).
If the massive pelvic exsanguination still does not stop, Figure 6.2 Two untied sutures were applied to adjacent
one can attempt using rectus abdominis muscle fragment tissue near the bleeding site. (From Harrison JL et al. Dis
welding to stop it. This technique involves harvesting a Colon Rectum 2003; 46:1115–7.)
DAMAGE CONTROL
Damage control laparotomy (DCL) was first described in the

trauma literature and refers to an abbreviated resuscitative
strategy where the primary goal is to rapidly control hemor-
rhage and contamination, restore normal physiology, and
remove/debride frankly necrotic tissue (19,20). Laparotomy
pads may be used to temporarily pack off and control mas-
sive hemorrhage. The abdominal wall is often closed tem-
porarily with the use of a negative pressure wound therapy
system, such as the wound V.A.C. (KCI) or ABThera system
(KCI). Patients are subsequently transported to the ICU
for continued resuscitation and correction of coagulopa-
Figure 6.3 Sutures tied over the free muscle flap to thy, hypothermia, and acidosis. Once the physiologic status
provide tamponade. (From Harrison JL et al. Dis Colon has improved and stabilized, the patient is returned to the
Rectum 2003;46:1115–7.) operating room for packing removal, reconstruction, defin-
itive surgery, and abdominal wall closure. This strategy is
now used by surgeons to treat patients in extremis needing
tube or a Bakri postpartum balloon has also been described. resuscitation and stabilization before definitive treatment.
Packing requires a trip back to the operating room in 24–36 The surgeon must use sound judgment in deciding which
hours for removal of the pads. At this time, most bleeding patients would benefit from DCL over continued time spent
will have stopped or become manageable. Another issue to in the operating room, as there is potential morbidity asso-
be addressed, especially in the pelvis, is whether or not to ciated with leaving the abdominal wall open and multiple
proceed with an anastomosis. The patient’s hemodynamic trips to the operating room. Previously proposed selection
status influences this decision. Shock, pressor requirement, criteria include the inability to achieve hemostasis due to
or multiple blood transfusions can lead to poor perfusion of coagulopathy, long surgical procedures in unstable patients,
the anastomosis resulting in a significantly increased risk inaccessible major venous injury, associated life-threaten-
of an anastomotic leak. Deterrents to an anastomosis in the ing injury in a second anatomical location, planned reas-
pelvis include ongoing hypothermia, hypotension, coagu- sessment (24–72 hours) of abdominal contents, inability to
lopathy, major blood loss, and increased time necessary to close the linea alba fascia due to visceral edema, or concern
create the anastomosis. for developing abdominal compartment syndrome (21–23).
With laparoscopic and robotic-assisted surgery, addi- The most common indications for DCL are related to
tional bleeding risk occurs at each trocar site. An abdominal hemorrhage, hemodynamic instability, and massive resus-
wall vessel can be injured during trocar placement. Injury citation. This is usually accompanied by the vicious cycle of
to the liver, spleen, aorta, inferior vena cava, and iliac ves- hypothermia, coagulopathy, and acidosis, along with mas-
sels can occur with trocar placement, as well. Major vascu- sive bowel wall edema. In the nontrauma venue, this may
lar injury rates of 0.03%–0.1% and major abdominal organ be the patient who has received large volume resuscitation
injury rates of 0.08%–0.14% have been reported (17). The for lower gastrointestinal or intraoperative hemorrhage or
most commonly injured abdominal wall vessels during lapa- who has returned to the operating room for postoperative
roscopic port placement are the inferior epigastrics, with an hemorrhage. During the operation, previously hemostatic
average incidence of approximately 0.1%. As the port can sites may begin to bleed, signaling the onset of coagulopa-
tamponade the bleeding, it is good practice to remove each thy, which could be dilutional, consumptive, and/or hypo-
port under direct laparoscopic visualization. Identifying and thermia related. In this bleak scenario, it may be reasonable
repairing an injury at this point prevents a subsequent return to pack the abdomen, apply an occlusive dressing, and take
to the operating room. Typically, the abdominal wall vessel the patient to the ICU for aggressive rewarming, ongoing
will be small and controlled with pressure or electrocau- resuscitation, and optimization. Once the patient is stabi-
tery alone. Should the bleeding persist, the incision may be lized and resuscitated, he or she is taken back to the oper-
enlarged and the vessel suture ligated. Alternatively, bleeding ating room within 24–72 hours. Likewise, a patient may
from a port site can be controlled laparoscopically by using escape the coagulopathic and hypothermic effects of large
a fascial closure device, such as an Endo Close (Covidien) or volume resuscitation, but massive edema may manifest as
a Keith needle (18). The stitch is passed externally through increased pulmonary pressures or hemodynamic compro-
the full thickness of the abdominal wall into the peritoneal mise from abdominal compartment syndrome when the
cavity, grasped, and laparoscopically passed back out of the fascia is closed (24).
abdominal wall and tied. This can be repeated as needed to Occasionally, massive bowel edema can prevent closure
stop bleeding from port sites. of the linea alba fascia. Attempting to close it in light of this
Damage control 53
edema can cause onset of abdominal compartment syn- The abdominal skin is prepared with a tincture of benzoin
drome. This can be encountered when operating for bowel and covered with an Ioban drape (3M) (28). The drains are
obstruction but can be worsened further when septic compli- kept to continuous wall suction. An alternative is to use a
cations accompany the obstruction. Even after the obstruc- negative-pressure wound therapy system device such as the
tion is relieved, the bowel remains edematous from both the V.A.C. (KCI), which may be associated with a higher rate of
obstruction and the resuscitation. Serosal injuries can occur primary delayed fascial closure (29) (Figures 6.4 and 6.5).
from massive dilation and edema. In this scenario, tempo-
rary closure of the abdominal wall with a negative pressure
wound therapy system can allow for resolution of the sys-
temic inflammatory response, edema, bowel dilation, and
patient stabilization. The patient can then be brought back to
the operating room at a later date, undergo assessment, any
anastomoses, and definitive abdominal wall closure.
Abdominal compartment syndrome is defined as
intraabdominal pressure (IAP) of 20 mm Hg or greater,
indirectly measured by urinary bladder pressure, with or
without abdominal perfusion pressure less than 50 mm Hg,
and with single/multiple organ system failure that was not
previously present (25). Abdominal compartment syn-
drome can be caused by increased intraabdominal volume,
retroperitoneal volume, and/or restriction of abdominal
wall expansion. As the IAP increases rapidly, physiologic
derangement in multiple organ systems occurs. Pulmonary
changes are usually the most prominent, with diaphragmatic
elevation leading to decreased pulmonary compliance with
decreased lung capacity, residual capacity, and volumes.
Cardiovascular changes include decreased filling secondary
to venous compression, decreased ventricular end-diastolic
volumes, increased afterload, decreased contractility, and
loss of cardiac output. Prerenal azotemia unresponsive to
Figure 6.4 Example of a damage control laparotomy
volume is characteristic, with oliguria leading to anuria before placement of temporary closure device.
due to decreased renal perfusion from low cardiac output,
decreased glomerular filtration rate, and increased retention
of sodium and water with renin production. Compression
of the splanchnic vasculature leads to ischemia and trans-
location of bacteria. Hepatic insufficiency can also result.
Intracranial pressure increases with decreased cerebral per-
fusion and decreased venous outflow (26).
Patients with high IAP and suspected abdominal com-
partment syndrome require a decompressive laparotomy.
If when attempting to close the abdominal wall there is a
sudden elevation in peak airway pressures, abdominal com-
partment syndrome should be suspected and definitive
abdominal wall closure delayed. The objectives of tempo-
rary closure with a negative pressure wound therapy system
are containment of the abdominopelvic viscera, control of
peritoneal secretions, maintenance of tamponade, and facil-
itation of future abdominal wall closure (27).
Different negative pressure wound therapy systems can
be employed. The surgeon can use a polyethylene sheet,
which is perforated multiple times and placed over the
abdominal viscera and underneath the abdominal wall
peritoneum. Laparotomy pads are placed on top of the poly-
ethylene sheet, and the sponge edges are tucked below the Figure 6.5 Example of a damage control laparotomy after
abdominal skin, fascia, and peritoneum. Jackson-Pratt or placement of temporary closure device, with application
similar closed suction drains are placed on the towels and of the V.A.C. (KCI) system. (Courtesy of Richard Fortunato,
tunneled beneath the skin to exit away from the wound edge. DO, Pittsburgh, PA.)
Patients should be brought back to the operating room are incised under direct visualization. The surgeon can then
for reassessment and possible abdominal closure within 72 visualize and palpate for any bowel adhesions before placing
hours (30). Careful planning, technique, and patient selec- the Hasson camera port. The abdomen is then insufflated
tion should minimize the colorectal surgeon’s encounters with CO2 gas. Another option is the closed Veress technique,
with damage control situations. However, when confronted where a Veress needle is inserted into the peritoneal cavity
with suspect options and a dubious outcome, a DCL can while lifting the abdominal wall for countertraction (32).
turn an uncontrolled situation into a controlled second- The peritoneal cavity is then insufflated with CO2 gas, and
look operation with more desirable options/outcomes. an optical viewing trochar, such as a Visiport (Covidien) or
Endopath Optiview (Ethicon), is then placed under direct
visualization. Alternatively, an optical viewing trochar can
be placed initially without first insufflating the abdomen
ADHESIVE DISEASE with a Veress needle. Ideally, the abdomen is entered in vir-
gin territory away from any adhesions, which is particularly
Abdominal and pelvic adhesions resulting from prior important in the two latter techniques described. One com-
abdominal surgeries or infections continue to be a source of mon location is in the left upper quadrant, also known as
frustration for patients and surgeons. Adhesions can make Palmer’s point (33). Some advocate using a “peek-port” in
an operation challenging by preventing safe entry into the patients who are high risk for conversion to an open proce-
abdomen, requiring conversion to a formal laparotomy, dure. A 7 cm incision is made and the abdomen is assessed.
and increasing the risk of hemorrhage and perforation (31). If the intraperitoneal conditions are favorable, the procedure
Particularly worrisome for adhesions are those patients with is done using hand-assisted laparoscopic surgery (HALS). If
a history of multiple prior abdominal surgeries, bowel per- not, the incision is converted to a formal laparotomy (34).
foration, enterocutaneous fistulas, or intraperitoneal mesh. During an open procedure, one should attempt to enter
An initial surgical objective is to safely enter the peri- the abdomen away from surgical scars. Upward retraction
toneal cavity without causing bowel or vascular injury. of the linea alba fascia can be done using Kocher clamps.
This can be challenging in patients who have had previous The linea alba fascia is then incised slowly and sharply with
abdominopelvic surgery. Sometimes, adhesions encoun- a scalpel. Exposure and visualization are important to avoid
tered during surgery are much less than initially anticipated, injury and can be achieved by good positioning of the oper-
and other times, they are much worse. Multiple scoring sys- ating room surgical lights, frequent suctioning, dabbing
tems for adhesions have been proposed. Adhesions can be with a laparotomy sponge, and use of appropriate handheld
categorized as shown in Table 6.1. retractors. Different techniques for adhesiolysis exist. Most
Patients who harbor significant adhesive disease are surgeons will lyse adhesions sharply with a scalpel or scis-
probably not candidates for laparoscopic or robotic surgery. sors. Thick adhesions close to bowel can be doubly clamped,
In these patients, one strategy may be to look inside the divided sharply, and tied off with sutures. Judicious use of
peritoneal cavity with a laparoscope and assess the adhe- electrocautery and other energy devices is undertaken to
sions. The decision whether to proceed with a laparoscopic/ minimize thermal spread and potential damage to adjacent
robotic procedure or convert to an open procedure can then bowel. Sharp scalpel dissection is especially useful when
be made. A limited number of laparoscopic instruments can very dense adhesions of the bowel to the abdominal wall
be initially opened to reduce costs until a final decision is exist. Sometimes a distinct plane does not exist, and small
made on the approach. pieces of peritoneum need to be excised sharply. More dif-
Placement of the initial camera port is considered criti- ficult or dense adhesions can be approached from different
cal in laparoscopic/robotic surgery, as it can be associated angles to help define the appropriate plane. Simpler adhe-
with gas embolism or injuries to bowel, bladder, or major sions can often be divided on either side or even behind the
vessels. Different options exist for abdominal entry dur- dense adhesion to help delineate the proper path of dissec-
ing minimally invasive surgery. One approach is the open tion. Placing one’s fingers on either side of the adhesion and
Hasson technique, where the rectus sheath and peritoneum palpating can help stretch the adhesion to allow for easier
division. The surgeon must use traction and countertrac-
tion to help expose adhesions and planes. One must be cau-
Table 6.1 Grading system for bowel adhesions tious about the amount of traction, as too much can lead to
Grade Description tears in the serosa and bleeding.
If an enterotomy occurs, it should be repaired immedi-
1 Thin filmy adhesions ately with absorbable sutures to minimize contamination.
2 Adhesions that can be divided by blunt Serosal tears should also be repaired immediately to pre-
dissection vent enlargement. If the case is difficult and additional sero-
3 Dense adhesions that require sharp division sal tears or enterotomies are predicted, temporary closure
4 Dense adhesions, the division of which results and tagging the bowel with long cut sutures can be done.
in bowel injury Once adhesiolysis is completed, the surgeon can run the
Source: Adapted from Fazio VW. Personal communication, 1998. bowel, easily identify the injured portion, and reexamine it.
Adhesive disease 55
A segment of bowel with extensive injuries may need to be

resected and undergo primary anastomosis. When a loop of
small bowel is firmly adherent to a vital structure such as a
ureter or iliac vessel, an enterotomy can be intentionally cre-
ated to separate the structures. The enterotomy can then be
primarily closed, leaving a small piece of bowel wall adhered
to the vital structure avoiding morbid injury. Mucosa left
behind should be desiccated with electrocautery to prevent
mucocele formation or malignant transformation.
The surgeon should always consider methods to mini-
mize and prevent adhesion formation. This minimizes the
risk of future bowel obstruction and makes future abdomi-
nal operations easier and safer. Adhesion formation is a local
response of the peritoneum and abdominopelvic viscera to
ischemia, desiccation, or trauma. They may form as a result
of the primary disease process or due to contact with surgi-
cal instruments, staples, sutures, gloves, sponges, or other
irritants. Several studies suggest that laparoscopic surgery
leads to less adhesion formation than open surgery, by lim-
iting bowel manipulation, lowering bleeding, and limiting
exposure to irritants (35–38). Further support is found in
laparoscopic-assisted ileocolic resection being associated
with a reduced rate of bowel obstruction compared to open
surgery (39,40). Adhesions to the anterior abdominal wall
are minimal or absent (41) (Figures 6.6 and 6.7). We know
that laparoscopic surgery does lead to adhesion formation,
but to a lesser degree than open surgery.
Despite laparoscopic surgery advantages, bowel obstruc-
tions continue to occur frequently but in a different mag-
nitude. In a report for the French Association for Surgical
Research, Duron and colleagues noted that only 33% of Figure 6.6 Laparoscopic images demonstrating lack of
postoperative bowel obstructions following various laparo- adhesions in a patient undergoing laparoscopic appen-
scopic surgeries were due to multiple adhesions, while 17% dectomy for appendicitis 2 years after hand-assisted
were due to a single band (42). Intestinal incarceration in an laparoscopic anterior resection sigmoid diverticulitis.
abdominal wall defect or port site (Figures 6.8 and 6.9) was (Courtesy of Thomas E. Read, MD, Pittsburgh, PA.)
responsible for another 46%.
A report from the Western Pennsylvania Hospital
describes unique mechanisms of bowel obstruction, such as closing the fascia is useful to avoid adhesions of bowel up to
an internal hernia, that are commonly associated with lapa- the abdominal wall. This maneuver may make subsequent
roscopic bariatric surgery (43). This is presumably due to reentry into the abdomen easier.
decreased scar formation between the newly apposed peri- Adhesion barriers can be used to reduce the inci-
toneal surfaces, which leaves defects open (44). A similar dence of adhesions (47). The most well-known and stud-
phenomenon may occur following laparoscopic colon resec- ied is Seprafilm (Genzyme), a bioabsorbable membrane of
tion. Obstructions from internal hernias are associated with sodium hyaluronate and carboxymethylcellulose (48–50). It
a high incidence of bowel ischemia. High suspicion must be significantly decreases the incidence and severity of adhe-
maintained and prompt surgical intervention undertaken. sion formation after abdominal and pelvic surgery (52).
Fortunately, many of these can be identified and managed These products should not be placed adjacent to the suture
laparoscopically. or staple line of newly created anastomoses as they may hin-
General principles to minimize adhesion formation der the bowel’s healing (51).
include handling tissue gently, ensuring adequate hemo- Special attention must be given to adhesions encoun-
stasis, minimizing bowel handling, avoiding desiccation, tered in patients with an abdominal or pelvic malignancy. If
minimizing irritants, and avoiding infection and ischemia. the adhesions are between a malignant tumor and another
Several studies have shown that routine closure of the peri- structure, they should be treated as an extension of the
toneum as a separate layer during open surgery, including malignancy and resected en bloc with the specimen. This
via midline laparotomy and Pfannenstiel incisions, did not may require partial resection of the abdominal wall or a
decrease the incidence of subsequent adhesions (45,46). loop of bowel. Not all adhesions encountered during cancer
Placing omentum over the bowel in the midline before operations are malignant. Attention should be paid to the
Figure 6.9 A port site hernia causing a bowel obstruction

and injury to the bowel.
tumor extent, such as tumor growth through the full thick-

ness of the bowel.
INTRAOPERATIVE LESION
IDENTIFICATION
Up to 22% of endoscopically unresectable colorectal tumors

with benign histology on initial biopsy harbor invasive can-
cer. Adhering to oncological principles for these neoplasms
is advised (52). Many of them will not be easily palpable dur-
ing open surgery and may be even more difficult to find lap-
Figure 6.7 Laparoscopic images demonstrating lack
of adhesions in a patient undergoing laparoscopic aroscopically. For operative planning, accurate preoperative
appendectomy for appendicitis 2 years after hand- localization of the tumor is imperative to ensure removal
assisted laparoscopic anterior resection for recurrent of the tumor-containing bowel segment (53). Preoperative
sigmoid diverticulitis. (Courtesy of Thomas E. Read, MD, estimation of tumor location via colonoscopy alone is poor,
Pittsburgh, PA.) unless the tumor is clearly noted to be in direct proximity
to an unmistakable landmark such as the ileocecal valve or
rectum (54,55).
Different methods to aid in the intraoperative localiza-
tion of a colon or rectal polyp/mass have been proposed. The
most commonly used technique, and now considered the
gold standard, is preoperative endoscopic tattooing (56).
The most commonly used dyes include India ink and SPOT
(GI Supply), both of which are durable and long lasting.
India ink needs to be diluted and sterilized before injecting.
SPOT is available in a prepackaged sterile dilute suspen-
sion. Other dyes, such as methylene blue, indigo carmine,
and indocyanine green, can be used, but these usually dis-
appear within days. Endoscopic injection of India ink or
SPOT in three or four quadrants of bowel adjacent to and
distal to, but not through the tumor, is safe and reliable, and
preferred in most centers (57,58) (Figure 6.10).
Another method used is preoperative colonoscopy with
mucosal endoclip placement followed by an abdominal
Figure 6.8 A port site hernia causing a bowel obstruction plain film (Figure 6.11). Preoperative endoclip placement
and injury to the bowel. can also be combined with intraoperative fluoroscopy or
Trouble with stoma creation 57
Vinci Firefly mode can be activated and the area injected

should fluoresce.
Due to the flexible nature of the colonoscope, the dis-
tance of the tumor from the anal verge cannot be accurately
determined on colonoscopy. When the tumor is obviously
within the colon or is palpable on digital rectal exam, this is
not an issue. Unfortunately, and not uncommonly, a tumor
reported by colonoscopy to be in the sigmoid colon is actu-
ally in the rectum. Surgeons should remember that a rigid
proctoscope is useful to accurately measure the distance
of the tumor from the anal verge. Accurate preoperative
localization of rectal tumors is essential as it defines the
operation to be performed and may influence the need for
neoadjuvant therapy prior to surgery (62).
Careful preoperative evaluation and planning is the
Figure 6.10 India ink injected endoscopically before lapa- best way to ensure that the appropriate segment of intes-
roscopy provides excellent lesion localization. tine containing the pathology is resected. The most notable
preventable cause is an assumption based on colonoscopic
determination of a site that is not within the direct proxim-
intraoperative ultrasound (59). These techniques can be
ity to an unmistakable landmark such as the cecum or rec-
inaccurate due to clip migration or dislodgement. They also
tum. Occasionally, despite best efforts, localization attempts
result in exposure to radiation and are user dependent and
fail to identify lesions intraoperatively in up to 12% of cases
resource intensive. Additional methods include preoperative
(63). Failure to visualize a tattoo can result from its disap-
barium enema and computed tomography colonography.
pearance, especially when products other than SPOT or
With increasing use of robotics in colorectal surgery,
India ink are used. Additionally, failure to inject the prod-
many surgeons are using the da Vinci Firefly high-definition
uct into the submucosal plane can result in dissemination
vision system to assess blood perfusion to the bowel when
of the ink, intraperitoneal injection, and imprecise localiza-
creating the anastomosis (60,61). This requires IV injection
tion. Techniques for proper injection have been described,
of indocyanine green (ICG) followed by switching to Firefly
including inserting the injection needle tangentially in the
mode to assess perfusion. ICG can also potentially assist
submucosa and injecting saline to develop the submucosal
in real-time localization of colon and rectal lesions dur-
plane before ink injection (64,65).
ing robotic cases. ICG can be injected into the submucosal
The surgeon must be prepared to deal with the case
plane around the tumor at the beginning of the case, in a
where localization efforts have failed. Blind resection is not
manner analogous to tattooing with India ink. Later, the da
recommended unless the surgeon is absolutely confident of
the lesion location from preoperative imaging. Mobilization
of the flexures and dissection of the omentum off the trans-
verse colon may reveal a hidden tattoo. During laparos-
copy, placement of a hand port to allow palpation of the
bowel wall can be used to try and palpate the mass. Even
then, the lesion may be too small to palpate. Under such
circumstances, intraoperative colonoscopy can allow for
lesion localization (66). Intraoperative colonoscopy using
CO2 insufflation will minimize bowel distention, as CO2 is
absorbed at a much faster rate than room air. A serosal clip
or suture can be placed laparoscopically at the localized site
once it is found with intraoperative colonoscopy. Regardless
of the localization technique, opening the resected speci-
men by the pathologist, or on a back table during the opera-
tion, is recommended to confirm that the lesion has been
appropriately resected.
TROUBLE WITH STOMA CREATION

Figure 6.11 A plain film of the abdomen after endoscopic
placement of clip (arrow) can provide valuable information
about the location of the lesion and aid in preoperative Surgeons often consider intestinal stomas a minor part of a
planning. major operation, but they have a major impact on the patient
postoperatively. The patient will have to learn to deal with its

physiology, function, appearance, and maintenance. Hence,
ideal stoma creation is essential. This involves choosing the
correct location, identifying the appropriate segment of
bowel to use, and correct construction. The surgical indi-
cation, patient’s lifestyle, clothing preference, occupation,
bowel function, body habitus, and surgical scars all influ-
ence the site chosen for stoma creation (67). Ideally, regard-
less of whether the stoma is temporary or permanent, a
stoma should be created so that it will last a lifetime.
In elective cases, it is important to discuss the possibility
of stoma creation with the patient and have the patient edu-
cated and marked by an enterostomal therapist (Wound,
Ostomy, and Continence Nurse [WOCN]). In urgent situa-
tions or when an enterostomal therapist is not available, the
surgeon must select and mark the stoma site, ideally before Figure 6.12 The “pseudo-loop” (end loop) colostomy can
surgery. In emergent cases, such as for perforated diverticu- be employed in rare cases when no other stoma can reach
litis with sepsis, preoperative marking may not be possible. the abdominal skin. (From Cataldo PA. Clin Colon Rectal
The surgeon must understand the “ostomy triangle,” Surg 2008;21:17–22.)
which is the area bounded by the anterior superior iliac
spine, the pubic tubercle, and the umbilicus. The stoma ide-
ally should be placed through the rectus muscle within the risk to creating a primary anastomosis with a diverting
ostomy triangle, slightly below the umbilicus. It should be loop ileostomy than to constructing a difficult and poorly
located 5 cm away from bony prominences, old scars, the placed colostomy. In rare situations, it may be impossible
belt line, midline incisions, and skin folds to allow proper to create a tension-free well-perfused colostomy despite the
stoma appliance fitting (67,68). The stoma marking should steps described above. In this case, an end-loop colostomy
take place with the patient sitting, standing, and laying can be created, where an oversized abdominal wall opening
supine. In obese patients, preoperatively marking a location is made and only a portion of the antimesenteric border
superior to the umbilicus will make creation of a stoma and of the colonic segment is incised and matured to the skin
its subsequent care much easier. The abdominal wall in the (Figure 6.12).
lower abdomen may be thicker due to a large pannus and Similar steps for assisting in the creation of a well-
obese patients may not be able to see their stoma to care perfused, tension-free loop ileostomy or end ileostomy in
for it. Patients with disabilities should be marked in the a difficult situation have been proposed: supraumbilical
position they spend the most time in to facilitate appliance placement (especially in obese patients), mobilizing the
fitting/care. small bowel mesentery to the base of the duodenum, liga-
The bowel brought up for the stoma has to have adequate tion of ileocolic artery at its origin, creation of an oversized
length to allow it to traverse the abdominal wall and be abdominal wall opening, creating windows in the small
matured without tension retracting it into the abdomen. It bowel mesentery overlying the superior mesenteric artery,
also has to be well perfused without compromise of the mes- bringing the stoma through an Alexis wound retractor
entery. Stoma creation can be challenging in many types of and then cutting out the wound retractor, and creation of
patients, particularly the obese, inflammatory bowel disease a noneverted ileostomy in an emergency case (69). Rarely,
patients with thickened and shortened mesenteries, and even with the above-mentioned steps, it may be impossible
those with edematous, dilated bowel. to create a viable ileostomy. An end-loop ileostomy can be
Steps proposed to facilitate creation of an end colostomy used in this situation, created in a fashion similar to that
or loop colostomy in a challenging situation include full described for an end-loop colostomy.
splenic flexure mobilization, complete division of the lateral When creating a temporary stoma, the surgeon should
peritoneal attachments, ligation of the inferior mesenteric plan strategically for the future and construct the tempo-
artery proximal to the origin of the left colic artery, high rary stoma so that it can be closed locally, if possible, with-
ligation of the inferior mesenteric vein, creating windows out a laparotomy. For example, a patient undergoing a right
in the peritoneum in the medial and lateral aspects of the colectomy for ischemia can have the terminal ileum and
mesocolon, transection of medial peritoneal attachments the antimesenteric corner of the stapled transverse colon
at the base of the mesocolon, using an enlarged abdomi- brought out via a single opening on the right abdominal
nal opening for the stoma, bringing the stoma through an wall. The terminal ileum and the antimesenteric transverse
Alexis wound retractor and then cutting out the retractor, colon corner staple lines are transected and matured. This
and choosing a stoma location superior to the umbilicus allows future creation of an ileocolic anastomosis without
(67–69). The surgeon should also consider if there is less the need for a formal laparotomy (Figure 6.13).
Trouble with stoma creation 59
(a)
(b)
(c)
Figure 6.13 End-loop stoma (Prasad). (a) The entire divided edge of the proximal limb and the antimesenteric corner
of the distal limb are gently drawn through the opening in the abdominal wall. After the abdomen has been closed,
the staple line of the proximal limb is excised completely and only the antimesenteric corner of the distal staple line is
removed. (b) The proximal limb is matured flush with the skin by suturing the deep dermal skin to full-thickness colon
with absorbable sutures. Transition sutures may be placed to help mature the mucous fistula, which has the appearance
of a “mini-stoma.” (c) Sagittal view of the completed end-loop colostomy. Note the portion of the distal staple line in the
subcutaneous tissue. (From Cataldo PA. Clin Colon Rectal Surg 2008;21:17–22.)
30. Abikhaled JA et al. Am Surg 1997;63:1109–12.

CONCLUSIONS 31. Maciver AH et al. Int J Surg 2001;9:589–94.
32. Taye MK et al. J Clin Diagn Res 2016;10(2):QC04–7.
The colorectal surgeon can be faced with any number of 33. Vilos GA et al. J Obstet Gynaecol Can 2007;29(5):
intraoperative challenges. Some of these can be anticipated, 433–47.
while others are only discovered during an operation. 34. Read TE et al. Surg Endosc 2009;23:477–81.
Proper preparation, calm reason, and sound surgical prin- 35. Gutt CN et al. Surg Endosc 2004;18:898–906.
ciples are the keys to optimizing the outcome. 36. Hiki N et al. Br J Surg 2006;93(2):195–204.
37. Kalff JC et al. Ann Surg 1998;228:652–63.
REFERENCES 38. Schwarz NT et al. Gastroenterology 2004;126:159–69.
39. Alabaz O et al. Eur J Surg 2000;166(3):213–7.
1. Ansell J et al. Chest 2004;126(3 Suppl): 204S–33S. 40. Bergamaschi R et al. Dis Colon Rectum 2003;46:
2. Schulman S et al. Circulation 2015;132:167–73. 1129–33.
3. Sunkara T et al. Health Serv Insights 2016;9:25–36. 41. Hasegawa H et al. Br J Surg 2003;90:970–3.
4. Ang-Lee MK et al. JAMA 2001;286(2):208–16. 42. Durron JJ et al. Arch Surg 2000;135(2):208–12.
5. Kwon AH et al. Eur J Surg 2001;167:316–8. 43. Papasavas PK et al. Surg Endosc 2003;17:610–4.
6. Holubar SD et al. Arch Surg 2009;144(11):1040–5. 44. Higa K et al. Obes Surg 2003;13:350–4.
7. Lange DA et al. J Trauma 1988;28(3):269–75. 45. Kapustian V et al. Am J Obstet Gynecol 2012;206:
8. Yorkgitis BK. Am J Med 2016;130(3):365e1–5. 56.e1–4.
9. Summary of Recommendations for Adult 46. Gurusamy KS et al. Cochrane Database Syst Rev
Immunization. The Advisory Committee on 2013;7:1–35.
Immunization Practices. Department of Health and 47. Rajab TK et al. J Surg Res 2010;161:246–9.
Human Services. Centers for Disease Control and 48. Becker JM et al. J Am Coll Surg 1996;183:297–306.
Prevention. Available at www.immunize.org/catg.d/ 49. Diamond MP. Fertil Steril 1996;66(6):904–10.
p2011.pdf. Accessed on January 7, 2017. 50. Nordic Adhesion Prevention Study Group. Fertil Steril
10. Hammerquist RJ et al. Am J Health Syst Pharm 2016; 1995;63(4):709–14.
73:e220–8. 51. Beck DE et al. Disc Colon Rectum 2003;46(10):
11. Oderich GS et al. J Vasc Surg 2004;39:931–6. 1310–9.
12. Lou Z et al. World J Gastroenterolog 2013;19(25): 52. Brozovich M et al. Surg Endosc 2008;22(2):506–9.
4039–44. 53. Larach SW et al. Disc Colon Rectum 1997;40:592–6.
13. Filippakis GM et al. Am Surg 2007;73(4):410–3. 54. Piscatelli N et al. Arch Surg 2005;140(10):932–5.
14. Celentano V et al. Ann R Coll Surg Engl 2014;96: 55. Yeung JM et al. Colorectal Dis 2009;11:527–30.
261–5. 56. Ibrahim-Abdelaziz I et al. J Clin Trials 2015;6:227–30.
15. Harrison JL et al. Dis Colon Rectum 2003;46:1115–7. 57. McArthur CS et al. Surg Endosc 1999;13(4):397–400.
16. Remzi FH et al. Dis Colon Rectum 2002;45:1109–11. 58. Nizam R et al. Am J Gastroenterol 1996;91(9):1804–8.
17. Uranues S et al. Langenbeck’s Arch Surg 2016;401: 59. Nagata K et al. Surg Endosc 2008;22(2):379–85.
909–12. 60. Hellan M et al. Surg Endosc 2014;28:1695–702.
18. Chatzipapas IK, Magos AL. Obstet Gynecol 1997; 61. Whealon M et al. Clin Colon Rectal Surg 2016;29:
90(2):304–6. 221–31.
19. Cotton BA et al. Ann Surg 2011;254(4):598–605. 62. McCormick JT, Gregorcyk SG. Surg Oncol Clin N Am
20. Germanos S et al. Int J Surg 2008;6:246–52. 2006;15(1):39–49.
21. Moore EE et al. World J Surg 1998;22:1184–91. 63. Feingold DL et al. J Gastrointest Surg 2004;8(5):543–6.
22. Loveland JA, Boffard KD. Br J Surg 2004;91(9): 64. Cho YB et al. World J Surg 2007;31:1491–5.
1095–101. 65. Park JW et al. Surg Endosc 2008;22(2):501–5.
23. Voiglio EJ et al. J Visc Surg 2016;153:13–24. 66. Zmora O et al. Surg Endosc 2002;16:808–11.
24. Offner PJ et al. Arch Surg 2001;136:676–81. 67. Cataldo PA. Clin Colon Rectal Surg 2008;21:17–22.
25. Sugrue M. Curr Opin Crit Care 2005;11:333–8. 68. Beck SJ. Clin Colon Rectal Surg 2011;24:259–62.
26. O’Mara MS et al. Am Surg 2003;69(11):975–7. 69. Williams RS, Fazio VW. Challenging stomas. In: Fazio,
27. Schreiber MA. Crit Care Clin 2004;20:101–18. Church, and Wu. (eds) Atlas of Intestinal Stomas, 2nd
28. Barker DE et al. J Trauma 2000;48:201–7. Edition. New York, NY: Springer Science+Business
29. Garner GB et al. Am J Surg 2001;182:630–8. Media, LLC, 2012, pp. 273–84.
7
Optimizing use of electronic health records
ALLISON B. McCOY AND GENEVIEVE B. MELTON-MEAUX
CHALLENGING CASE ELECTRONIC HEALTH RECORDS
A 64-year-old woman presents to your office with a The Centers for Medicare and Medicaid Services (CMS)
history of a symptomatic colon mass. define an EHR as “an electronic version of a patient’s medi-
cal history, that is maintained by the provider over time,
and may include all of the key administrative clinical data
relevant to that persons care under a particular provider,
CASE MANAGEMENT including demographics, progress notes, problems, medi-
cations, vital signs, past medical history, immunizations,
Review of the patient’s electronic health records laboratory data and radiology reports” (2). Computer-based
documents a colonoscopy with a partially obstructing hospital information systems first began to appear in the late
mass of the sigmoid colon and the pathology report 1960s, largely as text-based systems intended to facilitate
with a moderately differentiated adenocarcinoma. billing and reimbursement. In 1969, Lawrence Weed intro-
She has no other significant medical problems. As duced the problem-oriented medical record, suggesting
you initiate an order set for a colonic surgery using an that medical records should be organized primarily by the
enhanced recovery after surgery pathway, the system medical problem, and that diagnostic and therapeutic plans
identifies the patient’s penicillin allergy and alerts you should be linked to these problems (3). Over the next several
to modify your preoperative antibiotic selection. decades, information systems implemented in clinical set-
tings evolved to capture clinical information and improve
patient outcomes through better information management
and CDS. Since that time, EHRs have continued to evolve
INTRODUCTION and become increasingly vital to modern patient care.
Contemporary EHRs can be classified as having basic or
In recent years, the United States has invested billions of dol- comprehensive electronic functions (Table 7.1). Basic EHRs
lars and other resources through the Health Information include electronic clinical information; computerized pro-
Technology for Economic and Clinical Health (HITECH) act vider order entry (CPOE) for medications; and viewing of
to rapidly increase electronic health record (EHR) adoption lab reports, radiology reports, and diagnostic test results.
nationwide (1). Implementation of EHRs has been reported In addition to these, electronic functions required for com-
to improve patient safety and reduce health-care costs by prehensive EHRs include advance directives; CPOE for lab
providing better access to information, clinical decision sup- reports, radiology tests, consultation requests, and nursing
port (CDS), and more reliable communication. These systems orders; and viewing of radiology images, diagnostic test
require ongoing investment by health-care institutions and images, and consultant reports. As a result of the HITECH
by clinician stakeholders, particularly around new proto- act, including incentive programs such as the meaningful
cols and pathways, as well as the addition at the individual use program, implementation of both basic and comprehen-
or group practice level of personalized templates or other sive EHRs has increased in recent years. In 2015, 96% of U.S.
mechanisms to streamline documentation practices. In this nonfederal acute care hospitals reported possession of EHR
chapter, we present an overview of EHRs and their functions. technology, 83.8% reported implementation of a basic EHR,
We also describe evidence for the role of EHRs in improved and 40% reported implementation of a comprehensive EHR
outcomes through implementation and use of these func- (4). Another common framework is the HIMSS Analytics
tions with emphasis on colon and rectal surgery examples. Electronic Medical Record Adoption Model (EMRAM),
61
62 Optimizing use of electronic health records
Table 7.1 Electronic functions required for hospital Table 7.2 HIMSS analytics Electronic Medical Record
adoption of basic or comprehensive Electronic Health Adoption Model (EMRAM)
Records (EHRs)
Stage EMRAM cumulative capabilities
Basic Comprehensive
7 Complete electronic medical record, data
EHR EHR
analytics to improve care
Electronic clinical information 6 Physician documentation (templates), full clinical
decision support (CDSS), closed-loop
Patient demographics ✓ ✓
medication administration
Physician notes ✓ ✓
5 Full picture archiving and communication system
Nursing assessments ✓ ✓
(R-PACS)
Problem lists ✓ ✓
4 Computerized provider order entry; clinical
Medication lists ✓ ✓
decision support (clinical protocols)
Discharge summaries ✓ ✓
3 Clinical documentation, CDSS (error checking)
Advance directives – ✓
2 Clinical data repository (CDR), controlled medical
Computerized provider order entry vocabulary, clinical decision support (CDS),
Lab reports – ✓ health information exchange (HIE) capable
Radiology tests – ✓ 1 All three ancillaries installed—Lab, rad, pharmacy
Medications ✓ ✓
Consultation requests – ✓ Table 7.3 Top commercially available Electronic Health
Nursing orders – ✓ Record (EHR) vendors used for meaningful use attestations
Results management Meaningful use

View lab reports ✓ ✓ EHR vendor attestations
View radiology reports ✓ ✓ Epic Systems Corporation 166,215
View radiology images – ✓ Cerner Corporation 124,471
View diagnostic test results ✓ ✓ Allscripts 32,153
View diagnostic test images – ✓ athenahealth, Inc. 21,068
View consultant report – ✓ eClinicalWorks, LLC 16,818
Decision support GE Healthcare 15,440
Clinical guidelines – ✓ NextGen Healthcare 15,179
Clinical reminders – ✓ Greenway Health, LLC 8,469
Drug allergy results – ✓ Medical Information Technology, Inc. 5,344
Drug-drug interactions – ✓ (MEDITECH)
Drug-lab interactions – ✓
Drug dosing support – ✓ EHRs have increased, in part due to the extensive efforts
required by internal informatics personnel to meet criteria
which classifies EHRs using a similar paradigm around for the incentive programs introduced with the HITECH
functionality and capabilities available within their EHR act. In fact, all of the four benchmark institutions previously
and other integrated health information systems and imple- identified are fully transitioned or in the process of transi-
mented by the particular organization (Table 7.2) (5). tioning to a commercially available EHR. While hundreds of
Early adopters of EHRs were primarily large, academic, commercially available EHR products have been developed,
teaching facilities with locally developed, noncommercial sys- only nine vendors represent 40% of all attestations by provid-
tems. A systematic review of health information technology ers who had met requirements for meaningful use incentives
(HIT) and associated reported research in 2006 found that in 2016 (Table 7.3) (11).
24% of included studies came from four benchmark research
institutions with locally developed EHRs (6): the Regenstrief
Institute (Regenstrief Medical Record System) (7), Brigham
and Women’s Hospital/Partners Health Care (Brigham
EVIDENCE FOR ELECTRONIC HEALTH
Integrated Computer System/Longitudinal Medical Record) RECORD USE
(8), the Department of Veterans Affairs (Veterans Health
Information Systems and Technology Architecture [VISTA]) Since the advent of EHRs, extensive research has been con-
(9), and LDS Hospital/Intermountain Health Care (Health ducted to evaluate their impact and potential value on
Evolution through Logical Programming [HELP]) (10). More improving outcomes such as health-care processes and qual-
recently, the adoption and maturity of commercially available ity, patient morbidity and mortality, patient and provider
Challenges with electronic health record use 63
satisfaction, and costs. Much of the earliest research on EHRs optimal, efficient, and safe care, health-care organizations
included capabilities for identifying medication errors, which are increasingly recognizing the key role of EHRs as a cat-
occurs in 4%–6% of orders using traditional manual checks, alyst and lever to meet these goals. A 2011 study reported
thus supporting the value of these systems for prevention the cost of implementing EHRs to be $32,409 per provider
of medication errors particularly through CPOE and CDS through the first 60 days, in addition to $25,000 for one-time
functionality. For example, one study demonstrated that this hardware costs, $7,000 per provider for individual hardware
functionality was associated with a reduction of medication components, and $17,000 per provider annually for software
errors by 81% (12–18). Following these initial reports, most of and maintenance costs (29). Another report estimated the
the evidence for improved outcomes with EHRs has focused costs of implementation based on regional extension cen-
on health-care process and quality measures. CDS, in par- ter experiences at $33,000 up front and $4,000 annually
ticular, through the alerts and reminders, has been effective per provider for in-office EHRs and $26,000 up front and
at increasing preventive care and health maintenance com- $8,000 annually per provider for software as a service (i.e.,
pliance and publicly reported performance, clinical study Web-based) EHRs (28).
ordering, and therapy prescribing (19,20). Another frequently reported challenge with EHR use
Despite the great promise of improved safety through is poor fit with existing workflows (25–27). Specific work-
EHR use, evidence surrounding the effects on patient outflow issues that have been identified include altered pace,
comes, including length of stay, morbidity, mortality, quality sequencing, and dynamics of clinical activities; failure
of life, and adverse events, remains mixed. Many institu- to support all required activities for all clinical personnel
tions have reported positive effects, but many of these stud- particularly for specialty care; reduced clinical situation
ies have been limited by weak study designs, small sample awareness; and poor reflection of organizational policies
sizes, lack of significant findings, and limited generalizabil- and procedures (30). Further efforts by EHR vendors and
ity due to study setting and subtle or significant differences researchers as well as information technology functions
in EHR configuration, making generalizability question- within health-care organizations are necessary to ensure
able (19,20). Recent reports have called for improved assess- that EHRs meet the needs of providers and other members
ment and increased mechanisms to provide for scalability of the care team, ultimately to facilitate and improve patient
and external validity of EHR implementations to better care without causing additional burdens.
understand and ultimately optimize EHR implementations Beyond this, another important challenge with EHRs is
to ensure improved outcomes (21). that in some cases they have facilitated errors and inefficien-
Despite the large costs for implementation and main- cies through unintended consequences, including more/
tenance of EHRs, some evidence also exists to support a new work, workflow issues, increasing or different system
positive return on investment for EHRs, in particular with demands, altered communication, introduction of new kinds
implementation of CDS functionality to prevent costly of errors, systems causing power shifts, and user dependence
adverse drug events. One study found high costs involved on the system (31). A large survey found widespread occur-
with implementing CPOE but greater projected financial rence of these unintended consequences and highlighted the
benefit with a $28.5 million savings over 10 years, and CDS need for detection and management of these consequences
resulted in the greatest cumulative savings (22). A related to mitigate resulting errors. More recent efforts have con-
study evaluated the return on investment of vendor CPOE tinued to identify potential safety concerns or hazards with
systems implemented in four community hospitals and EHR implementations and called for new and improved
found modest returns, primarily due to lack of CDS (23). A approaches to identifying and reducing these safety issues
study of EHRs in the ambulatory setting also found a positive (21,32,33).
return on investment beyond reduced adverse drug events, In part due to the previously described challenges to
including savings in personnel and other indirect costs (24). implementation and use, provider satisfaction with EHRs,
especially after initial implementation, is low (34,35).
While it appears most providers would not want to return
to the paper record, the burden of use of EHRs is cited as
CHALLENGES WITH ELECTRONIC one of the greatest sources of provider burn-out and dis-
HEALTH RECORD USE satisfaction (36). Because health-care providers and other
clinicians such as nurses and other interdisciplinary staff
Although many have found a return on investment with are primary users of EHRs, their perceptions of EHRs are
EHR implementations, high initial costs of implement- key for ensuring the successful and optimal integration of
ing EHRs and ongoing high year over year maintenance EHRs effectively into clinical practice (37). In one study,
are frequently reported as barriers to adoption by many after implementing an EHR, clinicians’ perceptions of pro-
health-care organizations (25–27). Costs of EHR imple- ductivity, patient care, clinical decision quality, easy access
mentation vary but generally include hardware, software, to patient information, time for patients, computer access,
implementation assistance, training, and ongoing network adequate resources, and ease of use decreased significantly
fees and maintenance (28). With significant ongoing and (34). These findings support the hypothesis that improving
increasing regulatory requirements and the need to provide implementations and optimizations of EHRs to meet the
64 Optimizing use of electronic health records
needs of providers and reduce errors may increase satisfac- colon and rectal diseases, alerts often prompt primary care
tion and, ultimately, further increase health-care quality providers to order or remind patients about cancer screen-
and safety overall. The medicolegal implications of EHRs ing tests or to follow up with patients after abnormal screen-
are discussed in Chapter 15. ing test results (51–53).
EHR features around documentation, particularly opera-
tive notes, have also been described (39,54–58). This approach
for increased structured information and templated docu-
ELECTRONIC HEALTH RECORD USE
mentation with key elements based off of evidence-based
IN COLON AND RECTAL SURGERY care are termed “synoptic reports.” This documentation
approach has been used extensively in pathology, as well as
There are a number of examples of use of EHRs in surgi- to some extent in radiology, and is considered best practice.
cal settings and colon and rectal surgery in particular, Overall, studies around operative notes to date have been of
including CDS applications (38). A systematic review by marginal quality and have not been directly described with
Robertson and colleagues systematically describes the use colon and rectal surgery procedures specifically. However,
of EHRs for surgical care broadly including a number of it does appear that EHR capabilities and focused initia-
colon and rectal surgery examples (39). In the area of medi- tives around templated or synoptic operative reports can be
cation dosing support, EHRs have been demonstrated to be effective for improving the inclusion of critical information
highly effective in both improving patient safety and reduc- including closure details, type of anesthesia, use of antibiot-
ing associated costs, including important examples like ics, and type of dissection or approach.
compliance with the Surgical Care Improvement Project Information display ensures that clinicians have up-
for process metrics related to colon and rectal surgery care, to-date and necessary patient data to make decisions in
including prophylactic antibiotic dosing and early Foley providing care to the patients, such as showing recent labo-
catheter removal (40–43). For busy surgeons, order facilita- ratory test values during medication ordering. While spe-
tors including protocoled templates and preference lists can cific examples for colon and rectal surgery have not been
reduce the amount of time spent entering orders and leave extensively described in the literature, knowledge about lab
more time for providing patient care. Order sets, collections trends or cost of care is an important consideration for cli-
of orders that are grouped by a specific clinical purpose, nicians across all care settings.
are frequently used in the inpatient to ensure adherence to Expert systems apply advanced logic or computational
guidelines or protocols (44). In surgical settings, order sets methods to assist clinicians in ordering, diagnosing, treat-
include those for admission to the colon and rectal surgery ing, and interpreting elements with the EHR. One example
service or for postoperative care of patients (44,45). specific to colon and rectal surgery includes a study that
Another prominent example where EHRs around order applied statistical methods to predict outcomes for patients
facilitators like order sets are mechanisms around colon with diverticulitis (59). Another study evaluated a treat-
and rectal surgery enhanced recovery after surgery (ERAS) ment planning system that suggests volume resuscitation
programs. ERAS programs that include protocolized care and medication therapy, such as use of antibiotics and vaso-
like perioperative deep venous prophylaxis, early resump- pressors, for surgical patients identified as having sepsis,
tion of diet, and fluid restriction can be significantly aided and found that use of the CDS improved mortality in the
and improved with standard EHR content and functional- observed patients (60).
ity and result in decreased length of stay, lower costs, and Workflow support includes order routing, registry func-
other improved outcomes (46). In addition, EHRs enable tions, medication reconciliation, automatic order termination,
analysis, reporting, and registry implementation—often order approvals, free-text order parsing, and documentation
required by national flagship ERAS initiatives such as the aids. Registry functions are especially important for colon and
American College of Surgeons and Agency for Healthcare rectal surgery, where appropriate patient follow-up is impera-
Research and Quality Program for Improving Surgical Care tive for providing optimal care. For example, several studies
and Recovery (ISCR), which is currently focused primarily have described reminders sent to patients, through mailed let-
on colon and rectal surgery procedures (47,48). ters or telephone calls, to remind patients about completion of
Point-of-care alerts and reminders prompt clinicians fecal occult blood tests (61–63).
about drug-condition, drug-drug, and drug-allergy inter-
actions; remind clinicians to assess specific care items;
and notify clinicians about critical lab values or high-risk SUMMARY
states. These reminders can be passive alerts that display
additional text, change existing text colors, or show images, Adoption of EHRs continues with significant impact
without interrupting the workflow, and can also be inter- on the practice of medicine including colon and rec-
ruptive alerts, which require that providers acknowledge tal surgery. From most perspectives, it appears that
or respond to the alert before resuming order entry (49,50). this trend will continue, and as these technologies
Most examples of CDS usage in the literature for surgical mature and organizations learn and implement best
patients are related to antibiotic dosing and timing (39). For
References 65
27. Linder JA et al. AMIA Annu Symp Proc. 2006;2006:

practices, their potential benefits and value will hope- 499–503.
fully be increasingly realized, including improved 28. https://www.healthit.gov/providers-professionals/
outcomes. Utilization of these technologies has pro- faqs/how-much-going-cost-me. Accessed July 23,
duced improvements in patient care but at a cost as 2018.
described in this chapter. Several opportunities exist 29. Fleming NS et al. Health Aff (Millwood). 2011;30(3):
for EHRs to improve outcomes in colon and rectal 481–9.
surgery including best practices like ERAS. 30. Campbell EM et al. J Gen Intern Med. 2009;24(1):21–6.
31. Ash JS et al. J Am Med Inform Assoc JAMIA. 2007;
14(4):415–23.
REFERENCES 32. Singh H, Sittig DF. BMJ Qual Saf. 2016;25(4):226–32.
33. Sittig DF, Singh H. Qual Saf Health Care 2010;
1. Blumenthal D. N Engl J Med. 2010;362(5):382–5. 19(Suppl 3):i68–74.
2. https://www.cms.gov/Medicare/E-Health/EHealth 34. Krousel-Wood M et al. J Am Med Inform Assoc. 2018
Records/index.html. Accessed July 23, 2018. Jun 1;25(6):618–26.
3. Weed LL. N Engl J Med. 1968;278(12):652–7. 35. Hanauer DA et al. J Am Med Inform Assoc. 2016;
4. Henry J et al. Adoption of Electronic Health Record ocw077.
Systems among US Non-Federal Acute Care 36. Friedberg MW et al. Rand Health Q. 2014;3(4):1.
Hospitals: 2008–2015. Washington, DC: Office of 37. Zhou L et al. J Am Med Inform Assoc. 2009;16(4):
the National Coordinator for Health Information 457–64.
Technology, 2016. Report No.: 35. 38. McCoy AB et al. Clin Colon Rectal Surg. 2013;26(1):
5. HIMSS Analytics. 2017 [cited December 28, 2017]. 23–30.
Available from: http://www.himssanalytics.org/emram 39. Robinson JR et al. J Surg Res. 2016;203(1):121–39.
6. Chaudhry B et al. Ann Intern Med. 2006;144(10):742. 40. Jacques P S et al. Surg Infect. 2005;6(2):215–21.
7. McDonald CJ et al. Int J Med Inf. 1999;54(3):225–53. 41. O’Reilly M et al. Anesth Analg. 2006;103(4):908–12.
8. Teich JM et al. Int J Med Inf. 1999;54(3):197–208. 42. Schwann NM et al. Anesth Analg. 2011;113(4):869–76.
9. Brown SH et al. Int J Med Inf. 2003;69(2):135–56. 43. Thirukumaran CP et al. Health Serv Res. 2015;50(1):
10. Gardner RM et al. Int J Med Inf. 1999;54(3):169–82. 273–89.
11. https://www.healthdata.gov/dataset/cms-medicare- 44. Wright A et al. Int J Med Inf. 2012;81(11):733–45.
and-medicaid-ehr-incentive-program-electronic- 45. Hayman AV et al. Am J Surg. 2010;200(5):572–6.
health-record-products-used. Accessed July 23, 46. Carmichael JC et al. Dis Colon Rectum. 2017;60
2018. (8):761.
12. Bates DW et al. JAMA J Am Med Assoc. 1995;274(1): 47. Ban KA et al. J Am Coll Surg. 2017;225(4):548–557.e3.
29–34. 48. https://www.ahrq.gov/professionals/quality-patient-
13. Lesar TS et al. JAMA J Am Med Assoc. 1997;277(4): safety/hais/tools/enhanced-recovery/index.html.
312–7. Accessed July 23, 2018.
14. Bobb A et al. Arch Intern Med. 2004;164(7):785–92. 49. McCoy AB et al. AMIA Annu Symp Proc, Washington,
15. Reckmann MH et al. J Am Med Inform Assoc JAMIA. DC. 2008, pp. 1051.
2009;16(5):613–23. 50. Lo HG et al. J Am Med Inform Assoc JAMIA. 2009;
16. Ammenwerth E et al. J Am Med Inform Assoc. 2008; 16(1):66–71.
15(5):585–600. 51. McDonald CJ et al. Ann Intern Med. 1984;100(1):
17. van Rosse F et al. Pediatrics. 2009;123(4):1184–90. 130–8.
18. Bates DW et al. J Am Med Inform Assoc JAMIA. 1999; 52. McPhee SJ et al. Arch Intern Med. 1989;149(8):1866–72.
6(4):313–21. 53. Singh H et al. Am J Gastroenterol. 2009;104(4):942–52.
19. Bright TJ et al. Ann Intern Med. 2012;157(1):29–43. 54. Laflamme MR et al. AMIA Annu Symp Proc,
20. Jones SS et al. Ann Intern Med. 2014;160(1):48–54. Washington, DC. 2005, pp. 425–9.
21. Committee on Patient Safety and Health Information 55. Cowan DA et al. Dermatol Surg Off Publ Am Soc
Technology, Institute of Medicine. Health IT and Dermatol Surg Al. 2007;33(5):588–95.
Patient Safety: Building Safer Systems for Better Care. 56. Park J et al. J Am Coll Surg. 2010;211(3):308–15.
Washington, DC: The National Academies Press, 2012. 57. Hoffer DN et al. Int J Med Inf. 2012;81(3):182–91.
22. Kaushal R et al. J Am Med Inform Assoc JAMIA. 2006; 58. Ghani Y et al. Int J Surg Lond Engl. 2014;12(5):30–2.
13(3):261–6. 59. Chapman JR et al. Ann Surg. 2006;243(6):876–880.
23. Zimlichman E et al. Jt Comm J Qual Patient Saf. 2013; discussion 880–883.
39(7):312–8. 60. Moore LJ et al. Am J Surg. 2010;200(6):839–44.
24. Grieger DL et al. J Am Coll Surg. 2007;205(1):89–96. 61. Ornstein SM et al. J Fam Pract. 1991;32(1):82–90.
25. Poon EG et al. Health Aff Proj Hope. 2004;23(4):184–90. 62. Goldberg D et al. Am J Prev Med. 2004;26(5):431–5.
26. DesRoches CM et al. N Engl J Med. 2008;359(1):50–60. 63. Mosen DM et al. Med Care. 2010;48(7):604–10.
8
Postoperative anastomotic complications
TRACI L. HEDRICK
associated with comorbid conditions including prior radia-

CHALLENGING CASE tion or immunosuppression. Ultimately, much of the severe
morbidity associated with colorectal surgery can be attrib-
A 63-year-old morbidly obese woman undergoes a uted to the construction of an anastomosis. Anastomotic
left nephrectomy with partial colectomy and primary leak (AL) is, in fact, the most common cause of death fol-
anastomosis for a renal cell carcinoma involving the lowing colon and rectal surgery (1–3). Traditionally, it was
left colon through a left flank incision. On postopera- thought that the biggest threat to the integrity of an anasto-
tive day 3 she develops tachypnea and tachycardia mosis was tension and ischemia. However, modern research
requiring transfer to the intensive care unit. At that has called this into question as data continue to emerge cit-
time, an abdominal radiograph reveals dilated loops ing the local microbiome as a critical factor in the develop-
of bowel but no free air. Her white blood cell count ment of an anastomotic leak (4,5).
is 11, and the rest of her labs are normal. She subse- Along with esophageal anastomoses, the colorectal anas-
quently develops a fascial dehiscence after a deep tomosis is among the highest-risk gastrointestinal anas-
cough on postoperative day 4 requiring exploration tomoses (6). The incidence of anastomotic leak following
and closure with retention sutures. On postopera- colorectal surgery ranges from 3% to 7% for colon anasto-
tive day 6 she develops fever and worsening respira- moses and up to 20% for distal rectal anastomoses (6,7). The
tory failure necessitating intubation. A computed incidence has remained stable over time despite the advent of
tomography (CT) scan is performed demonstrating an minimally invasive surgery and improved perioperative care
anastomotic leak with contrast extravasation. (6). The burden of anastomotic complications is far reach-
ing, including effects on mortality, morbidity, costs, patient-
reported outcomes, and oncologic outcomes. Development
of an AL doubles the cost of index admission compared to
CASE MANAGEMENT those without a leak in patients undergoing gastrointestinal
anastomoses ($16,085 versus $30,409, p < 0.0001) and pro-
She is taken to the operating room where a trans- longs length of stay by up to 7.3 days (6,8). Anastomotic leak
verse loop colostomy is performed through a sepa- (AL) can also have a significant impact on subsequent bowel
rate right upper quadrant incision. A percutaneous function due to reduced distensibility and capacity second-
drain is placed under CT guidance into the fluid ary to fibrosis. Following an AL, patients report worse phys-
collection within the left paracolic gutter. She subse- ical and mental scores, emotional stability, social function,
quently improves and is discharged on postoperative and general health (9–12).
day 14. The development of an AL is also associated with higher
rates of colorectal cancer recurrence and worse disease-
free survival (13,14). Two large meta-analyses of 13,655 and
21,902 patients demonstrate significantly higher cancer-
INTRODUCTION/BACKGROUND specific mortality and local recurrences in patients who suf-
fered an AL (15,16). In addition to the alterations in the local
Colorectal surgery is associated with significant morbid- inflammatory milieu, the delay or omission of adjuvant che-
ity owing to a multitude of different factors. The operations motherapy is thought to mediate the deleterious effects on
are complex, involve multiple quadrants, and are frequently oncologic outcomes (17,18).
66
Anastomotic leak / Diagnosis/clinical presentation 67
shown to reduce surgical site infection (SSI) following elec-

ANASTOMOTIC LEAK tive colorectal surgery, including anastomotic leak (40,41).
However, more recent trials and meta-analyses failed to
ETIOLOGIES demonstrate improvement in SSI or anastomotic leak with
MBP (42,43). Based on these studies, MBP fell out of favor
Traditionally, it was thought that the biggest threats to the in the last decade (44).
integrity of an anastomosis were tension and ischemia. Compared to the initial efficacy trials during the 1970s,
Vignali et al. demonstrated that local ischemia was asso- the majority of recent trials included an MBP without the
ciated with AL (19), which can occur through division of addition of oral antibiotics. In the absence of oral antibiot-
mesenteric arcades and stretching or twisting of mesenteric ics, a mechanical cleansing results in liquid, bacteria-laden
vessels (20–23). However, Shakhsheer et al. (4) randomized stool that is more likely to contaminate the operative field
rats to undergo sham operation, segmental colon devas- (45). In a 2009 Cochrane Review, Nelson et al. (46) con-
cularization alone, colectomy alone, or segmental devas- cluded that oral antibiotics combined with a mechanical
cularization plus colectomy. The rats were sacrificed on cleansing reduced the rate of SSI. In addition, data from
postoperative day 6 and the anastomoses examined. There large risk-adjusted national database studies including the
was no difference in tissue hypoxia noted in the rats with National Surgical Quality Improvement Program and the
poor healing compared to those with well-healed anasto- Michigan Surgical Quality Collaborative-Colectomy Best
moses, suggesting that ischemia may not be the most criti- Practices Project demonstrate significant reduction in infec-
cal factor in the developing anastomotic leak. Tension is tious morbidity, including anastomotic leak, with the use of
intuitively associated with AL given that most distal rectal an MBP when combined with oral antibiotics (47–50).
ALs are posterior, corresponding to the point of greatest Proximal diversion reduces the septic complications and
mesenteric tension. However, there is very little research the need for surgical intervention following anastomotic
to actually demonstrate the deleterious effect of tension leak. However, it has also been shown to reduce the rate of
in the development of leak. Other important factors such AL. Meta-analyses show decreased rate of AL and need for
as choosing correct staple height relative to tissue thick- reoperation with creation of a protective diverting stoma at
ness (24), avoiding multiple intersecting staple lines (25), the index operation (3,51,52).
and minimizing the number of linear stapler firings on Intraoperative inspection of the rectal anastomosis has
the distal rectum (26–29) appear to play a greater role in evolved over time with the advent of modern technology.
AL. Additionally, data continue to emerge citing the local The basic air leak test with submersion of the anastomosis
microbiome as a critical factor in the development of an under fluid as air is insufflated transanally has been shown
anastomotic leak (4,5). Matrix metalloproteinases (MMPs) to reduce anastomotic leak rates (53). Whether the addition
and collagenases are part of the body’s normal response of endoscopy to visually inspect the anastomosis during the
to injury. However, these proteins have been implicated as air leak test improves outcomes has not yet been formally
important mediators of anastomotic leak. Animal studies studied. The use of fluorescence angiography has recently
have suggested that certain bacterial strains that produce been postulated as a predictor of anastomotic vascularity
these collagenolytic proteins (Enterococcus, Pseudomonas, that could theoretically reduce the risk of anastomotic leak.
or Serratia species) may contribute to the development of Several studies have suggested that it does affect surgical
AL (30–32). decision-making, but as of yet, none has demonstrated dif-
With regard to patient-related factors, location of the ferences in AL rates (54,55). There is currently a multicenter
anastomosis is one of the most important risk factors for AL prospective randomized controlled trial underway assess-
with distal rectal anastomoses having almost a five times ing the role of indocyanine green angiography in rectal
increased risk compared to resection for colon cancer (15). resections for cancer (PILLAR III) (http://www.clinicaltri-
In addition, male gender (33–36), chemoradiation (35,36), als.gov: NCT02205307).
hypotension (37,38), diabetes, smoking, and atherosclero-
sis are known risk factors. Radiation seems to be timing DIAGNOSIS/CLINICAL PRESENTATION
related, with the most significant risk in the first 3 weeks
following completion of radiotherapy (39). Malnutrition The classic presentation of an AL includes abdominal pain,
(defined as 10% or greater unintentional weight loss) is asso- fever, tachycardia, and a leukocytosis in the setting of peri-
ciated with increased postoperative morbidity. It is unclear tonitis. However, AL can present along a continuum from
whether preoperative nutrition support can negate these low volume wound drainage to multisystem organ failure.
effects but it may help with surgical planning with regard to Leaks present at varying time periods according to etiology
the construction of an anastomosis. and severity. Minor leaks generally take longer to recognize
than major anastomotic dehiscences (56). Leaks attributed
PREVENTATIVE MEASURES to technical issues are more likely to occur within the first
48 hours (57). The majority of ALs are diagnosed between
Dating back to trials from the 1970s, mechanical bowel the 7th and 12th postoperative days (58). Up to 12% are
preparation (MBP) combined with oral antibiotics has been diagnosed 3 weeks after the index operation (58–60).
68 Postoperative anastomotic complications
Early recognition of anastomotic leak can be difficult DEFINITION

owing to poor specificity of the common signs and symp-
toms of AL, including abdominal pain and tachycardia. Colorectal surgery outcomes have become surrogate mark-
Certain biochemical markers have demonstrated promise ers of institutional quality owing to the fact that colorectal
in early detection. C-reactive protein (CRP) was shown to surgery is a ubiquitous procedure that is performed at the
have a negative predictive value of 89%–97% for AL, with majority of hospitals regardless of size, affiliation, and loca-
an area under the curve of 0.811 (61,62). Warschkow et al. tion. However, there has been a general lack of standardiza-
(62) used CRP level (<135) as discharge criteria. In the tion surrounding the definition of AL (69). Frequently, the
PREDICS study, procalcitonin combined with CRP was Centers for Disease Control and Prevention’s definition for
highly predictive (area under the curve of 0.9) of AL on organ space infection is used to define AL. However, data
postoperative day 5 (63). suggest that organ space infection is an unreliable marker
of AL based on a study by Rickels et al. (70) looking at the
IMAGING capture rates of anastomotic leak in both National Surgical
Quality Improvement Program (which uses organ space
CT is the most common test employed in the diagnosis of infection to define AL) and a statewide registry, the Upstate
AL (Figure 8.1). While the specificity of CT is quite high New York Surgical Quality Initiative (UNYSQI). Only
(>84%), the sensitivity is only 68%–71% (64,65). Contrast 25% of ALs following colorectal surgery were captured by
extravasation is the most sensitive and specific sign, but it National Surgical Quality Improvement Program using the
is unfortunately not always preset (64). The most common organ space infection definition (70).
findings are extraluminal air and para-anastomotic col- Several groups have produced standard definitions, clas-
lections with AL detection rates between 80% and 100% sifying leaks into one of three grades based on severity and
(66). The use of rectal contrast significantly increases the impact on clinical management (71,72):
positive predictive value from 40% to 88%, p = 0.0009.
This should be administered very carefully through a soft, ●● Grade A—asymptomatic or minor symptoms (for
small-bore catheter by an experienced member of the sur- instance found on contrast enema prior to ileostomy
gical team (67). reversal or a small fluid collection (<3 cm) adjacent
Postoperative pneumoperitoneum is a very common to the anastomosis on CT. May respond to antibiotics.
finding in the postoperative period in the absence of com- Does not require procedural intervention.
plications. It can be found in up to 44% of patients within ●● Grade B—may have clinical signs of leak such as leu-
the first 3–6 days after surgery, up to 30% in the second and kocytosis or fever but are nontoxic with normal vital
third weeks following surgery, and up to 13% in the fourth signs. On CT scan there is a localized abscess that seems
postoperative week (68). In addition, fluid and inflamma- amendable to percutaneous drainage.
tory stranding are nearly always present in an early post- ●● Grade C—toxic in appearance with evidence of sepsis.
operative CT scan. For this reason, AL can be difficult to Requires surgical intervention and often diversion.
diagnose on CT in the first 3–4 days following surgery.
MANAGEMENT
Whatever the approach, the key to avoiding multisystem
organ failure and death following anastomotic leak is the
ability to rescue the patient early in the course so as to avoid
the development of septic shock. This requires a high index
of suspicion and an aggressive approach toward the man-
agement of anastomotic complications. Early recognition
and intervention are key prior to development of significant
contamination and subsequent fibrosis that will lead to sep-
sis and complicate the operative management. Management
will depend on the clinical presentation and the severity of
symptoms (Figure 8.2).
Antibiotics
In general, abscesses less than 3 cm in size can be managed
with antibiotics alone when the patient is clinically stable
(73,74). Antibiotic regimens should cover gram-negative
Figure 8.1 Demonstrates contrast extravasation from a rods and anaerobic organisms. Possible choices include
left-sided anastomotic leak after administration of rectal a β-lactam/β-lactamase inhibitor (piperacillin-tazobac-
contrast. tam) or the combination of cefazolin and metronidazole.
Anastomotic leak / Management 69
Clinical presentation Management inflammation. Traditional management of an anastomotic

leak dictated either exteriorization of the leaking anasto-
Grade Asymptomatic or No intervention or
A mosis or resection of the anastomosis with the creation of
minimal symptoms antibiotic therapy
an end stoma and Hartmann pouch or mucus fistula (80).
Percutaneous drainage
While certainly effective at controlling the leak, dissection
Fever, leukocytosis,
Grade *possible endoscopic around the anastomosis can prove difficult with the risk of
B abdominal pain nontoxic,
normal vital signs
techniques (stent,
sponge, clip)
injury to surrounding structures. In addition, restoration of
intestinal continuity under these circumstances requires a
Grade Toxic appearance with Surgical intervention
subsequent major operation to reverse, and as a result, many
C evidence of sepsis (Resection vs. Diversion) end stomas are never reversed (81).
While resection of the anastomosis may seem desirable, it
is not always feasible and, under certain circumstances, may
Figure 8.2 Management of anastomotic leak according to be deleterious. Under these circumstances, a more desirable
severity and clinical presentation. option may be drainage (either operatively or percutaneously)
and proximal diversion. With diversion alone, less opera-
Traditionally, patients were treated with 7–10 days of anti- tive stress is placed on patients who are already critically ill.
biotics for intraabdominal infections. However, the Study This “damage control” philosophy has proven successful in
to Optimize Peritoneal Infection Therapy (STOP-IT) trial trauma situations and may be appropriate after the traumatic
demonstrated similar efficacy between patients with compli- insult of an anastomotic dehiscence (82). This technique is
cated intraabdominal infection who had achieved adequate safe, has a high rate of anastomotic salvage, and can often be
source control randomized to a fixed course of antibiotics (4 reversed through a small peristomal incision (83,84). If the
days) versus antibiotics until 2 days after resolution of clini- index operation was performed laparoscopically, diversion
cal signs of infection (up to a maximum of 10 days) (75). It and drainage may be performed laparoscopically as well (85).
is imperative to note that adequate source control is of the We published our own experience utilizing loop ileos-
upmost importance, and there should be a very low thresh- tomy with drainage as a treatment for patients with an
old to repeat imaging following a course of antibiotic ther- anastomotic leak resulting in peritonitis or abscess (86).
apy for treatment of a small contained AL to assure that the Our study included a variety of both intraperitoneal (85%)
inflammation has resolved. If it persists, therapeutic options and extraperitoneal (15%) colon and rectal anastomoses.
should be considered, including percutaneous drainage or Proximal diversion and drainage were successfully utilized
surgery. in 70% of patients requiring surgery for an anastomotic
complication. Ultimately, 63% of patients treated with prox-
Drain imal diversion had restoration of their intestinal continuity,
a percentage that is similar to that reported by Matthiessen
For abscesses greater than 3 cm in size, percutaneous drain- et al. (59) and Eckmann et al. (87) following diversion for
age is a viable option with success rates up to 81%. For pelvic anastomotic leaks. Of the patients who underwent fluoro-
collections, this often requires a transgluteal approach. It is scopic evaluation for possible takedown, 92% had normal
important that the entire collection be drained, as incom- healing of their anastomosis. The algorithm in Figure 8.3
plete drainage is associated with high rates of failure (76). In may assist with surgical decision-making.
addition to the size and resolution of the fluid collection, the Surgery is critical not only in the management of patients
quality of the drainage is also important. Pus is obviously with overt sepsis and peritonitis, but also for patients with
more amendable to percutaneous drainage than stool, which an intraabdominal abscess that does not promptly respond
results in a controlled colocutaneous fistula through the to catheter drainage.
drain. Rarely, this can lead to a severe necrotizing infection
along the course of the drain. My personal bias leans toward Endoscopic techniques
diversion rather than percutaneous drainage alone for fluid
collections in the setting of contrast extravasation or when Based on the success in esophageal leaks, the use of covered
frank stool is encountered upon percutaneous aspiration. stents as treatment for colorectal leaks has been described
Some advocate placing a transanal drain, such as a small in several small case series (88–90). Counterdrainage of
Malecot catheter, through the anastomotic defect. The drain the accompanying abscess is required. There are different
remains in place until the abscess cavity closes to the size types of stents including self-expanding metal and plastic
of the drain upon instillation of contrast through the drain stents. They may be covered or uncovered. Stent migra-
(77–79). tion is the major issue. The colorectal anastomosis must be
an end-to-end anastomosis, and the distal end of the stent
Surgery must be at least 5 cm above the anal verge. Therefore, stents
are not effective for low anastomoses. Many of the stents
The operative management of anastomotic leak can be are not large enough for colorectal anastomoses. Some
very challenging due to gross contamination, and severe have described the use of endoclips to secure the stent in
Anastomotic leak
(Toxic or failed
conservative management)
Surgical
intervention
Anastomosis
accessible
Yes No
Large Small Drain

dehiscence dehiscence and divert
Hartmann's Drain
procedure and divert
Figure 8.3 Treatment algorithm for surgical management of anastomotic leak.
place, but that has not proven particularly effective. When this technique seems better for small leaks less than 1.5 cm
deployed, the stents are generally left in place up to 60 days in size in the absence of a pelvic collection. It has also been
and then removed once the anastomosis heals. The partially used in conjunction with the transanal sponge technique
covered stents seem to have less migration than fully cov- once that cavity is small enough (101).
ered stents. This technology has been described in patients
with or without a diverting stoma (88–90).
Another innovative technique for the treatment of anas-
tomotic leaks is endoscopic transanal vacuum-assisted rectal ANASTOMOTIC LEAK WITH
drainage (91–94). It was initially reported by Weidenhagen CUTANEOUS FISTULA
and Arezzo in separate reports with up to 79% success rate
(91,95). The technique involves placement of a sponge via It is not uncommon for an AL to manifest as a colocutane-
an introducer fitted over an endoscope through the anasto- ous fistula. The management of the colocutaneous fistula can
motic defect and into the presacral space. The placement of globally be organized into three principle phases. The initial
the sponge is verified endoscopically. It is exchanged every phase is the acute stabilization of the patient from a meta-
48–72 hours and is downsized as the cavity decreases in bolic and sepsis standpoint in the first hours to days of pre-
size. Treatment is stopped when the cavity is less than 1 cm sentation. This is followed by the development of an interim
in size. It seems to be more effective if placed early when the plan for wound care and nutrition. The final phase of treat-
rectum is more pliable prior to the development of associ- ment is the definitive management or closure of the fistula.
ated fibrosis. It is important to note that a visible vessel in
the presacral space is a contraindication to treatment. In CONTROL OF SEPSIS
addition, it has only been used in very distal anastomoses
and in the setting of proximal diversion (83,91). It is impor- Patients present along a continuum of severity depending
tant to note that although this technique has been described on the volume of fistula output. Presentations range from a
for several years, it has yet to become widely adopted. localized skin infection with a small underlying low-output
Endoscopic clip application as a method to reapproxi- fistula to septic shock with profound dehydration in patients
mate the anastomotic dehiscence has also been described with a proximal high-output fistula. Treatment of the fistula
(96–100). Although the data are limited to small case series, patient must be customized accordingly. In general, most
Anastomotic stricture 71
fistulas resulting from colorectal surgery are low output fis- significant advantages. Patients with low-volume (<200 cc/
tulas that can be supported conservatively. day) fistulas are unlikely to suffer electrolyte imbalance
Source control is imperative to prevent intraabdominal and volume depletion. Similarly, wound care is simpli-
sepsis. All patients who are suspected as having a fistula fied by effluent volume reduction. The majority of patients
should undergo CT. These patients do not always mani- with low-output fistulas can tolerate enteral nutrition (102).
fest the usual signs of infection. Rather, they may present Therefore, significant effort should focus on antimotility
with weight loss, hypoalbuminemia, or jaundice (102). therapy, including treatment with loperamide, diphenox-
Any intraabdominal fluid collections must be adequately ylate and atropine, tincture of opium, codeine, cholestyr-
drained either percutaneously or surgically. Contrast amine, and proton pump inhibitors.
injected into the cavity can give information regarding the
size of the cavity and identify the location of the fistulous DEFINITIVE MANAGEMENT
communication (102). At this point, the decision must be
made whether to operate or manage the patient nonopera- Once the patient has been stabilized with source control,
tively. This will depend on the timing of the leak. In gen- wound care, and nutritional support, a plan for definitive
eral, this is best managed operatively within the first 7–10 management of the fistula is imperative. Many low-volume
days postoperatively. After this, the decision to operate will colocutaneous fistula will close spontaneously over time
depend on the clinical presentation. Colocutaneous fistulas in the absence of complicating factors such as radiation or
representing anastomotic leak may require proximal diver- a distal obstruction. Therefore, patience is of the upmost
sion if possible to facilitate closure. importance in this situation. Once it has been determined
that surgery will be necessary, most surgeons wait at least
WOUND CARE 4–6 months from the inciting event. Definitive surgery is
fraught with difficulty owing to prior inflammation and the
The fistula must be isolated from surrounding tissue to common association with loss of abdominal wall domain
allow for nursing care, protection of the skin, or healing of (104). Tedious dissection upon entry to the abdomen is criti-
the surrounding open wound. This also allows for quantifi- cal to prevent inadvertent enterotomies, which can be aided
cation of the effluent, which is critical in the early stages of by using an alternative incision if possible or extending the
presentation for adequate replacement of the calculated vol- incision superior to the prior incision into virgin territory. In
ume loss. A stoma appliance can often be used to pouch most cases of a large abdominal wall defect, component separation
fistulas surrounded by intact skin. However, the enteroat- without or without biologic mesh is often necessary (104,105).
mospheric fistula can be extremely challenging to isolate.
A multidisciplinary team consisting of wound-care special-
ists, enterostomal therapists, and physicians is required to
identify an effective wound-care system. Various strategies ANASTOMOTIC STRICTURE
have been employed including wound-care managers, simi-
lar to very large stoma appliances that encompass the entire Anastomotic stricture is a common reason for failure of
wound bed. The fistula effluent is isolated with closed suc- stoma reversal after low anterior resection with diverting
tion drains, and the surrounding wound is managed with a loop ileostomy (106). The incidence of stricture has been
wet to dry dressing. reported in up to 15% of the patients following low anterior
The vacuum suction wound management system is uti- resection with diverting loop ileostomy (107). In most cases,
lized by many to isolate the fistula within the open wound, anastomotic strictures develop as the result of an anasto-
allowing for pouching of the fistula. However, caution must motic leak, which may or may not be clinically appreci-
be used as the development of an enterocutaneous fistula ated at the original operation. Therefore, the same general
has been described as a complication of wound vacuum principles for prevention and diagnosis apply. Anastomotic
management system in patients with open abdomens (103). stricture related to Crohn disease represents an entirely dif-
If vacuum-assisted therapy is utilized within an open ferent entity than sporadic postoperative strictures and is
intraabdominal wound, caution is advised, and precautions beyond the scope of this manuscript.
should be taken to protect the underlying bowel. Many strictures are found at the time of contrast enema
Regardless of the interim wound management scheme, in anticipation of ostomy reversal. Other patients will
once the adjacent tissue within the surrounding open have a delayed presentation, which may include changes
wound is granulated, skin grafting can be performed to in bowel habits, crampy abdominal pain, constipation or
allow for placement of a stoma appliance. diarrhea, incontinence, and obstipation. The presentation
can be insidious and surgeons must have a high index of
ANTIMOTILITY AGENTS AND suspicion to facilitate diagnosis and intervention prior to a
OCTREOTIDE complete obstruction. Diagnosis can be made endoscopi-
cally or radiographically on a contrast enema or CT scan
Volume reduction of effluent is not likely to facilitate spon- with rectal contrast. Management depends on the severity
taneous closure of the fistula. However, it does provide other of the stricture. Mild strictures, particularly on the right
side, where the stool is liquid in character, may be success- 2. Isbister WH. ANZ J Surg 2001;71(9):516–20.
fully managed with a low-residue diet. Balloon dilation is an 3. Matthiessen P et al. Colorectal Dis 2004;6(6):462–9.
effective solution (108). Biraima et al. (108) report a single 4. Shakhsheer BA et al. Int J Colorectal Dis 2017;32(4):
institution study of 76 patients over nearly 20 years with 539–47.
anastomotic stenosis. All patients were treated initially with 5. Krezalek MA, Alverdy JC. Curr Opin Clin Nutr Metab
balloon dilation. Balloon dilation was successful in 97% Care 2016. 23 [Epub ahead of print]
of the patients. The median number of dilations required 6. Turrentine FE et al. J Am Coll Surg 2015;220(2):
was three. Recurrences were 11%, 11%, and 25% at 1, 3, and 195–206.
5 years, respectively. Balloon dilation has also been com- 7. Phitayakorn R et al. World J Surg 2008;32(6):1147–56.
bined with endoscopic incision of the fibrotic anastomotic 8. Hammond J et al. J Gastrointest Surg 2014;18(6):
tissue with successful results in small case series (109,110). 1176–85.
Consideration can also be given to self-expanding metallic 9. Ashburn JH et al. Dis Colon Rectum 2013;56(3):
stents (90). Strictures that are refractory to balloon dilation 275–80.
require a surgical approach with revision of the anastomo- 10. Hallbook O, Sjodahl R. Br J Surg 1996;83(1):60–2.
sis, if possible (111). If an anastomosis cannot be salvaged, 11. Marinatou A et al. Dis Colon Rectum 2014;57(2):
the decision must be made whether to resect the anastomo- 158–66.
sis and bring up an end stoma or retain the anastomosis and 12. Nesbakken A et al. Br J Surg 2001;88(3):400–4.
simply divert above it. This decision obviously depends on 13. McArdle CS et al. Br J Surg 2005;92(9):1150–4.
each individual situation, and resection of the anastomosis 14. Smith JD et al. Ann Surg Oncol 2013;20(8):2641–6.
will prove impossible for some patients. However, if possi- 15. Mirnezami A et al. Ann Surg 2011;253(5):890–9.
ble, patients generally have fewer subsequent complications 16. Lu, ZR et al. Dis Colon Rectum 2016;59(3):236–44.
if the anastomosis can be resected and an end stoma can be 17. Espin E et al. Br J Surg 2015;102(4):416–22.
created. 18. Smith JD et al. Ann Surg 2012;256(6):1034–8.
19. Vignali A et al. Dis Colon Rectum 2000;43(1):76–82.
20. Boyle NH et al. J Am Coll Surg 2000;191(5):504–10.
21. Kashiwagi H. Surg Today 1993;23(5):430–8.
CHRONIC ANASTOMOTIC SINUS 22. Konishi T et al. J Am Coll Surg 2006;202(3):439–44.
23. Sheridan WG, Lowndes RH et al. Dis Colon Rectum
Chronic sinuses in low rectal anastomoses have been 1987;30(11):867–71.
reported in up to 36% of anastomotic leaks (112). Of these, 24. Chekan E, Whelan RL. Med Devices (Auckl) 2014;7:
8% are asymptomatic, found at contrast enema in evalua- 305–18.
tion for ileostomy takedown (113). These chronic sinuses 25. Zilling T, Walther BS. Dis Colon Rectum 1992;35(9):
represent a frustrating situation for both patients and sur- 892–6.
geons alike. Some will close spontaneously with the elapse 26. Ito M et al. Int J Colorectal Dis 2008;23(7):703–7.
of additional time. However, they can adversely affect 27. Kawada K et al. Surg Endosc 2014;28(10):2988–95.
bowel function due to fibrosis with reduction in com- 28. Park JS et al. Ann Surg 2013;257(4):665–71.
pliance (114). For patients with sinuses that do not heal 29. Qu H, Liu Y et al. Surg Endosc. 2015;29(12):3608–17.
spontaneously, many will result in the patient having a 30. Shogan BD et al. Surg Infect (Larchmt) 2014;15(5):
permanent stoma. 479–89.
Some advocate for early intervention to prevent the 31. Shogan BD et al. Sci Transl Med 2015;7(286):286ra68.
development of a chronic cavity with resulting fibrosis 32. Shogan BD et al. J Gastrointest Surg 2013;17(9):
(115). Marsupialization of the sinus has been reported using 1698–707.
either an endoscopic stapler or cautery. Theoretically, this 33. Kirchhoff P et al. Patient Saf Surg 2010;4(1):5.
allows complete drainage of the cavity and incorporation 34. Lipska MA et al. ANZ J Surg 2006;76(7):579–85.
of the sinus tract into the lumen of the bowel. Subsequently, 35. Midura EF et al. Dis Colon Rectum 2015;58(3):333–8.
with epithelialization of the cavity, the stoma can then the- 36. Trencheva K et al. Ann Surg 2013;257(1):108–13.
oretically be reversed (116,117). Transanal repair with flap 37. Kologlu M et al. Surgery 2000;128(1):99–104.
closure via an endorectal advancement flap has also been 38. Posma LA et al. Dis Colon Rectum 2007;50(7):1070–9.
described (118,119). If local measures fail, revising the anas- 39. Pettersson D et al. Br J Surg 2010;97(4):580–7.
tomosis is an option. However, this can be a technically 40. Nichols RL et al. Ann Surg 1973;178(4):453–62.
challenging operation with high rates of recurrent anasto- 41. Clarke JS et al. Ann Surg 1977;186(3):251–9.
motic problems (120). 42. Dahabreh IJ et al. Dis Colon Rectum 2015;58(7):
698–707.
REFERENCES 43. Rollins KE et al. World J Gastroenterol 2018;24(4):
519–36.
1. Demetriades H et al. Tech Coloproctol 2004;8(Suppl 44. Gustafsson UO et al. Clin Nutr 2012;31(6):783–800.
1):s72–5. 45. Mahajna A et al. Dis Colon Rectum 2005;48(8):1626–31.
References 73
46. Nelson RL et al. Cochrane Database Syst Rev. 83. Blumetti J, Abcarian H. World J Gastrointest Surg
2009;(1):CD001181. DOI: 10.1002/14651858. 2015;7(12):378–83.
CD001181.pub3 84. Longo J et al. J Contin Educ Nurs 2011;42(1):27–35.
47. Englesbe MJ et al. Ann Surg 2010;252(3):514–9, dis- 85. Joh YG et al. Dis Colon Rectum 2009;52(1):91–6.
cussion 519–20. 86. Hedrick TL et al. Dis Colon Rectum 2006;49(8):
48. Kim EK et al. Ann Surg 2014;259(2):310–4. 1167–76.
49. Moghadamyeghaneh Z et al. J Am Coll Surg 2015; 87. Eckmann C et al. Int J Colorectal Dis 2004;19(2):
220(5):912–20. 128–33.
50. Morris MS et al. Ann Surg 2015;261(6):1034–40. 88. Abbas MA. JSLS 2009;13(3):420–4.
51. Chude GG et al. Hepatogastroenterology 2008; 89. Lamazza A et al. Endoscopy 2013;45(6):493–5.
55(86–87):1562–7. 90. Lamazza A et al. Am J Surg 2014;208(3):465–9.
52. Huser N et al. Ann Surg 2008;248(1):52–60. 91. Arezzo A et al. Dig Liver Dis 2015;47(4):342–5.
53. Nachiappan S et al. Surg Endosc 2014;28(9):2513–30. 92. Glitsch A et al. Endoscopy 2008;40(3):192–9.
54. Kin C et al. Dis Colon Rectum 2015;58(6):582–7. 93. Kuehn F et al. J Gastrointest Surg 2016;20(2):237–43.
55. Jafari MD et al. J Am Coll Surg 2015;220(1):82–92 e1. 94. Smallwood NR et al. Surg Endosc 2016;30(6):
56. Paliogiannis P et al. Ann Ital Chir 2012;83(1):25–8. 2473–80.
57. Baker RS et al. Obes Surg 2004;14(10):1290–8. 95. Weidenhagen R et al. Surg Endosc 2008;22(8):
58. Hyman N et al. Ann Surg 2007;245(2):254–8. 1818–25.
59. Matthiessen P et al. Ann Surg 2007;246(2):207–14. 96. Mennigen R et al. World J Gastroenterol 2014;20(24):
60. Morks AN et al. Colorectal Dis 2013;15(5):e271–5. 7767–76.
61. Singh PP et al. Br J Surg 2014;101(4):339–46. 97. Mizrahi I et al. J Gastrointest Surg 2016;20(12):
62. Warschkow R et al. Ann Surg 2012;256(2):245–50. 1942–9.
63. Giaccaglia V et al. Ann Surg 2016;263(5):967–72. 98. Sulz MC et al. World J Gastroenterol 2014;20(43):
64. Kauv P et al. Eur Radiol 2015;25(12):3543–51. 16287–92.
65. Kornmann VN et al. Int J Colorectal Dis 2013;28(4): 99. Weiland T et al. Surg Endosc 2013;27(7):2258–74.
437–45. 100. Brunner W et al. Surg Endosc 2015;29(12):3803–5.
66. Daams F et al. World J Gastrointest Surg 2014;6(2): 101. Chopra SS et al. Surgery. 2009;145(2):182–8.
14–26. 102. Joyce MR, Dietz DW. Curr Probl Surg 2009;46(5):
67. Habib K et al. Int J Colorectal Dis 2015;30(8):1007–14. 384–430.
68. Chapman BC et al. J Surg Res 2015;197(1):107–11. 103. Bee TK et al. J Trauma 2008;65(2):337–42; discussion
69. Bruce J et al. Br J Surg 2001;88(9):1157–68. 342–4.
70. Rickles AS et al. Surgery 2013;154(4):680–7; discus- 104. Hodgkinson JD et al. Colorectal Dis 2017;19(4):
sion 687–9. 319–30.
71. Chadi SA et al. J Gastrointest Surg 2016;20(12): 105. Atema JJ et al. World J Surg. 2017;41(8):1993–9.
2035–51. 106. Haksal M et al. Ann Surg Treat Res 2017;92(1):35–41.
72. Peel AL, Taylor EW. Ann R Coll Surg Engl. 1991;73(6): 107. Kim MJ et al. Surgery 2016;159(3):721–7.
385–8. 108. Biraima M et al. Surg Endosc 2016;30(10):4432–7.
73. Elagili F et al. Dis Colon Rectum 2014;57(3):331–6. 109. Truong S et al. Endoscopy 1997;29(9):845–9.
74. Siewert B et al. AJR Am J Roentgenol 2006;186(3): 110. Tan Y et al. Int J Colorectal Dis 2016;31(5):1063–4.
680–6. 111. Maggiori L et al. Int J Colorectal Dis 2015;30(4):
75. Sawyer RG et al. N Engl J Med 2015;372(21): 543–8.
1996–2005. 112. van Koperen PJ et al. Colorectal Dis 2011;13(1):26–9.
76. Khurrum Baig M et al. Tech Coloproctol 2002;6(3): 113. Blumetti J et al. World J Surg 2014;38(4):985–91.
159–64. 114. Blumetti J et al. Colorectal Dis 2012;14(10):1238–41.
77. Thomas MS, Margolin DA. Clin Colon Rectal Surg. 115. Verlaan T et al. Colorectal Dis 2011;13(Suppl 7):18–22.
2016;29(2):138–44. 116. Stewart BT, Stitz RW. Dis Colon Rectum 1999;42(2):
78. Sirois-Giguere E et al. Dis Colon Rectum 2013;56(5): 264–5.
586–92. 117. Whitlow CB et al. Dis Colon Rectum 1997;40(7):
79. Thorson AG, Thompson JS. Dis Colon Rectum. 1984; 760–3.
27(7):492–4. 118. Fleshman JW et al. Int J Colorectal Dis 1988;3(3):
80. Alves A et al. J Am Coll Surg 1999;189(6):554–9. 161–5.
81. Parc Y et al. Dis Colon Rectum 2000;43(5):579–87; 119. Wexner SD et al. Dis Colon Rectum 1989;32(6):
discussion 587–9. 460–5.
82. Parr MJ, Alabdi T. Injury 2004;35(7):713–22. 120. Genser L et al. Dis Colon Rectum 2013;5:747–55.
9
General postoperative complications
J. MARCUS DOWNS AND KRISTINA K. BOOTH
CHALLENGING CASE with biologic mesh underlay. He recovered unevent-

fully, going home on his fifth postoperative day.
A 56-year-old man underwent surgery 3 years ago for He returned 2 days later with severe pain, nausea,
perforated diverticulitis, when a Hartmann procedure and vomiting with imaging suggesting small bowel
was performed. Six months later, he underwent colos- obstruction. Initial treatment with nasogastric suction
tomy closure. This operation was complicated by a failed to improve his pain, and he was returned to the
ureteral injury at the pelvic brim, which was repaired operating room the next day where he was found to
primarily over a double-J stent. Over the past year, he have obstruction related to an internal hernia defect
has had three admissions for small bowel obstruction, caused by his remaining momentum. This was cor-
which quickly resolved with nasogastric suction and rected, and the abdominal wall once again recon-
IV fluids. He has multiple hernia defects at his midline structed. His kidney has been left to involute.
wound, as well as a hernia at his colostomy site. He This patient represents a spectrum of postopera-
weighs 110 kg, and prior to this he had no medical tive complications commonly related to colon sur-
problems. He also has left hydronephrosis, which is gery: hernia, bowel obstruction, and ureteral injury.
on the side of his ureteral injury. He also represents the unfortunate truth that not all
repairs are successful.
CASE MANAGEMENT
This patient required surgery for his recurring small

INTRODUCTION
bowel obstruction felt to be related to his multiple
hernias. After induction of anesthesia, a cystoscopy With surgery too often come complications. As surgeons, we
was performed, and ureteral stent placement was decide upon whom to operate. We prepare our patients for sur-
attempted. This was unsuccessful, and retrograde ure- gery, operate, and care for our patients through recovery. Our
terogram showed a completely obstructed ureter. The approach to complications should be to minimize their occur-
planned hernia surgery was not performed, and he rence, recognize their presence, and optimize their treatment.
was awoken. The next day, he underwent a percutane- In this chapter, we discuss the postoperative problems of pain,
ous nephrostomy and ureterogram. The ureterogram bleeding, infection, deep vein thrombosis, nausea and vom-
showed only the proximal two-thirds of his ureter to iting, prolonged ileus, injury to normal organs, and retained
be patent. With the nephrostomy tube in place, it foreign bodies. In Chapter 10, enhanced recovery after surgery
was determined that the left kidney function was only (ERAS) protocols are presented. The elements of ERAS proto-
8% of normal. With this finding, ureteral reconstruc- cols typically include preoperative, intraoperative, and post-
tion was felt not to be indicated. He returned to the operative components. Often added is an additional bundle of
operating room 3 weeks later for hernia repair. At elements specifically addressing surgical site infections (SSIs).
that time, he was found to have a simple hernia at his As we discuss complications in this chapter, we break the dis-
stoma site, which was easily reduced, and multiple cussion of each complication into the same time frames.
midline defects, two of which had incarcerated loops Surgical outcomes are measured incessantly. Each of us
of small bowel. The hernias were repaired primarily receives quarterly, if not monthly, reports on our length of
stay, surgical volume, and SSIs. Patients having complications
74
Bleeding 75
have prolonged lengths of stay (1,2). We have all seen our complications and decrease the need for opioids postop-
own numbers go from high to low, from one quarter to the eratively. Minimizing incision length is one aspect that can
next, with no change in our treatments. Rather than using decrease trauma and thus decrease pain. This is an impor-
the following information to improve our averages, search tant benefit of minimally invasive surgery, which should be
each section for ways to improve the care of our next patient. pursued if possible. Injection of local anesthetic at the site
of port placement prior to incision is helpful. Intraoperative
lidocaine infusion has been shown to reduce postoperative
narcotic requirements. This has been demonstrated in both
PAIN MANAGEMENT open and laparoscopic colectomy (14,15). It is unclear what
the optimal duration of therapy should be (16).
Adequate pain control during the postoperative time frame Many conflicting reports have been presented surround-
can be both elusive to the patient and frustrating to the pro- ing various tools used to provide analgesia (13). Continuous
vider. Lack of pain control can lead to complications related preperitoneal analgesia was compared directly to epidural
to poor mobility, such as atelectasis and deep vein throm- analgesia and found inferior in many aspects, including pain
bosis (DVT); however, liberal use of opioids for pain control scores, length of hospital stay, and functional recovery for
also leads to respiratory depression, ileus, and constipation open surgeries (17). However, the benefits of epidural use in
(3). Each patient interprets pain differently and can enter laparoscopic surgery appear to be minimal and may actually
the clinical encounter with varying expectations on pain slow recovery. Patient-controlled analgesia (PCA) has been
management. Providers must balance easier opioid-based compared to epidurals with conflicting results (13,18). PCA
pain control options with their potential side effects includ- has the benefits of self-titration, immediate action, and high
ing prolonged length of stay. In addition to shorter lengths patient satisfaction scores, but the downside includes the sys-
of stay, reducing opioid prescriptions can prevent their temic effects of the opioid medication. Due to their localized
diversion and reduce narcotic abuse (4). While an opioid- site of action, epidurals do omit the effects of system opioids
free recovery may not be possible in all patients, a narcotic- from the equation. However, epidurals have a failure rate of
sparing regimen can provide good pain control and reduce about 12%, can induce hypotension, can add potential pro-
length of stay as part of enhanced recovery pathways (5–7). cedural-related complications, and can even delay discharge
Treating pain begins before surgery. The first preopera- in some instances (13,18). The use of epidurals in minimally
tive task is patient education. Patient’s expectations should invasive surgery does not seem to be justified given the risk
be addressed, and information provided. Patients should profile and limited benefit. The use of epidurals in laparo-
be coached that we want and expect them to be comfort- scopic colorectal surgery within an enhanced recovery path-
able and to be mobile, but that pain control does not mean way appears to prolong length of stay compared to PCA use
being pain free. Emphasis on preemptive pain control has (18). The use of transverses abdominis plane (TAP) blocks
mitigated the use of excess opioids in postoperative care has gained popularity as a reasonable alternative to epidur-
by providing a prophylactic barrier for painful stimuli. als and the cumbersome preperitoneal analgesia. TAP blocks
A multimodal approach can be started in preoperative care have been shown to reduce pain and length of stay (19,20).
and continued postoperative for a well-rounded strategy. The addition of a TAP block to an enhanced recovery path-
Nonsteroidal anti-inflammatory drugs and cyclooxygen- way results in further reduction in length of stay and can be
ase-2 (COX-2) inhibitors can be used preoperatively and delivered under laparoscopic guidance (19).
carried throughout the patient’s recovery. Care should be Ketamine is a dissociative anesthetic agent. It has also
taken to ensure adequate kidney function. There has been been used in postoperative pain management (21). Ketamine
some hesitance in their use due to concerns of bleeding, but has also been incorporated into ERAS pathways following
this seems not to be the case (8). There may be an association cholecystectomy (22).
between Ketorolac use and anastomotic leak, particularly in Adequate pain control is essential for optimizing patient
nonelective cases (9,10). In addition, Etoricoxib, a COX-2 outcome. Optimal pain control in colorectal surgery
inhibitor, has been associated with risk of complications requires a multiple modality approach and should minimize
after colorectal surgery (9). Gabapentin, used preopera- narcotic use. The optimal pharmacologic cocktail has yet
tively and scheduled postoperatively, can reduce opioid use to be defined. The addition of local anesthetics epidurally,
and improve analgesia (11). Gabapentin may be associated preperitoneally, locally, or via TAP will all reduce narcotic
with respiratory depression, and so intraoperative narcot- requirements, though epidural use may prolong length of
ics should be reduced. This problem is more prominent in stay. Additional information is presented in Chapter 10.
elderly patients (11,12). In patients with normal liver func-
tion, preemptive and scheduled acetaminophen can be most
helpful in decreasing pain levels during recovery (13). The
use of these medications appears to be synergistic, as they
BLEEDING
each contribute a different mechanism of pain control.
A team-based approach to minimizing opioid use is Minimizing bleeding complications begins well before
paramount. This again serves to decrease opioid-induced surgery. A thorough history and physical exam must be
76 General postoperative complications
completed on every patient. This includes specific inquiry such bleeding occurs, it is important to apply pressure, and
into the patient’s history of bleeding, easy bruising, known pause. This is the time to give anesthesia time to “catch up”
coagulopathy, medications expected to impair hemostasis, with fluids. This is not because the patient was behind in flu-
and family history of bleeding disorders. The patient evalu- ids, but that the occurrence of the bleeding often results in a
ation is then looked at in the context of the proposed surgi- massive, sudden blood loss. Now is the time to obtain blood,
cal procedure and the expected bleeding risk. Patients with if already crossed, or to send blood for crossmatch, if this
no identified risk factors undergoing procedures with low has not already been done. It is also a time to ensure needed
bleeding risks, such as colonoscopy or anorectal surgery, materials are in the room. Techniques often used include
require no laboratory evaluation. Low-dose aspirin may be electrocautery, topical hemostatic agents, sacral thumbtacks,
continued. and muscle fragment welding (Figure 9.1). Also reported is
Patients suspected of increased bleeding risk should have the use of tissue expanders, cyanoacrylate adhesives, and
preoperative lab testing to include complete blood count, endoscopic tacking devices. When these fail, or coagulopa-
platelet count, prothrombin time, partial thromboplastin thy occurs, the pelvis may be packed, the wound temporarily
time, and international normalized ratio. Abnormalities closed, and the patient taken to the intensive care unit for
that cannot be explained or corrected should trigger for- resuscitation, rewarming, and correction of coagulopathy.
mal hematologic evaluation, as should known or suspected The patient is then returned to the operating room in 24–48
coagulopathies. Anticoagulants should be stopped prior to hours, by which time the bleeding has usually ceased.
surgery. If this cannot be safely done, switching to heparin or The second area at risk of bleeding in colon surgery is the
enoxaparin prior to surgery and holding it the day of the pro- spleen. Meticulous dissection technique while mobilizing
cedure should be considered. Such decisions should be made the splenic flexure is the best way to avoid this complica-
in conjunction with the patient’s primary care physician. In tion. A useful technique in mobilizing this flexure is to initi-
short, coagulopathies should be corrected prior to surgery. ate the dissection medially at the ligament of Treitz, divide
Abdominal operations carry a higher risk of bleeding the inferior mesenteric vein, develop the plane posterior
than anorectal or endoscopic procedures. Each patient to the mesocolon out to the abdominal wall, and cephalad
should be typed and screened. Should problematic bleed- to the pancreas. This maneuver eases the identification of
ing occur, the blood bank then already has a sample for the proper plane as the peritoneal reflection and lienocolic
crossmatching. A positive antibody screen suggests that ligament are divided.
crossmatching blood will be more difficult, and should be The third area at risk of bleeding is at the base of the
followed with a discussion with the blood bank to ensure middle colic vessels, where anatomic variants can present
blood is available, if needed. Of course, not every abdominal increased bleeding risk, especially as the hepatic flexure is
operation carries the same bleeding risk. Patients anemic
to begin with, or when significant blood loss is expected,
may require crossmatching prior to surgery. Massive blood
loss may lead to coagulopathy and the need for blood com-
ponent therapy. Many hospital blood banks have treatment
protocols when massive transfusion is required (23). Work
with your blood bank to establish one if none yet exists at
your hospital.
While all benefit from the same preoperative evaluation,
endoscopic, abdominal, and anorectal cases present differ-
ing challenges with regard to intraoperative bleeding. It is
clear that increased transfusion leads to worse outcomes in
patients with colorectal cancer (24). The unique challenge
posed in endoscopy is the inability to apply direct pressure
to bleeding points. The tools available to control bleeding
endoscopically include injection of epinephrine, electro-
cautery, and application of hemostatic clips. Electrocautery
may be either monopolar or bipolar. Monopolar electro-
cautery should be used judiciously due to concern for full-
thickness injury. This risk is lessened with bipolar cautery
(Bicap). Clips are particularly useful when larger polyps are
removed.
Intraabdominal surgery presents several areas of bleeding
risks of particular importance in colorectal surgery. These
are pelvic bleeding, the spleen, and bleeding associated with
the middle colic vessels. Life-threatening bleeding in the Figure 9.1 Thumbtack occlusion of a bleeding basiverte-
pelvis is usually due to injury to the presacral veins. When bral vein.
Infection 77
mobilized medially. Bleeding here usually requires suture broken down into preoperative, intraoperative, and postop-
ligation, taking care to protect the superior mesenteric vein. erative elements.
Control of bleeding in anorectal surgery is rarely prob- One immediate problem is that of defining when SSI
lematic intraoperatively unless the patient has portal hyper- occurs. When surgeons use Centers for Disease Control
tension and/or a coagulopathy. These should be known and Prevention criteria to find infection, there is signifi-
before surgery as their presence may modify the operation cant interobserver variability (26). In another study, wound
performed. Patients with advanced liver disease or coagu- photographs were examined to determine infection rates.
lopathy and bleeding hemorrhoids should be treated with Experienced surgeons found rates ranging from 6.2% to
suture ligation rather than excisional hemorrhoidectomy to 14% looking at the same wounds (27). They recommend
minimize bleeding complications. using the ASEPSIS system for wound evaluation (28). The
Management of postoperative bleeding differs by type ASEPSIS score awards points for the presence of exudate,
of surgery. Bleeding after endoscopic procedures should erythema, purulent exudate, tissue separation, treatment
be managed initially as any lower gastrointestinal bleed. with antibiotics, incision and drainage, debridement, and
Ongoing bleeding demands repeating colonoscopy. isolation of bacteria to define wound infection (Table 9.1).
Reprepping is usually not required. The bleeding source is Using this tool, there was 96% agreement in defining infec-
usually apparent and controlled with application of hemo- tion (27,29).
static clips. If the site is found but covered with a blood clot, SSIs in colorectal surgery also vary by diagnosis. In a
the clot should be irrigated away to reveal the underlying National Surgical Quality Improvement Program study of
bleeding site, which should then be clipped. Rectal bleed- almost 25,000 colorectal operations, Pendlimari et al. found
ing following colectomy should be treated similarly, with infection rates for colorectal surgery ranging from 8.9% for
colonoscopy performed liberally. The two most common surgery for benign colon neoplasms to 17% for rectal cancer
sources of bleeding are at the anastomosis and ischemic surgery (30). In this study the rate of organ space infections
colitis. varied from 2.5%, also for benign colon neoplasms, to 7.7%
Postoperative bleeding in the abdomen often requires for ulcerative colitis. Operations involving additional organ
reexploration and evacuation of clotted blood. Often no resections increase the incidence of SSIs (31).
discrete bleeding site is found. Bleeding in the pelvis after Patients bring to us not only their colorectal problems,
a difficult dissection may be addressed by angiographic but their other health problems as well. Often these can-
embolization. not be modified. In the case of urgent surgery, little can be
Some bleeding after anorectal surgery is expected. changed. With the luxury of time, patients can be prepared
Excessive bleeding demands evaluation. This can be done for surgery. Three areas deserve particular attention: glu-
with either a proctoscope or anoscope, and irrigation of cose control, remote infections, and smoking cessation.
blood from the rectum. If a bleeding site is identified, it can Diabetic patients have a higher rate of SSIs than non-
be addressed with injection of epinephrine or oversewing. diabetics undergoing colorectal surgery. Ata et al. using
This may require returning to the operating room. National Surgical Quality Improvement Program data from
Patient outcomes are improved by reducing bleeding 2005–6, showed rates of 15.4% in diabetics compared to
complications. This is accomplished by the combination of 11% in nondiabetics (32). Improved glucose control results
a thorough preoperative evaluation, careful surgical tech- in fewer infectious complications. This can be assessed pre-
nique, and recognition and treatment when these complica- operatively by measuring hemoglobin A1C. If over 8% on
tions occur. preoperative lab testing, it can be improved with several
weeks of focused therapy, possibly improving postoperative
complications. Postoperative glucose levels should be kept
under 140 mg/dL. One mechanism of increased wound
INFECTION infections may be seen at the molecular level, where tissue
fluid sampled from surgical wounds showed reduced levels
Infection is an embedded element of colorectal surgery. of urokinase-type plasminogen activator as well as its recep-
While clean surgery regularly reports infection rates around tor in diabetic patients. Plasminogen helps wound healing
1%, colorectal surgery regularly experiences numbers closer by proteolytic degradation of extracellular matrices (33).
to 10%. One cannot operate in this field without an extensive Perioperative glycemic control requires modification
knowledge of surgical infections. If we are to improve our of regular medications. Glucose should be checked three
patient’s outcome, we must improve this dimension of care. times a day the day prior to surgery. Clear liquids contain-
Infections account for approximately one million hospi- ing sugar should be avoided. Basal insulin dosage should
tal days per year, and add $1.6 billion in direct costs (25). be reduced by two-thirds on the day prior to surgery. Stop
Improving patient outcomes in colon and rectal surgery mealtime insulin the day before and the day of the proce-
demands every effort to keep postoperative infections to a dure. Patients taking Metformin should continue this the
minimum. In this section, we discuss SSIs associated with day before and the day of surgery. Those taking sulfonylurea
abdominal surgery. Pneumonia and urinary tract infec- should stop this the day before and the day of the procedure.
tions are dealt with separately. As before, this discussion is Patients taking Humulin 70/30 should take one-half their
Table 9.1 ASEPSIS score
ASEPSIS wound score Proportion of wound affected

Wound characteristic 0 <20 20–39 40–59 60–79 >80
Serous exudate 0 1 2 3 4 5
Erythema 0 1 2 3 4 5
Purulent exudate 0 2 4 6 8 10
Separation of deep tissues 0 2 4 6 8 10
Points are scored for daily wound inspection.
Criterion Points
Additional treatment: Antibiotics 10
Drainage of pus under local anaesthesia 5
Debridement of wound (general anaesthesia) 10
Serous discharge* daily 0–5
Erythema* daily 0–5
Purulent exudate* daily 0–10
Separation of deep tissues* daily 0–10
Isolation of bacteria 10
Stay as inpatient prolonged over 14 days 5
Note: Category of infection: total score 0–10 = satisfactory healing; 11–20 = disturbance ot healing; 20–30 = minor wound infection;
31–40 = moderate wound infection; >40 = severe wound infection.
* Given score only on five of seven days. Highest weekly score used
normal dose the day prior to surgery. Carbohydrate loading (40). Nevertheless, oral antibiotic use declined. Two factors
as done in many enhanced recovery pathways should not be contributing to this decline were the nausea experienced by
done in diabetics (34). some patients with the antibiotics, leading to the substitu-
Remote infections increase the risk of SSIs and should tion of metronidazole for erythromycin, and to the favor-
be resolved prior to surgery, if possible. Particular atten- able reports of primary repair of traumatic colon wounds
tion should be paid to skin infection at the site of proposed using only intravenous antibiotics. This was followed by
incisions. articles questioning the need for mechanical bowel prepara-
Smoking cessation should be discussed with the patient tion (41,42). Later studies showed significant improvement
prior to surgery. This can lower the odds of both pulmo- in SSIs with the combination of oral and IV antibiotics
nary and wound complications. This should occur at least (43,44). Adding mechanical bowel preparation to IV anti-
4 weeks before surgery (35). biotics reduces ileus compared to IV antibiotics alone. The
Other factors increasing the incidence of SSIs include addition of oral antibiotics reduces the odds ratio of SSI
obesity, pulmonary comorbidities, obstruction, wound to 0.39, as well as lowers leak rate and decreases mortality
class, abdominoperineal resection, and prolonged operat- (45). This is the only regimen recommended in the ASCRS
ing time (31). Textbook of Colon and Rectal Surgery (46). An additional
Proper skin preparation lowers the incidence of wound discussion of bowel preparation was presented in Chapter 2.
complications. Hair should be clipped rather than shaved, Appropriate IV antibiotics should be administered
and this should be done the day of surgery. Cleansing with within 1 hour of the skin incision. Repeat dosing may be
chlorhexidine or soap the evening before surgery and the needed intraoperatively depending on antibiotic chosen
morning of is helpful. There does not seem to be any differ- and length of surgery. Antibiotic choices include cefazolin
ence between chlorhexidine and iodine povacrylex in SSIs plus metronidazole, cefoxitin, levofloxacin plus metronida-
following colorectal surgery, or any difference in preps with zole, and ertapenam. The antibiotic should be given within
or without isopropyl alcohol (36). 1 hour of the start of surgery, at which time adequate tis-
The topics of mechanical bowel preparation and nonab- sue levels should be present (47,48). There is no benefit to
sorbable oral antibiotics have come almost full circle in the extending IV antibiotics beyond the first 24 hours after
past 40 years. The combination of mechanical bowel prep surgery.
along with oral neomycin and erythromycin was popular- Intraoperative care is critical to minimizing postopera-
ized by Nichols and Condon (37). This combination was tive infections. Meticulous surgical technique will decrease
shown to reduce wound complications and overall infec- the chances of leaving behind devitalized tissue. In colorec-
tions from 43% down to 9% (38,39). The concentration of tal surgery, particular care is needed to avoid spillage of
bacteria in the gut was shown to undergo a 4–5 log decrease bowel contents and converting a clean contaminated wound
with oral antibiotics and mechanical bowel preparation to a dirty wound. Interestingly, wound contamination
Infection 79
does not always correlate with the development of SSI (49). within 48 hours (5,54,55). While it is clear that such bundles
Instruments contaminated should not be reused or replaced reduce SSIs, it is not clear which are essential elements (56).
on the clean back table. Wound protectors may be help- Within a given bundle, increasing compliance results in
ful (50–52). Results with topical antibiotics applied to the increased effectiveness (57).
wound are mixed. Not all care bundles have been effective in reducing SSIs
Closed suction drains are often used in abdominal sur- (58). Redraping, rescrubbing, and changing instruments
gery in attempts to prevent or control infection. They are make no difference in infection in the absence of additional
appropriately used when draining an abscess. They are inap- elements (59).
propriately used in attempts to drain the peritoneal cavity, Optimal postoperative care can also lessen infectious
or to protect an anastomosis. They should be used when complications. Maintaining normothermia both before and
ureteral or bladder surgery is performed in conjunction after surgery is helpful (60).
with colon resection. Leaving a drain in the pelvis following Oxidative burst function of neutrophils is one of the
pelvic dissection in order to evacuate accumulated blood primary defenses against SSI. As this is so, improving tis-
is reasonable, in which case the drain should be removed sue oxygenation should reduce SSIs. Measurement of oxy-
when drainage is clear. gen saturation at the thenar eminence using near-infrared
Diversion of the fecal stream from a distal anastomosis spectroscopy has been shown to correlate with SSI (61).
is commonly used in colorectal surgery in an attempt either Results of supplemental oxygen to reduce infections have
to prevent infection or to mitigate damage should infection been mixed. Belda et al. reported a 50% reduction in wound
or anastomotic leak occur. A loop ileostomy is most com- infection rate with increased FiO2 (62). When Kurz et al.
monly used. There are wide-ranging opinions regarding the randomly assigned patients to either 30% or 80% FiO2 dur-
use of diverting ileostomies. They are most favorably viewed ing surgery and for 1 hour following, no change in SSI rate
when performed at the time of creation of ileal pouch anal was found (63).
anastomosis and coloanal anastomosis. They are commonly After surgery, the focus evolves from preventing to iden-
used following proctectomy with low pelvic reconstruc- tifying and treating infectious complications. Organ space
tion. When used in this setting, the height of the anasto- infections, anastomotic leaks, superficial wound infec-
mosis from the anal verge, previous pelvic irradiation, and tions, and enteric infections are of particular importance
technical difficulty of the anastomosis are all factors to be in colorectal surgery. One tool emerging as possibly useful
considered in deciding to divert. Operations involving both in predicting postoperative infections is C-reactive protein
construction and closure of stomas carry with them an (CRP). This is an interleukin-6-dependent acute phase pro-
increased risk of wound infection, which must be weighed tein produced by the liver. It is increased in infections and
against the potential benefit of diversion. Of course, stoma in inflammatory diseases. Its level is dependent on its syn-
closure requires another operation, with additional cost and thesis rate and duration of the inflammatory stimulus (64).
risk to the patient. It has been evaluated as a tool for predicting infections fol-
There is some evidence to support glove, gown, and lowing surgery for colorectal cancer (65). An elevated level
instrument change for wound closure, and this is a com- on postoperative day 3 (POD3) (greater than 170 mg/L) pre-
mon element in treatment bundles intending to reduce SSI. dicted both increased length of stay and 30-day mortality
A wound vacuum-assisted closure applied over the closed (65). A separate study showed that CRP elevation on POD4,
skin may also reduce superficial wound infections. in this case, 125 mg/L, had a 82% sensitivity for predict-
Enhanced recovery pathways (ERPs) are covered in ing septic complications. Its negative predictive value was
Chapter 10 and are mentioned here because they result in 95.8% (66). It may be a useful tool for evaluating patients
reduced risk of SSI, among other benefits (53). The bundling for early discharge. CRP is a better tool than procalcitonin,
of care elements within a single pathway reduces unwanted another inflammatory marker, in predicting infection after
variability in patient care. In concert with ERPs, and spe- colorectal surgery (67).
cifically directed toward reducing infectious complications, Organ space infections are usually evident from 4 to 7
many institutions are adding specific SSI bundles of care. days after surgery and are found when the patient evidences
While these are not standardized, they share with ERPs a some combination of fever, white blood cell elevation, ileus,
division into preoperative, intraoperative, and postoperative and malaise. A computed tomography (CT) scan with oral
components. Common preoperative components include and IV contrast is the most useful test for identifying these
mechanical bowel prep with oral antibiotics, chlorhexidine complications, though their usefulness in the first week
skin wipes, patient education, appropriate selection and after surgery is limited due to the inability to distinguish
timing of antibiotics, and standardized catheter placement. postoperative fluid collections from purulence. Treatment
Intraoperative elements may include limiting operating is usually drainage of the abscess by interventional radi-
room traffic, a separate closing tray, wound protectors, main- ology, and antibiotics. Antibiotic choice is initially broad
tenance of body temperature, glove or gown change at clos- spectrum, with narrowed coverage as culture results are
ing, and euglycemia. Postoperatively attention remains on available. Reoperation is necessary when radiologic drain-
euglycemia and normothermia, patient and staff handwash- age cannot be performed. This can sometimes be done lapa-
ing, wound cleansing with Hibiclens, and dressing removal roscopically (68).
Superficial wound infections are common and are usu- Intraoperative measures to reduce postoperative pul-
ally treated by opening all or part of the wound and with monary complications include short-acting rather than
wound care. long-acting neuromuscular blockade and local or regional
Anastomotic leaks are covered in Chapter 8. anesthesia. Laparoscopic surgery is accompanied by a
Clostridium difficile infection (C-Diff) occurs with greater decreased incidence of pulmonary complications compared
frequency in colorectal surgery than in most other surgi- to open surgery (70,72).
cal procedures. It occurs in about 2% of patients following Postoperative maneuvers decreasing pulmonary compli-
colorectal surgery (69). It is more common following emer- cations include early ambulation and avoidance of nasogas-
gency procedures, in inflammatory bowel disease, and with tric tubes.
increased injury scores. It results in increased complications, A bundle of steps meant to reduce postoperative pulmo-
increased intensive care unit admissions, increased length of nary complications has been developed and can be included
stay, and increased 30-day readmissions. It can progress to in postoperative recovery pathways. This bundle includes
fatal colitis, and so it is a risk to the index patient. It is a lead- early ambulation, deep breathing and incentive spirometry,
ing cause of hospital-acquired infections and therefore pres- elevating the head of the bed, protocol-based pain control,
ents a public health risk as well. C-Diff should be considered and twice daily oral hygiene with chlorhexidine. These steps
in our patients when they have unexplained diarrhea, par- are reported to reduce postoperative pulmonary complica-
ticularly when accompanied by white blood cell elevation. tions by 81% (73).
When suspected, the patient should be placed in contact
isolation, and stool studies submitted. Prevention is depen-
dent on proper hygiene to avoid spreading it from patient
to patient. Handwashing between patient contacts and URINARY TRACT INFECTIONS AND
wearing gloves during patient exams are critical to contain- URINARY RETENTION
ment of this pathogen. Treatment varies with disease sever-
ity with metronidazole being the usual first-line treatment, Catheter-associated urinary tract infection (CAUTI) is con-
with fidaxomicin or vancomycin for recurrent or persistent sidered a preventable infection. The presence of a CAUTI is
disease. Fecal transplant is highly effective and increasingly now used as a negative quality measure, and hospital inci-
available. dence rates are reportable and are public information. A
CAUTI does have negative effects on the patient, as bactere-
mia, longer hospital stays, and higher mortality have all been
associated with CAUTIs. Preoperatively, symptomatic uri-
POSTOPERATIVE PULMONARY nary tract infections (UTI) should be addressed and treated.
COMPLICATIONS Intraoperative use of catheters is up to the discretion of the
surgeon, but most anorectal cases do not require catheter-
Postoperative pulmonary complications occur in about ization. The modifiable risk factor most closely related to
one in five patients undergoing abdominal surgery (70). CAUTI is length of catheterization. While it is unclear if pro-
Risk factors include American Society of Anesthesiologists longed catheterization is a cause of CAUTI, the presence of
(ASA) II or higher, chronic obstructive pulmonary dis- a catheter for more than 2 days is associated with increased
ease (COPD), congestive heart failure, smoking, dependent rates of UTI. Emphasis on early removal of urinary catheters
functional status, alcohol use, impaired sensorium, weight has decreased this occurrence. Implementation of ERAS and
loss, and sleep apnea. an SSI bundle, which includes removal of the catheter by
Pulmonary complications related to COPD may be POD2 has decreased UTI rates from 7.4% to 2.8% (5). Most
improved by preoperative antibiotic therapy to treat infec- patients can have the catheter removed in less than 24 hours,
tion, such as bronchitis, bronchodilator therapy, and gluco- but as pelvic surgeries have a higher rate of postoperative uri-
corticoids in selected patients (70). nary dysfunction, it is acceptable to leave the catheter in until
Preoperative assessment should include screening for POD2 or POD3.
obstructive sleep apnea. This should be considered in The downside of early catheter removal is the risk of
patients who are obese, have metabolic syndrome, or snore postoperative urinary retention. The rate of urinary reten-
(71). Neck circumference greater than 17 inches in men and tion for colon surgery is 2%–5%; however, this increases to
15 inches in women is also a risk factor. Patients with sleep 24%, and in some reports as high as 41%, when referring
apnea have an increased risk of intensive care unit stay after to rectal or pelvic surgery (74,75). Anorectal surgery also
surgery. Continuous positive airway pressure should be used carries a risk of urinary retention of 17% (76). Other risk
postoperatively in abdominal cases. Consideration should be factors include older age, male sex, lung disease, low rectal
given for overnight observation in anorectal cases (72). cancers, additional pelvic procedures, increased operative
Smoking clearly increases the risk of postoperative com- time, and poor pain control (74–76). One goal of surgeons
plication, and these can be reduced by smoking cessation should be to decrease the risk of urinary retention, which
prior to surgery. In order to be beneficial, smoking should will prevent repeated catheterization and prolonged cath-
be stopped at least 4 weeks prior to surgery (35). eterization, thus decreasing the risk for CAUTI (74,75). This
Retained surgical items 81
starts preoperatively and is carried through the operative

and postoperative recovery by minimizing the volume of
perioperative fluid administration; this has been shown to
lower the risk of urinary retention (74). This holds true for
both anorectal and abdominal surgeries. Larger amounts of
fluid resuscitation may correspond to increased bladder dis-
tention that results in poor contractility of the detrusor and
is followed by retention (77).
RETAINED SURGICAL ITEMS
Many resources and brainstorming have been used to help

keep surgical items from being retained in our patients. This
event is uncommon, occurring 1:8,801 but is still a source of
anxiety for all surgeons (78). The risk is not equal in every
case. Acknowledging and being aware of the cases that are
higher risk for retained foreign bodies can be the first step
in prevention. These risk factors are obesity, emergency sur-
gery, changes in the planned surgery, multiple surgical teams Figure 9.2 Radiograph demonstrating radiopaque mark-
being involved in the case, and both increased surgical time ers. Left to right. Laparotomy sponge, Ray-Tec sponge.
and increased blood loss (78,79). Sponges are the most com- The upper image is flattened, whereas the lower image
demonstrates the radiologic view when the item is
monly retained surgical item and thus garnish the majority of
crumpled.
the attention. The abdominal cavity is the operative site where
most incidences of retained foreign bodies occur (78,80).
Intraoperative prevention takes active participation of the
surgical teams and system policies focused on this potential
complication. Surgeon-specific practices to prevent leaving
a sponge in a wound are to avoid using smaller Ray-Tecs; in
large open cases, placing hemostats on the sponge; not using
radiolucent objects; audibly saying “lap in” and “lap out” to
keep the team accountable; and doing one more exploration
of the abdominal cavity before the final closure. An instru-
ment and sponge count should be completed before the
fascia is closed. The surgical count, which should be done
before the case starts, at close of fascia, at close of skin, and
anytime there is personnel change, should be documented
and communicated to the whole team. Ideally, those per-
forming the counting should be afforded the utmost con-
centration to avoid distraction.
It is important to remember that a correct surgical count
does not guarantee that no instruments were left behind,
and the surgeon must always be vigilant to the possibility
of leaving an object behind. In fact, up to 87% of cases with
retained surgical objects occur with correct counts (79,81).
If the count is wrong, several steps can help to look for the
missing object. Be sure to check the trash, the floor, and of
course the surgical field. Magnetic rollers can be used to
pick small needles off the floor. Intraoperative x-rays can be Figure 9.3 Radiograph of (left to right) Jackson-Pratt
used to visualize the abdominal cavity for any radiopaque drain, Penrose drain, nasogastric tube.
objects left behind (Figures 9.2 through 9.4). Many hospi-
tals have policies in place that these closing x-rays must be this problem, such as radiofrequency scanners and bar-
done for high-risk cases, such as incorrect count, two surgi- coded sponges.
cal teams operating, morbidly obese patients, and patients Retained surgical instruments can be devastating for the
returning to the operating room with an open abdomen. patient. These foreign bodies can be found incidentally on
Some hospitals have obtained other devices to help mitigate subsequent imaging or during investigation for infectious
(a) to surgery may also provide helpful information about ure-

teral anatomy. The ureters may be displaced due to previous
surgery, tumor, or inflammation. The collecting system may
be duplicated, or hydroureter may be present. Each of these
should be known prior to surgery. Consideration should be
given to placement of ureteral catheters in selected cases.
While not shown to reduce the incidence of injury, they may
be useful guides to dissection and may facilitate identifica-
tion of ureteral injury.
When ureteral injury is suspected, methylene blue can be
administered to allow easier visualization of a urine leak. The
most important thing one can do to improve surgical out-
comes when ureteral injury occurs is to recognize that injury
has occurred. When detected postoperatively, repair requires
additional surgery. Repair primarily over double-J stent, if
(b)
possible. For distal injuries, ureteroneocystostomy may be
required. Urology consultation should be obtained.
Ureteral injury identified postoperatively is usually evi-
denced by some combination of oliguria, abdominal disten-
sion, excessive drain output, or serum creatinine elevation.
When injury is suspected, a CT scan with IV contrast can
either demonstrate ureteral integrity or reveal the site of
injury. Management depends on site of injury. Stent place-
ment, either retrograde or prograde, can occasionally be
used for injuries other than transection. Reoperation with
either primary repair or reimplantation is the most com-
mon treatment. Less common treatments are intestinal
interposition and crossed ureteroureterostomy. Urinary
Figure 9.4 (a) CT scan of patient with a retained Ray-Tec diversion with percutaneous nephrostomy is appropriate if,
sponge. Image is the inferior cut of the study. The upper for some reason, immediate repair cannot be undertaken.
edge of the Ray-Tec marker is demonstrated as white Additional information on urologic injuries and their man-
dotted lines between the bladder (filled with contrast)
agement is presented in Chapter 41.
and the sacrum. (b) Pelvic radiograph of the same patient
demonstrating Ray-Tec marker in pelvis.
Splenic injury has been mentioned earlier under bleeding
complications. It is less likely to occur in laparoscopic cases
compared to open (84) but deserves particular mention due
or inflammatory symptoms. Finding this on imaging gen-
to its occasional occurrence during colonoscopy. Injury is
erally mandates surgical exploration to remove the objects,
almost never detected during the procedure. Symptoms can
though it is thought that some small needles may be left.
range from left upper quadrant pain from capsular tears to
This does carry significant morbidity to the patient. These
hypotension and shock for severe injuries. Patients should be
events are costly to the health-care system as well, with esti-
advised after colonoscopy to call if abdominal pain occurs.
mates that each incident costs over $200,000 in medical and
Evaluation should include exam, vital signs, and hemoglobin.
liability costs (81). With ample attention, retained surgical
CT should be performed if injury is suspected. Management
instruments should truly be a never event.
is as usual for splenic injury.
Injury to pelvic splanchnic nerves often occurs with
pelvic surgery and should be discussed with the patient if
INJURY TO NORMAL ORGANS rectal dissection is contemplated. Meticulous sharp dissec-
tion will decrease injury to the nerves. Symptoms of injury
Potential injury to normal organs is a common section in include urinary retention and alteration of sexual function,
obtaining informed consent. Particular areas of importance particularly in males.
in colorectal surgery are the ureters, spleen, and pelvic
nerves (82).
Ureteral injury, as sustained in our challenging patient
previously described, is reported to occur in 0.2% of colorec-
DEEP VEIN THROMBOSIS
tal resections (83). Our goal is to reduce the incidence of
this complication as much as possible. Preoperative imag- Sudden postoperative death from pulmonary embolism is
ing is almost always done prior to surgery for both malig- a dramatic and memorable event. This can be seen after an
nant and benign indications. Reviewing these studies prior uneventful course in a patient ready to go home. Years later
Ileus 83
we see patients wearing compressive stockings to control imaging availability. If there is a high degree of suspicion,
the swelling from postphlebitic syndrome. Each of these are begin parenteral anticoagulation while awaiting confirma-
sequelae from DVT, a complication our patients are at par- tion of DVT. For intermediate suspicion, withhold antico-
ticular risk of sustaining. Virchow triad defines thrombotic agulation until imaging, if that can be accomplished within
events as associated with stasis, hypercoagulability, and 4 hours. If the degree of suspicion is low, withhold anticoag-
endothelial damage. ulation if imaging will be done in 24 hours or less (91). This
The incidence of DVT in colorectal surgery patients is is usually initiated with heparin. Vena cava filter placement
1.2%. Pulmonary embolus occurs in 0.7% of patients under- should be considered in patients having massive emboli
going colorectal surgery. Of these, 40% are diagnosed within (in whom additional emboli could be fatal) or in patients
the first week after surgery. About one-third are diagnosed in whom anticoagulation is contraindicated. Duration of
after discharge. DVT prolongs length of stay from 6 days to treatment is dependent on the patient’s other medical prob-
17 days. It is more common in open than in laparoscopic lems but is generally for 6 months.
surgery, and with longer operations (85). In patients under- A related topic is that of dealing with the patient already
going surgery for cancer, the venous thromboembolism anticoagulated and needing either surgery or colonoscopy.
(VTE) risk is between 2.4% and 3.4% (86). In general, reversal of anticoagulation before either surgery
DVT is increased in prolonged surgery, pelvic surgery, or colonoscopy is indicated. In high-risk patients, enoxapa-
surgery for inflammatory conditions, and malignancy. Most rin is substituted prior to surgery and then held the day of
of our patients have one or more of these characteristics. surgery. Patients on low-dose aspirin do not need to hold
In addition, our patients may have the added risk factors of this prior to colonoscopy.
immobility, COPD, sepsis, anemia, and hypoalbuminemia. Portomesenteric venous thrombosis occasionally fol-
The Caprini VTE risk scoring system is a tool for assessing lows colorectal surgery. It is usually detected on CT and is
VTE risk (87). It assigns points for the above risk factors, and treated with anticoagulation. It is more common in younger
others. A score greater than or equal to five defines a high-risk patients, the obese, and following restorative proctocolec-
patient. This number is reached with a 45-year-old patient tomy. It is also more prevalent in patients with ulcerative
with no other risk factors who undergoes a laparoscopic par- colitis, or with postoperative thrombocytosis (92).
tial colectomy of more than 45-minutes duration. Low-risk
patients have a very low risk of VTE (88). In short, colorectal
patients undergoing major surgery are at significant risk of
DVT. Improving outcomes requires minimizing the inci- NAUSEA AND VOMITING
dence of DVT as well as its recognition and management.
We should recognize that we deal with patients at high Immediately following anesthesia up to 30% of patients
risk of DVT. Many risk factors our patients share cannot will experience troublesome nausea and vomiting that can
be changed. We can reduce the incidence of DVT by appro- greatly influence their initial recovery (93). The biggest pre-
priate prophylaxis and early ambulation. Prophylaxis in dictor of this problem is the presence of a history of post-
high-risk patients may be either unfractionated heparin, anesthesia nausea and vomiting. Many factors, including
low molecular weight heparin, or fondaparinux (87,89). surgical trauma, influence this occurrence. Routine place-
All patients undergoing abdominal surgery should have ment of a nasogastric tube does not mitigate this nausea and
compression stockings and/or prophylaxis. There has been is not recommended. Total IV anesthesia and avoidance of
debate regarding the timing of the first dose of heparin, with inhaled anesthetic agents reduces postoperative nausea and
concern raised about increased risk of intraoperative bleed- vomiting (94). The most headway in this subject has been
ing if given prior to surgery. The bleeding risk associated with prophylactically administering antiemetics to decrease
with preoperative dosing is low and should not prevent its this unpleasant side effect. Ondansetron can decrease post-
use in high-risk patients. Compression stockings should be operative nausea when given before induction of anesthesia
continued until the patient is fully ambulatory. Recent data (95). Dexamethasone, as a preemptive agent, has reduced
support the continued usage of heparin (or enoxaparin) for postoperative nausea and vomiting in multiple randomized
1 month after discharge in patients with either cancer or clinical trials (95,96). Dexamethasone and ondansetron can
inflammatory bowel disease. Antiplatelet therapy reduces be used together with potentially synergistic effects (93).
the risk of recurrent thromboembolism (90). Gabapentin is a common element of enhanced recovery
One difficulty with DVT is detection, as symptoms may pathways due to its role in lessening narcotic usage, but it
be minimal. Leg swelling and calf tenderness are signs of also has an antiemetic effect (97).
DVT. Duplex study of the lower extremity should be done
when DVT is suspected. Pulmonary embolism (PE) should
be suspected in patients with shortness of breath, hypoxia,
hypotension, or tachycardia. When suspected, CT angiog-
ILEUS
raphy is indicated.
Treatment for DVT and pulmonary embolism is anti- Traditionally, bowel function was expected to return around
coagulation. Timing depends on degree of suspicion and POD3 for laparoscopic surgery and POD5 for open surgery.
More recent experience has demonstrated that bowel func- unique to the case. This is the ideal time to ensure all needed
tion may be retained in some patients, especially when com- materials are available. Confirm availability about needed
ponents of ERAS are adopted. Ileus occurs when there is staplers, suture, stoma appliances, etc. A preoperative pro-
failure of bowel peristalsis by an expected amount of time. tocol including time out reduces the incidence of wrong
Patients will have distention, inability to tolerate intake by site surgery (103). The addition of a preoperative check list
mouth, abdominal discomfort, and nausea and vomiting. is associated with decreased mortality as well (104). Asking
Return of bowel function is often the limiting factor in hos- early will prevent waiting later. Preoperative history, con-
pital discharge. Many ERAS elements are directed toward sent, and imaging should be concordant, or discrepancies
the goal of restoring and preserving gastrointestinal motil- reconciled.
ity. Nevertheless, postoperative ileus occurs 3%–32% of the One area unique to colorectal surgery is localization of
time following colorectal surgery (98). neoplasia. When the lesion is in the right colon, this is rarely
Risk factors are inconsistently identified across studies a problem. Rectal lesions are sometimes found at surgery
but likely include older age, male patients, malnutrition, not to be where anticipated. This is more likely to occur
increased body mass index, larger wound size, increased when the endoscopy was not performed by the operating
opioid consumption, and increased blood loss or need for surgeon. A lesion in the rectum thought to be in the sigmoid
blood transfusion (3,98,99). Prevention should focus on raises obvious problems. This situation can be avoided by
optimizing the modifiable risk factors when the luxury of always performing a proctoscope exam on patients prior to
elective surgery presents itself. going to the operating room. Lesions between the rectum
Ileus can be influenced by the extent of surgical trauma; and hepatic flexure can also present problems with localiza-
thus, intraoperative prevention involves use of minimally tion. This is particularly so with large benign polyps, which
invasive surgery when possible, and gentle handling of tissues may not be palpable, and in minimal access surgery, where
to minimize surgical stress. Intraoperative normothermia palpation is not easily done. Tattooing can be used to help
may also help. Limiting IV fluids preoperatively can prevent localize the lesion intraoperatively, though this is less help-
ileus, as higher amounts of fluid have been linked to postop- ful with rectal lesions (105). Another tool that is useful in
erative ileus (99), likely by causing increased bowel edema. rectal lesions is preoperative endoscopic injection of indo-
Mitigating the effects of opioids by use of preemptive and cyanine green just distal to the lesion. SPY Fluorescence
multimodal pain control will again be vital to preventing yet Imaging System or Firefly illumination can then help local-
another complication (3). Early mobilization is tradition- ize the lesion and define the site of division (106). Lesions
ally felt to be important in stimulating bowel function. Early can also be marked with a clip at the time of colonoscopy. A
postoperative feeding decreases time to return of bowel func- abdominal x-ray can then clearly demonstrate the location
tion and decreases postoperative ileus. Sham feeding with of the lesion. This is particularly helpful in villous lesion and
gum will help to stimulate the autonomic reflexes to produce in minimal access surgery.
bowel motility (100). However, in a randomized controlled
trial, this utility lost its effect in patients who were fed early
(101). Alvimopan, a peripheral mu-opioid antagonist, can be
given immediately preoperative and scheduled postoperative
FUTURE DIRECTIONS
to decrease the incidence of ileus. Other prokinetic agents
have also been used with less success in colorectal surgery Enhanced recovery pathways provide a framework within
than in upper gastrointestinal surgery, such as metoclo- which we describe our plan for patient care. The path-
pramide and methylnaltrexone. ways contain many specific elements, and the frame-
When symptoms persist for more than 5–7 days, a pos- work enhances the probability of each patient receiving
sible etiology should be thoroughly sought. Postoperative the intended optimal care. Our task is to refine the path-
complications, particularly infection, can cause paraly- way and then to comply with pathway elements. As many
sis of the intestine. Dehydration, UTIs, pneumonia, and are fairly well established, they need only be applied. Our
intraabdominal abscess are often identifiable causes of ileus. patient’s outcome will improve to the extent we provide to
Other considerations are adhesions and strictures caus- each patient the care we intend for all. By lessening unin-
ing mechanical obstruction (102). Treating the underlying tended and nonbeneficial variation in care, we can expect
causes, intravenous fluid (IVF), nasogastric tube decom- our patients to do better. Internal audits within our hospi-
pression, and patience are the mainstays of treatment. tals are necessary to ensure we are actually doing that which
we intend for our patients.
TIME OUT AND WRONG SITE SURGERY

CONCLUSION
We are obliged to pause before beginning any invasive
procedure and confirm the patient, proposed procedure, Improving outcomes remains our perpetual goal. Choosing
administration of antibiotics, and any particular needs measurements to define improvement is more difficult. We
References 85
frequently use length of stay as a measure, assuming that 26. Hedrick TL et al. Surg Infect (Larchmt) 2014;15(4):
reduced length of stay while holding the line on emergency 372–6.
room return and readmission is an improvement in out- 27. Hedrick TL et al. Dis Colon Rectum 2015;58(11):
come. We also use return to normal activities as a defini- 1070–7.
tion of recovery with earlier return being synonymous with 28. Underwood P et al. Diabetes Care 2014;37(3):611–6.
improved outcome. One measure of recovery in oncology 29. Wilson AP et al. J Hosp Infect 1990;16(4):297–309.
patients is the time recovery has progressed to allow che- 30. Pendlimari R et al. J Am Coll Surg 2012;214(4):574–80;
motherapy to be administered. By this standard, recovery discussion 580–1.
after laparoscopic surgery is significantly improved over 31. Kwaan MR et al. Surg Infect (Larchmt) 2015;16(6):
open surgery. Return to intended oncologic treatment has 675–83.
been promoted as a measurement of recovery most impor- 32. Ata A et al. Am Surg 2010;76(7):697–702.
tant to the patient (107). However defined, to obtain the best 33. Akinci B et al. J Diabetes Complications 2014;28(6):
outcome for our patients, we need to be thoughtful in our 844–9.
approach, uniform in our intent, and meticulous in our 34. Thompson RE et al. Curr Diab Rep 2016;16(3):32.
technique. We must look to ourselves for technique, to our 35. Mills E et al. Am J Med 2011;124(2):144–154.e8.
hospitals for adherence to our planned care, and to our sur- 36. Kaoutzanis C et al. Dis Colon Rectum 2015;58(6):
gical community for continual refinement of the care plan 588–96.
we intend to follow. 37. Nichols RL, Condon RE. Surg Gynecol Obstet 1971;
132(2):323–37.
REFERENCES 38. Goldring J et al. Lancet 1975;2(7943):997–1000.
39. Clarke JS et al. Ann Surg 1977;186(3):251–9.
1. Keller DS et al. Surg Endosc 2014;28(1):74–9. 40. Bartlett JG et al. Ann Surg 1978;188(2):249–54.
2. Chand M et al. Int J Surg 2016;25:59–63. 41. Pineda CE et al. J Gastrointest Surg 2008;12(11):
3. Barletta JF et al. Ann Pharmacother 2011;45(7–8): 2037–44.
916–23. 42. Gravante G et al. Int J Colorectal Dis 2008;23(12):
4. Wick EC et al. JAMA Surg 2017;152(7):691–7. 1145–50.
5. Keenan JE et al. J Am Coll Surg 2015;221(2): 43. Fry DE. Surg Infect (Larchmt) 2008;9(6):547–52.
404–14.e1. 44. Bellows CF et al. Tech Coloproctol 2011;15(4):385–95.
6. Miller TE et al. Enhanced Recovery Study Group. 45. Kiran RP et al. Ann Surg 2015;262(3):416–25. discus-
Anesth Analg 2014;118(5):1052–61. sion 423–5.
7. Roulin D et al. Br J Surg 2013;100(8):1108–14. 46. Steele SR et al. The ASCRS Textbook of Colon and
8. Gobble RM et al. Plast Reconstr Surg 2014;133(3): Rectal Surgery. 3rd edition, New York: Springer
741–55. International Publishing, 2016.
9. Zittel TT et al. Dis Colon Rectum 2013;56(6):761–7. 47. Wilson SE et al. Surg Infect (Larchmt) 2008;9(3):
10. Hakkarainen TW et al. JAMA Surg 2015;150(3):223–8. 349–56.
11. Mishriky BM et al. Br J Anaesth 2015;114(1):10–31. 48. Ho VP et al. Surg Infect (Larchmt) 2011;12(4):255–60.
12. Cavalcante AN et al. Anesth Analg 2017;125(1):141–6. 49. Lauscher JC et al. Langenbecks Arch Surg 2012;
13. Chestovich PJ et al. Surg Clin North Am 2013;93(1): 397(7):1079–85.
21–32. 50. Cheng KP et al. Colorectal Dis 2012;14(6):e346–51.
14. Herroeder S et al. Ann Surg 2007;246(2):192–200. 51. Edwards JP et al. Ann Surg 2012;256(1):53–9.
Erratum in: Ann Surg 2009;249(4):701. Dijkgraaf, 52. Baier P et al. Int J Colorectal Dis 2012;27(9):1223–8.
Omarcel G W [corrected to Dijkgraaf, Marcel G W]. 53. Thiele RH et al. J Am Coll Surg 2015;220(4):430–43.
15. Ahn E et al. Int Surg 2015;100(3):394–401. 54. Cima R et al. J Am Coll Surg 2013;216(1):23–33.
16. Khan JS et al. J Clin Anesth 2016;28:95–104. 55. Wick E et al. J Am Coll Surg 2012;215(2):193–200.
17. Jouve P et al. Anesthesiology 2013;118(3):622–30. 56. Keenan JE et al. JAMA Surg 2014;149(10):1045–52.
18. Hübner M et al. Ann Surg 2015;261(4):648–53. 57. Waits SA et al. Surgery 2014;155(4):602–6.
19. Keller DS et al. J Am Coll Surg 2014;219(6):1143–8. 58. Anthony T et al. Arch Surg 2011;146(3):263–9.
20. Favuzza J et al. Surg Endosc 2013;27(7):2481–6. 59. Ortiz H et al. Arch Surg 2012;147(7):614–20.
21. Bell RF et al. Cochrane Database Syst Rev 2006;(1): 60. Wong PF et al. Br J Surg 2007;94(4):421–6.
CD004603. Review. Update in: Cochrane Database 61. Govinda R et al. Anesth Analg 2010;111(4):946–52.
Syst Rev 2015;7:CD004603. 62. Belda FJ et al. Spanish Reduccion de la
22. Zhu J et al. Int J Surg 2018;49:1–9. Tasa de Infeccion Quirurgica Group. JAMA
23. Johansson PI et al. Blood 2014;124(20):3052–8. 2005;294(16):2035–42. Erratum in: JAMA 2005
24. Halabi WJ et al. Am J Surg 2013;206(6):1024–32; December 21;294(23):2973.
discussion 1032–3. 63. Kurz A et al. Br J Anaesth 2015;115(3):434–43.
25. de Lissovoy G et al. Am J Infect Control 2009;37(5): 64. Guirao X et al. Surg Infect (Larchmt) 2013;14(2):
387–97. 209–15.
65. Platt JJ et al. Ann Surg Oncol 2012;19(13):4168–77. 86. De Martino RR et al. J Vasc Surg 2012;55(4):1035–
66. Ortega-Deballon P et al. World J Surg 2010;34(4): 1040.e4.
808–14. 87. Gould MK et al. Chest 2012;141(2 Suppl):e227S–
67. Facy O et al. Ann Surg 2016;263(5):961–6. e277S. Erratum in: Chest 2012 May;141(5):1369.
68. Vennix S et al. J Laparoendosc Adv Surg Tech A 2013; 88. Qadan M et al. Ann Surg 2011;253(2):215–20.
23(9):739–44. 89. Khorana AA. Oncologist 2007;12(11):1361–70.
69. Aquina CT et al. Dis Colon Rectum 2016;59(4):323–31. 90. Coleman DM et al. Curr Probl Surg 2015;52(6):
70. Sachdev G, Napolitano LM. Surg Clin North Am 233–59.
2012;92(2):321–44, ix. 91. Kearon C et al. Chest 2016;149(2):315–52.
71. Kaw R et al. Br J Anaesth 2012;109(6):897–906. 92. Robinson KA et al. Surg Endosc 2015;29(5):1071–9.
72. American Society of Anesthesiologists Task Force 93. López-Olaondo L et al. Br J Anaest 1996;76(6):
on Perioperative Management of patients with 835–40.
obstructive sleep apnea. Anesthesiology 2014;120(2): 94. Habib AS et al. Anesth Analg 2004;99(1):77–81.
268–86. 95. Gan TJ et al. Department of Anesthesiology,
73. Wren SM et al. J Am Coll Surg 2010;210(4):491–5. Duke University Medical Center. Anesth Analg
74. Kin C et al. Dis Colon Rectum 2013;56(6):738–46. 2003;97(1):62–71, table of contents.
75. Changchien CR et al. Dis Colon Rectum 2007;50(10): 96. Henzi I et al. Anesth Analg 2000;90(1):186–94.
1688–96. 97. Achuthan S et al. Br J Anaesth 2015;114(4):588–97.
76. Toyonaga T et al. Int J Colorectal Dis 2006;21(7): 98. Vather R, Bissett IP. Int J Colorectal Dis 2013;28(10):
676–82. 1385–91.
77. Joelsson-Alm E et al. Scand J Urol Nephrol 2009; 99. Vather R et al. Surgery 2015;157(4):764–73.
43(1):58–62. 100. Parnaby CN et al. Int J Colorectal Dis 2009;24(5):
78. Gawande AA et al. N Engl J Med 2003;348(3):229–35. 585–92.
79. Lincourt AE et al. J Surg Res 2007;138(2):170–4. 101. Lim P et al. Ann Surg 2013;257(6):1016–24.
80. Cima RR et al. J Am Coll Surg 2008;207(1):80–7. 102. Masoomi H et al. J Am Coll Surg 2012;214(5):831–7.
81. Regenbogen SE et al. Surgery 2009;145(5):527–35. 103. Algie CM et al. Cochrane Database Syst Rev 2015;(3):
82. Kaafarani HM et al. J Am Coll Surg 2011;212(6): CD009404.
924–34. 104. Haynes AB et al. Ann Surg 2017;266(6):923–9.
83. Halabi WJ et al. Dis Colon Rectum 2014;57(2):179–86. 105. Conaghan PJ et al. Colorectal Dis 2011;13(10):1184–7.
84. Isik O et al. Surg Endosc 2015;29(5):1039–44. 106. Handgraaf HJ et al. Minim Invasive Ther Allied Technol
85. Moghadamyeghaneh Z et al. J Gastrointest Surg 2016;25(1):48–53.
2014;18(12):2169–77. 107. Aloia TA et al. J Surg Oncol 2014;110(2):107–14.
10
Care paths and optimal postoperative
management
BENJAMIN D. SHOGAN AND HEIDI K. CHUA
INTRODUCTION TO CARE PATHS undervalued but necessary component of care. Although

care paths are made up in part by clinical guidelines or algo-
rithms, care paths significantly differ as they are used in a
Care paths refer to evidence-based structured multidis-
multidisciplinary fashion over the entire spectrum of all
ciplinary protocols allowing patients to progress through
phases of care. Prospective data recording of each element
a set pathway until discharge criteria are met. Care paths
of the care path allows for analysis of compliance, efficiency,
originated in the 1990s based on emerging evidence on the
and outcomes. This allows care paths to continually evolve
optimization of perioperative surgical care (1). Originally
and improve patient care over time, whereas algorithms
popular in Europe, they have rapidly spread and are increas-
remain static.
ingly becoming the standard of care across the globe.
Many synonyms exist for care paths: fast-track protocols,
enhanced recovery after surgery (ERAS), integrated path-
ways, multidisciplinary pathways of care, and care maps all
represent different flavors of care paths. Although there are BENEFITS OF CARE PATHS
no definitions that define a certain care path, specific spe-
cialties tend to use distinctive verbiage. For example, ERAS The goals in development of care paths are decreased post-
is generally used in reference to colorectal or gastrointes- operative complications and sooner return of bowel function
tinal surgery, whereas care maps are often used in nursing to promote a safe earlier hospital discharge without read-
or cardiac procedures. Regardless, for the purposes of this mission. Multiple randomized control trials have supported
review, despite the name used in the literature we consider these outcomes. Zhuang et al. published the results of a
all care paths similar. In clinical practice, the actual name meta-analysis and systematic review on 13 randomized con-
that defines the pathway is of little importance in compari- trolled trials comparing enhanced recovery protocols with
son to the elements that make up the protocol. traditional care in elective colorectal surgery. They reported
All care paths have four common elements: timeline, that enhanced recovery was significantly associated with a
categories of care and interventions, discharge criteria, and decrease in primary hospital stay (−2.44 days; p < 0.0001),
compliance recording. A timeline allows each patient to total hospital stay (−2.39 days; p = 0.0003), and total com-
undergo the specific intervention on a consistent schedule. plications (relative risk [RR] 0.71; p = 0.00006). There was no
Categories of care include preoperative, intraoperative, and difference in readmissions, surgical complications, or mor-
postoperative interventions. Different interventions have tality (2). A consequence of early discharge is hospital and
been described for each phase of care, and they may be hos- patient cost. Pritts et al. reported that nearly $7,000 was saved
pital specific and dependent on hospital resources, patient for every patient undergoing a small or large bowel resection
demographics, and institutional footprint (Table 10.1). after a care path was implemented compared with those who
Discharge criteria are a set of milestones each patient must underwent a traditional postoperative protocol (3).
achieve prior to hospital dismissal. Developing predeter- Care path protocols not only affect length of stay, but also
mined criteria for discharge allows for consistency across have been shown to be associated with decreased postop-
multiple providers. Importantly, discharge criteria do not erative complications. A recent meta-analysis and system-
need to differ compared to those used for conventional post- atic review of randomized trials of ERAS or FTS protocols
operative care (1). Last, protocol compliance recording is an focusing on health care–associated infections was recently
87
88 Care paths and optimal postoperative management
Table 10.1 Components of a colorectal surgery care path and the benefit of each individual component, allowing
each discipline to contribute ideas specific to their specialty.
Preoperative Intraoperative Postoperative
Health-care elements combine to form care paths.
Patient education Fluid Ambulation Separately, each element can have positive effects on patient
management outcomes, but when combined into a care path the impact
Optimizing overall Minimally Oral intake/ can be dramatic (9). It is not critical for every intervention in
health invasive nutrition the care path to have a beneficial impact. Polle et al. showed
Nutritional Temperature Fluid that despite patients only completing an average of 7.4 of
assessment control management 13 predefined fast-track modalities, the fast-track pathway
Labs/x-rays/tests Drain usage Pain still had a significant positive impact on length of stay and
management patient satisfaction (10). Even though the optimal arrange-
Bowel preparation Pain Discharge ment of elements to be included is yet to be fully elucidated,
management criteria we suggest that an optimal care path include elements that
Reduction of fasting cross all phases of care. As stated earlier, the number and
state and/or type of interventions used within a care path will be hospi-
nutritional tal specific and determined by local factors, such as hospital
supplementation resources, specialties present within the health-care system,
Pain management and the recommendations of the local multidisciplinary
team (Table 10.1). In the following sections, we discuss the
most common interventions within a care path (Table 10.2).
published (4). Compiling data from nearly 4,200 patients,
these authors reported that compared to conventional con- PATIENT EDUCATION
trols, ERAS/fast-track surgery (FTS) was associated with a
significant reduction in postoperative lung infections (RR We begin patient education during the preoperative clinic.
0.38), urinary tract infections (RR 0.42), and surgical site The concept of care paths should be explained to the patient.
infections (RR 0.75). Carli et al. investigated the effects of preoperative educa-
Why care paths are associated with better outcomes tion and counseling on patients undergoing laparoscopic
remains to be defined but is certainly multifactorial. colorectal surgery (11). During a clinic visit, patients were
Teamwork is a critical component of any care path proto- informed of milestone setting, multimodality analgesia,
col, and establishing a successful program requires buy-in early postoperative oral intake, early mobilization, and
from a diverse multidisciplinary team. Development and expected discharge on postoperative day three. Although
implementation of care path thus creates a culture of team-
work and safety that is independently associated with better
Table 10.2 Current enhanced recovery protocol
patient outcomes (5,6).
at Mayo Clinic
Emerging evidence has shown that care path protocols
may play a role in decreasing systemic stress and inflam- Preoperative
mation (7). In a prospectively collected cohort of patients • Education with surgeon and ostomy nurse
undergoing both open and laparoscopic colon resection, • Mechanical bowel prep with oral neomycin
patient care paths inhibited release of postoperative inflam- and metronidazole
matory mediators, therefore playing a protective role on • Gabapentin 600 mg
the immune system (8). Further investigation is needed to Intraoperative
determine if decreased systemic inflammation is the under- • Intrathecal opioid injection at discretion
lying cause of improved outcomes in care path protocols. of anesthesiologist
Studies attempting to delineate which parts of the proto- • Net even fluid balance
col are responsible for the blunted immune response are
Postoperative
needed.
• Early ambulation
• Low-residue diet starting immediately after surgery
• Intravenous ketorolac scheduled for 24 hours
ELEMENTS OF CARE PATHS transitioned to oral ibuprofen
• Oral oxycodone as needed
• Rate of 40 mL/h then saline lock at 8 a.m. on morning
Implementing an enhanced recovery protocol can be
after surgery
extremely daunting and often meet with resistance. Many
• Foley catheter removed at 8 a.m. on morning after
of the aspects of enhanced recovery may be novel to some
surgery
practitioners and in some instances directly counter current
• 400 mg magnesium oxide twice per day for 3 days
practice. We suggest an evidenced-based approach, explain-
starting on postoperative day 1
ing the rationale behind the concept of enhanced recovery
Elements of care paths / Bowel preparation 89
we cannot conclude that it was directly related to educa- of blinding in the majority of included studies significantly
tion, the authors reported that 95% of patients could be dis- limited their conclusion. Together, their remains only lim-
charged within the fourth postoperative day. ited evidence to suggest that preoperative carbohydrate
Preoperative ostomy teaching is mandatory. For all loading is a critical part of a care path.
patients undergoing surgery where creation of an ostomy The time to resume feeding postoperatively has been
is a possibility, a preoperative visit by a certified wound, extensively studied. Historically, oral intake was held until
ostomy, and continence (WOC) nurse is necessary. The pre- the passage of flatus or bowel movement. These recommen-
operative visit should not be limited to those undergoing dations were largely based on surgical dogma and were not
permanent ostomy, as many ostomies thought to be tem- scientifically sound. Over the last decade, investigations
porary at the time of creation are never reversed. During have proved that early feeding prior to clinical evidence of
the visit with the WOC nurse, expectations of life with an bowel function is safe and not associated with an increased
ostomy are delineated as well as care and troubleshooting. need for nasogastric decompression or increased minor or
Further, the WOC nurse marks the patient for the optimal major complications (19). Quite the opposite was, in fact,
stoma placement based on body habitus and patient flex- noted, with early feeding shown to be beneficial. In Dag
ibility. Preoperative visits with a certified WOC compared et al., patients were stratified into starting a liquid diet on
to other health-care providers are associated with a signifi- postoperative day one compared with waiting until the pas-
cantly higher postoperative quality of life (12,13). sage of flatus. Those patients in whom oral intake was initi-
ated early decreased time to tolerating solid food (2.48 versus
FEEDING: PREOPERATIVE 4.77 days) and decreased hospital length of stay (5.55 versus
CARBOHYDRATE TREATMENT AND 9 days) (20). Nearly 90% of patients in the early feeding group
EARLY FEEDING tolerated the early feeding schedule. Although routinely
done, initiating feeding with clear liquids and advancing
Preoperative fasting is the standard of care to prevent aspira- as tolerated to solid food showed no benefit in a random-
tion during general anesthesia and endotracheal intubation. ized trial on gastrointestinal and vascular patients compar-
Recently, it has been shown that postoperative insulin resis- ing clear liquids versus solid food as the first postoperative
tance induces postoperative hyperglycemia that is associated diet (21). This finding is of particular importance as elective
with overall complications, wound healing complications, colorectal patients prefer simple solid foods rather than clear
and increased length of stay, calling the long-standing “NPO fluids as their first diet (22). Besides patient safety and com-
after midnight” into question (14,15). To combat postopera- fort, early feeding has been shown to decrease complications
tive insulin resistance, practitioners have begun to explore and promote bowel function. In a meta-analysis of 15 studies
the use of preoperative administration of oral or IV high- comparing feeding during the first 24 hours postoperatively
dose carbohydrates instead of a conventional preoperative compared to a traditional feeding schedule, there was a 45%
fast. In 15 randomized patients, insulin resistance was better decrease in the relative odds of postoperative complications
maintained in patients given a preoperative carbohydrate- (odds ratio 0.55, CI 0.35–0.87; p = 0.01) in the early feed-
rich drink (16). Noblett et al. randomized 36 patients to three ing group (23). Furthermore, very early feeding may also be
groups: group 1 underwent conventional fasting, group associated with prevention of postoperative ileus. Patients
2 was given preoperative water, and group 3 was given a who were fed immediately after surgery had a significantly
high-carbohydrate drink (96 grams carbohydrate) the night decreased time to passage of flatus and passage of first def-
before and then again 3 hours prior to elective colorectal sur- ecation (24). In our practice, we routinely start every patient
gery (17). Results showed that the carbohydrate group had a on low residue diet the evening of surgery.
significantly lower length of stay (mean 7.5 days; p < 0.01)
compared to the water group (mean 13 days) or the fasting BOWEL PREPARATION
group (10). Time to first flatus, bowel movement, and com-
plication rate was similar between the three cohorts. Thus, it The utility of preoperative mechanical bowel preparation
is unclear the reason for the delayed discharge in the no-car- (MBP) has long been under question. Several recent stud-
bohydrate cohorts. Interestingly, in the fasting group, grip ies have failed to provide evidence of the benefit of MBP.
strength was significantly decreased compared to preopera- A 2011 Cochrane Systematic Review found no difference
tive values, whereas there was no significant difference in the in anastomotic leak rate, complication rate, or mortality in
carbohydrate or water group. patients undergoing colon or rectal excision (25). An update
A Cochrane Review of 1,976 patients in 27 trials compar- to this analysis was published 5 years later, again failing
ing preoperative carbohydrate treatment (at least 45 grams to show clear evidence of the benefit of MBP (26). As the
of carbohydrate) to placebo or fasting in elective surgery authors noted, significant limitations exist in analysis of
was recently performed (18). The authors noted that there this pooled dataset, specifically the variability of the type of
was a very small but significant reduction in length of stay bowel preparation and lack of reported details on location
(0.30 days, 95% confidence interval [CI] 0.56–0.04) with no of colon resection; therefore, they cannot rule out a mod-
significant change in the incidence of aspiration or postop- est 30%–50% benefit of MBP. Nevertheless, most care paths
erative complications. The authors did comment that lack have chosen to omit MBP as an intervention.
There are emerging data that anastomotic leak may have Despite the positive impact on morbidity, patients on a
an infectious cause rather than technical failure (27,28). restrictive fluid strategy are at an increased risk for revers-
Given this concept, there has been resurgence in the interest ible acute kidney injury. Because a strict restrictive fluid
in adding oral antibiotics to the traditional MBP. Using the strategy does not fully take into account patient hemo-
National Surgical Quality Improvement Program (NSQIP) dynamics, it may not be suitable for every patient. Goal-
database, Scarborough et al. reported that the addition of directed strategies have therefore been developed where
oral antibiotics resulted in a significantly decreased inci- the amount of fluid is tailored to reach certain goals such
dence of surgical site infection (3.2% versus 9%; p < 0.001), as stroke volume or cardiac output (34). Hemodynamic
anastomotic leak (2.8% versus 5.7%, p = 0.001), and hospital monitoring may be achieved by arterial line, central line,
readmission (5.5% versus 8; p = 0.03) compared to patients Swab-Ganz catheter, or esophageal echocardiography. In
who received no preparation (29). In our institution, we one such protocol, 108 patients were randomized to con-
have chosen to change practice to include MBP with oral ventional fluid administration left to the discretion of the
antibiotics (neomycin and metronidazole) in all patients anesthesiologist, while the other cohort’s fluid administra-
undergoing a bowel resection. This may be the one compo- tion was based solely on predetermined esophageal Doppler
nent of the care path that is most variable in application. perimeters. Goal-directed therapy resulted in a reduced
hospital stay, fewer postoperative complications, and early
FLUID MANAGEMENT toleration of full diet (35). A recent multicenter randomized
trial was performed comparing goal-directed via esopha-
Fluid administration during the intraoperative and postoper- geal Doppler to near maximal strove volume to a restrictive
ative period has become a cornerstone of current care paths. zero-balance strategy in patients undergoing elective colon
IV fluids are necessary to maintain tissue perfusion during surgery (36). Interestingly, goal-directed therapy provided
general anesthesia and postoperatively when oral intake is no additional benefit over a zero-balance strategy in terms
minimal. Traditional fluid administration focuses on replac- of major, minor, or cardiopulmonary complications.
ing all fluid loss in the operating room, including blood loss, Although clear that both goal-directed fluid strategies and
insensible losses, and third spacing. This fluid strategy is asso- restrictive strategies are better than traditional fluid adminis-
ciated with an increase of 3–6 kg of increased postoperative tration, the optimal fluid management remains to be defined.
weight, which, in turn, can lead to interstitial and pulmonary In an effort to develop a consensus on perioperative fluid
edema (17). Furthermore, supplemental fluid is associated management in colorectal surgery, the Perioperative Quality
with poor wound healing and anastomotic leak, potentially Initiative was formed, supported by the American Society
secondary to changes in tissue perfusion pressures (30,31). of Enhanced Recovery and Evidence-Based Perioperative
Two main strategies have been developed to limit the Medicine. In a consensus statement published in 2016, they
amount of perioperative fluids given: restrictive fluid recommend 13 evidence-based strategies for fluid manage-
therapy and goal-directed fluid therapy. Restrictive fluid ment (37). Among these are the recommendations to have
strategies attempt to replace only the fluid that is lost intra- access to clear fluids up to 2 hours prior to surgery to prevent
operatively and maintain a zero-fluid balance. Kressner dehydration and the use of hemodynamic monitoring equip-
et al. conducted a double-blinded randomized trial of ment when available to guide intraoperative fluid manage-
fluid restriction as part of a fast-track protocol in colorec- ment or perform a restrictive strategy. Currently, our strategy
tal surgery (32). In the restricted cohort, patients received is to maintain an intraoperative net zero balance with all IV
5% buffered glucose solution at 2 mL/kg/h from induction fluids stopped on the morning after surgery.
of anesthesia until skin closure, then 1 mL/kg/h until the
morning after surgery at which time IV fluids were stopped. ANALGESIA
The standard fluid group received 500–1,000 mL Ringer
lactate solution prior to the induction, 5 mL/kg/h of Ringer Postoperative pain control is critical to promote early
lactate solution during surgery, followed by 10% glucose ambulation, improve patient pain scores, and increase satis-
solution at 1 mL/kg/h until the morning after surgery. faction. Much has been learned regarding the mechanisms
Results showed that the restrictive cohort received signifi- and pathophysiology of acute pain, and a detailed review is
cantly less IV fluids (3,050 mL versus 5,775 mL; p < 0.001) outside the scope of this chapter. Prior to this understand-
and gained an average of two fewer kilograms over the first ing, many centers used single-modality opioid-based anal-
three postoperative days. Although the length of hospital gesia, usually with a patient-controlled device (PCA). It is
stay was similar between the two groups, the proportion of well known that opioids cause slowing of the gastrointesti-
patients with complications was significantly lower in the nal track leading to ileus, respiratory depression, confusion,
restrictive group (39% versus 57%; p = 0.027). A system- and sedation. Contemporary pain management used in care
atic meta-analysis confirmed these results (33). Compiling path is best undertaken in a multimodality approach with
3,861 patients, patients in the standard fluid therapy group the goal being to minimize the use of narcotics.
had a significantly higher risk of pneumonia (RR 2.2%, 95% Regional anesthesia can be given preoperatively as an epi-
CI 1–4.5), pulmonary edema (RR 3.8%, 95% CI 1.1–13), and dural or spinal and can be delivered with either an opioid or
increased hospital stay (2 days, 95% CI 0.5–3.4). local anesthetic. No preferred regimen currently exists, but
Care paths during emergency surgery 91
all regimens have shown a benefit in elective colorectal sur- all incisions. Postoperatively, patients receive acetamino-
gery. Six trials comparing the effect of epidural analgesia on phen 1,000 mg IV every 6 hours and ibuprofen 800 mg IV
laparoscopic colorectal surgery were reported in a meta-anal- every 8 hours until they are able to take oral medications.
ysis (38). Although there was no significant change in length Intermittent injections or PCA (morphine or dilaudid) are
of stay, time to return of bowel function and postoperative available for severe pain (scale 6–10) until the patients are
pain scores were significantly decreased with the epidural taking oral medications. When oral medications are toler-
pathway. The authors noted no significant adverse events in ated, the acetaminophen and ibuprofen are switched to oral
all the trials related to the regional anesthetic. Similarly, when forms, and oxycodone 10 mg orally every 4 hours is available
patients were given a spinal injection of bupivacaine and for moderate pain. As epidurals are time consuming, expen-
morphine versus IV morphine, those in the spinal anesthetic sive, and do not always work, their use in our institution for
cohort had significantly decreased opioid use and better pain abdominal colorectal surgery has been abandoned. A ret-
control (19). In our practice, we work collaboratively with a rospective review of our early experience (compared to his-
pain specialists service to determine if the patient is a candi- torical controls) demonstrated a shorter hospital stay, lower
date for a regional anesthetic, and if so we recommend its use. opioid consumption, and better pain scores (44).
Pharmacological adjuncts are the cornerstone of postop-
erative multimodality therapy. In a large meta-analysis of 52 EARLY AMBULATION AND USE
controlled trials comparing nonsteroidal anti-inflammatory OF URINARY CATHETERS
drugs (NSAIDs), acetaminophen, and cyclooxygenase-2
inhibitors to morphine, it was shown that the nonopioid Other common elements of care paths include early ambula-
medications were associated with a significant reduction on tion and early catheter removal. Early mobilization is signif-
morphine intake, pain intensity, and postoperative nausea/ icantly associated with a decrease in pneumonia and venous
emesis (39). Importantly, NSAIDs did increase the risk of thrombosis rates (45). Although routinely cited, it is not well
postoperative bleeding from 0% to 1.7% (number needed to supported that ambulation decreases the rate of postopera-
harm = 59). The addition of acetaminophen to an NSAID tive ileus. In fact, recently it was shown that adherence to an
regimen further decreases opioid dependence for pain c ontrol intensive ambulatory program consisting of dedicated staff
by up to 30% compared to either used independently (40). assistance compared to usual care of “targets” provided no
Other adjuncts may also play a role in reducing opioid benefit in terms of postoperative complication rate, return
use, although the data are less clear. Gabapentin and pregab- of bowel function, or length of stay (46). Nevertheless, there
alin, calcium channel blockers that can reduce opioid chan- is little harm in early ambulation, and we strive for at least
nel neurotransmitter release, have been studied in a variety five walks per day.
of perioperative settings. As described by Dauri et al., in 6 of Indwelling Foley catheters are significantly associated
10 randomized trials, gabapentin was shown to provide bet- with urinary tract infections (UTIs), and patients who have
ter postoperative analgesia when given preoperatively com- their catheters in place longer than 48 hours postoperatively
pared to placebo, but 14 trials did not show any benefit on are twice as likely to develop a UTI (47). Furthermore, an
postoperative nausea or emesis (41). Magnesium is a gluta- indwelling catheter is often a reason cited for decreased
mate receptor found on nerve cells and has also been stud- ambulation. In our practice, unless needed for strict fluid
ied in lowering opioid requirements. There is no standard management or the presence of a urogenital procedure, we
dosing for magnesium, and various studies have reported remove all Foley catheters in the morning of postoperative
both oral and/or IV dosing schemes. Nevertheless, it has day 1. If the patient is subsequently unable to void sponta-
been shown that perioperative magnesium reduces post- neously, he or she will undergo intermittent catheterization
operative pain and opioid consumption without increased until able to pass urine spontaneously.
toxicity (42). Looking at the entire body of evidence, it is
our current practice to give all patients gabapentin (600 mg
patients aged 18–59; 300 mg patients greater than 59) pre-
CARE PATHS DURING EMERGENCY
operatively, scheduled ketorolac for the first 24 hours after
surgery that is then transitioned to ibuprofen, scheduled
SURGERY
acetaminophen, and 400 mg magnesium oxide twice per
day for 3 days starting on postoperative day 1. While originally designed for elective procedures, care
At the editors’ institution (Ochsner Clinic Foundation paths have started to become routinely used during emer-
Hospital, New Orleans, Louisiana), a slightly different mul- gency colorectal surgery. Wisely et al. (48) showed that there
timodality approach to postoperative pain management is was a significant change in management toward ERAS prin-
used and includes the following elements (43). Preoperatively, ciples in emergency abdominal surgical patients following
acetaminophen 1,000 mg IV and ibuprofen 800 mg IV are introduction of an ERAS program in elective patients (46).
given. Intraoperatively, liposomal bupivacaine 266 mg and Specifically, post-ERAS patients received less IV fluids,
Marcaine 75 mg (0.25%) are infiltrated at the level of the spent less time with an indwelling Foley catheter, and spent
transversalis fascia just lateral to the rectus sheath with direct less time on a narcotic PCA. Emergency patients placed on a
or laparoscopic visual control and in the subdermal space of care path may also experience a similar benefit as compared
to elective patients. Although limited by a small sample size 16. Svanfeldt M et al. Br J Surg 2007;94(11):1342–50.
(n = 60), it was shown that a comprehensive care path in 17. Noblett SE et al. Colorectal Dis 2006;8(7):563–9.
the emergency setting is associated with shorter hospital 18. Smith MD et al. Cochrane Database Syst Rev 2014;
stays (5.5 days versus 7.5; p = 0.009) and decreased time to 14(8):CD009161.
first flatus (1.6 days versus 2.8; p = <0.001). Further studies 19. Wongyingsinn M et al. Br J Anaesth 2012;108(5):
are needed to define which subsets of emergency surgical 850–6.
patient care paths are safe and feasible. 20. Dag A et al. Clinics (Sao Paulo) 2011;66(12):2001–5.
21. Jeffery KM et al. Am Surg 1996;62(3):167–70.
22. Yeung SE, Fenton TR. Dis Colon Rectum 2009;52(9):
1616–23.
PUTTING IT TOGETHER 23. Osland E et al. JPEN J Parenter Enteral Nutr 2011;
35(4):473–87.
Care paths are multidisciplinary protocols used to optimize 24. Fujii T et al. Int Surg 2014;99(3):211–5.
patient care. When compared to traditional perioperative 25. Güenaga KF et al. Cochrane Database Syst Rev 2011;
care, care paths lead to decreased patient complications and 7(9):CD001544.
decreased hospital stays with no increased readmissions. In 26. Dahabreh IJ et al. Dis Colon Rectum 2015;58(7):
this chapter, we reviewed the relevant literature for the most 698–707.
commonly cited elements of care paths. As it remains unclear 27. Shogan BD et al. J Gastrointest Surg 2013;17(9):
which elements are most impactful, when designing a care 1698–707.
path, it is imperative to choose elements that are feasible and 28. Shogan BD et al. Sci Transl Med 2015;7(286):
reproducible at the specific institution. In this way, compli- 286ra68.
ance and efficiency can be documented, allowing for continu- 29. Scarborough JE et al. Ann Surg 2015;262(2):331–7.
ous recalibration to optimize future patient outcomes. 30. Arkiliç CF et al. Surgery 2003;133(1):49–55.
31. Shandall A et al. Br J Surg 1985;72(8):606–9.
REFERENCES 32. Abraham-Nordling M et al. Br J Surg 2012;99(2):
186–91.
1. Kehlet H, Wilmore DW. Br J Surg 2005;92(1):3–4. 33. Corcoran T et al. Anesth Analg 2012;114(3):640–51.
2. Zhuang CL et al. Dis Colon Rectum 2013;56(5): 34. Noblett SE et al. Br J Surg 2006;93(9):1069–76.
667–78. 35. Pearse RM et al. JAMA 2014;311(21):2181–90.
3. Pritts TA et al. Ann Surg 1999;230(5):728–33. 36. Brandstrup B et al. Br J Anaesth 2012;109(2):191–9.
4. Grant MC et al. Ann Surg 2017;265(1):68–79. 37. Thiele RH et al. Perioper Med (Lond) 2016;5:24.
5. Pronovost PJ et al. J Crit Care 2008;23(2):207–21. 38. Khan SA et al. Surg Endosc 2013;27(7):2581–91.
6. Bliss LA et al. J Am Coll Surg 2012;215(6):766–76. 39. Elia N et al. Anesthesiology 2005;103(6):1296–304.
7. Watt DG et al. Medicine (Baltimore) 2015;94(36): 40. Ong CK et al. Anesth Analg 2010;110(4):1170–9.
e1286. 41. Dauri M et al. Curr Drug Targets 2009;10(8):716–33.
8. Wang G et al. J Gastrointest Surg 2012;16(7): 42. De Oliveira GS et al. Anesthesiology 2013;119(1):
1379–88. 178–90.
9. Wilmore DW, Kehlet H. BMJ 2001;322(7284):473–6. 43. Beck DE. Perioperative care. In Beck DE, Wexner SD,
10. Polle SW et al. Dig Surg 2007;24(6):441–9. Raferty J, Jayne D (eds). Gordon and Nivatvong’s.
11. Carli F et al. Can J Anaesth 2009;56(11):837–42. Principles and Practice of Surgery for the Colon,
12. McKenna LS et al. J Wound Ostomy Continence Nurs Rectum, and Anus. 4th Edition. Thieme. In press.
2016;43(1):57–61. 44. Beck DE et al. Ochsner J 2015;15(4):408–12.
13. Maydick D. Ostomy Wound Manage 2016;62(5): 45. Wren SM et al. J Am Coll Surg 2010;210(4):491–5.
14–24. 46. Fiore JF et al. Ann Surg 2017;266:223–31.
14. Ljungqvist O et al. Proc Nutr Soc 2002;61(3):329–36. 47. Wald HL et al. Arch Surg 2008;143(6):551–7.
15. van den Berghe G et al. N Engl J Med 2001;345(19): 48. Wisely JC, Barclay KL. ANZ J Surg 2016;86(11):
1359–67. 883–8.
11
Limitations of anorectal physiology testing
THOMAS E. CATALDO AND SYED G. HUSAIN
balance of a number of competing factors and requires

CHALLENGING CASE intricate correct performance of enteric and colonic physi-
ology, rectal, anal, and pelvic sensory and motor nerves, as
A 65-year-old woman and her 30-year-old daughter well as anatomically intact and functioning anal and pelvic
both present to your office with complaints of fecal musculature. Disruption of any of these factors may result
incontinence. Both are gravida 3 para 3, all vaginal in fecal incontinence. On the other end of the spectrum, the
deliveries, and have had at least one delivery of a child patient may suffer difficult, painful, or incomplete evacu-
of 3.6 kg or more. Both have required an assistance ation. Anorectal dysfunction is often devastating to the
device for delivery of one child. Both report fecal patient resulting in emotional distress and social isolation.
soiling for the last year. The older woman reports pro- Fecal continence is defined as the ability to defer defeca-
gressive uncontrolled passage of flatus and occasional tion until a socially appropriate time and place. Incontinence
identification of stool in her undergarments that she has a number of definitions, from simply involuntary pas-
was unaware of having passed. The daughter reports sage of stool to inability to control passage of solid, liquid,
incontinence to moderate amounts of stool despite or gas. In a 2001 consensus conference report, fecal incon-
attempts to delay defecation. There is additional tinence is defined as, “recurrent uncontrolled passage of
incontinence associated with athletic activity. fecal material for at least one month in an individual with a
developmental age of at least four years” (1). Reported prev-
alence varies from 1.4% to 18%, with rates as high as 45% in
CASE MANAGEMENT elderly, debilitated, or psychiatrically impaired institution-
alized adults (1–3). The numbers are generally accepted as
Both patients are examined. The younger patient has underreported due to patients’ unwillingness to come for-
reduced tone and squeeze. An anterior defect is found ward due to associated social and cultural stigma.
but the sphincter muscles appear to contract with Constipation is as difficult to define. It may be as subjective
squeeze. The older patient also has poor sphincter as any difficulty or infrequency in passing stool as perceived
tone and no squeeze. Both patients are offered a trial by the patient. The Rome II criteria define constipation as two
of increased fiber and Imodium, which fails to resolve or more of the following for at least 3 months: straining more
their symptoms and are scheduled for physiologic test- than 25% of the time, hard stools more than 25% of the time,
ing. Anal manometry in the younger patient confirms incomplete evacuation more than 25% of the time, or two or
the physical exam and intra-anal ultrasound confirms fewer bowel movements in a 7-day period (4).
an anterior sphincter defect. The patient is offered an A problem inherent to all anorectal physiology (ARP)
overlapping sphincteroplasty. Manometry on the older testing is the scarcity of “normal” values for comparison.
patient confirms low resting and squeeze pressures. There is relative paucity of literature describing ARP test-
Ultrasound does not identify any sphincter defects. ing on the normal population, and almost all of the available
The patient is offered a sacral nerve stimulator. studies are composed of a small group of subjects.
INTRODUCTION PHYSIOLOGY OF FECAL CONTINENCE
Normal function of the anus and rectum resulting in com- Normal fecal continence relies on a number of mechanisms.
fortable passage of stool under voluntary control is a complex The first of which is normal enteral and colonic motility and
93
94 Limitations of anorectal physiology testing
fluid transport physiology resulting in manageable stool

volume and consistency. Stool consistency may be the most
important characteristic that influences fecal continence
(5). Some patients may be continent to solid stool but not
to liquid or gas. The rectum needs to maintain adequate
reservoir capacity. In addition, the rectum, anus, and pelvic
musculature must have adequate ability to sense and dif-
ferentiate the presence of solid, liquid, and gas and to dif-
ferentiate between these. Additionally, it is postulated that
the vascular cushions or hemorrhoids create a controllable
seal as a rectal “corpus cavernosum.” This is supported by
the identification of leakage in some patients after otherwise
uncomplicated hemorrhoidectomy (5). On a mechanical
level, defecation occurs when the pressure within the rec-
tum exceeds the pressure or resistance provided by the anus.
For normal defecation, this relies on the controlled increase
in rectal pressure combined with simultaneous relaxation Figure 11.1 Typical anorectal physiology lab with manom-
of the anus and straightening of the rectum through relax- etry, transanal ultrasound, pudendal nerve terminal motor
ation of the pelvic muscles. latency testing, biofeedback, storage, and equipment for
sterilization.
For normal defecation to occur, the rectum must also pos-
sess correct mechanical properties of capacity and distensibil-
ity. It must be able to sense the need to empty and the qualities arranged radially or in a spiral orientation. Water within the
of the contents within it. Disease processes or injuries that channels was either static or continuously perfused. Initial
limit the ability of the rectum to distend to accept stool and continuous recordings were made with pen and ink on poly-
air from the sigmoid colon will alter the urge to defecate, and graph devices. State-of-the-art manometers contain solid-
the ability to defer defecation. Conversely, a chronically dis- state micropressure transducers mounted within the catheter
tended capacious rectum may lose the ability to sense when itself (Figure 11.2a,b). In addition to more reliable and repro-
it is full and thereby overflow. Both cases might present as ducible data, the catheters are more easily and reliably cleaned
incontinence. Alternatively, if the body cannot differentiate from patient to patient. Pressure data are recorded continu-
between solid, liquid, or gas, or if the mechanism by which ously to computer-based software that assists in creating the
this is sampled is altered, the result is often fecal soiling. interpretation and the report. One critical observation of the
A battery of devices and tests has been developed to water-perfused systems is that the patient may react to the
investigate many aspects of normal and altered defeca- sensation of water dripping from the anus with increased
tion. Much work remains to fully elucidate the source and tone. Scrupulous technique may avoid this.
thereby the solutions to disordered defecation. Because various technologies and methods exist for mea-
suring anal canal pressures, no universally accepted set of
normal values exists. Simpson et al. addressed this issue in
THE ANORECTAL PHYSIOLOGY LAB
a study comparing five different catheters and techniques
A battery of devices and tests have been developed to inves- of manometry in both normal and incontinent patients
tigate many aspects of normal and altered defecation. Many (6). Although their sample size was small, 10 normal and
centers have collected the equipment to accomplish these 11 patients with incontinence, the authors found no signifi-
tests (Figure 11.1). Work remains to be done to fully elu- cant difference between five commonly employed devices.
cidate the source and thereby the solutions to disordered They were a water-perfused end-hole catheter, a catheter
defecation. water perfused with four radially arranged side holes, a
water-filled microballoon, a microtransducer, and an air-
filled portable microprocessor-controlled device.
INVESTIGATIONS FOR INCONTINENCE:
MANOMETRY
SPHINCTER PRESSURE MEASUREMENT
Manometry is a technique to measure the pressures that exist
within the anal canal and the pressures the anus is capable Although written consent is not required as the patient is
of achieving voluntarily. Over many years a variety of cathe- fully awake, at our institution we obtain full informed con-
ters, pressure detectors, and recording apparatuses have been sent and confirmation of patient identity, condition being
developed. In addition, different operator techniques have evaluated, and the patient’s understanding of the tests he or
been developed making standardization of results difficult. she is about to undergo. The patient takes one or two small-
Throughout the 1960s a variety of catheters were developed volume cleansing enemas at home prior to the exam. Anal
with different numbers of open-tipped channels and micro- canal pressures are measured with the patient lying comfort-
balloons. The number of channels varied, and they were ably in the left lateral decubitus position with knees and hips
Physiology of fecal continence / Sphincter pressure measurement 95
(a) (b)
(c)
Figure 11.2 (a) Manometry catheter, with balloon. (b) Manometry catheter, detail microtransducers. (c) Manometry
tracing, computer display.
flexed 90°. Some emphasis is placed on comfort and relax- it extremely difficult to compare data obtained at differ-
ation as anxiety, talking, and anything that increases the ent centers. The range of accepted normal values is wide.
intraabdominal pressure may affect the results. We employ Furthermore, normal values change for gender, parity, age,
stationary pull-through technique. The catheter is placed and numerous other factors. Despite the fact that the newer
transanally with the measuring points (balloons, holes, catheters are more comfortable and easier to maintain, the
or microtransducers) to a distance of 6 cm above the anal test remains mildly invasive and uncomfortable for the
verge. Measurements are taken in the anterior, posterior, patient.
and left and right lateral positions (Figure 11.2c). Pressures There are several technical caveats that may lead to
are recorded at relaxation and at maximum “squeeze” for considerable alteration in results. Patients with megarec-
10 seconds. The patient must be instructed to try to isolate tum may require a higher volume to illicit RAIR and may
squeeze of the anus and not employ the gluteal and other be falsely labeled as RAIR negative if the usual volume of
accessory muscles. The catheter is repositioned 1 cm dis- 30–40 cc is used to illicit RAIR (7). The balloon material
tally, and the process is repeated. The process is repeated in can influence the results as latex balloons tend to deform
step-wise fashion until the entire canal has been tested. An along their axis, resulting in a falsely elevated rectal compli-
alternative to this “station pull-out” technique is recording ance (8). Rectal compliance testing depends entirely upon
pressures during a continuous pull-out of the catheter at a the patient’s input; thus, the patient’s psychological status
controlled steady rate. Some modern catheters have trans- plays a very important role in data acquisition during this
ducers extending the length of the anal canal that allow for test (7). Furthermore, the results of rectal compliance may
simultaneous pressure measurements and obviate the need differ if the test is performed on “prepared,” that is, after
for withdrawing the catheter. The following parameters are enema evacuation, versus unprepared rectum (7). The rate
recorded: length of the anal high-pressure zone, mean rest- at which water is injected into the balloon may also affect
ing tone, and maximum squeeze pressure. the rectal sensitivity testing (9). Thus, it is recommended
In an effort to study the symmetry and detailed overall that slow filling should be accomplished at a rate of 1 mL/s
pressure profile of the anal sphincter, pressure vectography, (7). Whatever the method used, the same technique should
a technique that provides graphical representation of the be applied to all patients in order to obtain reproducible and
radial pressure profile of the anal canal, was developed. comparable results.
Caution should be exercised while making treatment
THE RECTOANAL INHIBITORY REFLEX decisions based on manometric findings, as normal or
abnormal values in incontinent patients do not necessarily
The presence or absence of the rectoanal inhibitory reflex correlate with severity of symptoms. In a large prospective
(RAIR) is identified by rapid distention of the rectum by study, Lieberman et al. evaluated 90 incontinent patients,
insufflation of the balloon at the tip of the catheter with including 6 males with a specific goal at determining what
10 cc of air. Simultaneous recordings taken in the middle of impact physiology testing including manometry had on
the anal canal high-pressure zone are made for 10 seconds. treatment and outcome. After appropriate history and
If the RAIR is present, a slight contraction of the external physical exam, patients were selected for medical or sur-
sphincter followed by a reflex relaxation of internal sphinc- gical management. Following this determination, they
ter and resultant decrease in anal canal pressure should be underwent ARP testing including manometry, pudendal
observed. Balloon insufflation may be repeated with more nerve terminal motor latency (PNTML), and transanal
air at 10 cc increments up to 60 cc until a reflex is observed. ultrasound (TAUS). Overall, only 9 (10%) had a change
in their management plan. Based on the results of these
RECTAL CAPACITY AND SENSATION tests, 5 of 45 patients initially assigned to medical man-
agement were offered surgery instead. Three of 45 patients
The balloon at the end of the catheter may also be filled with assigned to undergo surgical treatment were switched to
water in an incremental fashion to assess rectal sensation the medical group. Almost all of these alterations in man-
and compliance. Measurements are made at the minimum agement were based on TAUS. Manometry was found to be
volume of first rectal sensation, the volume required to abnormal in one-third of both management groups, and
produce a sustained feeling of the need to defecate, and a there was no correlation between manometric results and
maximum volume that creates significant discomfort or an change in management plan. There did not appear to be
irresistible need to defecate. an association between manometry, TAUS, and PNTML
results (10). In an elaborate study of 350 patients includ-
VALUE AND LIMITATIONS OF ing 80 controls, Felt-Bersma et al. (11) found that the most
MANOMETRY FOR INCONTINENCE significant difference between continent and incontinent
patients was maximum squeeze pressure. However, the
Anal manometry has become a staple in the evaluation of authors surmised that continent function could not be
fecal incontinence. Though routinely performed in many predicted based on anal manometry alone and suggested
centers, manometry lacks standardization of technique, that these results should only be interpreted in conjunc-
data collection, and methods of interpretation. This makes tion with other tests.
Physiology of fecal continence / Pudendal nerve terminal motor latency 97
Although it creates a rather striking and impressive

graphical representation of anal canal pressure profile, pres-
sure vector diagrams have been shown to be of question-
able clinical value for sphincter evaluation (12). A study by
Yang et al. (12) could not demonstrate any correlation when
vectoral analysis was compared to needle electromyography
(EMG) and ultrasonography.
Anorectal manometry remains of value for objective pre-
operative documentation of anal tone function or muscle
weakness. It is also useful in excluding patients from sur-
gery (2). Perhaps the biggest merit of manometry is its abil-
ity to diagnose short-segment Hirschsprung disease and
pelvic floor dyssynergia.
ELECTROMYOGRAPHY
EMG is the measurement of the electrical activity generated Figure 11.3 Anal plug, surface EMG electrode for
by muscle fibers during contraction or at rest. In 1930 Beck biofeedback.
first described anal sphincter EMG (13). Specifically, the
EMG measures activity in a motor group or those muscle polyphasic signal. The test is performed by taking multiple
cells innervated by a single axon. Muscles whose nerves have readings requiring multiple skin punctures around the anus.
been damaged will demonstrate altered activity. Myography
has been used to map the perianal area for muscular activity Surface EMG/biofeedback
and thereby detect sphincter defects. EMG is also used to
demonstrate nerve conduction and appropriate activation Measurement of muscular activity through the insulation of
and relaxation used in biofeedback therapy. the skin is far more imprecise but less painful than needle
EMG. Surface EMG is valuable for documentation of overall
Concentric needle EMG activity, especially during attempted voluntary rest, inhibition,
or contraction of a muscle. Surface EMG is helpful to docu-
A concentric needle electrode is two insulated electrodes, ment paradoxical sphincter activity as part of the diagnosis
one within the other. With the needle inserted into the of disordered defecation. Two self-adhering surface electrodes
muscle to be observed, in this case the external sphincter or can be applied on opposite sides of the anus over the subcu-
the pelvic floor, the electrical potential from one electrode taneous portion of the external sphincter, with a grounding
to the other is recorded. Information collected includes electrode placed at a distance on the patient. Alternatively, a
amplitude, duration, and frequency, as well as the number of plug electrode is employed within the anal canal (Figure 11.3).
phases. Amplitude is proportional to the number of muscle Surface measurement of muscle activity is more valuable if the
fibers activated. Normal values are an amplitude of less than muscle is being artificially activated by stimulating the nerve.
600 µV and duration less than 6 microseconds (14,15). Longer When the time of nerve stimulation is known and the time
duration or spreading of the signal can indicate dispersion of of muscle activity is measured, nerve conduction velocity can
the motor unit potential (MUP). This may represent dener- be assessed. A specific application of nerve stimulation and
vation or demyelinization, or simply aging. The sum of the surface EMG is measurement of the PNTML.
activity of many muscle cells creates a shape to the MUP.
Normal MUPs are bi- or triphasic. In general, more phases PUDENDAL NERVE TERMINAL MOTOR
within the action potential indicated denervation and rein- LATENCY
nervation. However, four or more phases have been reported
in normal muscle in up to a quarter of the time. The pudendal nerve arises from the second, third, and
fourth sacral nerve roots bilaterally and passes along the
Single-fiber EMG inferior pubic rami through Alcock’s canal. Prolonged labor
or the use of forceps for delivery may injure the pudendal
Individual muscle fiber action potentials can be recorded nerve as it exits from the canal. The conduction time of
with a single-fiber EMG. The recording area of the needle is the nerve can be measured by stimulating the nerve tran-
much smaller, 25 µm. In normally innervated external anal srectally and observing the time to electrical activity of the
sphincter muscle, only a few fibers will be activated by a sin- external anal sphincter. A St. Mark’s electrode attached to a
gle motor group axon. However, when damage occurs, dener- gloved finger provides both stimulation and measurement
vated muscle fibers are recruited by surviving axons. The (Figure 11.4a–c). An absent trace may indicate injury to the
number of muscle fibers and thereby signal density within nerve, whereas a prolonged PNTML is interpreted to indi-
the recording area of the needle increases, resulting in a more cate nerve injury and repair.
(a) LIMITATIONS OF ELECTROMYOGRAPHY

IN INCONTINENCE
The EMG delineation of an anatomic sphincter defect has been
largely supplanted by imaging studies such as ultrasound and
pelvic magnetic resonance imaging (MRI). Concentric needle
EMG and single-fiber EMG testing are uncomfortable, or in
some cases, frankly painful for the patient. The equipment is
expensive and difficult to master. Results are variable based on
the cooperation of the patient, the experience of the examiner,
and the patience of both (16). Surface EMG is mildly uncom-
fortable to the patient and technically challenging to perform.
Identification of the nerve trace can be subjective. Pudendal
nerve latency testing is operator dependent. Since PNTML
measures the fastest remaining fibers, a normal latency time
does not exclude injury. The latency values obtained are also
(b) affected by the distance between the electrode and the puden-
dal nerve; the shortest latencies are obtained by placing the
electrode as close to the nerve as possible (9). This is usually
accomplished with subtle movements of the electrode-bear-
ing finger inside the anal canal while observing waveforms
for the shortest latency thus generated in response to repeated
electrical stimuli. This trial-and-hit method results in sig-
nificant patient discomfort in many cases. There is some
bilateral crossover innervation of the sphincter; therefore, a
unilaterally abnormal test does not preclude normal function.
Earlier studies indicated that significantly abnormal bilateral
results were predictive of poor outcome with sphincter repair
(17–19). However, other authors have not found PNTML to
be helpful in this regard (10,20). Increased pudendal nerve
terminal velocities have been previously associated with
patients with idiopathic incontinence (21). Newer literature,
(c) however, suggests that this association might not be entirely
true. Ricciardi et al. (22) showed that only a small percentage
of patients with idiopathic fecal incontinence had associated
pudendal neuropathy.
Transanal ultrasonography
High-quality circumferential images of the anal sphincter
complex can be obtained using TAUS. Although a number
of probes are available, the most commonly used for evalu-
ation of the anal sphincter is a rotating probe that creates
a 360°, two-dimensional (2D) transverse image. The trans-
ducer generally used is a combined 7 or 10 MHz transducer,
rotating within a water-filled rigid cap covered with a bal-
loon or condom. Newer probes are fully self-contained.
They still require protection with a condom and some type
of interface media such as gel or water (Figure 11.5).
In many outpatient ARP labs, the procedure is performed
in the left lateral decubitus position in conjunction with anal
manometry. For a patient scheduled to undergo multiple ARP
studies the same day, we follow a policy of performing manom-
Figure 11.4 (a) St. Mark’s electrode. (b) Electrode on etry initially followed by other investigations as the sphincter
gloved finger to be placed transanally. (c) St. Mark’s stretch induced by 12 mm sonogram probe may produce erro-
electrode is inserted into anus with the finger tip directed neous manometric findings. As such, the patient may have had
toward Alcock’s canal. limited preparation with a small volume enema. This is not
Physiology of fecal continence / Limitations of ultrasound in incontinence 99
Global deficiencies or thinning of the sphincters rather

than defects are more difficult to define with TAUS. The
internal sphincter is normally between 2 and 4 mm. Since
it is more distinct on TAUS, excessive thickness or thinness
can be identified. One elusive objective is to identify atrophy
of the external sphincter, as this correlates with poor out-
come from sphincter repair (26).
Three-dimensional (3D) axial endosonography is now
available. The probe spirals and moves through the sphinc-
ter at a fixed rate collecting a 3D block of echo-data that can
be represented on a computer screen and evaluated through
any plane through the block.
In a study involving 33 women with suspected sphincter
injury, two different observers compared 2D TAUS with 3D
evaluation. There was an identifiable improvement in the
confidence of the examiner in detecting sphincter defects
with 3D evaluation over 2D images. Interobserver correla-
tion was also improved by 3D evaluation but not to a sig-
nificant degree (27). Nevertheless, 3D TAUS has not been
indisputably demonstrated to be more sensitive or specific
than transverse planar TAUS.
LIMITATIONS OF ULTRASOUND
IN INCONTINENCE
Figure 11.5 B&K 3D self-contained transanal ultrasound. Ultrasound is the most important test in the evaluation of
Computer console and probe. fecal incontinence with few limitations. Anorectal ultra-
sound entails a significant learning curve (28), and results
required for TAUS alone. Some authors prefer the prone or are operator and experience dependent. The external sphinc-
lithotomy positions, feeling that the lateral position deforms ter is less distinct than internal sphincter and smaller, <90°,
the anatomy (23). The clinical significance of this is unclear. defects are harder to demonstrate (29). Patients with mini-
TAUS can distinguish the internal and external sphincters mal symptoms and limited defects may not require surgery;
individually, with an intact internal sphincter representing a therefore, the clinical significance of a defect is determined
continuous hypoechoic band. The external sphincter is more by the combination of physical exam, ARP testing, and
heterogenous but distinctly more hyperechoic. Although TAUS. The presence of atrophy of the external sphincter is
images can be taken throughout the anal canal, images are similarly hard to prove. This is due to the fact that atrophic
traditionally documented and preserved at proximal, mid, external sphincter becomes replaced with fat, making sono-
and distal anal canal. Defects in either the internal or exter- graphic delineation of the sphincter from the surrounding
nal sphincters are identified as a disruption in the continuous fat tissue more difficult (30). Many investigators believed
ring. The external sphincter naturally splits proximally as it that 3D ultrasound, by virtue of its superior resolution, may
extends to the levator sling and the pelvic floor musculature. result in improved identification of external sphincter atro-
Disrupted tissue heals with a scar that appears amorphous, phy. However, a comparative study showed no correlation
more echogenic than the internal sphincter, but less so than between 3D TAUS and MRI in 18 incontinent women with
the external. It is seen bridging the gap in the defect between MRI evidence of sphincter atrophy (31).
the disrupted ends of the sphincters. The presence of a sphinc-
ter defect on TAUS correlates well with a history of obstetrical Biofeedback for fecal incontinence
trauma, as well as with physical exam and manometric find-
ings (10,20,24). Interobserver agreement is excellent, and when Biofeedback is a process by which the patient is given an
an anatomic defect is present, TAUS sensitivity approaches auditory or visual representation of anorectal information,
100% (25), specifically for internal anal sphincter defects in the pressure, or muscle activity, which he or she cannot other-
mid anal canal. Different techniques have been employed to wise perceive or correctly interpret. Techniques of biofeed-
either improve or make easier definition of the anal anatomy. back have successfully been used in the treatment of fecal
Some authors claim anal squeeze and relaxation improve the incontinence for over 25 years. The practice parameters of
yield of the sonographic exam, while others have no benefit the American Society of Colon and Rectal Surgeons give a
(23). A finger placed in the posterior wall of the vagina used to grade “B” recommendation for its use as a first-line therapy
measure the thickness of the perineal body has been shown to and in patients who have incomplete success after sphincter
aid in the evaluation of anterior sphincter defects (25). repair (2).
LIMITATIONS OF BIOFEEDBACK FOR include neurologic or endocrine dysfunction, for example,

INCONTINENCE Parkinson disease, diabetes mellitus, and hypothyroidism.
Finally, dysfunction of the enteric nervous system seen in
Despite some encouraging earlier reports describing suc- Hirschsprung and Chagas diseases and psychological fac-
cess of biofeedback in the management of incontinence, tors may also play an important role in the pathogenesis.
the Cochrane system review of treatments for incontinence Refractory constipation that fails to respond to dietary
in 2006 did not support its use (32). The authors reported, modification and conservative management warrants a
“The 11 trials reviewed were of very limited value because formal workup. From a management perspective, constipa-
they were generally small, of poor or uncertain quality, and tion is usually referred to as either slow transit constipation
compare different combinations of treatments.” Overall or obstructed defecation. The initial history and physical
success with biofeedback varies from 65% to 89% (2,33,34). examination, in most cases, are able to indicate if the patient
Two large randomized controlled trials included more than is experiencing slow transit constipation versus obstructed
100 subjects. Both concluded that biofeedback provided no defecation. Colonic transit studies are usually the first tests
additional benefit over office counseling therapy, such as to be ordered in cases where slow transit is suspected to be
advice, education, dietary modification, digital guidance, the underlying etiology, whereas in patients with obstructed
and medication. Despite the lack of demonstrated benefit, defecation, a defecogram should be offered as the initial
both trials showed improvement in severity of symptoms, diagnostic study. However, studies have shown that there
fecal incontinence scores, and quality of life (35–37). These is little, if any, correlation between these two diagnostic
benefits were seen in both the treatment and control groups, modalities, and the clinical picture does not necessarily
indicating the role of patient motivation and ongoing medi- reciprocate the radiological findings (40).
cal involvement in the treatment of fecal incontinence. The
most important predictors of success were completion of DEFECOGRAPHY
the program and age over 60 years. Higher body mass index Since the 1960s, continuously recorded fluoroscopy has been
was associated with a worse outcome. used to evaluate the dynamic function of the pelvic floor.
Defecating proctography or cinedefecography is a method
Summary: Value and limitations of ARP whereby semisolid radiopaque contrast material is placed
testing for evaluation of fecal incontinence retrograde into the rectum and lateral images are obtained
in real time. Creating a realistic “pseudo stool” has been a
The perceived value of ARP testing in the evaluation and challenge. A commercially available product was available
management of fecal incontinence varies greatly depending but was transiently taken off the market. Many institutions
on the perspective of the examiner and the expectations of create their own contrast as needed using a combination
the patient. The most frequently employed tests include anal of barium and potato starch. At the authors institution, we
manometry, TAUS, and PNTML. Techniques and normal use a unique recipe based on breadcrumbs. The material
values are not universally accepted. Abnormal results do must be thick enough to simulate stool but able to be passed
not equate with specific disease, injury, or symptomatology. transanally.
TAUS and MRI provide excellent anatomic definition to aid Once the enema is administered, the subject is seated in a
in the planning of surgical intervention. At best, manometry lateral orientation on a radiolucent commode. Prior to defe-
serves for documentation of preoperative function and may cation, measurements are made of the angles of the proximal
assist in patient selection for surgery. PNTML is still contro- and distal rectum. In women the vagina may be delineated
versial as to its role in the treatment of fecal incontinence. with a tampon soaked with water-soluble contrast, and in
certain circumstances the small bowel is opacified with oral
Investigations for constipation contrast. If further delineation is required, sterile water-
and disordered defecation soluble contrast can be place intraperitoneally to define the
lower peritoneal reflection; furthermore, in patients with
Constipation is one of the common ailments presented to suspected cystocele, instillation of dye into the bladder may
the colorectal surgeon. It usually entails unsatisfactory def- increase the diagnostic yield of the study.
ecation resulting from decreased frequency of defecation or The anorectal angle is created in part by the tone and
difficulty in passing stools or both. Prevalence in the general function of the puborectalis muscle. Measurements are
population in the United States has been reported to be as taken as the patient is sitting at rest, during forced contrac-
high as 2%–15% (38,39). Women are affected two to three tion, straining without defecation (Valsalva maneuver),
times more commonly than men with incidence increasing and during defecation. Perineal descent is defined as the
with age. As with incontinence, the etiology is multifactorial change in distance of the line drawn perpendicularly from
and complex. Etiologic factors associated with constipation the anorectal junction to the pubococcygeal line. This line
include lifestyle issues and medications, especially narcot- is drawn from the tip of the coccyx to the posterior-inferior
ics, antidepressants, and calcium channel blockers. Pelvic margin of the pubic ramus. In addition, perianal skin can be
outlet obstruction (puborectalis dysfunction or rectocele) marked with a metal marker and the motion or descent of
is also a common underlying abnormality. Other causes the perineum measured.
Physiology of fecal continence / Limitations of defecography in constipation 101
Normal reported values vary widely. One author offers abnormalities on defecograms (50); thus, the clinical rel-
a broad range, 70°–140° at rest, 100°–180° defecating, and evance of these findings is questionable, and therapeu-
75°–90° squeezing (41,42); where another is more specific, tic decisions should be based on clinical rather than mere
92° ± 1.5° resting and 137° ± 1.5° straining (5). The change abnormal findings on radiological studies.
in the angle may be more important than the absolute num-
bers. In our practice, the test is of most value if the surgeon Colonic transit studies
reviews the study with the radiologist while the test is being
performed. Colonic transit studies play a pivotal role in the assessment
Abnormal findings include perineal descent of more of constipation. The majority of colorectal surgeons agree
than 3 cm while resting or more than 3 cm while strain- that transit studies supply the most pertinent information
ing. Paradoxical contraction of the puborectalis and disor- out of all the physiology testing modalities available for con-
dered defecation is indicated by an observed ascent of the stipation (51). The most widely accepted technique involves
perineum or a static or more acute anorectal angle during ingestion of a commercially available capsule containing
attempted defecation. Additional findings may include 24 radiopaque rings followed by x-ray at days 3 and 5. A
internal intussusception to frank prolapse, rectocele, or normal study entails passage of more than 80% of the rings.
enterocele. Small, <2 cm rectoceles are commonly seen in The mean colonic time has been shown to be 31 hours in
asymptomatic patients and are regarded as a normal finding. males and 39 hours in females (15). Based on the location
of retained rings, abnormal studies may be labeled as “out-
LIMITATIONS OF DEFECOGRAPHY let obstruction” if 20% or more rings are retained at day 5
IN CONSTIPATION in the rectosigmoid region or “colonic inertia” if more than
20% rings are dispersed throughout entire colon. Clinical
It must be remembered that defecography is not a “physi- efficacy of colonic transit studies to detect segmental bowel
ologic” study, as the study is not performed in response to a motility remains controversial (9). No bowel prep is admin-
natural desire to defecate; instead patients are asked to evac- istered prior to the study, and patients are directed to avoid
uate in a rather alien, uncomfortable environment. Among using laxatives and promotility agents, including dietary
other criticisms regarding defecography are poor interob- fiber.
server agreement (43,44). To complicate issues further,
abnormal defecographic findings are common in asymp- Small bowel transit studies
tomatic patients (45,46). The significant degree of overlap
between defecographic findings in patients with constipa- Since it is generally accepted that slow transit constipation
tion and asymptomatic controls raises questions regarding is overwhelmingly attributed to colonic dysfunction, small
the cause-and-effect relationship between clinical symptoms bowel transit studies are infrequently requested. However,
and defecographic findings. One of the radiological signs when clinical suspicion exists, such as patients with gastro-
frequently documented during these studies is contrast paresis and dilated small bowel on plain x-rays, small bowel
retention within the rectoceles. The clinical significance of motility studies should be undertaken before undertaking
this “barium trapping” seen has also been questioned (47). a surgical intervention. Several techniques are available to
In a study by Shorvon et al., one-half of asymptomatic assess small bowel transit. Nondigestible carbohydrates are
subjects had some aspect of mucosal prolapse and intussus- broken down into hydrogen and fatty acids upon reaching
ception, and 17 of 21 women demonstrated some degree of the colon. Hydrogen and fatty acids are then absorbed into
rectocele (48). In addition, prior to that work, no work had the bloodstream. Therefore, the interval between inges-
employed normal, healthy volunteers as controls, and “nor- tion of substrate and increments in exhaled hydrogen levels
mal” was determined retrospectively by lack of anatomic estimate small bowel transit. Similarly, orally administered
abnormality. Other studies were performed in patients sulfasalazine is broken down by colonic bacteria into
undergoing barium enemas for other, nonanorectal condi- mesalazine and sulfapyridine and then absorbed. Colonic
tions (45,48,49). Anorectal angle assessment and its inter- transit can be measured by serum detection of sulfapyri-
pretation should be performed with utmost caution. As dine. Radionucleotide scintigraphy has also been used to
eluded to earlier, there is a wide variation in normal values assess small bowel transit function. However, the clinical
for anorectal angle, and many investigators believe that it application of these tests is limited by their complexity and
is the change in angle rather than the absolute values that variation in bacterial flora in different subjects.
serve as a useful guide to therapy (7).
Patients with urge incontinence may frequently show MAGNETIC RESONANCE IMAGING
increased threshold for urge to defecate. It is unclear if this MRI of the pelvic floor is the newest addition to the diag-
finding is the result rather than the cause of constipation nostic armamentarium available for pelvic floor evaluation.
(9), and the clinical implication of this finding remains MRI obviates the exposure to radiation. One technique
uncertain. Abnormal puborectalis function noted at defe- involves filling the rectum with ultrasound gel. Images can
cography has also been a topic of considerable debate. Many be obtained in a “static” manner or in the form of dynamic
normal individuals have been shown to have puborectalis pelvic MRI, which involves having the patient perform
maneuvers similar to those performed during conventional been shown to be of importance in differentiating between
defecography. During these maneuvers, multiple images are constipation caused by slow transit from that caused by
obtained, which are then viewed as a cine loop. MRI pro- pelvic floor dyssynergia (57).
vides excellent spatial orientation of the sphincter complex
and provides superior delineation of the surrounding struc- Biofeedback for constipation
tures. MRI appears to be superior to ultrasonography in dis-
cerning external sphincter abnormalities (30). Additionally, Biofeedback training is widely utilized to teach relaxation
dynamic MRI defecography appears to be superior to con- of the pelvic floor in patients with pelvic floor dyssyner-
ventional defecography in the evaluation of descending gia. A critical review of the available literature by Heymen
perineum syndrome as it provides excellent spatial assess- et al. (58) including 38 studies showed that mean success
ment of pelvic floor musculature (52). rate with pressure biofeedback was 78% compared to mean
success rate of 70% seen with EMG feedback. The authors
LIMITATIONS OF MRI IN CONSTIPATION surmised that despite the reported success rates, quality
research is lacking.
Dynamic pelvic floor MRI shares similar limitations as The most controversial area involving biofeedback train-
conventional MRI: cost, claustrophobia, and availability. ing for constipation is the questionable durability of the
There are, however, some specific limitations related to the results. Ferrara et al. (59) reported a clear loss of benefits
diagnostic modality. Studies comparing dynamic MRI with over time despite initial success.
conventional defecography have yielded conflicting results.
Healy et al. (53) found significant correlation between PATIENT PERSPECTIVE
dynamic MRI findings and defecography in 10 patients Inherent to the evaluation of fecal incontinence is the
examined employing both techniques. On the contrary, patient’s feelings of shame, embarrassment, and discomfort.
Matsouka et al. (54), in their study of 22 patients, reported These sensations are felt by the incontinent patient, result-
defecography to be more sensitive than dynamic MRI and ing in depression and social isolation. A number of quality-
recommended against the routine use of this expensive of-life tools have been developed to quantify the results of
modality. Most centers perform pelvic floor imaging with evaluation and treatment of fecal incontinence. No one tool
patient in supine position. Patients are asked to strain in is universally accepted, and these tools have been difficult to
a position that is far from physiologic and raises concerns validate (60,61).
regarding the reliability of the test. The influence of patient In addition, the testing incontinent patients are subjected
positioning has been investigated. Bertschinger et al. (55) to may be embarrassing and uncomfortable. Deutekom
performed a prospective comparison of 38 patients who et al. conducted a cohort study of 240 consecutive patients
underwent closed MRI in supine position followed by open undergoing evaluation of fecal incontinence in 16 Dutch
MRI in a sitting position. Four rectal descents, two entero- centers. Each patient underwent manometry, defecography,
celes, four small cystoceles, and four small anterior recto- TAUS, PNTML, and MRI. Two hundred forty of the 270 self-
celes were missed at supine MRI. The clinical significance administered questionnaires were returned. Patients were
of these findings, however, remains questionable. As men- asked to evaluate anxiety, discomfort, embarrassment, and
tioned earlier, the lack of “normal controls” makes it dif- pain. Answers were scaled from 1 (not 0), none, to 5, severe.
ficult to assess the efficacy of this test. Results were also summarized as total test burden. Overall
test results were surprisingly low, with average scores in
Balloon expulsion test each category not exceeding 2. Overall, MRI was the most
preferred and least uncomfortable test. Defecography was
The balloon expulsion test is an infrequently used method the most inconvenient and uncomfortable. Anorectal com-
to test motor defecatory function of the rectum. There is a bined testing, manometry, PNTML, and TAUS, also scored
complete lack of standardization of methods used in vari- low for discomfort and overall test burden but more so than
ous anorectal manometry laboratories. Various size bal- MRI (62).
loons have been used for this purpose. Commonly, 50–100 cc
deformable balloons are used. Alternatively, smaller, more
rigid balloons may also be employed. The impact of size and
compliance of balloon on the final interpretation of the test CONCLUSION
is unclear. In general, it is easier to evacuate larger balloons
(56). Many investigators believe that the volume of the bal- Physiological studies of anorectal function can provide
loon should be individualized to induce a constant desire to valuable information in carefully selected cases. While
defecate. Consequently, the use of lower volumes may result performing these studies, one should be cognizant of the
in false-positive results (57). There is a wide variation in fact that these procedures can be embarrassing, and at best
what is considered to be a normal test. The inability to expel are far from the patient’s usual habits. It is unnerving for
the balloon in a sitting position within 30–60 seconds is many patients to perform the act of defecation in the pres-
considered abnormal in most centers. Balloon expulsion has ence of an audience, and it is conceivable that “performance
References 103
anxiety” may lead to results that are not truly representative 24. Nazir M et al. Dis Colon Rectum. 2002;45:1325–31.
of actual patient status. Thus, these studies should be inter- 25. Zetterstrom JP et al. Dis Colon Rectum. 1998;41:
preted with a grain of salt. Despite the plethora of litera- 705–13.
ture available, the clinical usefulness of these tests remains 26. Briel JW et al. Br J Surgery. 1999;86:1322–7.
vague, and there is limited evidence that anorectal imaging 27. Christensen AF et al. Brit J Radiol. 2005;78:308–11.
guides management in pelvic floor disorders (63). There are 28. Badger SA et al. Int J Colorectal Dis. 2007 October;
multiple well-designed studies, which, unfortunately, report 22(10):1261–8.
conflicting results. We therefore recommend that the most 29. Dobben AC et al. Int J Colorectal Dis. 2007;22:783–90.
decisive factor governing treatment decisions is history and 30. Terra MP, Stoker J. Eur Radiol. 2006;16:1727–36.
physical exam. Physiology testing should always be used as 31. West RL et al. Int J Colorect Dis. 2005;20:328–33.
an adjunct rather than a primary determinant. 32. Norton C et al. Cochrane Database Syst Rev.
2006;(3).
REFERENCES 33. Jensen L, Lowry A. Dis Colon Rectum. 1997;40:
197–200.
1. Whitehead WE et al. Dis Colon Rectum. 2001;44: 34. Heyman S et al. Dis Colon Rectum. 2001;44:728–36.
131–144. 35. Solomon MJ et al. Dis Colon Rectum. 2003;46:703–10.
2. Tjandra JJ et al. Dis Colon Rectum. 2007;50: 36. Pager CK et al. Dis Colon Rectum. 2002;45:997–1003.
1497–1507. 37. Norton C et al. Gastroenterology. 2003;125:1320–9.
3. Person B et al. Surg Clin N Am. 2006;86(4):969–86. 38. Stewart WF et al. Am J Gastroenterol. 1999 December;
4. Thompson WG et al. Gut. 1999;45(Suppl 2):ii43–7. 94(12):3530–40.
5. Gordon PG. Anatomy and physiology of the ano- 39. Sonnenberg A, Koch TR. Dis Colon Rectum. 1989;32(1):
rectum. In Fazio VW, Church JM, Delaney CP (eds). 1–8.
Current Therapy in Colon and Rectal Surgery, 2nd 40. Infantino A et al. Dis Colon Rectum. 1990 August;
Edition. Philadelphia: Elsevier-Mosby, 2005, pp. 1–9. 33(8):707–12.
6. Simpson RR et al. Dis Colon Rectum. 41. Moieira H, Wexner SD. Anorectal physiologic test-
2006;49:1033–8. ing. In Beck DE, Wexner SD (eds). Fundamentals of
7. Wexner SD et al. Setting up a colorectal physiology Anorectal Surgery, 2nd Edition. Philadephia, PA: WB
laboratory. In Corman ML (ed.). Colon and Rectal Saunders, 1998, pp. 37–53.
Surgery, 5th Edition. Philadelphia, PA: Lippincott, 42. Finlay IG et al. Int J Colorectal Dis. 1998;3:67–98.
Williams and Wilkins, 2005, pp. 129–67. 43. Penninckx F et al. Int J Colorectal Dis. 1990 May;5(2):
8. Madoff RD et al. Int J Colorectal Dis. 1990;5(1):37–40. 94–7.
9. Barnett JL et al. Gastroenterology. 1999;116(3):732–60. 44. Ferrante SL et al. Dis Colon Rectum. 1991 January;
10. Liberman H et al. Dis Colon Rectum. 2001;44:1567–74. 34(1):51–5.
11. Felt-Bersma RJ et al. Dis Colon Rectum. 1990;33(6): 45. Bartram CI et al. Gastrointest Radiol. 1988;13(1):72–80.
479–85. 46. Turnbull GK et al. Dis Colon Rectum. 1988;31(3):190–7.
12. Yang YK, Wexner SD. Int J Colorectal Dis. 1994;9(2): 47. Halligan S, Bartram CI. Dis Colon Rectum. 1995;38(7):
92–5. 764–8.
13. Beck A. Phlugers Arch. 1930;224:278–92. 48. Shorvon PJ et al. Gut. 1989;30:1737–49.
14. Ferrara A et al. Tech Coloproctol. 2001;5:13–8. 49. Roe AM et al. J Roy Soc Med. 1986;79:331–3.
15. Smith LE, Blatchford GJ. Physiologic testing. In 50. Jones PN et al. Dis Colon Rectum. 1987;30(9):667–70.
Wolff BG, Fleshman JW, Beck DE et al. (eds). The 51. Karulf RE et al. Dis Colon Rectum. 1991;34(6):464–8.
ASCRS Textbook of Colon and Rectal Surgery. New 52. Healy JC et al. Br J Surg. 1997;84(11):1555–8.
York, NY: Springer, 2007, pp. 40–56. 53. Healy JC et al. AJR Am J Roentgenol. 1997;169(3):
16. Timmcke AE. Limitations of anal physiologic testing. 775–9.
In Hicks TC, Beck DE, Opelka FG, Timmcke AE (eds). 54. Matsuoka H et al. Dis Colon Rectum. 2001;44(4):571–6.
Complications of Colon and Rectal Surgery. Baltimore, 55. Bertschinger KM et al. Radiology. 2002;223(2):501–8.
MD: Williams and Wilkins, 1996, pp. 419–30. 56. Azpiroz F et al. Am J Gastroenterol. 2002;97(2):232–40.
17. Jacobs PPM et al. Dis Colon Rectum. 1990;33(6): 57. Minguez M et al. Gastroenterology. 2004;126(1):57–62.
494–7. 58. Heymen S et al. Dis Colon Rectum. 2003;46(9):1208–17.
18. Laurberg S et al. Br J Surg. 1998;75:786–8. Review.
19. Wexner SD et al. Dis Colon Rectum. 1991;34:22–30. 59. Ferrara A et al. Tech Coloproctol. 2001;5(3):131–5.
20. Buie WD et al. Dis Colon Rectum. 2001;44:1255–60. 60. Wexner SD et al. Dis Colon Rectum. 1993;36:139–45.
21. Kiff ES, Swash M. Br J Surg. 1984 August;71(8):614–6. 61. Rockwood TH et al. Dis Colon Rectum. 2000;43:9–17.
22. Ricciardi R et al. Dis Colon Rectum. 2006 June;49(6): 62. Deutekom M et al. Brit J Radiology. 2006;79:94–100.
852–7. 63. Bharucha AE, Fletcher JG. Gastroenterology. 2007;
23. Frudinger A et al. Abdom Imaging. 1998;23:301–3. 133(4):1069–74.
12
Limitations of colorectal imaging studies
ANDREW C. MATTHEWS AND CHARLES C. MATTHEWS
CHALLENGING CASE CONVENTIONAL IMAGING OF THE

ABDOMEN AND PELVIS
A 53-year-old women presents to the emergency
room with fever and left lower quadrant abdominal Classic radiographic examination of the colon includes
pain and tenderness. Her temperature is 39°C, and plain radiographs of the abdomen and single and double
her white blood cell count is 17,000 cells per cubic barium contrast enemas under fluoroscopy. The fluoro-
milliliter. What is the best radiologic test to confirm scopic examinations are almost an art form, requiring
her diagnosis? real-time interaction including the patient, the fluoro-
scope, the instillation of barium contrast, and the radiolo-
gist. Both of these modalities are projection techniques,
CASE MANAGEMENT rendering a three-dimensional (3D) object into a two-
dimensional (2D) image, and have limited soft tissue con-
A computed tomography (CT) scan of the abdomen
trast resolution (1).
and pelvis will evaluate her to confirm the diagnosis
Plain radiographs of the abdomen have limited diagnos-
of acute diverticulitis. In the absence of acute diver-
tic value. In a study of 1,000 consecutive emergency depart-
ticulitis, it may very well provide another explanation
ment patients, Ahn et al. evaluated the diagnostic finding
for her symptoms.
of plain radiographs and CT examinations in the setting of
nontraumatic abdominal pain. This retrospective review of
the radiology reports excluded 129 patients in whom a final
diagnosis was not reliably established. The review included
871 patients with abdominal radiographs and 188 patients
INTRODUCTION with abdominal CT scans. The overall findings on plain
radiographs were nonspecific in 68%, normal in 23%, and
Diagnostic imaging examinations are an important com- abnormal (specific diagnosis) in 10%. The radiographs had
ponent of the overall evaluation of diseases of the colon the highest sensitivity for detection of foreign bodies (90%)
and rectum. Today, multiple modalities are available to and bowel obstruction (49%). The radiographs had 0% sen-
examine the abdomen and colon including plain radio- sitivity for appendicitis, pyelonephritis, pancreatitis, and
graphs, contrast studies under fluoroscopy, nuclear medi- diverticulitis. The overall findings on CT scans were normal
cine (NM) scintigraphy, computed tomography (CT), in 20% and abnormal (specific diagnosis) in 80%. CT had
positron emission tomography (PET), magnetic resonance the highest sensitivities for bowel obstruction (75%), uroli-
imaging (MRI), and ultrasound. This chapter focuses on thiasis (68%), and pancreatitis (60%). These results demon-
two cross-sectional modalities—CT and MRI—with par- strate greater diagnostic yield with a CT exam (2) (Figures
ticular application to specific clinical problems. Diagnostic 12.1 and 12.2).
imaging examinations have variable accuracy—some find- Multiple clinical trials have evaluated the performance
ings are diagnostic while others are helpful in the differ- of double-contrast barium enema (DCBE), CT colonogra-
ential diagnosis but are not specific. Understanding the phy (CTC), and optical colonoscopy (OC) for the detection
strengths and limitations of each exam is important for of polyps and masses in the colon. Rockey et al. reported
the clinical physician to properly select what is best for the a series of 614 patients who had all three examinations.
patient. Their data indicated sensitivity for the detection of colonic
104
Cross-sectional imaging 105
Figure 12.1 Radiograph. Cecal volvulus. Markedly dis-

tended cecum with pneumatosis in wall.
Figure 12.3 Radiograph. Double-contrast barium enema.

Decubitus view demonstrates layering of barium liquid
and air distending the colon.
both CTC and OC. Their data indicated a sensitivity of 90%

for CTC for colonic lesions 10 mm or greater. Sensitivity
decreased with decreasing size of lesion (4) (Figure 12.3).
CROSS-SECTIONAL IMAGING
Cross-sectional imaging modalities include ultrasound,

PET, CT, and MRI. All of these came into clinical practice
in the 1970s and 1980s. CT and MRI were recognized as
such great advances in medicine that the inventors of each
were awarded the Nobel Prize in Medicine (Cormack and
Hounsfield for CT in 1979 and Lauterbur and Mansfield for
Figure 12.2 CT. Small bowel obstruction, S/P Roux-en-Y MRI in 2003) (5).
gastric bypass. Coronal image demonstrates fluid-filled
Cross-sectional imaging such as CT demonstrates addi-
and obstructed biliary limb in mid-abdomen.
tional anatomic information with improved soft tissue con-
trast providing the advantage of showing pathology in the
lesions 10 mm or greater as follows: DCBE, 48%; CTC, 59%; colonic wall and adjacent structures. The routine transverse
and OC, 98% (3). Because of the low sensitivity of DCBE, images can be supplemented with multiplanar reformations
its continued use to screen for polyps is controversial (1). in the coronal and sagittal planes. Today, the images are
Others have similarly reported very high sensitivity for OC routinely acquired at 5 mm or thinner sections. Nonionic
but better sensitivity for CTC. For example, the National CT iodinated contrast agents are routinely given intravenously
Colonography Trial of the American College of Radiology to enhance solid organs and vascular structures; the images
Imaging Network reported their results of the multicenter are typically acquired around 70 seconds postinjection for
trial involving 2,531 asymptomatic patients undergoing maximal enhancement of solid organs such as the liver.
106 Limitations of colorectal imaging studies
Gastrointestinal (GI) contrast is also routinely given by details vary among published protocols (9–12). Excellent
mouth to opacify the stomach and intestines. GI contrast vascular enhancement is performed with 125–150 mL non-
agents are usually 2% preparations of barium sulfate or ionic IV contrast at a flow rate of 4 mL/s. The timing of the
nonionic iodinated contrast; they are given 60–90 minutes scan varies from 30 (arterial phase) to 70 (portal venous
before the exam to allow transit through the intestines. The phase) seconds following the start of the injection. Many
basic CT scan is performed after the IV contrast injection, protocols use a single phase between these times, the enteric
but additional series may be done before IV contrast and phase at 45 seconds (10). High-level vascular enhancement
after several minutes of delay. CT evaluation of the GI tract is necessary to demonstrate mural hyperenhancement,
has high sensitivity for detection of pneumoperitoneum wall stratification (“target sign”), and engorged vasa recta
and leaks of oral contrast from the GI tract. Significant CT (“comb sign”) (9). The neutral density oral contrast is given
limitations include low sensitivity for early and superficial to distend the bowel lumen and aids visualization of muco-
mucosal abnormalities of inflammatory bowel disease and sal enhancement; this would be obscured by the traditional
depth of tumor invasion in the colon wall (6). high-density oral contrast. A large volume of oral contrast
is administered, typically 450 mL at 60 and 40 minutes and
INFLAMMATORY AND INFECTIOUS 225 mL at 20 and 10 minutes prior to the exam for a total of
BOWEL DISEASES 1350 mL. A commonly used agent is VoLumen by E-Z-EM
(11). Images are typically reconstructed from a helical
Inflammatory and infectious diseases of the colon have acquisition at 2.5 or 3 mm thickness; thin sections improve
numerous possible etiologies. These are usually regarded as the demonstration of small structures (Figure 12.4).
various types of “colitis.” The principal pathologic processes Paulsen et al. from Mayo Clinic reported their expe-
are infection, inflammatory bowel disease, and ischemia. rience with over 700 cases of CT enterography. The most
Other inflammatory processes that are not usually consid- common indication was suspected Crohn disease, and this
ered colitis are diverticulitis, epiploic appendagitis, stercoral was confirmed in about half of the patients. They report
colitis, and radiation enteritis. Infectious colitis may require that mural hyperenhancement, mural stratification, bowel
a stool culture for accurate diagnosis. Inflammatory bowel wall thickening (>3 mm), stranding in the mesenteric fat,
disease has two major forms—Crohn disease and ulcerative and engorged vasa recta correlate with active inflammation.
colitis. The typical presentation is cramping abdominal A minority of their patients also demonstrated associated
pain, fever, leukocytosis, and change in bowel habit, usually pathology outside the bowel, such as fistula and abscess
diarrhea. The prevalence in the general population in the or both (12). Another study from Mayo Clinic evaluated
United States is low—approximately 10–12 per 100,000 for small bowel enhancement characteristics and sensitivity
ulcerative colitis and 20–40 per 100,000 for Crohn disease. and specificity by comparing to endoscopic and histologic
For both diseases, most patients are young adults (7). results. The study involved 42 patients undergoing CT
Several general observations and patterns are helpful in enterography. Exams were performed with 150 mL of IV
the CT diagnosis of bowel diseases. The length of diseased contrast delivered at 4 mL/s with a 45-second delay. Their
colon is an important feature. Focal disease may represent
neoplasm, diverticulitis, epiploic appendagitis, and infec-
tion. Segmental disease usually indicates some type of coli-
tis—Crohn disease, ischemia, infection, or ulcerative colitis.
Diffuse disease usually indicates infection or ulcerative coli-
tis. Certain locations have a predilection for involvement by
certain processes: examples include amebiasis and Crohn
disease in the cecal region, ischemia in the splenic flexure
region, and ulcerative colitis and stercoral colitis in the rec-
tum. Enhancement patterns are also helpful—the target or
double halo pattern is typically benign with possibilities
including infection, inflammatory bowel disease, edema,
and ischemia (7,8).
CT AND MRI ENTEROGRAPHY

More advanced CT exam protocols have been developed to
better evaluate inflammatory bowel disease. The technique
commonly used today is termed CT enterography. The
major features of the technique are (1) excellent vascular
enhancement with IV contrast, (2) bowel distension with
low or neutral density oral contrast, and (3) thin-section
reconstructions in the transverse and coronal planes. The Figure 12.4 CT enterography. Normal small bowel.
Cross-sectional imaging / CT and MRI enterography 107
study concluded that jejunal attenuation is greater than of the disease and the typical young age of the patients
ileal attenuation, and collapsed bowel has a greater attenu- (13). A review in the New England Journal of Medicine by
ation than distended bowel. The most sensitive appearance Brenner and Hall reports a dramatic increase in the wide-
of Crohn disease was mural enhancement (73%–80%) fol- spread utilization of CT scans in the U.S. population from
lowed by mural thickening. The most specific findings of 1980 to 2006, estimated at 62 million scans in 2006. They
active disease were the comb sign of engorged vasa recta and estimated the lifetime risk of death from a single abdominal
increased attenuation of perienteric fat (10). Similar find- CT exam for a 50-year-old person at around 0.01%. The risk
ings were also reported by Baker et al. from the Cleveland for a 20-year-old is about six times higher (14).
Clinic study of 630 patients undergoing CT enterography DelGaizo et al. from Mayo Clinic reported multiple tech-
(11) (Figures 12.5 and 12.6). niques to reduce effective dose in CT enterography. These
Concerns about radiation exposure have been raised include the following: (1) reduce scan length to shortest
in the evaluation of patients with suspected or confirmed possible, (2) reduce tube voltage to 100 kV, (3) reduce tube
Crohn disease, particularly considering the chronic nature current based on patient size, and (4) utilize automated
(a)
(a)
(b)
(b)
Figure 12.5 CT enterography. Crohn disease with

obstruction. (a) Mural thickening and enhancement, Figure 12.6 CT enterography. Crohn disease with fistula
engorged vasa recta, and stranding of adjacent fat. and abscess. (a) Mural enhancement and thickening of
(b) Segmental stricture causing obstruction. terminal ileum. (b) Adjacent abdominal wall abscess.
exposure control. These techniques will reduce the total

exposure, noting that image noise will increase with reduced
dose (13). A trade-off or balance has to be reached to lower
the dose as low as possible while maintaining diagnostic
image quality. In recent years, new methods of image recon-
struction termed adaptive statistical iterative reconstruction
(ASIR) have allowed a lower-dose CT exam with equivalent
image quality. Kambadakone et al. from the Massachusetts
General Hospital investigated the use of ASIR technology
to achieve a low-dose CT enterography exam using an ASIR
setting of 30%. The images obtained were evaluated as opti-
mal diagnostic quality, and the radiation dose was reduced
to 34% (15). In our practice, we use more aggressive dose
reduction with ASIR 40%.
With a goal of reducing the risk of radiation exposure,
MRI enterography has been developed and evaluated in
comparison to CT. The major benefits of CT enterogra-
phy are better spatial resolution, fewer motion artifacts,
increased availability, lower cost, and shorter exam time.
The principal detraction for CT enterography is radiation
exposure. MRI enterography advantages include better
Figure 12.7 MRI enterography. Crohn disease.
tissue contrast resolution, better evaluation of perianal
Precontrast T2-weighted coronal image demonstrates
disease, better distinction between acute and chronic dis- thickening of terminal ileum.
ease, and no radiation exposure. Similar to CT, the MRI
enterography techniques utilize oral contrast for bowel
distension and IV contrast (gadolinium based) for bowel
wall enhancement. Most MRI protocols also use anti-
peristaltic agents. Multiplanar, multisequence images are
obtained prior to IV contrast; exact sequences are vari-
able and machine dependent. Most protocols use a liver
acquisition volume acceleration (LAVA) type sequence
around 15 and/or 60 seconds following injection of IV
contrast (16).
Comparing CT and MRI enterography, Siddiki et al.
from Mayo Clinic reported on a series of 30 patients with
suspected Crohn disease. MRI and CT enterography
showed excellent and similar sensitivities, 91% and 95%,
respectively (p = 0.32). CT was evaluated with higher
image quality. The authors concluded MRI enterography
may be the preferred imaging exam for patients with small
bowel Crohn disease who may be undergoing multiple
exams (17).
A more recent study on MRI and CT enterography by
Masselli et al. reported on 150 patients with suspected
small bowel disease. Their findings indicated sensitivity of
MRI enterography for all small bowel disease was signifi-
cantly higher compared to CT at 93% and 76%, respectively Figure 12.8 MRI enterography. Crohn disease of terminal
ileum. Postcontrast T1-weighted image shows thickened
(p = 0.0159). The sensitivities for detecting inflammatory
wall with increased enhancement and mural stratification.
bowel disease, however, were not significantly different
(MRI, 100%; CT, 96%; p > 0.99). MRI performed signifi-
cantly better for the detection of neoplastic disease (18) under age 40 increasing to >65% by age 85. Acute diverticu-
(Figures 12.7 through 12.9). litis is the most common complication of the disease, begin-
ning as a localized infection in a segment involved with
DIVERTICULITIS diverticulosis. The inflammation extends to the adjacent
pericolic area. Complications include perforation, abscess,
Diverticular disease is a common condition in the United and fistula formation (19). Zaidi and Daly found in their
States with increasing frequency by age, approximately <5% series of patients at an urban medical center in the United
Cross-sectional imaging / Diverticulitis 109
(a) diverticulitis on helical CT in their series of 312 patients at

the Massachusetts General Hospital. They found the most
sensitive findings for diverticulitis were bowel thickening
(96%) and fat stranding (91%). Less frequent but highly spe-
cific signs were fascial thickening and inflamed diverticula.
The diverticulitis was located in the sigmoid colon in 55%,
in the junction of the descending and sigmoid colon in 17%,
and in the descending colon in 16% (22). Abdominal CT is
generally considered the preferred imaging test for evalu-
ating patients with possible acute diverticulitis. Besides its
high accuracy for diverticulitis, CT is also an accurate test
for diagnosing alternative conditions including appendicitis
(19) (Figure 12.10).
(a)
(b)
(b)
Figure 12.9 MRI. Crohn disease with perianal fistula.

(a) Short T1 inversion recovery (STIR) sequence shows fis-
tula from perianal area to ischiorectal fossa. (b) In sagittal
plane, shows fistula extending to buttock cleft.
States that a high percentage of patients, 54%, with acute

diverticulitis were 50 years of age or younger and that 21%
were 40 or younger. Their conclusion suggests a rising fre-
quency of diverticulitis in younger and obese patients in the
United States (20).
Before the development of CT, barium enema was the
imaging test used to confirm the clinical diagnosis. The
ability of CT to diagnose diverticulitis was recognized early
in the development of this diagnostic tool. Hulnick et al. in Figure 12.10 CT. Acute diverticulitis. (a) Axial image
1984 described the findings and advocated the routine use shows marked thickening of descending colon with inflam-
of CT in patients with signs and symptoms of diverticuli- mation of adjacent fat. (b) Coronal image shows segmen-
tis (21). Kircher et al. reported on the individual signs of tal involvement of descending colon.
Abscess development is a recognized complication of was found in 14 patients for a prevalence of 2.1%. The same
diverticulitis that may be treated with antibiotics, inter- authors calculated from Surveillance, Epidemiology, and
ventional radiology percutaneous drainage, and/or sur- End Results (SEER) Program data a prevalence of colon
gery. Siewert et al. found abscesses in 30 (17%) of their 181 cancer in the general population of U.S. adults older than
patients with CT diagnosis of diverticulitis. Twenty-two 55 years at 0.68%. They concluded the broad recommen-
(73%) of those 30 patients had abscesses less than 3 cm in dation of follow-up colonoscopy to exclude colon cancer is
size and were successfully treated with antibiotics alone not justified (25). deVries et al. reviewed pooled data from
(p < 0.001). Patients with abscesses larger than 4 cm can nine published series including 2,490 patients with diver-
be managed with CT-guided abscess drainage followed by ticulitis. Follow-up colonoscopy was performed in 1,468
referral for surgical treatment. The majority of patients with patients (59%); colorectal cancer was found in 17 patients
the larger abscesses underwent surgery after resolution of for a prevalence of 1.16%. They concluded routine colonos-
symptoms (23) (Figure 12.11). copy follow-up is not necessary unless there are clinical
In the CT assessment of diverticulitis, some features signs of colon cancer or a patient over 50 years of age has
overlap with findings of colon cancer. Chintapalli et al. not had screening colonoscopy (26). In another review arti-
reported their retrospective and prospective results in a cle on radiologic and endoscopic imaging in the diagnosis
series of 58 patients with diverticulitis (n = 27) or colon and follow-up of colonic diverticular disease, Flo et al. also
cancer (n = 31). They found the most specific findings for concluded the role of colonoscopy after acute diverticuli-
diverticulitis were pericolonic inflammation (p < 0.01) tis remains controversial. While many practice guidelines
and length of segment greater than 10 cm (p < 0.012). from professional societies advise follow-up colonoscopy,
The most specific signs for colon cancer were presence of this is mainly supported by expert opinion and a lack of
lymph nodes (p < 0.001) and luminal mass (p < 0.003) robust scientific evidence. Systematic reviews show that
(24). As the CT features of acute diverticulitis can over- colonoscopy is generally not necessary (19).
lap those of colon cancer, professional societies such as
the American College of Gastroenterology have previ- GASTROINTESTINAL BLEEDING
ously recommended patients have follow-up colonoscopy
to rule out cancer. Sai et al. performed a literature review Acute and chronic GI bleeding has a variety of etiologies
for patients with a CT diagnosis of acute diverticulitis who and involves multiple medical specialties in diagnosis,
had surgery, colonoscopy, or barium enema within 24 management, and treatment. In 2009, the hospitalization
weeks. From the articles in the literature, the pooled data rate for GI bleeding in the United States was approximately
identified 771 patients that met the criteria. Colon cancer 60.6/100,000 for upper GI bleeding and 35.7/100,000
for lower GI bleeding. The overall hospitalization rate
decreased in the 2000s (27). The ligament of Treitz divides
the classification into upper GI bleeding proximal to this
level and lower GI bleeding distal to it. The most common
causes of upper GI bleeding are erosions or ulcers and vari-
ceal bleeding. The most common causes of lower GI bleed-
ing are diverticular disease, angiodysplasia, and neoplasm
(28). Investigation of GI bleeding is routinely performed
with esophagoduodenoscopy (EGD) and colonoscopy.
When neither of these endoscopic examinations has pro-
vided a diagnostic cause for the bleeding, it is considered
obscure GI bleeding (29). Patients typically become symp-
tomatic with tachycardia and hypotension when the acute
blood loss exceeds approximately 500 mL in a day. Most
bleeds will stop spontaneously with supportive measures
(28). In the era prior to the development of CT, catheter-
directed angiography and nuclear medicine (NM) were
used to diagnose causes of obscure GI bleeding. In a paper
published in 1971, Rosch et al. reported on 21 patients with
acute GI bleeding that were treated with selective arterial
infusion of the vasoconstrictors epinephrine and pitressin.
The GI bleeding was successfully controlled in the majority,
though not all, of patients (30). In another sentinel paper
Figure 12.11 CT. Acute diverticulitis with abscess. Acute from the 1970s, Bookstein et al. reported their results from
diverticulitis (white arrow) with adjacent 3 cm abscess transcatheter embolization in seven patients with lower GI
(black arrow) and inflammation of bladder dome. Patient bleeding. They used Gelfoam and Ivalon for the embolic
was successfully treated with IV antibiotics. agents to achieve control of bleeding in five of the patients;
Cross-sectional imaging / Gastrointestinal bleeding 111
the other two re-bled from angiodysplasia after several (a)

months (31). In a report from a larger and more contem-
porary series of 102 patients with acute nonvariceal GI
bleeding, Koo et al. achieved a clinical success rate in 77%
(78/102) using N-butyl-2-cyanoacrylate (NBCA) as the
embolic agent. Bowel infarction was a complication experi-
enced by two patients. Cancer-related bleeding was associ-
ated with increased clinical failure. NBCA is widely used as
an embolic agent in many other clinical settings besides GI
bleeding (32).
In recent years, CT has developed as a mainstream tool
for the noninvasive diagnosis of acute GI bleeding, particu-
larly for obscure bleeding. Two exam protocols have been
used—CT enterography and CT angiography (29,33,34).
Lee et al. reported a sensitivity of 55% and positive predic-
tive value of 100% for identifying the source of GI bleeding
with CT enterography. Diagnostic yield was higher in mas-
sive bleeding. CT enterography requires 8 hours of patient (b)
preparation time for fasting and the administration of a
neutral type of oral contrast agent (29). However, CT angi-
ography requires no special preparation and is suited for use
in the emergent setting. CT angiography is performed as a
three-phase examination with initial noncontrast series,
then postcontrast arterial and portal venous phases. This
exam needs to be performed on a modern scanner, typically
a 64 detector helical unit with thin sections approximately
1 mm, depending on vendor specifications. Importantly, no
oral contrast is given. IV contrast is administered as a dose
of 100–125 mL at a high flow rate of 4–5 mL/s to achieve
high levels of vascular enhancement. GI bleeding is diag-
nosed when extravasation of contrast is identified in the
bowel lumen. Noncontrast images are very important to
reveal opaque material that is often present in the GI tract
and which may mimic extravasation of contrast. The venous
phase helps confirm GI bleeding with increased volume (c)
and dispersion of the extravasated contrast in the bowel
lumen compared to the arterial phase. CT angiography
has a reported lower limit for the detection of bleeding at a
rate of approximately 0.5 mL/min. Of course, false-negative
exams can occur in the setting of intermittent bleeding (34)
(Figures 12.12 and 12.13).
Nuclear medicine scintigraphy with Technetium-99m-
labeled red blood cells (RBCs) is also a mainstream test
for the detection of GI bleeding with reported sensitivity
to detect bleeding at 0.1 mL/min. Strengths of the exam
include noninvasive modality and prolonged observation
time to detect intermittent bleeding. Weaknesses of the
exam include less accessibility for the test in off-hours and
inaccuracy in localizing the site of bleeding (35). In a retro-
spective study of 45 CT angiography and 90 RBC exami-
nations for suspected lower GI bleeding, Feuerstein et al.
found that CT angiography and RBC scintigraphy showed
Figure 12.12 CT angiogram. Rectal bleeding. (a) Initial
active bleeding in 38% of cases each. However, the site of noncontrast image shows fluid in rectum. (b) Arterial
bleeding was accurately localized in 30% of RBC scintigra- phase shows extravasation of contrast at anterior wall of
phy cases compared to 53% of CT cases (p = 0.008). Time rectum. (c) Coronal image confirms active bleeding with
to complete the exam after placement of the order was sig- pooling of contrast in rectal lumen. Colonoscopy diag-
nificantly shorter for CT at 1 hour 41 minutes compared to nosed a bleeding ulcer at this site.
(a) (b)
(c) (d)
Figure 12.13 CT angiogram; IR embolization. Bleeding pseudoaneurysm in duodenum. (a) Precontrast image. (b) Arterial
phase image shows pseudoaneurysm of gastroduodenal artery without active bleeding. (c) Catheter angiogram dem-
onstrates the pseudoaneurysm (white arrow) with active bleeding (black arrow) into duodenal lumen. Massive blood loss
required a 13-unit blood transfusion during procedure. (d) Angiogram shows successful embolization with microcoils.
3 hours 9 minutes for RBC scintigraphy (p < 0.001). These RECTAL CANCER STAGING
two advantages for CTA have important implications in the
treatment of the bleeding source (36). Advances in surgical techniques for the treatment of pri-
Today, a variety of embolic agents are now in clini- mary rectal cancer have generated a need for equal advances
cal use for transcatheter treatment of GI bleeding. These in preoperative imaging. High-resolution MRI has become
include microcoils, polyvinyl alcohol particles, Gelfoam, the essential imaging modality in the preoperative evalu-
NBCA, and embucrylate. A combination of agents is often ation of all rectal tumors, with MRI found to be superior
employed. Reported technical (angiographic cessation of to transrectal ultrasound (TRUS) and CT in preoperative
bleeding) and clinical (no bleeding for 30 days) success rates imaging (37). TRUS and MRI have comparable accuracy
are variable. Some reported results for clinical success rates in terms of T and N staging, but the limited field of view
range from 68% to 82% for upper GI bleeding and 81% to and operator dependence of TRUS yields MRI as the bet-
91% for lower GI bleeding (28) (Figure 12.14). ter alternative (38,39). CT lacks the detailed evaluation of
Cross-sectional imaging / Imaging interpretation 113
(a) IMAGING TECHNIQUE

Adequate evaluation of rectal tumors relies on high-reso-
lution T2-weighted images and a detailed clinical history
including tumor location (low, mid, or high). Although
3.0 Tesla magnets yield improved spatial recognition and
signal-to-noise ratio, studies have not yielded significant
differences in diagnostic accuracy between 1.5 T and 3.0 T
magnets (37). Typically there is no significant patient prepa-
ration prior to the examination, although antispasmod-
ics can be used to reduce bowel motion. Similarly, rectal
gel is another optional component to help define small or
polypoid tumors. Body or endorectal coils can be used for
image acquisition. Endorectal coils typically generate more
detailed evaluation of the rectal wall, but cost and patient
factors limit their use.
The essential portions of the examination are the high-
resolution T2-weighted sequences. Typically, this involves
a T2-weighted fast spin echo sequence with 3 mm slices.
(b) Initial sagittal T2-weighted images from pelvic sidewall
to sidewall are acquired for planning the orthogonal slices
(Figure 12.15a). It is important to know the tumor location
to accurately identify and plan the tumoral axis in which
the orthogonal slices will be based. The tumoral axis will be
used to plan axial orthogonal images perpendicular to the
tumor. Proper angulation is crucial to avoid volume aver-
aging and correctly identify tumor borders (Figure 12.15c).
Additional coronal T2-weighted images are obtained, which
provide valuable information on the relationship of low rec-
tal tumors to the anal sphincter (41).
Additional but nonessential sequences include T1,
diffusion-weighted, and postgadolinium sequences. T1
sequences are useful for the evaluation of incidental osseous
lesions in the pelvis such as metastases. Diffusion-weighted
images have shown promise in detecting colorectal tumors
(42) and in detecting lymph nodes, although they are not
highly specific in differentiating malignant versus hyper-
plastic benign nodes (43). The use of IV gadolinium con-
Figure 12.14 CT angiogram; IR embolization. (a) CT
trast has not typically been included in rectal tumor MR
angiogram shows active bleeding from jejunal ulcer. (b) IR evaluation, and current studies suggest no added benefit of
embolization image demonstrates embolization with contrast material in assessing tumor penetration through
microcoils. the rectal wall and into the MRF (44). Other studies have
suggested a benefit of gadolinium enhancement in detect-
the rectum and surrounding structures, as provided by the ing tumors and malignant lymph nodes (45,46). Currently,
other modalities. however, it appears gadolinium is not crucial or necessary
Although MRI can suggest the initial diagnosis of cancer, in the evaluation of rectal cancers.
its utility is the evaluation of previously histologically diag-
nosed rectal tumor to preoperatively assess depth of tumor IMAGING INTERPRETATION
invasion in regard to the muscularis propria (T1–3), local
involvement of adjacent structures (T4) including the meso- The radiologist’s duty is to convey numerous important
rectal fascia (MRF), and the morphology of pelvic lymph findings to the colorectal surgeon to guide surgical plan-
nodes. MRI acts to guide surgical planning and/or indicates ning or neoadjuvant treatment. The multitude of facts per-
the use of neoadjuvant treatments as a means to achieve taining to the lesion is best conveyed through a structured
curative treatment with negative margins on surgical resec- report, as recent studies have shown the benefit of a struc-
tion. Its full potential as an imaging tool requires a detailed tured report in improving the quality of MRI reports in
clinical history, prior colonoscopy, and details of the rectal rectal cancer staging (47). Information on the report should
tumor (40). include tumor location, tumor type (mucinous or not),
(a) (b)
(c)
Figure 12.15 MRI. (a) Sagittal T2-weighted image of the pelvis demonstrates planing of the tumoral axis with an arrow run-
ning down the length of the rectal tumor. Axial and coronal orthogonal images will be based off this new “tumoral axis.”
(b) Arrows depict the extent of the tumor, which straddles the peritoneal reflection as the tumor is identified above the
level of the urinary bladder. (c) Coronal orthogonal T2-weighted image shows the rectal tumor in plane.
relationship to the muscularis propria and MRF, involve- via colonoscopy/sigmoidoscopy as low (within 5 cm of the
ment of adjacent organs, and morphology and location of anorectal verge), mid (between 5 and 10 cm from the ano-
pelvic lymph nodes. The structured report at our institution rectal verge), or high (between 10 and 15 cm from the ano-
was developed as a collaborative effort among the radiology rectal verge). Although likely diagnosed histologically, the
and colorectal departments to ensure inclusion of pertinent lesion should first be characterized as mucinous or not, as
information (Figure 12.16). mucinous tumors usually present at a more advanced stage
(48). This is simply accomplished as mucinous tumors pres-
T-STAGING ent with higher intrinsic T2 signal intensity than nonmu-
cinous tumors (49). MRI is then utilized to determine the
The utility of MRI over other modalities, namely, CT and T-stage of the lesion (Figure 12.17).
TRUS, is the detailed evaluation of the rectum and sur- A rectal tumor, if not mucinous, typically demonstrates
rounding structures, such as the MRF, in determining initial an intermediate signal between the generally mildly hyper-
local extent of disease. Prior to imaging, the location of the intense submucosa and hypointense muscularis propria
tumor has usually been established by direct visualization on T2-weighted images (40). Diffusion-weighted imaging
Cross-sectional imaging / T-staging 115
PELVIS MRI RECTAL MASS PROTOCOL
Technique: Multiplanar FSE T2 sequences were performed.

Overall image quality:[Adequate Suboptimal Nondiagnostic]
Results:
Tumor Location: Tumor location (from anal verge):

[Low(0–5.0 cm) Mid (5.1–10.0 cm) High (10.1–15.0 cm)]
Relationship to anterior peritoneal reflection: [Above At or straddles Below Not able to assess]
Tumor Characteristics:
Circumferential extent/location (clock face): [ ]

Tumor Length : [ ] cm
Mucinous: [No]
Tumor Extent:
Does [not ]extend beyond muscularis into perirectal fat. [ ]
There is [not ]spiculation of the periectal fat. [ ]
There is [not ]invasion of adjacent structures. [ ]
The distance of [tumor] to the mesorectal fascia is [ ] cm. [ ]
[For low rectal tumors only: Is the lower extent of the tumor at or below the upper border of the puborectalis sling? If YES, for the most pen-
etrating component of the tumor below the upper border of the puborectalis sling, note:
● Possible confinement to the submucosa; no definite involvement of internal sphincter (suspected T1)
● Confined to the internal sphincter; no involvement of intersphincteric fat or external sphincter (early T2)
● Through the internal sphincter and intersphincteric fat; possible or definite involvement of the external sphincter (advanced T2)
● Through the external sphincter and into surrounding soft tissue; no organ involvement (T3)
● Through external sphincter and possible involvement of the adjacent organs (i.e., prostate, vagina) (T3/T4)
● Through external sphincter and definite involvement of adjacent organs (i.e., prostate, vagina) (T4)]
Lymph Nodes:
There are [no] perirectal lymph nodes [greater than 5mm]in the mesorectal fat. [The distance of the closest lymph node to the mesorectal
fascia is [ ] cm.] [ ]
There are [no] extra-mesorectal nodes lymph nodes in the visualized pelvis. [ ]
There are [no] lymph nodes that are suspicious based on [heterogeneous] [speculated] morphology. [ ]
There is [no] evident vascular invasion. [ ]
Impression:
[]
Figure 12.16 Structured report template for rectal cancer staging at our institution.
can also be used reliably to assist in the detection of rectal be confused with small penetrating vessels disrupting the
cancers (42) (Figure 12.18). It is usually not practical and muscularis propria (50). T3 tumors are further subdivided
typically unreliable to differentiate between T1 (isolated to by their depth of invasion (mm) past the muscularis propria
mucosa and submucosa) and T2 (involves muscularis pro- and whether the tumor involves the MRF. The MRF is the
pria) rectal tumors by MRI (Figure 12.19). The real utility of potential surgical resection plane, so involvement with the
MRI is differentiating T2 from T3 and T3 from T4 tumors. MRF (considered tumor within 1 mm) needs to be closely
T3 rectal tumors involve extension of the tumor beyond evaluated. MRF involvement requires therapy to downstage
the muscularis propria and into the surrounding peri- the lesion prior to surgery.
rectal soft tissues. The low rectum is surrounded by high In the more superior rectum toward the sigmoid colon,
signal intensity mesorectal fat encased by the MRF, which the peritoneum begins to cover the anterior portion of the
is represented as a thin hypointense line on T2-weighted rectum at the anterior peritoneal reflection. Superior to this
images (Figure 12.20). T3 involvement represents invasion point, the rectum becomes gradually enveloped until it is
into this surrounding fat. Invasion has been described as completely encircled by the peritoneum. It is important to
“broad based bulging or nodular configuration in continu- identify the peritoneal reflection in mid and high rectal
ity with the intramural portion of the tumor” (50) (Figure tumors, which is depicted as a thin hypointense line on
12.21). Tumor extension may be confused with a similarly sagittal T2-weighted images connecting the superior aspect
appearing benign desmoplastic reaction to underlying of the bladder to the anterior rectum (Figure 12.15b and c).
tumor, although differentiation may not be possible on Tumoral involvement into the peritoneum upstages the
MRI (40) (Figure 12.22). T3 involvement should also not lesion to T4. T4 rectal tumors are further classified as T4a,
Seminal
vesicle Bladder
Uterus
Prostate
T4
Muscularis
propria
T1
T3
Submucosa T2
Figure 12.19 MRI. Oblique coronal T2 image of the pelvis

demonstrates a rectal tumor (white arrow) that appears
Figure 12.17 Tumor staging in rectal cancer. Stage T1 confined to the submucosa and muscularis propria (T1/
tumors are confined to the submucosa; stage T2 tumors T2). Note the submucosa is relatively hyperintense, and
involve the muscularis propria (arrows), which consist the muscularis propria is hypointense. Adjacent prominent
of inner circular and outer longitudinal muscle layers; lymph node (black arrow).
stage T3 tumors extend beyond the muscularis propria;
and stage T4 tumors involve adjacent organs and/or the
peritoneum. (Republished image with permission from the
RSNA. Kaur H et al. Radiographics. 2012;32:389–409.)
Figure 12.20 MRI. Oblique axial T2 image of the pelvis

demonstrates a single prominent lymph node (white
Figure 12.18 MRI. Axial diffusion-weighted image (DWI) arrow) within 1 mm of the MRF, deemed MRF involve-
of the pelvis depicts a hyperintense rectal tumor. DWI can ment. MRF (black arrow) is depicted as a thin hypointense
be utilized for tumoral localization. line on T2-weighted MRI.
Cross-sectional imaging / T-staging 117
Figure 12.21 MRI. Oblique axial T2 image of the pelvis Figure 12.23 MRI. True axial T2 images of the pelvis show
shows spiculated extension (white arrow) of the underly- lobulated extension (white arrow) of the rectal tumor to
ing tumor into the adjacent mesorectal fat indicating T3 involve the vagina, indicating T4 disease.
involvement.
comparable ability in detecting lymph node involvement

as TRUS (51), but the ability of MRI to differentiate benign
from malignant nodes has been progressing. Location and
number of nodes are important to consider in preopera-
tive planning. Nodes located within the mesorectal fas-
cia will be removed during transmesorectal excision, but
those outside or involved with the MRF (defined as a node
within 1 mm of the MRF) may indicate further treatment
prior to surgery (Figure 12.20). An increasing number
of involved nodes also has portended a poor prognosis
and may also require neoadjuvant treatment. The abil-
ity to determine which if any of these nodes represents
malignant involvement will further guide preoperative
planning. Certainly the size and location of local nodes
have some prognostic abilities. Lymph nodes greater
than 5–8 mm tend to purport malignancy (37). The most
frequently encountered nodes in rectal cancer are those
in the surrounding mesorectal fat with the majority of
malignant nodes within 5 cm of the tumor (52) and typi-
cally located in the lateral and dorsal mesorectum at or
above the level of the tumor (53).
Figure 12.22 MRI. True axial T2 image of the pelvis shows Morphologic nodal characterization with MRI is also
hypointense spiculations (white arrow) extending into the a factor in determining malignant from benign nodes.
surrounding mesorectal fat from the anterior rectum to Irregular border morphology and mixed signal intensity
suggest T3 invasion or benign desmoplastic reaction. The increase accuracy in diagnosing malignant nodes (52).
two processes are not reliably distinguishable by MRI. Diffusion-weighted imaging (DWI) has also shown some
utility in detecting pelvic lymph nodes, but limited util-
involving the surface of the visceral peritoneum, or T4b, ity in determining their involvement with tumor (54,55).
involving adjacent organs or structures (Figure 12.23). Recent advances in lymph node–specific contrast agents
Another important factor in initial rectal cancer such as ultrasmall superparamagnetic iron oxide, an iron-
staging is the evaluation of regional pelvic lymph nodes based nanoparticle, and gadofosveset, a gadolinium chelate
including mesorectal and internal iliac nodes. MRI has that binds albumin, have shown promise in their ability to
detect malignant from nonmalignant nodes and appear to demonstrated similar per patient sensitivities in diagnosing
be viable methods of doing such (56,57). hepatic metastatic colorectal cancer (62). FDG-PET utilizes
Although not part of the TNM staging criteria, extramu- fluorinated deoxyglucose to identify areas of hypermeta-
ral vascular invasion is a significant finding that should be bolic activity, presumed areas of tumor activity (Figure
commented on during a rectal cancer staging examination. 12.24). FDG-PET is not without its limitations, especially
Vascular invasion has been shown to be a significant risk in the liver, as the baseline hypermetabolic activity in the
factor in local recurrence and disease relapse (58,59). MRI liver limits evaluation for small areas of increased uptake.
has proven reliable in identifying vascular tumoral involve- To that end, several studies have validated this result. In
ment as defined by irregularity and expansion of adjacent comparing CT and FDG-PET, CT was found more sensi-
vessels (59) warranting inclusion in rectal cancer staging tive in detecting hepatic metastases smaller than 1.5 cm
workups. (63). Studies have also shown that neoadjuvant chemother-
MRI is an invaluable tool in the initial workup of known apy lowers the sensitivity of FDG-PET in the detection of
rectal cancer. High-resolution T2 images provide valuable hepatic metastatic disease (64). When compared to MRI,
information to accurately determine tumor involvement however, CT and FDG-PET have shown to be inferior. In
and stage similar to transrectal ultrasound. MRI, however, a direct comparison with contrast-enhanced CT, MRI was
has the added benefit of a greater field of view to more accu- shown to have increased sensitivity to lesions less than
rately assess lymph node involvement and local metastatic 1 cm (65). When compared directly against FDG-PET,
disease to inguinal nodes or the pelvic osseous structures. MRI demonstrated greater sensitivity on a per lesion basis,
While providing details of nodal involvement, studies have and the same study concluded that MRI had the most data
indicated a role for MRI in differentiating benign versus supporting its use in detecting hepatic colorectal cancer
malignant nodes, which has the potential to be a tremen- over all modalities (62).
dous advantage in predicting surgical success or recurrent Metastatic lesions on MRI are typically iso- to hypoin-
disease. Continued advances in magnetic resonance tech- tense on T1-weighted images and iso- to hyperintense
nology should only further the utility MRI has in guiding on T2-weighted images (66), and further lose signal as
surgical workup and increasing the surgeon’s ability to pro- compared to hemangiomas and cysts on increasingly
vide tumor-free resection margins. T2-weighted sequences (67) (Figure 12.25a and b).
Specifically, colorectal metastases tend to show central
hypointensity secondary to necrosis with a surrounding
HEPATIC METASTATIC COLORECTAL rim of viable, hyperintense tumor (68). Contrast-enhanced
CANCER images with gadolinium-based contrast agents typically
Close to a quarter of all patients diagnosed with colon and

rectal cancer present with metastatic disease (60), and the
most common location for colorectal metastases is the liver
(61). Surgical treatment of hepatic metastatic disease has
proven to be a successful option in a small subset of patients,
but accurate diagnosis of the size, number, and extent of
disease is needed before potential surgical intervention or
neoadjuvant treatment can occur. Hepatic metastases can
be evaluated by a variety of imaging modalities including
ultrasound, CT, MRI, and fluorodeoxyglucose-positron
emission tomography (FDG-PET); but MRI has repeatedly
been shown as the superior imaging modality for the evalu-
ation of metastatic disease and specifically colorectal meta-
static disease.
A meta-analysis comparing the variety of imaging
modalities in diagnosing hepatic metastatic colorectal
disease was compiled by Floriani et al. and demonstrated
similar sensitivities and specificities of ultrasound and CT
in the diagnosis on a per patient basis (62). Ultrasound does
not utilize ionizing radiation but is limited by its small field
of view and is significantly operator dependent. CT can be
a valuable screening tool as it has the ability of a large field Figure 12.24 FDG-PET/CT. Axial image through the liver
of view acquired in a relatively short time period, but the demonstrates two hypermetabolic foci in the liver (white
repeated doses of ionizing radiation raise concern. In the arrows) indicating metastatic disease in this patient with
same study, the evaluation of FDG-PET was somewhat lim- colon cancer. Note the significant background activity of
ited by small sample size, but when compared to CT, it also the liver, which could potentially mask a smaller lesion.
Cross-sectional imaging / Hepatic metastatic colorectal cancer 119
show relative hypoenhancement of metastatic colorectal The increased sensitivities MRI affords have to be bal-
lesions, which are more conspicuous on the portal venous anced against the costs of the examination, which tend to
phase (Figure 12.25c). In addition to conventional MRI be significantly higher than CT or ultrasound. Another
sequences, more recent advances in diffusion-weighted factor to consider is the patient’s ability to hold his or
imaging and hepatobiliary contrast agents such as gadox- her breath for the acquisition (Figure 12.26). Although
etic acid (Gd-EOB-DTPA) for the evaluation of metastatic faster acquisition techniques have been generated, signifi-
liver lesions have only further increased sensitivity (69,70). cant patient motion during the examination can severely
In fact a recent study demonstrated that higher sensitivities degrade image quality and sensitivity. MRI is also contra-
were generated when applying both diffusion and post-hep- indicated in patients with certain implantable devices such
atobiliary contrast sequences (71). Hepatobiliary contrast as pacemakers or defibrillators. In such cases, other imag-
agents are taken up by hepatocytes and excreted in the bili- ing modalities must be pursued. Despite its limitations,
ary system. Colorectal metastases or other non-hepatocyte- current evidence supports MRI as the imaging modality
derived lesions will demonstrate relative hypoenhancement, most suited for the detection and quantification of colorec-
allowing visualization. tal liver metastases.
(a) (b)
(c)
Figure 12.25 MRI. (a) Axial noncontrast T2 image through the liver demonstrates an intermediate-intensity liver lesion
(white arrow) compatible with metastatic disease in contrast to an adjacent hyperintense cyst (arrowhead). Note the thin
rim of T2 hyperintensity around the metastatic lesion (white arrow) to suggest viable tumor surrounding central necrosis.
(b) Coronal noncontrast T2 image through the liver again shows the intermediate-intensity metastatic liver lesion (white
arrow). (c) Axial postcontrast T1 fat-saturated image through the liver demonstrates the relative hypoenhancement of the
metastatic colorectal lesion (white arrow). Also note the hypoenhancement of the adjacent cyst (arrowhead).
24. Chintapalli KN et al. Radiology. 1999;210:429–35.

25. Sai VF et al. Radiology. 2012;263:383–90.
26. de Vries HS et al. Surg Endosc. 2014;28:2039–47.
27. Laine L et al. Am J Gastroenterol. 2012;107:1190–5.
28. Laing CJ et al. RadioGraphics. 2007;27:1055–70.
29. Lee SS et al. Radiology. 2011;259:739–48.
30. Rosch J et al. Radiology. 1971;99:27–36.
31. Bookstein JJ et al. Radiology. 1978;127:345–9.
32. Koo HJ et al. AJR. 2015;204:662–8.
33. Graca BM et al. RadioGraphics. 2010;20:235–52.
34. Artigas JM et al. RadioGraphics. 2013;33:1453–70.
35. Zink SI et al. AJR. 2008;191:1107–14.
36. Feuerstein JD et al. AJR. 2016;207:578–84.
37. Beets-Tan RG et al. Eur Radiol. 2013 Sep;23(9):2522–31.
38. Bipat S et al. Radiology. 2004;232(3):773–83.
39. Fernandez-Esparrach G et al. GIE. 2011;74(2):347–54.
40. Jhaveri KS, Hosseini-Nik H. AJR. 2015;205:W24–55.
41. Brown G et al. Br J Radiol. 2005;78(927):245–51.
42. Ichikawa T et al. AJR. 2006;187:181–4.
43. Figueiras RG et al. AJR. 2010;195:54–66.
44. Vliegen RF et al. Radiographics. 2005;234:179–88.
Figure 12.26 MRI. Axial postcontrast T1 fat-saturated
45. Alberda WJ et al. Int J Colorectal Dis. 2013;28(4):
image through the liver demonstrates significant motion
related to the patient’s inability to hold the breath. 573–80.
Despite the motion blurring, several large hypoenhancing 46. Heijnen LA et al. Eur Radiol. 2014;24(2):371–9.
metastatic lesions are visible in this patient with colon 47. Sahni VA et al. AJR. 2015;205:584–8.
cancer. 48. Younes M et al. Cancer. 1993;72(12):3588–92.
49. Hussain SM et al. Radiology. 1999;213(1):79–85.
50. Taylor FG et al. AJR. 2008;191:1827–35.
REFERENCES 51. Bipat S et al. Radiology. 2004;232(2):773–83.
52. Koh DM et al. Eur Radiol. 2005;15(8):1650–57.
1. Sahani DV, Samir AE (eds). Abdominal Imaging. Mary 53. Engelen SM et al. Eur J Surg Oncol. 2008;34(7):776–81.
land Heights, Missouri: Saunders, 2011, Chapter 51. 54. Heijnen LA et al. Eur Radiol. 2013;23(12):3354–60.
2. Ahn SH et al. Radiology. 2002;225:159–64. 55. Cho EY et al. Eur J Radiol. 2013;82(11):e662–8.
3. Rockey DC et al. Lancet. 2005;365:305–11. 56. Koh DM et al. AJR. 2010;194(6):W505–13.
4. Johnson CD et al. N Engl J Med. 2008;359:1207–17. 57. Lambregts DM et al. Abdom Imaging. 2013;38(4):
5. Nobel Prize. Available from https://www.nobelprize. 720–7.
org/nobel_prizes/ 58. Dresen RC et al. Eur J Surg Oncol. 2009;35(10):
6. Sahani DV, Samir AE (eds). Abdominal Imaging. Mary 1071–7.
land Heights, Missouri: Saunders, 2011, Chapter 52. 59. Smith NJ et al. Br J Surg. 2008;95(2):229–36.
7. Sahani DV, Samir AE (eds). Abdominal Imaging. Mary 60. van der Geest LG et al. Clin Exp Metastasis. 2015;
land Heights, Missouri: Saunders, 2011, Chapter 55. 32(5):457–65.
8. Balthazar EJ. AJR. 1991;156:23–32. 61. Riihimaki M et al. Sci Rep [Internet]. 2016 [cited January
9. Elsayes KM et al. RadioGraphics. 2010;30:1955–74. 12, 2017];6:Article 26975 [about 26 screens]. Available
10. Booya F et al. Radiology. 2006;241:787–95. from: http://www.nature.com/articles/srep29765
11. Baker ME et al. AJR. 2009;192:417–23. 62. Floriani I et al. J Magn Reson Imaging. 2010;31(1):19–31.
12. Paulsen SR et al. RadioGraphics. 2006;26:641–62. 63. Ruers TJ et al. J Clin Oncol. 2002;20(2):388–95.
13. Del Gaizo AJ et al. RadioGraphics. 2013;33:1109–24. 64. Lubezky N et al. J Gastrointest Surg. 2007;11(4):472–8.
14. Brenner DJ, Hall EJ. N Engl J Med. 2007;357:2277–84. 65. Niekel MC et al. Radiology. 2010;257(3):674–84.
15. Kambadakone AR et al. AJR. 2011;196:W743–52. 66. Saini S, Nelson RC. Radiology. 1995;197:575–7.
16. Towbin AJ et al. RadioGraphics. 2013;33:1843–60. 67. Namasivayam S et al. Cancer Imaging. 2007;7:2–9.
17. Siddiki HA et al. AJR. 2009;193:113–21. 68. Sica GT et al. Clin Liver Dis. 2002;6(1):165–79.
18. Masselli G et al. Radiology. 2016;279:420–31. 69. Wu LM et al. Eur J Cancer. 2013;49(3):572–84.
19. Flor N et al. AJR. 2016;207:15–24. 70. Chen L et al. PLOS ONE [Internet]. 2012 [cited
20. Zaidi E, Daly B. AJR. 2006;187:689–94. January 13, 2017];7(11):e48681 [about 10 screens].
21. Hulnick DH et al. Radiology. 1984;152:491–5. Available from: http://journals.plos.org/plosone/
22. Kircher MF et al. AJR. 2002;178:1313–8. article?id=10.1371/journal.pone.0048681
23. Siewert B et al. AJR. 2006;186:680–6. 71. Vilgrain V et al. Eur Radiol. 2016;26(12):4595–615.
13
Transanal endoscopy
AARON L. KLINGER AND BRIAN R. KANN
indications. Knowledge of potential complications and their

CHALLENGING CASE management is critical to successful patient management.
A 50-year-old man with a history of hypertension pres-

ents for screening colonoscopy. A 2 cm sessile polyp
is found in the cecum and removed with a “hot” snare. BOWEL PREPARATION
He presents to the emergency department 2 days
later with fevers to 101.5°F and abdominal pain with
INTRODUCTION
localized guarding over the right lower quadrant.
Although significant debate continues regarding the benefit
of preoperative bowel preparation for surgery, it is univer-
sally accepted that adequate bowel cleaning is crucial to
CASE MANAGEMENT proper colonoscopy. Despite this up to 25% of all colonos-
copies are reported to have inadequate bowel preparation
You see and examine the patient. Blood is drawn (1). If adequate mucosal examination cannot occur, a repeat
for complete blood count (CBC), basic metabolic procedure with additional costs and risks must occur. An
profile, and C-reactive protein (CRP). His white blood ideal preparation must be safe, tolerable, and effective.
cell (WBC) count returns at 14 × 103/mL, and CRP Various preparations and methods of consumption exist,
is 30 mg/dL. An upright chest x-ray does not show but consensus to the optimal regimen is lacking (2). The U.S.
pneumoperitoneum, and a computed tomography Multi-Society Task Force (USMSTF) on Colorectal Cancer
(CT) scan with rectal contrast does not show extralu- (CRC) published its recommendations for precolonoscopy
minal air or contrast extravasation. You admit the bowel cleaning in 2014. They concluded that the selection of
patient, start him on broad-spectrum antibiotics, a bowel-cleaning regimen should be determined based on a
and order bowel rest. Serial abdominal exams are patient’s history, other medications, and response to prior
performed to watch for generalized peritonitis. After bowel preparations (3).
3 days, the patient’s pain resolves and WBC normal-
izes. He is started on a clear liquid diet and advanced
POLYETHYLENE GLYCOL
slowly. He is discharged home on hospital day 5.
Polyethylene glycol (PEG) is an inert osmotically active
polymer commonly used for bowel preparation. It is gen-
erally mixed with an electrolyte lavage solution (ELS) to
INTRODUCTION a volume of 4 L and functions as an electrolyte-balanced
colonic lavage solution. “Low-volume” PEG regimens also
Endoscopic evaluation of the lower gastrointestinal (GI) exist, consisting of 2 L of PEG-ELS with an adjunct medica-
tract is a vital tool for the colorectal surgeon and should tion such as Bisacodyl or ascorbic acid.
be viewed as an extension of the physical exam. Modalities
include anoscopy, rigid proctoscopy, flexible sigmoidos- SODIUM PHOSPHATE
copy, and colonoscopy. Each procedure has its own benefits
and limitations. It is important for a colorectal surgeon to be Sodium phosphate (NaP) formulations are hyperosmotic
comfortable performing each technique and to know their compounds that draw water into the GI tract to create
121
122 Transanal endoscopy
bowel cleaning through a purgative action. Proponents of elective colonoscopy with a same-day regimen as an accept-
NaP refer to better patient tolerance compared to PEG with- able alternative especially for afternoon endoscopy. They also
out sacrificing bowel cleaning (4,5). Given its mechanism emphasize the importance of patient education for increased
of action, NaP can result in significant fluid and electrolyte compliance and quality of cleansing and recommend the use
shifts. Patients must take a significant quantity of fluids of both oral and written instructions as well as additional sup-
orally if using this preparation to prevent dehydration. For port measures to ensure correct preparation use (3).
this reason, NaP should not be used in elderly patients or
those with congestive heart failure, decompensated cirrho-
sis, renal failure, or electrolyte disorders (6).
Reports of chronic renal failure or insufficiency from ANOSCOPY
acute phosphate nephropathy led the U.S. Food and Drug
Administration (FDA) to issue an alert regarding its use INDICATIONS
in 2006 (7). Last, NaP can cause mucosal lesions including
aphthous ulcers and should be avoided in patients being Anoscopy is used for examination of the anal canal and
evaluated specifically for inflammatory bowel disease (8). distal rectum following digital rectal exam (DRE). This
When used in a carefully selected patient, NaP remains a procedure allows for evaluation of the anoderm, distal
good option for precolonoscopy bowel cleansing. However, rectal mucosa, dentate line, and anal pathology, including
the limitations discussed previously have led manufacturers internal/external hemorrhoids, fistulas, fissures, and anal
to withdraw many of the NaP products from the market (9). masses. The device consists of a scope, often with a beveled
end and an obturator. Light can be provided by a built-in
light source, by a fiber-optic light cable, or from a separate
IRRITANT LAXATIVES (SODIUM light source. Disposable plastic anoscopes with built-in light
PICOSULFATE) sources are commercially available and are useful for bed-
side or emergency room examination.
Sodium picosulfate is a contact stimulant laxative that can
either be used as a treatment for constipation or to prepare
the large bowel before colonoscopy or surgery. Sodium TECHNIQUE
picosulfate is a prodrug, and has no significant direct physi-
The patient does not need any special preparation for ano-
ological effect on the intestine. Instead, it is metabolized
scopic examination. The patient is placed in a prone jack-
by gut bacteria into the active compound 4,4′-dihydroxy-
knife or Sims’ position. Following DRE, the anoscope is
diphenyl-(2-pyridiyl) methane, which increases peristalsis
lubricated and slowly inserted with the examiner’s thumb
in the gut. Sodium picosulfate is typically prescribed in a
supporting the obturator while the patient bears down to
combined formulation with magnesium citrate, which is an
assist in relaxation of the anal musculature. The obturator
osmotic laxative. The use of sodium picosulfate can be asso-
is then removed, and a circumferential examination is per-
ciated with clinically significant hyponatremia and hypo-
formed. Prior to turning the anoscope the obturator should
kalemia, and patients should be encouraged to drink large
be reinserted to avoid pinching the anoderm. As the ano-
amounts of clear fluids as well as rehydrate to reestablish
scope is withdrawn, the patient is again asked to bear down
the electrolyte balance. In the United States, this drug com-
to evaluate for any prolapsing hemorrhoids.
bination is marketed as Prepopik (Ferring Pharmaceuticals,
Parsippany, NJ).
COMPLICATIONS
SINGLE VERSUS SPLIT DOSING Complications following anoscopy are rare. Bleeding may
occur from irritation of hemorrhoids or from accidental
Whatever preparation formulation is chosen, there is good tears to the anoderm.
evidence to support using a split-dose protocol. A 2014 meta-
analysis found good preparation in 85% of patients using a split
dose compared to 63% in patients taking a single dose. They
found the most important factor to be the “runway time”—the RIGID PROCTOSCOPY
time between the last dose of preparation and the beginning of
the colonoscopy. The benefit of split dose is lost if runway time INDICATIONS
exceeds 5 hours (2). Consistent with this finding, a 2012 sin-
gle-blinded study evaluated split-day preparation with same- Rigid proctoscopy (to be referred to as simply “proctoscopy”
day preparation for afternoon colonoscopies. They found a for the remainder of this chapter) allows for visualization
same-day regimen (doses at 7:00 and 10:00 a.m.) to result in of the anus, rectum, and possibly sigmoid colon. Like the
improved mucosal cleansing with greater patient satisfaction anoscope, it is made of a scope, generally 25 cm long, with
than a two-day regimen (10). As a result, the USMSTF’s guide- an obturator, a light source, a viewing lens, and a bellows
lines strongly recommend the use of split-dose cleansing for to allow for gentle air insufflation. Additional instruments
Flexible sigmoidoscopy/colonoscopy / History, introduction, and indications 123
such as suction, cotton-tipped swabs, and biopsy tools can flexible endoscopy. In a recent study, the average difference
be passed through the scope. Like anoscopes they come in a between rigid and flexible measurements of tumors was
variety of adult and pediatric sizes and materials. Disposable found to be 3.1 cm in the upper rectum and 5 cm at the rec-
scopes, made of clear plastic, may allow for additional visu- tosigmoid junction. The addition of proctoscopy to flexible
alization and do not need to be cleaned following use. endoscopy changed the treatment options in up to 38% of
rectal/rectosigmoid cancer patients (14).
TECHNIQUE
COMPLICATIONS
A full bowel preparation is not typically required for proctoscopy.
Rather one or two enemas should be given within 1–2 hours of Serious complications of proctoscopy are rare. The most
the procedure to clear stool for easy visualization and passage of common complications are bleeding, especially at biopsy
the instrument. Enema use earlier than this period may allow or polypectomy sites. Patients may have vasovagal reac-
stool from the more proximal bowel to migrate to the rectum. tions during the procedure. If a patient becomes faint or
The procedure can generally be performed without anesthesia, unresponsive, the scope should be removed and the patient
although some patients may require sedation. should be taken out of the jackknife position, placed supine
The patient is placed in a prone jackknife or Sims’ posi- on the table, and oxygen should be provided. Perforation is
tion. Patients may have significant anxiety about this proce- very rare and should not happen if careful technique is uti-
dure, and it is important to explain that it will not hurt and lized. In a review of 20 years of experience and over 300,000
that each step will be explained before being performed. proctoscopies at the Mayo Clinic, only four bowel injuries
Following DRE, a well-lubricated scope is inserted into the occurred and only one involved bowel perforation (15).
anal canal, aiming toward the umbilicus with the opera-
tor’s thumb holding the obturator. Once past the sphincter
muscles, the obturator is removed and the scope is reposi-
tioned to face the sacrum and is inserted into the rectum.
FLEXIBLE SIGMOIDOSCOPY/
Suctioning or swabbing to clear rectal contents is performed COLONOSCOPY
with the examiner’s right hand while the left hand holds and
advances the scope. If stool cannot be cleared with these HISTORY, INTRODUCTION, AND
techniques, the rectum can be cleared by inserting 50 mL of INDICATIONS
tap water, closing the lens, and providing insufflation.
At about 14 cm, the rectum turns anteriorly and the Documented attempts to use scopes to visualize the interior
lumen may be difficult to follow. At this point the viewing of body cavities have been made since the early 1800s. Early
lens can be closed and gentle insufflation can be provided to efforts were limited by lack of a reliable light source. Within
distend the bowel to allow further navigation. It is important just a few years of Edison’s introduction of the electric light
to recognize that although most proctoscopes are 25 cm in bulb, the technology was used for endoscopic purposes. The
length, it is not always possible to advance to this distance, introduction of modern flexible endoscopy came with the
and attempts to do so may cause the patient serious harm. advent of fiber optics in the 1960s and videoendoscopy in
In fact, a 1980 review of 1,000 proctoscopes performed at the 1980s (16).
the University of Minnesota found the average depth of Flexible lower endoscopy (flexible sigmoidoscopy and
insertion to be 19.5 cm in males and 18.6 cm in females (11). colonoscopy) is a commonly used procedure that can diag-
If a fixed angle is encountered or the scope will not pass eas- nose and treat conditions of the colon, rectum, and distal
ily, advancement should stop. Blanching mucosa should be small intestine. It should be viewed as an extension of the
taken as a “danger sign.” Attempts to force the scope further colorectal physical examination and is an indispensable tool
may result in pain or complications (12). for the colorectal surgeon. These techniques allow for direct
Although lesions encountered on insertion should be inspection of the mucosal surface, allowing the endoscopist
noted, full mucosal inspection should be done as the scope to visualize and treat neoplasms, polyps, bleeding lesions,
is withdrawn. All areas of the bowel wall should be exam- and inflammatory bowel disease. Successful colonoscopy is
ined, suctioning as needed, as the scope is withdrawn in a dependent on adequate sedation, bowel preparation, patient
spiral motion. The endoscopist must be mindful of polyps, selection, and operator experience.
inflammation, tumors, or other abnormalities. Flexible sigmoidoscopies are 35–65 cm in length (colono-
With the widespread availability of flexible sigmoid- scopes in comparison are 120–160 cm long). They are able
oscopy, the utility of proctoscopy has been questioned. to evaluate more bowel length than rigid sigmoidoscopy
Opponents note decreased rates of polyp detection, patient (although less than full colonoscopy), thereby increasing the
discomfort, ease of biopsy, and the ability of flexible scopes chances of finding a lesion. In addition, given the flexible
to be inserted further (13). Proctoscopy has proven itself nature of the scope and smaller diameter, it is generally more
invaluable in the localization of rectal tumors, however. comfortable for the patient. Disadvantages compared to
A proctoscope has length markings on its outer surface rigid proctoscopy include the higher cost of the instruments,
and offers a more accurate estimate of lesion location than the need to clean after use, and the higher tendency for the
instruments and their fiber optics to break. Unlike full colo- benzodiazepines (midazolam and diazepam), which are
noscopy, this technique can be performed without full bowel generally given in conjunction with an opiate such as fen-
preparation and generally does not require sedation. tanyl (21). The most common alternative regimens consist of
The indications for colonoscopy vary by condition and hypnotics such as propofol, which provides deep sedation.
are often debated even among experts. Screening for CRC The use of propofol has increased steadily in recent years
remains a primary use for the technique, and various appro- and has surpassed midazolam in some populations (22).
priateness guidelines for screening colonoscopy have been Advantages of propofol include reduced procedure time,
published. In 2008, both the USMSTF on CRC and the reduced recovery time, and higher physician satisfaction.
American College of Gastroenterology (AGA) published Furthermore, patients report greater satisfaction with its
their guidelines for CRC screening (17,18). Both suggest use compared to midazolam with fentanyl and are less likely
screening of average-risk individuals starting at age 50 to to experience nausea or “hangover” symptoms (22). Unlike
detect and prevent CRC. Although other screening tools benzodiazepines and opioids, there are no reversal agents
exist, the AGA notes a 70%–90% reduction in CRC for available for propofol, but it has a short (4–8 minute) dura-
patients undergoing colonoscopy and polypectomy com- tion of action (19).The major disadvantage of propofol is its
pared to reference populations and the association of colo- narrow therapeutic index. The FDA therefore recommends
noscopy use with earlier and more favorable stages of CRC it only be administered by a trained anesthesia specialist.
presentation. Furthermore, they refer to its wide availability, A recent meta-analysis, however, found propofol to have a
ability to examine the entire colon, and longer test interval similar risk of cardiopulmonary adverse events compared
than other screening options. For these reasons, colonos- to other sedatives. When used in simple endoscopic proce-
copy remains the AGA’s “preferred” screening strategy and dures, it is actually associated with a decrease in complica-
should be the test offered to patients, saving other modali- tions. In advanced endoscopic procedures, its use does not
ties for patients who decline colonoscopy (17). result in increased rates of complications compared to seda-
tion with benzodiazepines/narcotics (23).
TECHNIQUE A major perceived disadvantage of propofol is an
increased cost of its use compared to traditional sedatives.
Sedation: Conscious sedation versus Both propofol and midazolam are inexpensive drugs, and
propofol: Comment on cost efficiency propofol has been shown to reduce costs drastically by short-
ening recovery times (24). Due to concerns of increased risks
While select colonoscopies can be performed without the of propofol use and the FDA’s recommendations regarding
use of sedation, most patients are given IV medication to its use, endoscopists are less likely to self-administer pro-
reduce discomfort and facilitate the procedure (19). If seda- pofol than midazolam (19). The use of a trained anesthesia
tion is to be used, it is important that the patient be provided provider for endoscopy results in additional professional
with oxygen and has routine monitoring of blood pressure, fees, which can be as high as $450 per procedure (25).
heart and respiratory rate, pulse oximetry, and electro- Despite FDA recommendations and endoscopist con-
cardiogram, as well as postprocedural monitoring in the cerns, a growing body of literature supports the use of endos-
recovery area. There are four generally recognized stages copist-directed propofol (EDP) administration. A review of
of sedation, which range from minimal sedation to general over 220,000 cases of EDP over 5 years found no instances
anesthesia. Colonoscopy is generally performed under mod- of death and only one reported instance of intubation (26).
erate sedation (“conscious sedation”), though some patients The body of supporting data led the American Society
may require deep sedation or even general anesthesia. for Gastrointestinal Endoscopy (ASGE), the American
Prior to the administration of any sedation, it is Gastroenterological Association, and the American College
important to evaluate the patient for preexisting medi- of Gastroenterology to release a joint statement endorsing
cal conditions or physical exam findings that may lead to EDP so long as the provider had been adequately trained in
compilations of sedation. The patient’s American Society of the drug’s use (27).
Anesthesiologist (ASA) classification should be determined.
Higher-risk patients (ASA class IV–V) should be evaluated
by an anesthesia specialist prior to administration of seda- Maneuvers for difficult colonoscopy:
tives. Women of childbearing age should be screened for Change position (supine/prone), abdominal
pregnancy. If endoscopy is necessary during pregnancy, pressure, etc.
attempts should be made to postpone until the second tri-
mester if possible, and an obstetrician should be consulted Colonoscopy is a technically challenging procedure, and
(20). Patients should fast for at least 2 hours following clear expertise requires adequate training, practice, patience, and
liquids to reduce the risk of aspiration (19). knowledge of maneuvers. Technical challenges can occur
Ideal drugs for endoscopic evaluation have rapid onset, due to redundant or floppy bowel, sharp angulation, or fixa-
short duration of action, do not alter hemodynamics, and tion from previous surgery.
have low side-effect profiles. The most commonly used It is important for an endoscopist to constantly use both
medications for conscious sedation in the United States are hands during a procedure. While the left hand controls
Technical complications 125
the direction of the tip with control knobs, the right hand endocarditis and known GI tract infections that may include
works to both advance and withdraw the scope and provide enterococci. The guidelines also recommend prophylaxis for
torque. Rotation of the scope not only changes the viewing active peritoneal dialysis patients, but they note that quality
angle but also acts to help prevent or reduce loops. evidence for this recommendation is lacking (30).
Dithering refers to rapid back-and-forth jerking move-
ments, where the endoscope is advanced and slowly with- Patients on antithrombotic medications
drawn. This can result in pleating of the bowel over the
scope rather than pushing it away and results in a shortened An endoscopist should expect to encounter patients on
colon length. This technique can be especially useful in the antiplatelet or anticoagulation medications, especially as
descending and transverse colon and should be used even more antithrombotic medications become available and
during portions of the procedure that are not difficult. more patients are being treated with them. It is important to
External pressure can be applied by an assistant or the screen for use of these medications as well as the associated
endoscopist to the abdomen in attempts to splint areas of indication prior to scheduling a procedure. Consideration
redundant bowel and help prevent or reduce loop forma- must be given to holding aspirin, clopidogrel bisulfate and
tion. The presence of a formed loop can help guide pres- other antiplatelet agents, and anticoagulants prior to colo-
sure placement. In cases when no loop can be felt, pressure noscopy, especially if a higher-risk therapeutic procedure
should generally be applied from the right upper quadrant (such as polypectomy or mucosal resection) is planned. It
toward the left lower quadrant. This maneuver can stabi- is important to recognize that each medication has its own
lize the redundant sigmoid colon, but care must be given to duration of action, and therefore, “hold times” vary from
keep pressure gentle to avoid patient discomfort. Significant drug to drug.
pressure may also contribute to gastroesophageal reflux and The 2016 ASGE guidelines (available at www.asge.org)
potentially aspiration. include a table of commonly used antithrombotic agents,
When other maneuvers fail and the endoscope cannot their durations of action, and reversal agents if available.
be inserted further, repositioning the patient may assist Given the wide range of medications and indications,
with scope advancement. Colonoscopy generally starts with they do not provide a “one-size-fits-all” approach to hold-
the patient in a left lateral decubitus position. Turning the ing antithrombotic medications. The endoscopist along
patient to a supine position may assist with further naviga- with the patient’s cardiologist or primary care provider
tion. During turns, the endoscopy team should move the must weigh the risk of endoscopy-induced bleeding to the
patient together while the endoscopist maintains control patient’s own cardiovascular risk. Generally speaking, the
of the endoscope and attempts to keep the scope in the morbidity associated with thrombotic events outweighs
middle of the bowel lumen. In some cases, especially with that of bleeding. When possible, elective procedures should
obese patients, turning the patient to a prone position can be delayed until temporarily antithrombotic drug courses
assist with advancement. Failure to be able to comfortably have been completed.
advance the scope despite the use of these maneuvers should In general, per ASGE guidelines, patients with low throm-
be taken as an indication to abort the procedure. boembolic risk and high bleeding risk should have their
medications held for the appropriate drug-specific interval.
Need for prophylactic antibiotics Heparin bridging therapy to be held in the immediate peri-
procedural period should be considered for those at high
Cases of infectious endocarditis following colonoscopy have risk of bleeding and thrombotic events. Antithrombotic
been reported, and given the high concentration of gram- medications should be resumed as soon as possible after the
positive bacteria in the colon, some argue that prophy- procedure (30).
laxis should be given to select high-risk patients including The most recent guidelines from the American College
elderly patients, cancer patients, and immunocompromised of Chest Physicians are similar to the ASGE guidelines.
patients (28). However, the most recent (2008) guidelines They suggest using heparin bridging therapy for patients at
from the American College of Cardiology and American high risk of thromboembolism (to be held 4–6 hours before
Heart Association do not recommend endocarditis prophy- the procedure) and continuing aspirin or holding it at the
laxis for any GI tract procedures including colonoscopy. time of surgery (as opposed to 7–10 days prior) for moder-
They acknowledge that transient bacteremia may occur ate to high cardiovascular risk patients (31) In contrast, low
but that there are few cases of infectious endocarditis fol- cardiovascular risk patients should have aspirin held 7–10
lowing colonoscopy and that evidence that antibiotics can days prior to their procedure.
prevent endocarditis in these cases is lacking (29). Also, the
risk of Clostridium difficle following antibiotic usage appears
higher than the potential risk of bacteremia. The 2015 guide-
lines from the ASGE recommend against routine antibi- TECHNICAL COMPLICATIONS
otic prophylaxis as well. They do consider pre-endoscopy
prophylaxis for patients with high-risk cardiac conditions Although colonoscopy is very commonly performed and
such as new prosthetic heart valves or previous infectious considered a low-risk procedure, it must be remembered
that it is an invasive procedure and does carry an associated healthy patients with well-prepped bowel following polypec-
risk. The most commonly described serious complications tomy, it may be safe to admit the patient and treat nonopera-
of colonoscopy are bleeding, perforation, infectious com- tively. It should be mentioned, however, that patients treated
plications, missed lesions, and cardiopulmonary events. operatively within 24 hours of perforation are less likely to
A Centers for Disease Control and Prevention publication develop peritonitis, require bowel resection or stomas, and
estimates the incidence of serious complication at 2.38 per have complications (35,36). Perforations from diagnostic
1,000 screening colonoscopies with rates in previous studies endoscopy are more likely to be larger and thus less likely to
ranging from 0.79 to 8.4 per 1,000 procedures (32). be successfully treated with conservative measures.
Patients being treated conservatively should be admitted,
HEMORRHAGE started on antibiotics, have a nasogastric tube placed, and be
put on bowel rest. Development of peritonitis is an indica-
Bleeding is the most common serious complication of colo- tion for surgical exploration. Likewise, patients presenting
noscopy with a reported incidence as high as 3%, though late after colonoscopy are more likely to require operative
most recent publications report the incidence of serious intervention. Reports of successful repairs of perforations
bleeding at less than 3/1,000 (32,33). Hemorrhage usually at the time of endoscopy using clips have been made as well
comes from an intraluminal source, but it is also possible (37,38). The most common signs and symptoms of perfora-
for bleeding to occur from mesenteric or splenic lacerations. tion are abdominal pain, tachycardia, and leukocytosis (39).
Postpolypectomy bleeding occurs an average of 6 days after A CT scan with rectal contrast can help to confirm the diag-
the procedure and can present up to 2 weeks later, though nosis and may help guide operative intervention.
most bleeding is recognized and controlled at the time of
the procedure. Risk factors for bleeding include polypec- POSTPOLYPECTOMY SYNDROME
tomy, which increases with the size of the polyp, and anti-
coagulation use (34). Postpolypectomy syndrome (sometimes called postpoylpec-
Management of post-colonoscopy hemorrhage depends tomy coagulation syndrome or transmural burn syndrome)
on the etiology of the bleeding source and knowledge of describes a variety of symptoms that can occur following
what was done at the time of the initial procedure. If postcolonoscopic polyp removal with electrocoagulation. This
polypectomy bleeding is suspected, repeat colonoscopy is more likely to occur following polypectomy in the thin-
after a rapid prep allows for direct intervention if the bleed- walled cecum and usually occurs following removal of ses-
ing site can be identified. Bleeding from a polypectomy site sile polyps >2 cm in diameter (40). Hypertension has also
can be managed with endoclips, direct cautery, argon bean been shown to be an independent risk factor (41). Symptoms
coagulation, and epinephrine injection. Unprepped colo- include abdominal pain (often with rebound and/or guard-
noscopy in the setting of acute bleeding is often limited by ing), fever, and leukocytosis without frank perforation.
lack of visualization due to a combination of stool, blood, Patients may present within hours of the procedure to as
and clots within the lumen of the colon, interfering with late as 5–6 days postprocedure.
visualization. Alternatively, mesenteric arteriography with Workup of suspected postpolypectomy syndrome should
embolization of a bleeding source may be an effective means start with blood work and plain x-rays or a CT scan to rule
of dealing with post-polypectomy bleeding. Bleeding from out pneumoperitoneum. Leukocytosis and elevated CRP
an extramucosal source, such as the mesentery or spleen, are suggestive of the syndrome. Patients can then be treated
usually requires direct operative intervention. with broad-spectrum antibiotics and bowel rest and are fol-
lowed with serial abdominal exams to assure the patient
PERFORATION does not progress from localized to frank peritonitis (40,41).
Perforation is a rare but serious complication of colonos- INFECTIOUS COMPLICATIONS

copy, with recent studies estimating the rate at far less than
1/1,000 procedures (34). In fact, the USMSTF on CRC set A rare but potentially serious complication of colonos-
quality improvement targets for perforation of <1/1,000 for copy is the spread of transmittable diseases from improp-
therapeutic examinations and <2/2,000 for screening proce- erly cleansed equipment or examiner exposure. Standard
dures (33). Multiple mechanisms can cause perforation dur- precautions should be taken with gowns, gloves, and face
ing colonoscopy. These include postpolypectomy and other shields. Mechanical cleaning protocols, when followed cor-
procedural perforation, direct trauma from instrument pas- rectly, should prevent the transmission of HIV and viral
sage (most commonly in the sigmoid colon), shearing injury hepatitis (40). There have been no reported cases of endos-
caused by lateral bowel pressure from a bowing loop, and copy-transmitted HIV to date, and reports of transmis-
barotrauma from air insufflation (most commonly in the sion of viral hepatitis are limited to single digits. Reports
cecum). of transmission of microorganisms such as Pseudomonas
Treatment of postcolonoscopic perforation depends on aeruginosa and Salmonella spp. by colonoscope have been
the patient’s condition and presentation, time of perforation made as well, including those with multidrug resistance. The
discovery, and the suspected nature of the perforation. In true incidence of endoscopy-related infection is unknown
References 127
but has been estimated to be approximately 1 in 1.8 million hernia. A “pulley” technique has been described where a
procedures (42). Despite the low incidence, it is important large loop is intentionally created within the hernia sac,
for the endoscopist to ensure that proper cleaning protocols grasped manually, and withdrawn over the “pulley hand”
have been followed prior to any endoscopy. limb by limb (52). Others report removal of the scope under
x-ray guidance (53,54); however, an inability to remove the
POSTCOLONOSCOPY COLITIS (FROM scope with gentle traction necessitates operative interven-
DISINFECTANT) tion (55).
Disinfectant colitis or glutaraldehyde-induced colitis is

a condition first described in 1986 by Castelli (43). It is
attributed to retention of a commonly used disinfectant,
SUMMARY
glutaraldehyde, in the colonoscopic channel. This condi-
Adequate training, knowledge, and experience as
tion typically presents with abdominal pain, tenesmus, and
well as technical skill are vital to perform safe and
bloody diarrhea within 48 hours of colonoscopy, though
effective endoscopy and to limit complications. It is
milder cases can occur without bleeding. Laboratory
important for the endoscopist to know the applica-
workup can reveal leukocytosis and elevated CRP, and
tions and limitation to the various modalities. Early
endoscopic features are similar to ischemic colitis. CT find-
recognition, workup, treatment, and understanding
ings are also nonspecific and can include mural thickening
of potential complications are vital to minimize the
or pericolonic fat stranding (44–46). Treatment is non-
risk of morbidity and mortality.
specific and includes bowel rest; the disease course is self-
limited. Incidence is estimated at 0.1%–4.7%, though given
the difficulty in diagnosis and its self-limited nature, rates
may actually be higher (47). Diagnosis remains one of exclu- REFERENCES
sion, but it is important for the clinician to be aware of this
potential complication and to consider it when other causes 1. Harewood GC et al. Gastrointest Endosc. 2003;58(1):
of pain or bleeding have been ruled out. 76–9.
2. Bucci C et al. Gastrointest Endosc. 2014;80(4):
INCOMPLETE COLONOSCOPY 566–76.e2.
3. Johnson DA et al. Gastrointest Endosc. 2014;80(4):
A complete colonoscopy will not be possible in a certain 543–62.
number of patients. Reasons commonly include the pres- 4. Curran MP, Plosker GL. Drugs. 2004;64(15):1697–714.
ence of an obstructing mass, excess redundancy, poor prep- 5. Cohen SM et al. Dis Colon Rectum. 1994;37(7):
aration, medical instability, or lack of patient cooperation. 689–96.
6. Parente F et al. Dig Liver Dis. 2009;41(2):87–95.
MISCELLANEOUS COMPLICATIONS 7. Markowitz GS et al. J Am Soc Nephrol. 2005;
16(11):3389–96.
Although the most common complications of colonoscopy 8. Rejchrt S et al. Gastrointest Endosc. 2004;59(6):
are bleeding and perforation, a wide body of literature exists 651–4.
describing other rare but potential complications. Splenic 9. Marshal JB. Clin Gastroenterol Hepatologu. 2014;12:
injury following colonoscopy was first reported in 1974 and 1522–4.
is becoming increasingly recognized (48). The injury is pro- 10. Longcroft-Wheaton G, Bhandari P. J Clin Gastroenterol.
posed to occur from excessive traction of the splenocolic 2012;46(1):57–61.
ligament rather than from therapeutic maneuvers (40). 11. Nivatvongs S, Fryd DS. N Engl J Med. 1980;303(7):
Incidence is estimated at around 1 in 100,000 procedures, 380–2.
and it carries an overall mortality rate of about 5%, making 12. Ellis DJ, Bevan PG. Br Med J. 1980;281(6237):435–7.
prompt diagnosis and treatment vital. Traditionally treat- 13. Rao VS et al. Colorectal Dis. 2005;7(1):61–4.
ment consisted of laparotomy with splenectomy. Although 14. Schoellhammer HF et al. Am J Surg. 2008;196(6):
splenectomy remains a mainstay of treatment, there are 904–8. discussion 908.
recent trends toward embolization or conservative man- 15. Beck DE. Handbook of Colorectal Surgery, 2nd
agement (49). Edition, Vol. xiv. New York, NY: Marcel Dekker, 2003.
Other rare complications include acute appendicitis, first 16. Achord JL. The history of gastrointestinal endos-
reported in 1988 (50). Although rare with only about 30 copy. In Ginsberg GG (ed). Clinical Gastrointestinal
reported incidences, one reported patient died despite Endoscopy. St. Louis, MO: Elsevier Saunders, 2012,
appendectomy, and it is important to recognize appendicitis pp. 2–11.
as a potential complication of lower endoscopy (51). Another 17. Rex DK et al. Am J Gastroenterol. 2009;104(3):
rare but potentially serious complication is the incarcera- 739–50.
tion of the colonoscope within an inguinal or ventral 18. Levin B et al. Gastroenterology. 2008;134(5):1570–95.
19. Wiggins TF et al. Clin Colon Rectal Surg. 2010;23(1): 38. Trecca A et al. Tech Coloproctol. 2008;12(4):315–21.
14–20. discussion 322.
20. Shergill AK et al. Gastrointest Endosc. 2012;76(1): 39. Avgerinos DV et al. J Gastrointest Surg. 2008;12(10):
18–24. 1783–9.
21. Childers RE et al. Gastrointest Endosc. 2015;82(3): 40. Hicks TC. Transanal endoscopy. In Whitlow CB (ed).
503–11. Improved Outcomes in Colon and Rectal Surgery.
22. Schroeder C et al. Dis Colon Rectum. 2016;59(1): London, UK: Informa Healthcare, 2010, p. 132–9.
62–9. 41. Cha JM et al. Endoscopy. 2013;45(3):202–7.
23. Wadhwa V et al. Clin Gastroenterol Hepatol. 42. Kovaleva J. Best Pract Res Clin Gastroenterol. 2016;
2017;15(2):194–206. 30(5):689–704.
24. Sipe BW et al. Gastrointest Endosc. 2002;55(7): 43. Castelli M et al. Am J Gastroenterol. 1986;81:887.
815–25. 44. Yen HH, Chen YY. Endoscopy. 2006;38(Suppl. 2):E98.
25. Hassan C et al. Endoscopy. 2012;44(5):456–64. 45. Mohamad MZ et al. Am J Emerg Med. 2014;32(6):
26. Clarke AC et al. Med J Aust. 2002;176(4):158–61. 685.e1-2.
27. Lichtenstein DR et al. Gastrointest Endosc. 2008; 46. Shih HY et al. Kaohsiung J Med Sci. 2011;27(12):
68(5):815–26. 577–80.
28. Patane S. J Cardiovasc Transl Res. 2014;7(3):372–4. 47. Stein BL et al. Can J Surg. 2001;44(2):113–6.
29. Wilson W et al. Circulation. 2007;116(15):1736–54. 48. Wherry DC et al. Med Ann Dist Columbia. 1974;43(4):
30. Committee ASOP et al. Gastrointest Endosc. 2015; 189–92.
81(1):81–9. 49. Jehangir A et al. Int J Surg. 2016;33(Pt A):55–9.
31. Douketis JD et al. Chest. 2012;141(Suppl. 2): 50. Houghton A, Aston N. Gastrointest Endosc. 1988;
e326–50S. 34(6):489.
32. Reumkens A et al. Am J Gastroenterol. 2016;111(8): 51. Kuriyama M. Clin J Gastroenterol. 2014;7(1):32–5.
1092–101. 52. Koltun WA, Coller JA. Dis Colon Rectum. 1991;34(2):
33. Rex DK et al. Am J Gastroenterol. 2002;97(6):1296–308. 191–3.
34. Castro G et al. Cancer. 2013;119(Suppl. 15):2849–54. 53. Kume K et al. Endoscopy. 2009;41(Suppl. 2):E172.
35. Orsoni P et al. Endoscopy. 1997;29(3):160–4. 54. Fan CS, Soon MS. Endoscopy. 2007;39(Suppl. 1):
36. Castellvi J et al. Int J Colorectal Dis. 2011;26(9): E185.
1183–90. 55. Tas A et al. Endoscopy. 2015;47(Suppl. 1 UCTN):
37. Kim JS et al. Surg Endosc. 2013;27(2):501–4. E125–6.
14
Laparoscopy for colorectal disease
BRADLEY R. DAVIS AND KEVIN R. KASTEN
performance of an anastomosis. The greater complexity

CHALLENGING CASE of laparoscopic colectomy has been associated with lon-
ger operative times and a longer learning curve. For these
A 64-year-old woman with a history of arteriosclerotic reasons, despite its advantages, laparoscopic colectomy
peripheral vascular disease is undergoing a laparo- has taken many years to gain traction. In the early days
scopic right colectomy for an early cancer. After the of colorectal laparoscopy, it was estimated that only 3% of
ileocolic artery and vein are divided with an energy- surgeons had performed more than 50 laparoscopic colec-
sealing device, pulsatile blood is noted from the tomies (2). The complexity of the procedure necessitated
proximal divided vessel. technological solutions that have evolved over the past 20
years so that surgeons are increasingly utilizing laparos-
copy to treat colorectal disease. Some of the most recent
estimates suggest about 40%–50% of colectomies are being
CASE MANAGEMENT
performed laparoscopically, while about 5% are being per-
formed with robotic assistance (4,5). While technical chal-
The proximal stump is grasped with a grasper, which
lenges have hampered its growth, other challenges remain
controls the bleeding. A vessel loop is inserted and
including access to enough cases to overcome the learn-
using a second grasper is placed around the vessel
ing curve, the perceived lack of benefit to the patient, as
and tightened. This controls the bleeding.
well as concerns regarding safety and oncologic outcomes.
Some questions remain to be answered, but laparoscopy
has now been shown to be equivalent in terms of oncologic
outcomes in colon cancer, and the benefit to the patient
INTRODUCTION has been clearly demonstrated. As more general surgery
and colorectal trainees gain access to cases during their
Although laparoscopy has been in clinical use since the turn residencies, the penetration of minimally invasive tech-
of the twentieth century, its use in general surgery was lim- niques will continue to grow so that the day is approach-
ited. That changed in the 1980s with the recognition that ing when the majority of colorectal resections will be done
laparoscopic cholecystectomy could be performed with sig- laparoscopically.
nificant reduction in operative morbidity, hospital length The Surgical Care and Outcomes Assessment Program
of stay, and recovery. Now a host of complex surgeries are evaluated the use of laparoscopy in 48 hospitals in the United
being performed preferentially using minimally invasive States between 2005 and 2010. Of the 9,705 patients under-
techniques with excellent outcomes. going elective colorectal operations, 38% were performed
Laparoscopic colon and rectal resections were demon- laparoscopically (17.8% of the procedures were converted to
strated to be feasible in the mid-1990s (1,2), but the rate open). The use of laparoscopic procedures increased from
of adoption has been slower when compared to other 23.3% in 2005 to 41.6% in 2010 (6). The authors found that
minimally invasive procedures such as cholecystectomy hospital characteristics (urban location and less than 200
(3). Laparoscopic colonic surgery is a significantly more beds), diverticular disease, and right hemicolectomies were
challenging operation as it frequently involves mul- factors associated with the use of laparoscopy. They also
tiple abdominal quadrants, identification and ligation found the greatest increase in the total number of colorectal
of named vascular pedicles, mobilization and transec- operations among hospitals with the highest laparoscopy
tion of the bowel, retrieval of the surgical specimen, and adoption rates.
129
130 Laparoscopy for colorectal disease
In another recent study using the University Health 2.7%, respectively. Anastomotic leakage was the only fac-
System Consortium administrative database, which includes tor that predicted organ/space SSI (p < 0.01). Independent
more than 300 academic hospitals, laparoscopic colorectal risk factors of incisional SSI included blood transfusion
resection was attempted in 36,228 (42.2%) out of 85,712 (p = 0.047) anastomotic leakage (p < 0.01), and open
patients, with 15.8% requiring conversion to open surgery. colorectal resection (p = 0.037) (16). In another study of
The authors concluded that there is a trend of increasing 3,701 patients, 2,518 (68%) underwent colon surgery and
use of laparoscopy in colorectal surgery, across hospitals in 1,183 (32%) rectal surgery. In colon surgery, the overall SSI
the United States in recent years with acceptable conversion rate was 16.4%, and the organ/space SSI rate was 7.9%, while
rates (7). in rectal surgery the rates were 21.6% and 11.5%, respectively
(p < 0.001). Independent risk factors for organ/space SSI in
colon surgery were male sex (odds ratio [OR] 1.57; 95% CI,
1.14–2.15) and ostomy creation (OR 2.65; 95% CI, 1.8–3.92),
ADVANTAGES while laparoscopy (OR 0.5; 95% CI, 0.38–0.69) and oral
antibiotics combined with IV antibiotic prophylaxis (OR
Laparoscopic surgery in general has many advantages com- 0.7; 95% CI, 0.51–0.97) were protective factors (17). Open
pared to open surgery, including smaller incisions, which are surgery was shown to be a risk factor for superficial SSI
expected to translate into decreased postoperative pain and after elective rectal surgery (18). Following elective rectal
offer a superior cosmetic result. More importantly, however, surgery, 8,880 patients were evaluated, and superficial SSIs
are a decreased risk of postoperative complications both long were diagnosed in 861 (9.7%) patients. Multivariate analysis
and short term, including ileus, surgical site infections, and demonstrated the following risk factors: male gender, body
adhesion formation, as well as incisional hernia formation. mass index (BMI) >30, current smoking, history of chronic
Postoperative pain has been evaluated in a number of pro- obstructive pulmonary disease (COPD), American Society
spective, randomized trials demonstrating a reduction in of Anesthesiologists III/IV, abdominoperineal resection
narcotic use following laparoscopic surgery (8–10). Further (APR), stoma formation, open surgery (versus laparo-
pain reduction can be seen with the use of multimodal pain scopic), and operative time >217 minutes. The benefit of
management that has become common in enhanced recov- minimally invasive surgery with respect to SSIs has carried
ery programs, including the use of transversus abdomi- over to hand-assisted procedures as well (19). A limitation of
nis plane blocks (11) and multimodal drug therapy (12). many of the studies reviewed is the selection bias associated
Laparoscopy has been associated with a significant reduc- with selecting patients for a laparoscopic approach over an
tion in both incisional hernia and small bowel obstruction open procedure.
(SBO). In a review of 11 randomized controlled trials (RCTs) Postoperative ileus (POI) results in greater hospital
and 14 non-RCT comparative studies (6,540 patients), lapa- length of stay and comes at a significant cost to the health-
roscopy was associated with a significant reduction in the care system (20). In a meta-analysis of 54 studies compris-
occurrence of SBO (relative risk [RR] 0.57; 95% confidence ing 18,983 patients, the incidence of POI was 10.3%. A lower
interval [CI], 0.42–0.76) and incisional hernia (RR 0.60; 95% incidence of POI after laparoscopic resections was identi-
CI, 0.50–0.72). No significant difference between laparos- fied with an incidence of 6.4% (95% CI, 3.5%–11.5%) after
copy and open surgery was found when the analysis was lim- laparoscopic resection and 10% (95% CI, 6.2%–15.8%) after
ited to studies with conversion rate: >15% demonstrating the open colorectal resection (21). Similar studies have demon-
importance of minimizing conversions. Length of follow-up strated the protective effect of laparoscopy on POI following
did not substantially impact the results (13). colorectal resections (22–24).
The incidence of both superficial and deep/organ space
surgical site infections (SSIs) has decreased with the pen-
etration of laparoscopy for colorectal disease (14). In a study
looking at the National Surgical Quality Improvement DISADVANTAGES
Program, 10,979 patients undergoing colorectal surgery
were analyzed (laparoscopy [LAP] 31.1%, open 68.9%). While the advantages of laparoscopy are significant, they
The SSI rate was 14% (9.5% LAP versus 16.1% open, mostly favor patient outcomes at the expense of longer
p < 0.001). On multivariate analysis age, American Society operative times and resource utilization in the operat-
of Anesthesiologists (ASA) stage ≥ 3, smoking, diabetes, ing room. In addition, there is a physical cost to the sur-
o
perative time >180 minutes, appendicitis or diverticulitis, geon with as many as 74% experiencing musculoskeletal
and regional enteritis diseases were found to be significantly disorders attributable to performing minimally invasive
associated with high SSI while the LAP approach was asso- surgery (25,26).
ciated with a reduced SSI rate (15). In a prospective study of The conversion from laparoscopic to open surgery for
SSIs in 1,011 patients undergoing elective colorectal resec- colorectal disease is a predictable part of any practice. The
tions, the overall rate of incisional SSI and organ/space SSI proper patient selection and increasing expertise and expe-
was 4.8% and 1.7%, respectively. Rates of incisional SSI in rience, as well as improvements in equipment, will reduce
open and laparoscopic colorectal resection were 5.7% and the need for conversion and its overall impact.
Disadvantages 131
Laparoscopy for colorectal disease is among the most available instruments. It is important to be familiar with
c omplex minimally invasive procedures surgeons will per- more than one device, as all of them can and will fail at some
form. The challenge as in all laparoscopic procedures is the point. Staplers, vessel clips, sealant devices, as well as vessel
translation of a surgeon’s expertise in open surgery to mini- loops are the standard tools used to achieve intracorporeal
mally invasive surgery. While good judgment and critical hemostasis. Early in a surgeon’s experience, extracorporeal
thinking in the operating room are common to both tech- vessel ligation may be preferred.
niques, the skill sets and even the presentation of the anatomy While laparoscopy for patients with a BMI over 30 has
are often quite different. Improvements in hand port technol- been safely performed, increasing size will add to the chal-
ogy have the potential to bridge the gap between open and lenge as is true with open surgery. Women tend to have
minimally invasive procedures and afford the surgeon an much less intraabdominal adipose deposits, while obese
opportunity to gradually acquire the necessary expertise. men tend to have heavy, thick omentums with bulky mesen-
There is a wide range of conversion rates published in the teries and large appendices epiploic. Therefore, laparoscopic
literature from as low as 3% to 30% (27,28). In early trials colon resections in morbidly obese men tend to be more dif-
assessing the safety of laparoscopy in treatable colorectal ficult and should be avoided early in a surgeon’s experience.
cancer, all trial participants had to have completed a mini- Obesity plays an important role in outcomes following lapa-
mum of 20 laparoscopic colon resections for benign disease, roscopic surgery for colorectal disease. In a meta-analysis
and yet in the CLASICC trial there was a decrease in the of 43 studies evaluating the impact of obesity, it was found
conversion rate from 38% to 16% during the 6-year study, that BMI was associated with significantly longer opera-
presumably reflecting an ongoing learning curve beyond 20 tive time (p < 0.001), greater blood loss (p = 0.01), and
cases (29). While it is certainly possible that 20 cases will higher incidence of conversion to open surgery (p < 0.001).
be adequate for some surgeons, the majority will require Moreover, BMI was a risk factor for overall complication
more—likely 50–70 cases—to fully master the technique rates (p < 0.001), especially for ileus (p = 0.02) and events
(30). Much of this will depend on the choice of procedures— of the urinary system (p = 0.03). Significant association was
right-sided versus left-sided colectomies—the disease pro- identified between higher BMI and risk of SSI (p < 0.001)
cess, and the laparoscopic technique (31). In a more recent and anastomotic leakage (p = 0.02). Higher BMI might
review of the published literature on learning curves for also lead to a reduced number of harvest lymph nodes for
colorectal laparoscopy, it was found that the definition of patients with colorectal cancer (p = 0.02) (36).
proficiency was subjective, and the number of operations While one study demonstrated that timely conversion
to achieve it ranged from 5 to 310 cases (32). It can be dif- to open surgery does not place a patient at increased risk
ficult for many surgeons in practice to achieve this volume of perioperative morbidity and mortality (37), most dem-
of cases, and several options are available to facilitate the onstrate that conversion results in worse outcomes com-
learning process. Simulators are still fairly crude but are pared to both laparoscopic and open procedures (38,39). In
being developed and improved upon constantly. Eventually a retrospective study using the Premier Prospective data-
a surgeon may be able to master a procedure through simu- base to evaluate predictors and outcomes of conversions
lation prior to taking on a live case. While the numbers of for left-sided colectomies, 41,417 patients were evaluated;
laparoscopic cases being performed by general surgery resi- 63% of these cases were attempted laparoscopically with the
dents is increasing, a recent review of graduating residents incidence of conversion of 13.3%. Length of stay (LOS) in
demonstrated an average case volume of just 20 laparoscopic days was significantly lower for the Lap-Successful group
colorectal resections, a number that when averaged over 5 (4.9 compared with Lap-Conversion 6.8 and Open-Planned
years of training is probably inadequate to achieve compe- 7.0), but Lap-Conversion and Open-Planned had similar
tency (33,34). The number of cases performed by graduating LOS. Adjusted mean cost was higher for Lap-Conversion
colorectal surgery residents was 80 cases per resident, which $20,165 compared to Open-Planned $18,797, but this differ-
reflects the increasing penetration of minimally invasive ence was smaller than the cost savings for Lap-Successful
surgery techniques in colorectal practice (33). $16,206 ± $219. Open-Planned had lower odds of anasto-
Previous abdominal surgery does not preclude safe lapa- motic leak when compared to converted cases. Conversion
roscopy, but it is totally unpredictable. Open trocar inser- risk factors included obesity, inflammatory bowel dis-
tion should be considered to avoid unintended bowel injury. ease, and left hemicolectomy versus sigmoid colectomy.
The decision to proceed should be made early without a lot Colorectal specialists performing the procedure were asso-
of time spent in laparoscopic adhesiolysis. With no ability to ciated with 38% decreased odds of conversion (40). With
pack the abdomen, exposure is often obtained using gravity, regard to long-term oncologic outcome, overall and disease-
with the patient at steep inclinations. Omental and small free survival in the case of converted patients undergoing
bowel adhesions will often prevent this, precluding safe laparoscopic colorectal cancer surgery seems to be worse
visualization. As is true with all laparoscopic cases, feasibil- than in those in whom the surgery was successfully com-
ity should be decided as quickly as possible to prevent long pleted minimally invasively. However, it remains difficult
operative times and unnecessary morbidity (35). to draw a proper conclusion due to the heterogeneity of the
If named vessels are to be taken laparoscopically, the sur- current studies and reported outcomes as well as due to the
geon will need familiarity with a variety of commercially inclusion of both colon and rectal cancer patients (41).
Converting from laparoscopy to open is usually accom- to open surgery was 21%. There have been several other
plished through a midline laparotomy; however, for surgeons well-designed prospective randomized multicenter trials
with experience using hand-assist techniques, alternatives to that have demonstrated no differences in the incidence of
performing a midline laparotomy exist. For many left-sided metastasis in the surgical wound as well as in oncological
and pelvic procedures, a Pfannenstiel incision can provide outcomes when the laparoscopic approach was compared
adequate to excellent exposure when used in combination to open surgery (52–55).
with hand-assisted laparoscopic techniques. However, the While some studies have suggested better oncologic out-
surgeon should be certain that the lower transverse incision comes favoring laparoscopy, particularly in patients stage
will adequately accomplish what needs to be addressed. III colon cancer (51,56), the results have not been confirmed
in all studies, but without question the use of laparoscopy
for the management of colon cancer is currently accepted
worldwide (57).
TREATABLE CONDITIONS As a direct result of this data, the American Society
of Colon and Rectal Surgeons as well as the Society of
COLON CANCER American Gastrointestinal and Endoscopic Surgeons have
formulated a position statement in support of laparoscopy
The impact of laparoscopy on long-term oncological out- for curable colon cancer with the caveat that surgeons
come was a subject of controversy for many years and was should have performed a minimum of 20 laparoscopic col-
a hurdle that had to be crossed for laparoscopy to become ectomies for benign disease or metastatic cancer before a
widely accepted. Early reports of trocar site cancer recur- curable cancer resection can be performed. Individual sur-
rence and concerns regarding lower number of lymph nodes geons need to be certain that they can perform an adequate
retrieved (42,43) led to a self-imposed moratorium on lapa- lymphadenectomy with negative margins and create an
roscopy for curable cancer, and the procedure lost some of anastomosis before they take on curable cancer resections
its momentum further delaying widespread utilization. In laparoscopically.
fact, some series reported port site metastases and perito-
neal dissemination in 10%–20% of patients (44,45). Many of RECTAL CANCER
the benign conditions (e.g., diverticular disease) tended to
be more difficult to do laparoscopically, and surgeons were With respect to rectal cancer, only one of these early tri-
cautious to remove polyps as several of them contained ade- als reported on the outcomes, and the incidence of positive
nocarcinoma on final pathological analysis. The relation- radial margin was higher in the laparoscopic arm with much
ship between different factors related to the laparoscopic higher conversion rates noted (52). More recent trials have
technique (pneumoperitoneum) to the tumor (manipu- assessed the safety and outcomes of patients undergoing a
lation, degree of differentiation, and stage) and the host laparoscopic approach to rectal cancer, and the outcomes
(immune and inflammatory factors) were investigated in have been mixed. The need to perform a total mesorectal
several experimental studies (46,47). The “chimney effect,” excision in a deep and narrow pelvis increases the techni-
referring to leakage of carbon dioxide from trocars and cal complexity of this procedure and the risk of oncological
aerosolizing of tumor cells, has been proposed as a causative compromise.
factor (48). Based on these results, a series of recommenda- In a recent single-center randomized trial, Lujan et al.
tions were made to avoid port site metastasis (49,50). These (58) compared surgical outcomes after laparoscopy and
included avoiding manipulation of the tumor to prevent open surgery in patients with mid and low rectal can-
exfoliation of malignant cells, using povidone-iodine solu- cers. Blood loss was significantly greater for open surgery
tions, emptying the CO2 through the trocars, using a device (p < 0.001), and operating time was significantly greater for
to protect the wall incision, and closing all trocar holes. laparoscopic surgery (p = 0.020), while return to diet and
To assess the outcomes of laparoscopy for colon cancer, hospital stay were longer for open surgery. Complication
a randomized multicenter trial was established to assess rates and involvement of circumferential and radial mar-
if laparoscopy was equivalent in terms of oncologic out- gins were similar for both procedures, but the number of
comes when compared to traditional open techniques (51). isolated lymph nodes was greater in the laparoscopic group
The results of the COST trial were published in 2004. The (mean 13.63 versus 11.57; p = 0.026). There were no differ-
trial enrolled 872 patients with curable colon cancer and ences in local recurrence or disease-free or overall survival.
randomized them to open versus laparoscopic resection. In the COLOR II trial, 1,044 patients were random-
The median follow-up was 4.4 years. There was no differ- ized to undergo laparoscopic or open rectal resections for
ence in oncological outcomes, and patients in the laparo- adenocarcinoma (within 15 cm from the anal verge—699
scopic arm enjoyed a faster return of bowel function and in the laparoscopic surgery group and 345 in the open sur-
earlier discharge. Surgeons who participated in the trial gery group) (59). At 3 years, the locoregional recurrence rate
were selected based on proven expertise as decided by a was 5% in the two groups (90% CI, −2.6 to 2.6). Disease-
review of a videotaped laparoscopic colectomy. Despite free survival rates were 74.8% in the laparoscopic surgery
this proven ability, the conversion rate from laparoscopy group and 70.8% in the open surgery group (95% CI, −1.9
Treatable conditions / Inflammatory bowel disease 133
to 9.9). Overall survival rates were 86.7% in the laparoscopic 206 patients (87%) in the laparoscopic surgery group and 216
surgery group and 83.6% in the open surgery group (95% patients (92%) in the open surgery group (p = 0.06).
CI, −1.6 to 7.8). As with many of these trials, surgeons had Several systematic reviews and meta-analyses have
to demonstrate expertise in laparoscopic total mesorec- recently confirmed the short-term benefits and oncological
tal excision through submission of unedited video prior safety of the minimally invasive approach for rectal can-
to participation in the study. Conversion to open surgery cer surgery (62,63). While patients with rectal cancer ben-
was 16%. In the laparoscopic surgery group, the operating efit from laparoscopic approaches, surgeons must be very
time was 52 minutes longer, bowel function returned 1 day mindful of the results of the randomized trials and how
earlier (p < 0.0001), and the hospital stay was 1 day shorter patients are informed of the possibility that an open tech-
than in the open surgery group (p = 0.036) (60). There were nique may be a better option. Individual surgeons will need
no significant differences in the rates of anastomotic leak- to track their results, and it is likely that minimally invasive
ing, complication, or death. There were no significant dif- approaches to rectal cancer should be reserved for higher
ferences between the groups with respect to macroscopic volume centers with access to more cases to achieve profi-
completeness of the mesorectum, involved circumferential ciency in these techniques.
resection margins, or distal resection margins (median, 3
cm in the two groups). INFLAMMATORY BOWEL DISEASE
The ACOSOG Z6051 trial randomized patients with rec-
tal cancer within 12 cm of the anal verge to undergo either Ulcerative colitis
open (n = 222) or laparoscopic (n = 240) total mesorectal
excision. The trial was a noninferiority design, and the pri- Pooled data from 1975 to 2007 demonstrate overall morbidity
mary outcome was a composite of circumferential radial and mortality of 40.1% and 1.8%, respectively, following open
margin greater than 1 mm, distal margin without tumor, surgery for ulcerative colitis (UC) (64). Studies report 10%–
and completeness of total mesorectal excision (61). 45% of UC patients ultimately require surgery, even as that
Conversion to open resection occurred in 11.3% of number declines due to a paradigm shift in medical treat-
patients. Operative time was significantly longer for lapa- ment using biologic agents (65–67). These agents are keeping
roscopic resection (mean, 266.2 versus 220.6 minutes; more patients out of the operating room; however, a signifi-
p < 0.001). Length of stay (7.3 versus 7 days), readmission cant volume of higher acuity patients still require surgical
within 30 days (3.3% versus 4.1%), and severe complications resection (68). Despite contradictory data regarding the effect
(22.5% versus 22.1%) did not differ significantly. of biologics on surgical complications, providers continue to
Successful resection occurred in 81.7% of laparoscopic strive for reduction in risk while maximizing quality of life
resection cases and 86.9% of open resection cases and did (69). Minimally invasive surgery helps in risk reduction.
not support noninferiority. Quality of the total mesorectal Specific patient populations have been shown to benefit
excision specimen in 462 operated and analyzed surgeries from laparoscopic surgery for UC. While obesity is inde-
was complete (77%) and nearly complete (16.5%) in 93.5% pendently associated with conversion to open procedure,
of the cases. Negative circumferential radial margin was no differences were seen between laparoscopic and open
observed in 90% of the overall group (87.9% laparoscopic procedures in overweight or obese patients compared with
resection and 92.3% open resection; p = 0.11). The distal normal BMI patients (70). When comparing abdominal
margin result was negative in more than 98% of patients adhesions following laparoscopic and open ileal pouch-anal
irrespective of type of surgery (p = 0.91). anastomosis (IPAA), the intraabdominal adhesion score
The ALaCaRT trial randomized patients with T1–T3 was significantly lower in the laparoscopic group (median
rectal cancer within 15 cm of the anal verge to either open score 0 versus 4). Additionally, the adnexal adhesions score
(n = 237) or laparoscopic (n = 238) total mesorectal exci- was significantly lower at 5.2 versus 20. These findings were
sion (TME) in a European multicenter trial with a similar independent of two- or three-stage procedures (71). With
composite outcome measure and noninferiority design. a median age of IPAA near 27 years and 65% of female
Conversion to open surgery occurred in 9% of patients. patients experiencing infertility when undergoing an open
Operative times were longer in the laparoscopy group, while procedure due to adhesions of the fallopian tubes (71),
length of hospital stay (8 days) and complication rates were laparoscopy may benefit females interested in future chil-
not different among the groups. A successful resection was drearing. Studies of laparoscopic IPAA show significantly
achieved in 194 patients (82%) in the laparoscopic surgery reduced infertility rates (27%–45%) that are comparable
group and 208 patients (89%) in the open surgery group, to UC patients undergoing laparoscopic appendectomy
which once again did not support noninferiority. The cir- (65,72). Reduced adhesions following laparoscopic surgery
cumferential resection margin was clear in 222 patients for UC also benefits subsequent procedures. Operative
(93%) in the laparoscopic surgery group and in 228 patients time is reduced in subsequent procedures due to less time
(97%) in the open surgery group (p = 0.06), the distal margin required for adhesiolysis, and there are lower rates of bowel
was clear in 236 patients (99%) in the laparoscopic surgery obstruction (71). A conversion rate of 5% during second-
group and in 234 patients (99%) in the open surgery group and third-stage procedures is likely due to significantly
(p = 0.67), and total mesorectal excision was complete in fewer abdominal and interloop adhesions, even in patients
with severe colitis (64,73,74). This is important to remember they favor an open versus laparoscopic approach (79). Data
as surgeons become more aggressive in performing urgent on laparoscopic TAC and IPAA support a minimally inva-
minimally invasive colectomy for severe colitis. Newer pub- sive approach as safe and effective, with fewer complications
lished indices are being increasingly used by gastroenter- and higher quality of life. Surgeons are encouraged to dis-
ologists and surgeons to determine patients at increased risk cuss all options, utilizing decision aids where available (66).
for surgical intervention. The simplest index is the Oxford Tailoring an approach based on patient condition, expecta-
Index that shows an 85% likelihood of requiring colectomy tions for quality of life, and surgeon comfort with procedure
during that admission if the patient has stool frequency are tantamount to success.
>8/day or stool frequency of 3–8/day with C-reactive pro-
tein (CRP) >45 mg/L on hospital day 3 (75). Such indices Crohn disease
support decision-making algorithms that improve laparo-
scopic surgical outcomes for severe colitis. Despite advances in medical management, 15%–20% of
When evaluating laparoscopic total abdominal colec- patients undergo surgical resection within the first year
tomy (TAC) (straight lap, hand assisted laparoscopic sur- of diagnosis, increasing to 65% within 10 years of diag-
gery [HALS], single incision laparoscopic surgery [SILS]) for nosis. A multidisciplinary approach with gastroenterol-
severe colitis, a 40% overall complication rate, 7.5% reopera- ogy is necessary as 40%–50% of operative patients require
tion rate, 17.2% readmission rate, and 0.5% mortality rate repeat procedure within 10 years (80). Unlike UC, disease
were demonstrated. The most common complications were pathophysiology helps determine the role for laparoscopic
stump leak (15%), bowel obstruction (10%), and wound intervention in Crohn disease (CD). While most patient
infection (9%), with no differences between types of laparo- populations benefit from a minimally invasive procedure,
scopic approach (64). Studies also show significantly lower rates of complications and conversion to open surgery are
rates of infectious complications following laparoscopic predicated on uncomplicated versus complicated disease.
procedures in this patient population (73). At Washington For uncomplicated disease and nonrecurrent strictures, a
University, the use of laparoscopy for stage 1 total abdomi- minimally invasive approach is recommended with resul-
nal colectomy yielded less narcotic usage, faster return of tant decreased length of stay, decreased costs, lower com-
bowel function, shorter postoperative length of stay, and plication rates, conversion rates below 10%, and long-term
no difference in complication rates. Further, when the first recurrence rates similar to open procedures (80). For com-
stage was done laparoscopically, patients received comple- plicated and recurrent Crohn disease, intraoperative or
tion proctectomy and IPAA an average of 49 days earlier CT findings of abscess or fistula are independently asso-
than patients with open abdominal colectomy. Subsequent ciated with conversion and septic complications (80–82).
ileostomy closure occurred a mean of 17 days earlier. Interestingly, prior open procedure is not associated with
Laparoscopic approach was the only factor independently higher rates of conversion, although adhesive disease is the
associated with decreased elapsed time to completion of all most common indication for conversion. Further, iterative
surgical procedures (74). procedures in complicated Crohn patients are not associ-
A range of studies, including two RCTs, showed equiv- ated with higher rates of complications (80,82–84). As such,
alence of laparoscopic and open IPAA in quality of life, minimally invasive surgery for both uncomplicated and
mortality, morbidity, return of bowel function (ROBF), and complicated Crohn disease in most cases is a reasonable
postoperative LOS (76). In contrast, a Cochrane Review option, understanding that some patients are at higher risk
demonstrated shorter ROBF, shorter LOS, better cosmesis, for conversion to an open procedure.
longer operative time, higher operative cost, and lower total More than 95% of fistulizing Crohn disease patients
hospital cost for laparoscopic versus open IPAA (76,77). require surgery, with 27% undergoing two procedures and
From a long-term standpoint, studies show similar out- 55% requiring more than two procedures (85). These opera-
comes between open and laparoscopic procedures, aside tions are technically demanding when performed laparo-
from cosmesis and body image scores. Further, long-term scopically, but the patient benefit is clear. Despite rates of
IPAA function at 20 years (regardless of surgical approach) conversion nearing 40%, the postoperative complication
demonstrates a mean of six bowel movements per day with rate is significantly reduced (12% versus 43%). This includes
one at night. Continence is maintained in greater than 70% lower rates of SSI, anastomotic leak, and intraabdominal
of patients, with more than 80% able to defer defecation bleeding. Decreased complication rates result in shorter
for 30 minutes. Overall pouch failure is 4.7%, with rates postoperative length of stay. Measurement of serum CRP
decreasing since 2000. And male sexual dysfunction rates and procalcitonin in these patients indicates significantly
are 2%–3%, with lower rates achieved when dissection is reduced postoperative stress response in laparoscopic
close to the rectal wall. Interestingly, up to 40% of women patients versus their open counterparts (85). Further, mul-
complain of sexual dysfunction following TAC/IPAA. tiple studies support minimally invasive procedures for
Overall costs are no different between approaches (78). Crohn disease due to lower rates of pulmonary dysfunction,
Published practice parameters from the American shorter postoperative length of stay, lower complication
Society of Colon and Rectal Surgeons do not favor one sur- rates, and similar long-term results compared with open
gical procedure over another for management of UC, nor do procedures (86).
Treatable conditions / Inflammatory bowel disease 135
Most surgical interventions for Crohn disease involve group with results demonstrating a 15.4% (25% versus 9.6%)
terminal ileal disease and small bowel strictures. However, absolute reduction in major complications in the laparo-
a certain number of patients will require surgery for Crohn scopic group. Interestingly, there was a 10% leak rate in
disease or indeterminate colitis. In the 5%–10% of Crohn open cases and 6% rate in laparoscopic cases. There was a
patients requiring an urgent or emergent colectomy, lapa- 19.2% conversion rate (9.6% to hand-assist, 9.6% to open),
roscopy may be considered. Total abdominal colectomy with half of converted procedures due to adhesions and 15%
with end ileostomy is advised in these cases due to safety due to obesity. While operative time was significantly lon-
and speed of procedure, combined with knowledge that only ger in the laparoscopic group, there was less blood loss, no
50% of severe colitis cases can be defined as either Crohn difference in return of bowel function, and a shorter length
disease or UC. In fact, some authors recommend saving of stay (5 versus 7 days). No differences in mortality were
the distal sigmoid and bringing it out as a mucous fistula seen between groups. Less reported pain in the laparoscopic
for possible anastomosis in the future if proctectomy is not group produced higher quality of life scores. In a 6-month
required (87). Laparoscopy in most cases for Crohn disease follow-up analysis, there was no difference in late complica-
is safe, producing fewer complications, shorter postopera- tions or recurrence between laparoscopic and open groups
tive length of stay, and lower costs. Based on significant (92). Subsequent financial analysis of a large subset of
literature on the subject, the use of minimally invasive pro- SIGMA patients demonstrated a significantly higher opera-
cedures for inflammatory bowel disease patients should be tive cost in laparoscopic colectomy, which was negated by
the preferred approach, understanding that higher rates of lower hospitalization, blood product, and emergency room
conversion are expected but not detrimental to the patient. visit expenditures (93). The cost of 1% reduction in compli-
cation rate attributable to laparoscopic colectomy was only
Diverticulitis 31 euros ($39), making minimally invasive colectomy for
diverticulitis a bargain.
Surgical treatment of diverticulitis remains controversial. A subsequent study in 2010 with similar design showed
Recent studies indicate <5% of patients present with recur- no difference in complication rates between laparoscopic
rent episodes worse than their sentinel presentations, most (13.5%) and open (9%) groups, longer operative time, reduced
patients requiring abscess drainage do not require resection, perception of pain with less morphine narcotic usage, and
and more than 90% of patients who develop diverticulitis shorter length of stay (36 hours) in the laparoscopy group
will resolve with conservative treatment (88,89). A review (94). Follow-up evaluation of long-term outcomes and qual-
of recent literature is important to provide patients with the ity of life was published in 2011 with a median follow-up of
best options for management of this disease process with- 30 months (95). Only one patient in each group developed
out undue risk of a surgical procedure. When resection is recurrent diverticulitis treated with antibiotics. Incisional
deemed necessary (stricture, inability to exclude malig- hernias were demonstrated in 9.8% of open cases and 12.9%
nancy, or ongoing sepsis), minimally invasive colectomy of laparoscopic cases, and were not significantly different.
is encouraged for elective resection, emergent resection, Of note, four of the seven laparoscopy group patients with
elective closure of an end colostomy, and urgent or emer- hernias had their first case converted to open. Quality of
gent washout of purulent peritonitis. In fact, more studies life scores were the same between groups, except for higher
continue to demonstrate lower overall, minor, and major satisfaction with cosmesis in the laparoscopic group. Total
morbidity when laparoscopy is used (90). Additionally, calculated costs were similar between groups. The authors
reports indicate staying close to the colon wall for benign cautioned that converted patients experienced worsened
disease results in fewer complications with resection onto outcomes, so choose patients carefully and attempt conver-
the upper rectum necessary to reduce recurrence rates (88). sion to hand-assist instead of open whenever possible.
Understanding where and when to utilize each technique
reduces complications while providing the best patient care. EMERGENT OR URGENT PROCEDURES FOR
DIVERTICULITIS
ELECTIVE RESECTION FOR DIVERTICULITIS In 2016, results from the DILALA randomized trial com-
In the past decade, a few RCTs evaluating laparoscopy paring laparoscopic lavage (LL) to open Hartmann proce-
for diverticulitis were completed. The SIGMA trial was a dure (OHP) for perforated diverticulitis were published (96).
multicenter, double-blind, parallel-arm study that blinded Patients were randomized following confirmed Hinchey III
patients and staff for 5 days; allowed discharge decision- classification by diagnostic laparoscopy. LL involved 3 L
making by blinded, independent physicians; and sought or more of body-temperature saline solution instilled and
to determine if laparoscopy produced fewer postopera- aspirated from all four abdominal quadrants until clear. A
tive complications (91). For study inclusion, patients had drain was left in the pelvis for at least 24 hours, with post-
confirmed diagnosis of diverticulitis via imaging (CT or operative management comparable to those undergoing
barium enema) and colonoscopy, diagnosis of two recur- OHP procedure. A higher 30-day mortality was seen in the
rent attacks, or prior attack with complication (i.e., abscess, LL group (7.7% versus 0%); however, 90-day mortality was
fistula, stricture, or bleeding requiring blood transfusion). similar (7.7% versus 11.4%). Higher readmission occurred
From 2002 to 2006, 104 patients were randomized to each in the OHP group (0% versus 5.7%), although no differences
were shown in number (52% lavage versus 40% OHP) or the creation of stomas, which has the added value of no
severity of complications. Additionally, no difference was additional incisions beyond the stoma (103).
demonstrated in the rate of 30-day reoperation, suggesting
lavage was adequate for controlling the Hinchey III pro- RECTAL PROLAPSE
cess. Length of stay was 6 days in the LL group, significantly
shorter than 9 days in the OHP group. Abdominal approaches to rectal prolapse are associated with
Another prospective RCT published in 2015, the fewer recurrences when compared to the perineal approach.
SCANDIV trial evaluated LL versus laparoscopic or open Operative approaches include resection rectopexy, suture
resection with or without primary anastomosis (97). A total rectopexy, or mesh rectopexy from either a ventral or pos-
of 101 LL and 98 colonic resection patients were compared terior approach. All of the abdominal approaches have been
with rates of 90-day severe postoperative outcomes as the described using laparoscopic techniques with comparable
primary endpoint. This endpoint was met in 30.7% of LL results in terms of recurrence and function when compared
patients and 26% of resection cases, which was not statis- to open surgery (104), but with fewer postoperative compli-
tically different. Mortality at 90 days was similar between cations and shorter hospital stay than open rectopexy (105).
groups (13.9% LL versus 11.5% resection). In this study, A review of 321 prolapse operations (laparoscopic rectopexy
reoperation rate was significantly higher in the LL group 126 patients, open rectopexy in 46, and resection rectopexy
(20.3%) compared to the resection group (5.7%). Also of in 21 patients) with a median follow-up of 5 years demon-
note, four sigmoid cancers were missed on LL, indicating strated a 4% recurrence following laparoscopy. There was no
a strong need for follow-up colonoscopy in all patients not significant difference between groups in terms of recurrence
undergoing resection at time of presentation. In a meta- and postoperative complications (106). In contrast to the
analysis of the DILALA, SCANDIV, and LADIES trials, posterior approaches, the ventral mesh rectopexy (VMR)
the main conclusion was LL produced a significantly lower avoids posterolateral rectal mobilization and thereby mini-
12-month reoperation rate (OR 0.32), at a cost of increased mizes the risk of postoperative constipation. Because of a
90-day morbidity (OR 1.7) attributable mostly to a higher low overall recurrence rate, good functional results, and low
rate of intraabdominal abscess formation (OR 3.5) (98). Of mesh-related morbidity in the short to medium term, VMR
note, LL failed in 17% of patients, but no differences in mor- has been popularized in the past decade. Laparoscopic-
tality or morbidity were noted in patients when LL failed assisted VMR is now being progressively performed, and
and required resection. several articles and guidelines propose the procedure as the
For patients undergoing Hartmann procedure for per- treatment of choice for rectal prolapse (107).
forated diverticulitis, laparoscopy is a viable alternative to
open. In a cost-effectiveness study from 2013 utilizing pro-
pensity-matched cohorts in the NSQIP data set, no differ-
ences in 30-day morbidity or mortality were seen between TECHNICAL CONSIDERATIONS IN
laparoscopic and open Hartmann procedure. Statistical MINIMALLY INVASIVE COLECTOMY
modeling was used to account for inherent selection bias
in patients undergoing laparoscopic Hartmann procedure Minimally invasive surgery provides a technically equiva-
for emergent condition, which may explain why operative lent outcome to open colorectal surgery, but with signifi-
times between groups were no different (99). While effective cant reductions in complication rate, pain, hospital length
in the right hands, remember that, "In all cases, the adop- of stay, and adhesion formation. Myriad approaches include
tion of laparoscopic lavage in emergent settings, abdominal straight laparoscopic colectomy (SLC), hand-assisted lapa-
exploration for generalized peritonitis, pelvic dissection in roscopic surgery (HALS), and single-site laparoscopic sur-
inflammatory conditions and possible suture of a diseased gery (SILS), each geared toward utilizing surgeon strengths
colon require the surgeon have a minimum of colorectal and to maximize patient outcomes. Being comfortable with
minimally invasive skills” (100). When utilized, minimally these approaches enables surgeons to tailor surgical inter-
invasive techniques for elective and emergent management vention to individual patients.
of diverticulitis greatly benefit the patient. Maximizing success in minimally invasive colectomy
begins with port placement and incision planning. It is
STOMA CREATION possible to utilize 5 mm trocars for all minimally invasive
procedures until confirmation that intracorporeal stapling
The creation of a stoma can be easily accomplished using will occur. A 5 mm trocar is placed circumumbilical for all
laparoscopic techniques and is an excellent way for sur- colectomies, allowing for SLC, HALS, SILS, or a hybrid pro-
geons to gain experience in laparoscopic colorectal surgery. cedure following diagnostic laparoscopy. Trocars are placed
Several studies have shown that laparoscopic stoma creation based on target anatomy and likelihood of a hand port being
is a viable alternative to an open approach, with benefit in used. In selected cases, a hand port through a Pfannenstiel
both morbidity and mortality (101,102). There is also evi- incision 2 cm above the top of the pubic symphysis can be
dence to support the use of single-incision laparoscopy for placed early in the procedure. Using this incision requires
Technical considerations in minimally invasive colectomy / Straight laparoscopic colectomy (SLC) 137
understanding that converting to open surgery via mid- Understanding limitations reduces risk of conversion
line laparotomy carries significant risk of wound infection to open technique and the resultant worsened patient out-
and incisional hernia formation. If doubts exist regarding comes. Patient selection is of utmost importance. Factors
the ability to complete a HALS procedure via Pfannenstiel associated with risk of conversion include BMI, surgical
incision, a midline incision is utilized. The importance of expertise, intraabdominal adhesions, stricture or fistula,
trocar placement cannot be overstated as poor triangula- and severity of diverticulitis/inflammation on pathologic
tion, awkward surgeon stance, and poor visualization from examination (88). Review all available diagnostic films and
inadequate traction lead to worsening technical outcomes endoscopy reports. Consider preoperative (or intraopera-
and higher rates of conversion. tive) ureteral stent placement for help with identifying anat-
One of the earliest opportunities for injury occurs during omy and reducing injury rates (88,115).
trocar insertion into the abdomen. The initial trocar inser-
tion will provide the greatest opportunity for a bad outcome, STRAIGHT LAPAROSCOPIC
and as such, several different techniques have been described COLECTOMY (SLC)
(108). These include an open technique, a blind technique
using the Veress needle, and an optical technique using spe- Following diagnostic laparoscopy, further 5 mm ports are
cialized trocars that allow visualization of the layers of the placed and triangulated toward patient pathology. Fogging
abdominal wall using the laparoscope. Mastery of one these of the laparoscopic lens, splatter of irrigation fluid, blood,
techniques is essential for safe access to the abdomen, and and bodily fluids are among those factors that affect a sur-
knowledge of more than one is essential to keep the oper- geon’s ability to maintain a clear operating field, which is
ating surgeon and the patient out of trouble; however, no exacerbated by the use of a 5 mm camera. Aerosolization
one technique has been shown to be superior with potential of fat and other debris will quickly diminish the optics and
pros and cons of each (109,110). A recent review including 28 preclude safe visualization. Condensation on the lens due to
RCTs with 4,860 individuals undergoing laparoscopy dem- temperature discrepancies will also degrade the optics and
onstrated no advantage using any single technique in terms is perhaps the most common reason a surgeon will remove
of preventing major vascular or visceral complications. the laparoscope during surgery. Several commercial prod-
Using an open-entry technique compared to a Veress needle ucts exist to help mitigate the detrimental effects of smoke
demonstrated a reduction in the incidence of failed entry and condensation on the tip of the laparoscope, and a thor-
(OR 0.12; 95% CI, 0.02–0.92). There were three advantages ough evaluation of these technologies is appropriate for all
with direct-trocar entry when compared with Veress needle surgeons considering laparoscopy as part of their colorectal
entry, in terms of lower rates of failed entry (OR 0.21; 95% practice. The most commonly used product is the fog reduc-
CI, 0.14–0.31), extraperitoneal insufflation (OR 0.18; 95% CI, tion/elimination device (FRED, US Surgical, North Haven,
0.13–0.26), and omental injury (OR 0.28; 95% CI, 0.14–0.55). Connecticut), which consists of less than 15% isopropyl
Advocates of the open technique regard this as the safest alcohol, 2% surfactant, and more than 85% water (116).
and most effective means to place the initial trocar (111), Advantages include ease of use, widespread availability, and
although there are some limitations. It is very difficult to low cost; the main disadvantage is the need to remove the
keep the skin incisions smaller than 1 cm, and larger trocars laparoscope and the cooling that occurs upon removal of
(11 or 12 mm) are generally needed to prevent loss of pneu- the laparoscope, leading to further condensation and wors-
moperitoneum during the case. It is not always desirable or ening view. A newer product is the Clearify (Medtronic
necessary to have a 12 mm trocar, particularly in the mid- Minneapolis, Minnesota), a compact device that both heats
line, and placing a trocar using an open technique off mid- the tip of the laparoscope and applies defogging solution
line is difficult in all but the thinnest patients. Additionally, at the same time. While it is still necessary to remove the
while acute and chronic herniation can occur through tro- laparoscope to utilize this device, the warming that occurs
cars as small as 5 mm (112), it is generally acceptable to close greatly eliminates the need for repeat cleanings. Another
the fascial defect of trocars larger than 12 mm (113,114) effective strategy to maintain the laparoscopic image is the
resulting in longer overall times using an open technique. use of heated insufflation tubing, which can also be used
Complications associated with the open technique include to humidify the gas as it enters the abdomen. This has also
enterotomy and vascular and solid organ injury, as well as been shown to reduce postoperative hypothermia in clinical
acute and chronic herniation, which in the immediate post- trials (117). The suction-irrigator can be used effectively in
operative period can result in a bowel obstruction and need short bursts to keep the operative field free of smoke, espe-
for emergent repair. In the obese patient, it can be very diffi- cially when working in the narrow confines of the pelvis.
cult to visualize the fascia through a small skin incision, and On rare occasions when pulsatile bleeding strikes the cam-
if necessary, it should be enlarged to ensure an adequate clo- era, the operative field will be totally obscured, creating a
sure. It is probably best to avoid this technique altogether in situation that is particularly unnerving. It is important to
the significantly obese patient, as the abdominal wall thick- determine the significance of the bleeding and to deal with
ness will preclude adequate fascial visualization without a it as quickly as possible. Often the camera operator is the
generous skin incision. least experienced surgeon or student involved in the case,
and the senior surgeon must quickly take control of the in morbidly obese patients with voluminous mesenteric fat
situation. Blood in the trocar will frustrate any attempts obscuring visualization, extensive intraabdominal adhe-
at good visualization, and if it cannot be cleared quickly, sions from prior surgery, or when operative time is at a
then an alternate trocar should be chosen for the camera premium due to patient status. For cases when HALS is
as long as it provides good exposure to the bleeding vessel. expected from the outset, a hand port can be placed in a
Alternatively, a 5 mm trocar can be upsized to accommo- variety of places depending on the anatomy—midline for a
date a 10 mm laparoscope, which will be less temperamental right colectomy and a Pfannenstiel incision for totals, lefts,
in the face of blood and debris. Once the operative view has and ileal pouch surgery.
been restored, an assessment of the bleeding can be made After insufflation, it is of utmost importance to move all
and dealt with appropriately. nonpathologic bowel out of the operative field. Take advan-
While segmental colectomy of the right or left can be tage of the hand port and insert lap pads as necessary to
completed with only two working trocars, the addition of a facilitate packing the bowel. Be sure to utilize patient bed
third trocar in the contralateral abdomen greatly enhances positioning to assist in bowel positioning, but also for sur-
overall visualization and dissection. For intracorporeal geon comfort while performing HALS. Do not hesitate to
anastomosis following right colectomy, an isoperistaltic use both left and right hands during different portions of
anastomosis with stapling achieved through a left upper the operation. Patients can be placed on split-leg tables to
quadrant 12 mm trocar is an effective way to restore intes- facilitate operating from between the legs without fear of
tinal continuity. The common channel can be closed with nerve injury. The ability to work from all sides during a
either a 9″ 2.0 V-lock absorbable suture or a staple load HALS procedure ensures a technically precise dissection
fired transversely. A recent meta-analysis including 484 with limited surgeon strain. Finally, once mobilization has
patients undergoing laparoscopic right colectomy, 272 with been completed, the surgeon should utilize the hand port
intracorporeal anastomosis and 212 with extracorporeal incision for extraction and anastomotic creation. After
anastomosis, reported the best outcomes associated with appropriate mobilization, each anastomosis can be per-
intracorporeal anastomosis, especially in terms of return formed via this incision, so there is little reason to struggle
of bowel function, length of hospital stay, and cosmetic with a laparoscopic anastomosis.
results. However, the meta-analysis did not show a signifi-
cant difference between the two techniques for anastomotic SINGLE-SITE LAPAROSCOPIC SURGERY
leaks or overall short-term morbidity (118). In another
review of this topic, 12 nonrandomized comparative studies This technique has gained only limited traction in the perfor-
were included in the analysis with a total number of 1,492 mance of colorectal surgery with most studies demonstrat-
patients. No significant change in mortality was found, ing parity to traditional multiport laparoscopy (120). Studies
while short-term morbidity decreased significantly in favor indicate the main benefit of SILS as cosmesis from a small
of intracorporeal anastomosis (OR 0.68, 95% CI, 0.49–0.93). incision hidden in the umbilicus. In a meta-analysis of stud-
Length of stay was also decreased significantly. Subgroup ies using SILS, there was no significant reduction in length
analysis for papers published after 2012 resulted in an even of hospital stay with SILS. Most patients selected for colonic
larger decrease in short-term morbidity and decrease in SILS had a low BMI, nonbulky tumors, and were operated on
length of stay (119). by experienced laparoscopic surgeons (121). When utilized,
Left colectomy is achieved in one of two ways. The first several products are available for the SILS port, which can be
involves distal transection via an enlarged Pfannenstiel inci- used with a 5 mm 30° laparoscope and curved instruments
sion with anastomosis achieved under direct visualization to prevent collision and maximize tension for dissection.
using a circular stapler. The alternative involves intracor-
poreal transection through a right lower quadrant 12 mm REFERENCES
trocar with specimen extraction via smaller Pfannenstiel
incision and placement of stapling anvil. After reduction 1. Scott HJ, Spencer J. Surg Laparosc Endosc. 1995;
of conduit back into the abdomen, the anastomosis is com- 5(5):382–6.
pleted under direct laparoscopic visualization. 2. Jacobs M et al. Surg Laparosc Endosc. 1991;
1(3):144–50.
HAND-ASSISTED LAPAROSCOPIC 3. Moloo H et al. Can J Surg. 2009;52(6):455–62.
SURGERY 4. Davis BR et al. JSLS. 2014;18(2):211–24.
5. Keller DS et al. Surg Endosc. 2017;31:1855–62.
Perhaps the most underrated minimally invasive approach 6. Kwon S et al. J Am Coll Surg. 2012;214(6):909–18.e1.
to colectomy, HALS consistently provides the benefits 7. Simorov A et al. Ann Surg. 2012;256(3):462–8.
of open surgery (tactile feedback, speed of dissection and 8. Milsom JW et al. J Am Coll Surg. 1998;187(1):46–54;
decreased operative time, and shorter learning curve for discussion 5.
most surgeons) with the beneficial outcomes of a laparo- 9. Stage JG et al. Br J Surg. 1997;84(3):391–6.
scopic approach (reduced pain, reduced hospital length of 10. Morneau M et al. Can J Surg. 2013;56(5):297–310.
stay, and lower complication rate). HALS can be beneficial 11. Oh TK et al. Surg Endosc. 2017;7:903–8.
References 139
12. Helander EM et al. J Laparoendosc Adv Surg Tech 49. Franklin ME, Jr. et al. Dis Colon Rectum. 1996;39(10
Part A. 2017;27(9):903–8. Suppl):S35–46.
13. Pecorelli N et al. Surg Endosc. 2017;31(1):85–99. 50. Lacy AM et al. Surg Endosc. 1995;9(10):1101–5.
14. Hennessey DB et al. Int J Colorectal Dis. 2016;31(2): 51. Clinical Outcomes of Surgical Therapy Study Group.
267–71. N Engl J Med. 2004;350(20):2050–9.
15. Kiran RP et al. J Am Coll Surg. 2010;211(2):232–8. 52. Guillou PJ et al. Lancet. 2005;365(9472):1718–26.
16. Poon JT et al. Ann Surg. 2009;249(1):77–81. 53. Buunen M et al. Lancet Oncol. 2009;10(1):44–52.
17. Gomila A et al. Antimicrob Resist Infect Control. 54. Hazebroek EJ Surg Endosc. 2002;16(6):949–53.
2017;6:40. 55. Nakamura T et al. Hepato-Gastroenterology. 2006;
18. Sutton E et al. J Surg Res. 2017;207:205–14. 53(69):351–3.
19. Zhang X et al. Medicine. 2017;96(33):e7794. 56. Lacy AM et al. Lancet. 2002;359(9325):2224–9.
20. Iyer S et al. J Managed Care Pharmacy. 2009;15(6): 57. Bencini L et al. World J Gastroenterol. 2014;20(7):
485–94. 1777–89.
21. Wolthuis AM et al. Colorectal Dis. 2016;18(1):01–9. 58. Lujan J et al. Br J Surg. 2009;96(9):982–9.
22. Vather R et al. Surgery. 2015;157(4):764–73. 59. Bonjer HJ et al. N Engl J Med. 2015;372(14):1324–32.
23. Wolthuis AM et al. Int J Colorectal Dis. 2017;32(6): 60. van der Pas MH et al. Lancet Oncol. 2013;14(3):
883–90. 210–8.
24. Sugawara K et al. J Gastrointest Surg. 2018;22: 61. Fleshman J et al. JAMA. 2015;314(13):1346–55.
508–15. 62. Arezzo A et al. Surg Endosc. 2015;29(2):334–48.
25. Alleblas CCJ et al. Ann Surg. 2017;266(6):905–20. 63. Chen K et al. Int J Surg. 2017;39:1–10.
26. Dalager T et al. Surg Endosc. 2017;31(2):516–26. 64. Gu J et al. Colorectal Dis. 2013;15(9):1123–9.
27. Luglio G et al. Ann Med Surg (2012). 2015;4(2):89–94. 65. Bartels SA et al. Ann Surg. 2012;256(6):1045–8.
28. Zelhart M, Kaiser AM. Surg Endosc. 2018;32(1): 66. Cohan JN et al. Dis Colon Rectum. 2016;59(6):520–8.
24–38. 67. Mao EJ et al. Aliment Pharmacol Ther. 2017;45(1):
29. Lee JK et al. Ann Surg Innovation and Research. 3–13.
2012;6(1):5. 68. Abelson J et al. J Am Coll Surg. 2016;223(4):S31–2.
30. Dincler S et al. Dis Colon Rectum. 2003;46(10):1371–8; 69. Coquet-Reinier B et al. Surg Endosc. 2010;24(8):
discussion 8-9. 1866–71.
31. Tekkis PP et al. Ann Surg. 2005;242(1):83–91. 70. Krane MK et al. J Am Coll Surg. 2013;216(5):986–96.
32. Barrie J et al. Annals Surg Oncology. 2014;21(3): 71. Hull TL et al. Br J Surg. 2012;99(2):270–5.
829–40. 72. Beyer-Berjot L et al. Ann Surg. 2013;258(2):275–82.
33. Shanker BA et al. JSLS. 2016;20(3). 73. Bartels SA et al. Br J Surg. 2013;100(6):726–733.
34. Malangoni MA et al. J Surg Education. 2013;70(6): 74. Chung TP et al. Dis Colon Rectum. 2009;52(1):4–10.
783–8. 75. Hindryckx P et al. Nat Rev Gastroenterol Hepatol.
35. Feigel A, Sylla P. Clin Colon Rectal Surg. 2016;29(2): 2016;13(11):654–64.
168–80. 76. Hata K et al. Surg Today. 2015;45(8):933–8.
36. He Y et al. Dis Colon Rectum. 2017;60(4):433–45. 77. Ahmed Ali U et al. Cochrane Database Syst Rev.
37. Casillas S et al. Dis Colon Rectum. 2004;47(10): 2009(1):CD006267.
1680–5. 78. Buskens CJ et al. Best Pract Res Clin Gastroenterol.
38. Clancy C et al. Colorectal Dis. 2015;17(6):482–90. 2014;28(1):19–27.
39. Gouvas N et al. J Laparoendosc Adv Surg Tech A. 79. Ross H et al. Dis Colon Rectum. 2014;57(1):5–22.
2018;28:117–26. 80. Tavernier M et al. J Visc Surg. 2013;150(6):389–93.
40. Etter K et al. JSLS. 2017;21(3). 81. Mino JS et al. J Gastrointest Surg. 2015;19(6):
41. Allaix ME et al. World J Gastroenterol. 2016;22(37): 1007–14.
8304–13. 82. Shigeta K et al. Surg Today. 2016;46(8):970–8.
42. Martinez J et al. A review. Int Surg. 1995;80(4): 83. Aytac E et al. Surg Endosc. 2012;26(12):3552–6.
315–21. 84. Brouquet A et al. Surg Endosc. 2010;24(4):879–87.
43. Berends FJ et al. Lancet. 1994;344(8914):58. 85. Ren J et al. J Surg Res. 2016;200(1):110–6.
44. Lacy AM et al. Surg Endosc. 1998;12(8):1039–42. 86. Neumann PA et al. Int J Colorectal Dis. 2013;28(5):
45. Vukasin P et al. Dis Colon Rectum. 1996;39(10 Suppl): 599–610.
S20–3. 87. Maggiori L, Panis Y. Best Pract Res Clin Gastroenterol.
46. Bouvy ND et al. Ann Surg. 1996;224(6):694–700; 2014;28(1):183–94.
discussion 1. 88. Collins D, Winter DC. Best Pract Res Clin Gastroenterol.
47. Watson DI et al. Arch Surg. 1997;132(2):166–8; dis- 2014;28(1):175–82.
cussion 9. 89. Hall JF et al. Dis Colon Rectum. 2011;54(3):283–8.
48. Whelan RL, Lee SW. J Laparoendosc Adv Surg Tech A. 90. Cirocchi R et al. Colorectal Dis. 2012;14(6):671–83.
1999;9(1):1–16. 91. Klarenbeek BR et al. Ann Surg. 2009;249(1):39–44.
92. Klarenbeek BR et al. Surg Endosc. 2011;25(4):1121–6. 110. Deffieux X et al. Eur J Obstet, Gynecol Reprod Biol.
93. Klarenbeek BR et al. Surg Endosc. 2011;25(3):776–83. 2011;158(2):159–66.
94. Gervaz P et al. Ann Surg. 2010;252(1):3–8. 111. McKernan JB, Champion JK. Endosc Surg Allied
95. Gervaz P et al. Surg Endosc. 2011;25(10):3373–8. Technol. 1995;3(1):35–8.
96. Angenete E et al. Ann Surg. 2016;263(1):117–22. 112. Moreaux G et al. J Minim Invasive Gynecol. 2009;
97. Schultz JK et al. JAMA. 2015;314(13):1364–75. 16(5):643–5.
98. Ceresoli M et al. World J Emerg Surg. 2016;11(1):42. 113. Chiong E et al. Urology. 2010;75(3):574–80.
99. Turley RS et al. Dis Colon Rectum. 2013;56(1):72–82. 114. Yamamoto M et al. JSLS. 2011;15(1):122–6.
100. Daher R et al. World J Gastrointest Surg. 2016;8(2): 115. Coakley K et al. American Society of Colon and
134–42. Rectal Surgeons Annual Scientific Meeting; June
101. Oliveira L et al. Surg Endosc. 1997;11(1):19–23. 10–14, 2017; Seattle, WA.
102. Liu J et al. Tech Coloproctol. 2005;9(1):9–14. 116. Material Safety Data Sheet: FRED Anti-Fog Solution.
103. Miyoshi N et al. World J Gastrointest Endosc. 2016; North Haven CUSS. Material Safety Data Sheet:
8(15):541–5. FRED Anti-Fog Solution. North Haven, CT: United
104. Cadeddu F et al. Tech Coloproctol. 2012;16(1):37–53. States Surgical 2004. [Available from: http://www.
105. Tou S et al. Cochrane Database Syst Rev. 2015(11): autosuture.com/imageServer.aspx?contentID=6591
CD001758. &contenttype=application/pdf.]
106. Byrne CM et al. Dis Colon Rectum. 2008;51(11): 117. Ott DE et al. JSLS. 1998;2(4):321–9.
1597–604. 118. Carnuccio P et al. Tech Coloproctol. 2014;18(1):
107. van Iersel JJ et al. World J Gastroenterol. 2016;22(21): 5–12.
4977–87. 119. van Oostendorp S et al. Surg Endosc. 2017;31(1):
108. Varma R, Gupta JK. Surg Endosc. 2008;22(12): 64–77.
2686–97. 120. Gibor U et al. J Laparoendosc Adv Surg Tech Part A.
109. Ahmad G et al. Cochrane Database Syst Rev. 2012;2: 2018;28(1):65–70.
CD006583. 121. Fung AK, Aly EH. Br J Surg. 2012;99(10):1353–64.
15
Medical legal issues
TERRY C. HICKS AND DAVID E. BECK
to provide practical information that can be used by medi-

CHALLENGING CASE cal care providers.
A 60-year-old woman with a strongly positive family

history of colorectal cancer undergoes a colonos-
copy. She has a 1.5 cm pedunculated polyp snared CURRENT STATUS OF MEDICAL
from the transverse colon. Five days after the pro- LIABILITY AND INSURANCE
cedure, she presents to the emergency room with a
lower gastrointestinal (GI) bleed. She is hemodynami-
cally stable, and you admit her for observation. She The cost of medical malpractice insurance began to rise in
remains stable and is discharged 2 days later with no the early 2000s after a period of essentially flat process. Rate
further bleeding episodes. The hospital risk manager increases were precipitated in part by the growing size of
calls you to discuss this case. claims, particularly in urban areas. Among other factors
driving up prices was a reduced supply of available cover-
age as several major insurers exited the medical malpractice
business because of the difficulty of making a profit (1). The
CASE MANAGEMENT number of claims filed has dropped over the last few years,
and the cost of claims has moderated. Malpractice payouts
When you meet with the risk manager, you inform her in 2016 were over $3.8 billion (2). However, medical mal-
that you had seen the patient in your office prior to the practice costs have outpaced other tort areas. Regardless of
procedure. During this office visit, you had discussed whether a case is won, going to court is expensive.
with the patient her risk factors, indications for the pro- A study from the New England Journal of Medicine (3)
cedure, details of the procedure, and potential risks. found that 7.4% of all physicians could expect a medical
This conservation was documented in your office note, malpractice claim to be filed against them in any given year,
and the patient signed a consent for the procedure. but only 1.6% of physicians would be subject to a claim that
The procedure was performed in the usual fashion. would lead to a payment.
Postpolypectomy is a recognized complication of the Many states have considered tort reform, but limits on
procedure. You feel that you have a good relationship noneconomic damages have been found unconstitutional in
with the patient, and the records are well documented. some states. States with tort limits have seen reductions in
Although any untoward outcome could lead to litiga- suits filed (1). Despite these efforts, the cost of liability insur-
tion, the risk manager agrees that you have taken the ance is leading some doctors to retire prematurely, relocate
appropriate actions to minimize your risk. their practices to nonlitigious areas, practice without insur-
ance, or drop risky procedures.
The evidence is clear that there remain serious medical
Despite efforts at tort reform in the United States, medi- malpractice issues in the United States, and multiple grass-
cal liability remains a serious issue. This chapter briefly roots efforts continue to address this on a local as well as on
reviews important aspects of the U.S. medical liability situ- a national level. The current environment endangers care in
ation and then addresses some risk-prevention techniques rural areas and among low-income, inner-city populations.
for colorectal surgeons. This includes a general overview of Based on comprehensive jury verdict research, there is lit-
the legal process pertaining to medical malpractice issues tle doubt that soaring jury verdicts are serious, ongoing prob-
and tips to help prevent and defend such cases. It is intended lems. At present, half of the jury awards in medical liability
141
142 Medical legal issues
cases exceed $1 million, and the average award is $4.7 million fragmentation often decreases the opportunity to com-
(4). The number of mega-awards has skyrocketed, especially municate effectively with patients, who have also become
in states with no limits on noneconomic damages. For the much more demanding consumers, increasingly aware of
past several years, juries have awarded lottery-size verdicts of their “rights” through media and lawyer advertising. Health
$80 million, $90 million, or even $100 million (5). insurers contribute to the problem, not only by creating
Many physicians feel the medical liability crisis is very incentives that discourage referrals to a specialist but also by
straightforward. They note that medical liability costs are placing restrictions on the specialist, once referral is made,
soaring faster than the rate of overall health-care costs and that can impede opportunities to establish rapport with the
the rate of inflation, leading directly to increasing insurance patient. Under such circumstances, it is important to make
premiums for doctors. In short, their position is that the liti- the most of each opportunity to listen to the patient, remem-
gation system generates too many lottery-size verdicts and ber and use the patient’s name, explain procedures in lay
encourages too many meritless cases. As a result, insurance terms (avoid medical terminology), and take the time neces-
companies are fleeing the market, making it more difficult sary to answer any and all questions. Remember that listen-
for doctors to obtain liability coverage at any price. The ing to a patient’s questions and complaints will be much less
U.S. Department of Health and Human Services concluded: time consuming than defending a malpractice claim.
“The excess of a litigation system raises the cost of health- Still one of the best books for improving communi-
care for everyone, threatens Americans access to care, and cation and relationships is Dale Carnegie’s How to Win
impedes efforts to improve the quality of care” (6). Friends and Influence People (10). For a more practical guide
Other major impacts of the malpractice crisis are the with a medical orientation, one should read Malpractice
practice of defensive medicine and a negative impact on the Prevention and Liability Control for Hospitals, by Orlikoff
young physicians in training. In a patient American Medical and Vanagunas (11).
Association survey, 48% of the students in their third and The frequency of medical malpractice claims has been on
fourth years of medical school indicated the liability situa- the rise since the early 1970s (12). As long as the contingency
tion was a factor in their specialty choice. fee system exists and there is not a loser pay provision, the rise
It is of interest to note that overall, 75% of medical liability in suits against physicians will likely continue. Accordingly,
claims in 2004 were closed without payment to the plaintiff, it is incumbent on the well-educated and well-trained spe-
and of the 7% of the claims that went to a jury verdict, the cialist to be aware of areas of treatment in colorectal disease
defendant won 83% of the time. Unfortunately, physicians that present an increased risk of malpractice claims.
that win at trial still have large fees to pay for their defenses.
The average cost is $93,559 per case where the defendant pre-
vailed at trial. In all cases where the claim was dropped or HIGH–RISK AREAS IN COLORECTAL
dismissed, the cost of the defendants averaged $18,774 (7). TREATMENT
Until medical liability issues are resolved, physicians will
be forced to continue to deal with the present medical legal
climate, and it is our hope that the following information The following circumstances associated with increased risk
will provide some guidelines to lower their exposure to med- for malpractice claims in colorectal disease have been iden-
ical legal risks by utilizing proactive risk management steps. tified (13):
In today’s litigious society, physicians who practice good
medicine, exercise effective communications skills, establish 1. Delay in diagnosis of colon and rectal cancer and
rapport with the patient, and accurately document care have appendicitis
the best chance of averting malpractice claims. Even when 2. Iatrogenic colon injury (e.g., colon perforation)
physicians do all of these, however, a bad outcome may still 3. Iatrogenic medical complications during diagnosis or
result in the patient’s filing of a claim for malpractice (8). treatment
Research appears to support the position that a patient who 4. Sphincter injury with fecal incontinence resulting from
perceives the physician as having good interpersonal skills anorectal surgery
and communication is less likely to sue (9). There are ways to 5. Lack of informed consent
conduct a medical practice that deter patients from making
The colorectal physician who is aware of these potential
claims and, even after one is made, can enhance the chances
high-risk conditions can use risk-prevention strategies to
of winning the case.
avoid litigation.
PHYSICIAN–PATIENT RELATIONSHIP INFORMED CONSENT
Medicine has changed dramatically in the last few decades Physicians should be mindful that consent and informed
because of extraordinary technologic advances that have consent are quite different concepts. Consent implies
resulted in specialization, such as colorectal surgery. This permission. Informed consent is assent given based on
Documentation / Charting 143
information provided or knowledge of the procedure and cannot be overemphasized in deterring lawsuits; however,
its inherent risks, benefits, and alternatives. complete and accurate documentation of patient care is
Courts have long recognized that “Every human being of invaluable to a defense of claims. In addition, good docu-
adult years and sound mind has a right to determine what mentation may well nip in the bud a potential claim when the
shall be done with his own body” (14). The law of informed plaintiff ’s attorney who is considering filing suit reviews the
consent may vary to some degree from state to state, but record and care is fully documented. Plaintiff attorneys are
regardless of the law of the state, each patient should be more likely to bring suit when the case is poorly documented,
allowed an exchange of information with the physician because they can more easily argue that what happened in the
before a procedure is done. Informed consent is not satisfied care of the patient was sinister and improper. Where docu-
by merely having the patient sign a form. It is satisfied when mentation is clear and accurate, the plaintiff ’s attorney may
consent was obtained after full disclosure of the risks, ben- be deterred from filing suit, because what happened is easily
efits, and alternatives of the procedure. proved from the record. Thus, judgment becomes the issue
Many states use the “reasonable practitioner standard” to when documentation is accurate, and judgment used by phy-
judge whether informed consent was obtained. This standard sicians in most cases is easier to successfully defend than a
focuses on what a reasonable physician would disclose. The vague, evasive, and poorly documented chart.
physician’s duty is not to disclose all risks but primarily those The following are some time-honored rules for charting
that are significant or material. A risk is material depend- that help defend against malpractice claims.
ing on its likelihood of occurrence or the degree of harm it
presents. The focus is on whether a reasonable person in the CHARTING
patient’s position probably would attach significance to the
specific risk. This is the “reasonable patient standard” that 1. Thorough and accurate charting is your primary shield
some state courts apply. to liability.
Moreover, to prevail on a claim for lack of informed con- 2. If an event in which you are involved gives rise to litiga-
sent, in most states the patient must still prove causation (i.e., tion, chances are your testimony will not be taken for 1
that he or she would not have consented to the procedure if or 2 years after the event. Accordingly, your chart will
informed of the risk). As a practical matter, it is difficult for a provide the content and guidelines for your testimony.
patient to persuade a judge or jury that even though the sur- 3. Most important: If it is not charted, it was not done, nor
gery was needed to relieve pain or disease, he or she would was it observed, administered, or reported. In Smith v.
not have consented if told of the risk of, for example, perfora- State through Dept of HHR (15), the court stated:
tion of the colon. This is particularly true when a patient is
The experts concluded that decedent’s condition
told of much more severe risks such as death or paraplegia
required continued monitoring and that chart-
and agrees to the surgery. In that regard, the question to be
ing should have been done on a regular basis. The
answered by the judge or jury on an issue of informed con-
experts also agreed that the lack of documentation
sent is whether a reasonable patient in the plaintiff’s posi-
indicated that no one was properly observing the
tions would have consented to the treatment or procedure
decedent, based on the standard maximum “not
even if the material information and risks were disclosed.
charted, not done.”
The following points should always be discussed with the
The evidence indicates that the decedent was
patient:
not adequately monitored in this case. The nurses
●● The general nature of the proposed treatment or did not specifically recall the patient, and thus the
procedure best evidence of their actions would have been the
●● The likely prospects for success of the treatment (but no documentation of the chart (emphasis added).
guarantee) 4. General guidelines
●● The risks of failing to undergo the treatment a. If you are the treating or primary physician, make a
●● The alternative methods of treatment, if any, and their daily entry on the chart.
inherent risks b. Chart at the earliest possible time.
c. If the situation prevents you from charting until
Suffice it to say that good rapport with the patient cou- later, state why and that the recorded times are best
pled with accurate and complete charting are the best tools estimates and not fully accurate.
to deter suits based on informed consent and to provide a d. Always record the time (designate a.m. or p.m.) and
heavy shield in defending them. the date of every entry.
e. Chart all consultations.
f. Never black out or white out any entry on a chart.
DOCUMENTATION Should you make a mistake in charting, place a single
line through the erroneous entry and label the entry
The importance of good communication and rapport with “error in charting.” However, if a hospital policy
patients (i.e., treating patients as you would like to be treated) exists that governs errors in charting, follow it. An
addendum is acceptable if placed properly in sequence

with the date and time it is made. An addendum ELECTRONIC HEALTH RECORD
squeezed between progress notes is inappropriate.
g. Write legibly. The federal government has encouraged adoption of elec-
h. Spell correctly. tronic health records (EHRs), which have four major func-
i. Chart professionally; do not impugn or insult the tions: documentation of clinical findings, recording of test
patient. and imaging results, computerized provider order entry,
j. Never alter the medical records. and clinical decision support. Basic EHRs include access
k. Do not insult, impugn, or criticize colleagues, to clinical information, provider notes, and order entry
coworkers, or support staff. for medications. Comprehensive EHRs include these func-
l. Always designate the dose, site, route, and time of tionalities as well as more extensive order entry (labs and
medication. tests) and clinical decision support. The legal implications
m. Sign your entries on the chart. of EHRs extend beyond changes in medical liability and
n. Do not chart an incident report in your notes. include liability related to privacy, disputes over ownership
o. Chart objectively, not subjectively; do not use of health data, and heightened vulnerability to Medicare or
ambiguous terms (examples follow). Medicaid fraud (16). Additional advantages and limitations
of the electronic medical record are discussed in depth in
Subjective Objective Chapter 7. This section highlights aspects related to medical
legal liability. Implementation of electronic medical records
Patient doing well Patient denies any complaints. can be divided into three phases: initial implementation,
Awake, alert, and oriented. mature systems in place, and widespread use.
Vital signs stable: BP, During initial implementation, transition from paper
100/70: P, 72; R, 18 to electronic records may create documentation gaps.
Respirations regular and Inadequate training on EHR systems may create new error
unlabored pathways such as incorrect or missing data or gaps in com-
Breath sounds clear and equal munication. System failure or bugs in the system could
bilaterally on auscultation. adversely affect clinical care. As the system matures in place,
No rales or rhonchi noted e-mail advice multiplies the number of patient interactions
Breath sounds within Pedal pulses noted bilaterally. and provides the potential to offer advice with inadequate
normal limits (WNL) Nail beds blanch quickly investigation or examination of the patient. Increased docu-
Circulation check WNL and toes warm to touch mentation creates more discoverable evidence for plaintiffs,
Patient denies any pain or including metadata. Copy and pasting of history, physicals,
tingling and notes risks missing new information and perpetuates
previous mistakes. Failure to respond to patient e-mails in
a timely fashion could constitute negligence. Information
p. Document use of all restraints and safeguards, and overload could cause clinicians to miss important informa-
patient positioning (extremely important in surgery). tion. Departure from clinical decision support guidelines
q. Document all patient noncompliance. could bolster a plaintiff’s case.
r. Document all patient education and discharge Despite these potential limitations, successful imple-
instructions, and patient responses. mentation of EHR systems may reduce errors in care and
s. Always document patient status on transfer or adverse events. Better documentation may improve the abil-
discharge. ity to defend against malpractice when care was appropriate.
t. Record the patient’s name on each page of the medi- Compliance with clinical decision support care guidelines
cal chart. provides evidence that the legal standard of care was met,
u. Use accepted medical abbreviations. Finally, better messaging may improve patient satisfaction
v. Do not chart in advance. and communication, reduce adverse events, and reduce
5. Guidelines for charting in the ambulatory setting claims or a propensity to sue.
a. Always chart the return visit date and that the date As EHRs become widespread, better access to clini-
was provided to the patient. cal information could lead to legal duties to act on infor-
b. Always chart all cancelled and missed appointments. mation. Widespread use of clinical decision support may
c. Document all telephone conservations and their solidify standards of care that might otherwise be subject to
content. debate. Rise of health information exchanges may increase
d. Chart all prescriptions and refills, as well as patient a clinician’s duty to search for information generated by
teaching regarding prescriptions. other providers. Finally, failure to adopt and use electronic
e. Chart all follow-up and discharge instructions. If technologies may itself constitute a deviation from the
possible, have the patient or his or her representative standard of care. These potential liabilities are balanced by
cosign these instructions. the situation where adherence to clinical decision support
Anatomy of a malpractice suit / Deposition pitfalls 145
recommendations may protect providers from liability. The cooperation with the defense attorney in preparation is
rise of health information exchanges may facilitate sharing fundamental. Above all, a physician must be his or her own
of information, leading to better care and fewer claims. person.
In summary, EHRs have the potential to improve care and Thorough preparation will assist physicians in giving a
may increase or decrease medical liability. Clinicians must deposition with which they will be perfectly comfortable
understand their systems and use them to their full potential. when they see the printed transcript, that is, one that will
be easily defended, should any part of it later be challenged.
The following suggestions for giving testimony in depo-
sitions can be helpful to the physician:
ANATOMY OF A MALPRACTICE SUIT
1. Tell the truth; you must testify accurately.
INITIAL PHASE 2. Do not guess or speculate. If you do not know the
answer to a question, say so.
Once a patient initiates a claim for medical malpractice, the 3. If you are not certain of what the attorney is asking,
physician should immediately place a call to the risk man- ask that the questions be clarified or repeated. Do not
ager or the malpractice insurance carrier. An attorney will attempt to rephrase the question for the interrogator
usually be selected, and the physician should insist that the (e.g., “If you mean such and such”).
appointed counsel be experienced and have a well-estab- 4. Keep your answers short and concise. Do not volunteer
lished reputation in the handling of malpractice cases. information. Answer only the question posed.
Physicians should work closely with the defense attorney 5. Be courteous. Avoid jokes and sarcasm.
to review and analyze the allegations of the suit, with par- 6. Think about each question that is posed. Listen to each
ticular focus on the strengths and weakness of the case. This work. Formulate an answer, and then give the answer.
team effort can often substantially enhance the strength of Do not permit yourself to become hurried.
the defense by educating the attorney on the medical aspects 7. Do not argue with opposing counsel. If an argument is
of the case. necessary, your attorney will do it for you.
8. If you realize that you have given an incorrect answer
PRETRIAL DISCOVERY to a previous question, stop at that moment and say so,
and then correct your answer.
During this stage, each side will discover the facts and opin- 9. Be aware of questions that involve distances and time.
ions in the case. Written questions, or interrogatories, can usu- If you make an estimate, make sure everyone knows it
ally be propounded to obtain written responses. Depositions is an estimate.
usually follow the written discovery and are important to the 10. Do not lose your temper, no matter how hard pressed.
overall outcome of the case. Before testifying by deposition This may be a deliberate ploy; do not fall for it.
or otherwise, it is advisable that the physician be thoroughly 11. Do not anticipate questions. Be sure to let the attor-
familiar with the facts, including previous and subsequent ney completely finish the question before you begin to
medical care of the patient and the allegations against the respond.
physician. This requires careful review of medical records, 12. Do not exaggerate or brag.
other depositions, and all medical data related to the case. A
conference should be held with the attorney before the physi- DEPOSITION PITFALLS
cian’s deposition. They physician should allow ample time to
confer with the defense attorney before testifying. Remember Testing your memory of the case. You have the right to refer
that the judicial system is adversarial, and the purpose of the to the chart or hospital records whenever you wish. Your
deposition is not to convince the plaintiff’s attorney to under- memory is usually a composite of events you recall as jogged
stand the case is frivolous. The physician is there to answer by your records. Watch for generalities, ploys, and tricky
the questions and defend the care administered, not to edu- questions by the plaintiff’s attorney during the deposition.
cate the plaintiff’s attorney. Generalities. Often the plaintiff’s attorney will begin
The deposition is simply the physician’s testimony, given with general questions, such as, “Doctor, how do you treat
under oath, before a court reporter, in an informal setting. a patient when you suspect he has X disease?” In all likeli-
Attorneys for both defendant and plaintiff are present. Any hood, the lawsuit to which you are a party involves X dis-
party to the lawsuit may be present, but often the physician ease or involves the plaintiff’s attorney trying to make it X
is the only party present. The testimony is taken down in disease. You really cannot answer this question, and you
question-and-answer form. Under the laws of discovery, the should say just that. X disease probably occurs in various
plaintiff’s attorney has the right to ask the defendant phy- forms, and you have been given no particular informa-
sician proper questions. The physician is present simply to tion—no patient complaints, no patient history, no findings
discharge a legal obligation to answer proper questions. on physical examination, no results of laboratory studies, no
The physician’s deposition is most important. A good clinical impression—all factors you must know to diagnose
effort is essential for an effective presentation. Close and treat intelligently. The question is simply too general.
A similar question might be, “Doctor, what are the stan- Doing things differently. Almost all malpractice cases
dards for making a diagnosis of X disease?” Again, you involve the “retro spectroscope” or Monday morning quar-
should advise that this question is too broad and defies ratio- terbacking to suggest the physician knew things beforehand
nal response because no details have been given. You, as a that were only learned later or that the physician has 100%
physician, do not immediately diagnose X disease or any control over the healing process.
other disease. You evaluate all the data in light of your formal
training and clinical experience in considering or making a
diagnosis. Patient signs and symptoms are innumerable. You Question: “Doctor, is there anything you would do
must have specifics. For example, in one doubtful clinical differently now if you had Mrs. White’s case to treat
presentation, you may have to order a particular set of labo- again?”
ratory studies; in another, the evidence of a certain disease Appropriate response: “My recommendations to Mrs.
process may be more definitive and clear-cut from the his- White were based on her complaints, her history, and
tory and clinical examination. findings at the time and on my clinical impression at
A proper question is, “Doctor, what are the characteris- that time. The course I recommended was appropri-
tics of X disease?” Particularly if your case involves X dis- ate on the basis of those factors.
ease, you should know its characteristics, but you should
also point out that they are general characteristics and most
certainly will vary in specific instances.
The point is, you must avoid generalities. You must demand Question: “Doctor, you did not intend for Mrs. White
specifics. Try to make the questioner stick to the specific case. to have this complication, did you?”
Appropriate response: “Of course, no harm to Mrs.
PLOYS White was intended. At the time of my recommenda-
tions, there were good prospects for a good result.
The procedure (or regimen) does have known com-
Question: “Doctor, you have no memory of events
plications, and that is why the risks were explained to
independent of your records, do you?”
her beforehand.”
Ploy: “Doctor, if an event is not noted in your records
or in the hospital records, is it fair to say that event
did not occur?”
Many other factors are involved in preparing for and
Appropriate response: “That is incorrect. It is impossi- successfully testifying by deposition or at trial (17). Suffice it
ble for a physician to note everything that occurs. My to say that effective and sincere testimony is critical to a suc-
records are for my own use, to jog my memory. Thus, cessful defense in malpractice cases. Ineffective testimony
I note pertinent highlights, which when later reviewed can render a defensible case indefensible. Many tricks and
give me the complete picture at the time in question.” ploys may be used by the plaintiff attorney, and the physi-
cian who is prepared with a basic understanding of how
to answer such questions can substantially enhance the
Remember that physicians treat patients, not charts. You defense.
may properly testify to the following:
1. What you actually recall TRIAL

2. What you recall with the assistance of your records
3. What is recorded After pretrial discovery, the physician should have a clear
4. What your routine or standard procedure is, even when understanding of the evidence and witnesses, the experts in
such is not recalled and not recorded particular, to be used against him or her at trial. Working
with the defense attorney to rebut this evidence and to assist
Tricky questions. Many plaintiff attorneys will use ques- with selection of expert witnesses to testify for the defense
tions cleverly phrased to evoke a response that can later be is strongly advised and helps the physician to prepare the
used against the physician. defense.
Possibilities. Questions phrased in terms of possibility At the trial, the physician is carefully observed at all times
invite speculation and are improper. The criterion is reason- by the judge and jury, and the physician’s trial testimony,
able medical probability. mannerisms, and behavior are critical to a favorable verdict.
A well-trained and educated physician who portrays a sin-
cere, conscientious, and caring attitude about the patient’s
Question: “Doctor, isn’t such and such possible?” or well-being greatly increases the chances of a favorable jury
“Couldn’t such and such have happened?” verdict, even where severe complications have occurred and
Appropriate response: “Most improbable.” there may be questions of the appropriateness of the course
of treatment chosen.
References 147
6. Addressing the New Health Care Crisis. https://aspe.

CONCLUSION hhs.gov/basic-report/addressing-new-health-care-
crisis-reforming-medical-litigation-system-improve-
The defense of medical malpractice claims is similar to the quality-health-care. Accessed, July 23, 2018.
defense of criminal cases. The physician stands accused, 7. American Medical Association Medical liability
and his or her reputation is usually an issue of great impor- reform June 2006. https://www.ama-assn.org/sites/
tance. The emotional costs to the physician are sometimes default/files/media-browser/premium/arc/mlr-now.
staggering. The physician should recognize that until addi- pdf. Accessed July 23, 2018.
tional tort reform is enacted, these cases will likely continue 8. Entman SS et al. JAMA 1994:272:1588–91.
to increase and should be dealt with as a regrettable aspect 9. Hickson GB et al. JAMA 1992:267:1359–63.
of the practice (18). Under these circumstances, it is best to 10. Carnegie D. How to Win Friends and Influence
accept the reality of the medicolegal arena and use the best People. New York, NY: Simon and Schuster, 1936.
means available to aggressively defend and win the mal- 11. Orlikoff J., Vanagunas A. Malpractice Prevention and
practice case (19). Liability Control for Hospitals. Chicago, IL: American
Hospital Association, 1988.
REFERENCES 12. Danzon PM. Law Contemp Probl 1986;49:57–84.
13. Kern K. Dis Colon Rectum 1993;36:531–9.
1. Medical malpractice payout analysis. http://www. 14. Schloendorff v. Society of New York Hospital, 2112
diederichhealthcare.com/the-standard/2017- NY125, 105 NE 92,93;1914.
medical-malpractice-payout-analysis/ Accessed 15. Smith v. State, through Dept of HHR, 517 SO2d 1072
October 26, 2017. App 3d Cir;1987.
2. Medical malpractice update. https://www.iii.org/ 16. Mangalmurti SS et al. NEJM 2010:363:2060–7.
issue-update/medical-malpractice Accessed 17. Taraska JM. The Physician as Witness. In Legal Guide
October 26, 2017. for Physicians. New York, NY: Matthew Bender, 1994,
3. Jena AB et al. N Engl J Med. 2011;365:629–63. pp 1–56.
4. Manhattan Institute Malpractice maladies. https:// 18. Taraska JM. Tort reform. In Legal Guide for
www.manhattan-institute.org/pdf/TLI-IL.pdf. Physicians. New York, NY: Matthew Bender, 1994,
Accessed July 23, 2017. pp 1–64.
5. Gay CF, Hicks TC. Medical legal issues. In: Whitlow 19. Gay CE. Medicolegal issues. In: Hicks TC, Beck DE,
CB, Beck DE, Margolin DA, Hicks TC, Timmcke AE. Opelka FG, Timmcke AE. (eds). Complications of
(eds) Improved Outcomes in Colon and Rectal Surgery. Colon and Rectal Surgery. Baltimore, MD: Williams
London, UK: Informa Healthcare, 2010:148–53. and Wilkins, 1996, 468–77.
16
Miscellaneous conditions
ALEXANDER T. HAWKINS AND M. BENJAMIN HOPKINS
CLINICAL PRESENTATION
CHALLENGING CASE
Most patients report perianal itching and burning that is
A 28-year-old construction worker is referred to your exacerbated during hot, humid weather or after exercise. On
office for evaluation of a 9-month history of anal physical exam, the affected area can vary from mild ery-
itching. He has oscillating constipation and diarrhea. thema and excoriations to marked skin thickening, crack-
Physical exam reveals excoriations and macerated ing, and lichenification (Figure 16.1). Excessive scratching
anoderm. Digital rectal exam is normal. Anoscopy is or vigorous cleansing of the affected area in an attempt to
unremarkable. alleviate symptoms can instead exacerbate the condition
and lead to a downward spiral of the disease.
CASE MANAGEMENT ETIOLOGY AND DIAGNOSIS

Empiric treatment for pruritus ani was initiated with Pruritus ani can be classified as either primary (idiopathic,
increased dietary fiber and meticulous anal hygiene accounting for 50%–90% of cases) or secondary (due to a
with instructions to minimize anal trauma. Punch biopsy separate condition) (5). The causes of secondary pruritus ani
is deferred at this time. Differential diagnosis includes can be divided into several broad categories: dermatologic,
anal condyloma, Paget disease, HIV-associated anorec- infectious, systemic disease, neoplastic, and anorectal causes
tal lesions, and other sexually transmitted disease. (2,4). We first discuss common etiologies of secondary pru-
ritus ani, as these are important to rule out in evaluation.
Many dermatologic conditions may present as pruritus
ani. The most common is perianal eczema, but the differen-
PRURITUS ANI tial diagnosis is wide and includes psoriasis, seborrheic der-
matitis, atopic dermatitis, contact dermatitis, lichen planus,
Pruritus ani is a dermatologic condition with a wide range lichen sclerosis, and local malignancies. Proper diagnosis
of etiologies and potential diagnoses. It is characterized by requires a comprehensive history, complete skin examina-
itching or burning at the perianal area. Despite the marked tion, and full thickness skin biopsy, preferably containing
impact on a patient’s quality of life, it remains an underdiag- both normal and abnormal skin. Recent data have called
nosed condition. Most patients choose to self-medicate and attention to the prevalence of systemic contact dermatitis
do not seek medical care (1). The incidence of pruritus ani is in patients with pruritus ani, advocating for patch testing
estimated to range from 1% to 5% in the general population (6). In one study of 150 patients with perianal dermatoses
with men more affected than women by 4:1. It is most com- and pruritus ani, 40% had a positive reaction, and 20% of
monly diagnosed in the fourth through sixth decades of life tests revealed relevant allergens to be avoided (7). Treatment
(2–4). Because pruritus ani often has a multifactorial etiol- should be focused on the condition identified.
ogy and high chronicity, most patients report symptoms Infectious causes of pruritus ani are multiple and can
and self-medication for a number of years. Pruritus ani can be fungal, bacterial, viral, and parasitic in origin. Bacterial
be caused by a spectrum of conditions, the most common of agents include ß-hemolytic streptococci, Staphylococcus
which is perianal eczema. Due to the wide range of causes, aureus, and Corynebacterium (8). Perianal streptococcal
the disease can be a challenging condition to treat for both dermatitis occurs in adult patients more often than reported.
the patient and clinician. It is mainly caused by group B ß-hemolysing Streptococcus.
148
Anal condyloma 149
reached. Etiology can range from poor hygiene, to poorly

absorbent or ventilated clothing, to excessive or improper
cleaning, to dietary intolerances. Fecal soilage is a strong
caustic substance leading to skin irritation. Causes of soil-
age can include incomplete wiping due to skin tags, loose or
tenacious stool, and poor anal sensation or sphincter tone.
A physiologic basis has been forwarded via the observation
that patients with pruritus ani demonstrated a greater rise in
rectal pressure associated with decreased anal pressure and
longer duration of internal anal sphincter relaxation (17).
Tight fitting and nonaerating underwear can exacerbate the
problem. Foods such as caffeinated beverages, chocolate,
tomatoes, and citrus fruit have been associated with pru-
ritus ani (10,18,19). A dose response is seen in coffee con-
sumption, possibly due to decreased internal sphincter tone
similar to that found in relaxation of the lower esophageal
sphincter in patients with gastroesophageal reflux disease.
TREATMENT
Treatment of secondary causes of pruritus ani should focus
Figure 16.1 Pruitis ani with excoriations. on resolution of the underlying etiology. For patients with
idiopathic pruritus ani, simple reassurance is the best initial
Its diagnosis is important because it can cause serious sys- treatment. Lifestyle changes should include improved clean-
temic infections, especially in the elderly and in newborns. liness, changes in clothing, and diet modification with avoid-
Antibiotics resolve the condition in a high proportion of ance of the above-mentioned foods. Patients should cleanse
patients (9). Though thought to be rare, data suggest that themselves several times a day and avoid alcohol-based prod-
fungal infections may be more prevalent in patients than ucts and excessive wiping. Patients should be instructed to
originally suspected (10,11). Fungal infections can account dry the areas with a hair dryer when possible. If the patient
for 10%–43% of infectious causes of pruritus ani (8,10). The reports excessive moisture in the perianal region, sprinkling
most common fungi identified is Candida albicans (12). the area with baby powder and placing a dry cotton ball on
Pinworms are a common cause of nocturnal and postdef- the anal verge can help alleviate symptoms, as well as provide
ecation symptoms, especially in children. HIV and sexually a reminder not to scratch the area. Hydrocortisone cream can
transmitted diseases are reviewed later in this chapter. be trialed but only for 2 weeks. Barrier creams, especially the
Though exact causation remains unknown, a number of zinc oxide–based types, can be helpful in protecting the skin.
systemic conditions are associated with pruritus ani includ- For patients who fail conservative measures after 4–6
ing diabetes mellitus, liver disease, pellagra, renal fail- weeks, a more detailed examination for a secondary cause
ure, hyperthyroidism, and vitamin deficiencies (A and D) should be undertaken. If no secondary cause can be found,
(2,4,8). Dermatologic neoplasms, including condyloma acu- more extreme options are available. An intradermal appli-
minata, Paget disease, and Bowen disease can present with cation of 1% methylene blue solution has been associated
pruritus ani. Biopsy is essential in securing these diagnoses. with a positive effect on idiopathic pruritus ani with mild
Condyloma acuminata and Paget disease are discussed later side effects related to sensory cutaneous innervation in all
in this chapter. Bowen disease, also known as intraepithe- patients within the first 4 weeks following the procedure
lial squamous cell carcinoma in situ of the anus, is rare but but with only a 20% 5-year success rate (20). Another study
can often present with pruritus ani (13). Given the indolent suggests that tacrolimus 0.1% ointment may be an effective
natural history, treatment has shifted away from aggressive treatment for idiopathic pruritus ani, resulting in a symp-
wide local excision to surveillance with targeted biopsies tom reduction in 68% of the patients after 2 weeks of treat-
and destruction of discrete tissue (14). ment (21). Long-term data are lacking.
A number of anorectal conditions can lead to pruri-
tus ani, either on their own or via leakage and soiling.
Hemorrhoids, skin tags, and chronic anal fissures all have
been associated with pruritus ani. Anoscopy is usually suf- ANAL CONDYLOMA
ficient to make the diagnosis, and treatment of the underly-
ing condition often corrects the pruritus (15,16). Anal condyloma, or anal warts, present as growths in the
When a comprehensive investigation fails to identify perianal skin or anal mucosa. Human papillomavirus
associated disease conditions that cause secondary pruritus (HPV) is the causative pathogen in condyloma. The condi-
ani, the diagnosis of primary or idiopathic pruritus ani is tion affects nearly 20 million sexually active adults with 5.5
150 Miscellaneous conditions
million new cases occurring each year worldwide (22). The such, it should be applied only to the wart itself. Patients
virus is spread by close contact with an infected individual, need to return to the office every 7–10 days for reapplica-
and autoinoculation to other body surfaces is possible. tion. Clearance rates reported in randomized trials range
HPV-induced genital warts are the most common anorec- from 56% to 81% (28). Podophyllin is a topical agent that
tal infection among homosexual men. In this population, can be applied either in the office or by the patient at home.
intraanal lesions are especially common as are fistula in ano Again, great care must be taken to apply the agent only to
(23,24). Patients who are HIV positive or otherwise immu- the condyloma. Podophyllin is a destructive agent that leads
nosuppressed are (1) more likely to develop anal condyloma, to necrosis. Pregnancy is an absolute contraindication for
(2) less likely to respond to treatment, and (3) more likely to the use of topical podophyllin. Treatment success is as high
experience recurrence of anal condyloma. Patients can pres- as 50%, but recurrence is almost inevitable, requiring repeat
ent with pain, pruritus, discomfort, bleeding, or even par- administration. Complications of trichloroacetic acid and
tial obstruction, depending on the location and size of the podophyllin can include skin necrosis, fistula in ano, and
condyloma. Diagnosis is achieved through visual inspec- anal stenosis. Imiquimod is a newer agent that not only
tion and anoscopy. High-risk HPV types are more likely to has destructive properties but also stimulates the innate
progress to invasive squamous cell carcinoma of the anus. and cell-mediated immune response to clear HPV-infected
Oncogenic HPV infection is often associated with having cells. A systematic review of randomized trials found
a high lifetime number of either female sexual partners or imiquimod superior to placebo for achieving complete and
male anal-sexual partners (25). partial regression of anogenital warts but did not identify
superiority for recurrence rates or new wart development
TREATMENT (31). The cream is applied to the wart and left in place for
8 hours before being washed off. A single treatment course
Treatment options include excision, destruction, and topi- involves application three times a week for up to 16 weeks.
cal therapy. Excision of the condyloma allows for tissue Imiquimod appears to have decreased incidence of skin
diagnosis as well as typing of the causative papillomavirus necrosis and fistula in ano compared to the other two topi-
(26). Due to the risk of malignant transformation, histo- cal agents. It should be noted that all topical agents have
pathological examination is recommended for all patients decreased efficacy in treating highly keratinized warts.
undergoing treatment. Excision usually is performed in the Giant condyloma acuminatum (Buschke-Löwenstein
operating room using monitored anesthesia care, although tumor) of the anorectal region is a highly aggressive tumor
treatment in an office setting is possible in the case of dis- with the propensity for recurrence and malignant trans-
crete lesions. Care must be taken to leave the underlying formation but without metastatic potential. A high rate of
musculature, as well as normal skin and mucosa, intact. recurrence is seen in patients with long duration of the dis-
Complications of intraanal excision include strictures of ease. Salvage of patients with recurrences can be achieved
the anal canal. Unfortunately, surgical excision has a high successfully with radical surgery (32). There are case reports
recurrence rate ranging from 9% to 46% (27,28). of primary rectal adenocarcinoma presenting as a giant
Destructive techniques used in the treatment of con- perianal mass mimicking giant condyloma acuminatum,
dyloma include electrocautery, cryotherapy, and laser. and this diagnosis should be considered when treating (33).
Fulguration of condyloma using electrocautery is an effec- HIV-positive and immunosuppressed patients have a
tive tool in treating condyloma. It can be performed in the higher risk of developing invasive squamous cell carci-
ambulatory setting with monitored anesthesia care. Deep noma of the anus and should be screened yearly to detect
burns, which can damage and scar both surrounding skin precursor lesions. This is commonly performed with anal
and the sphincter complex, must be avoided. Curettage of Pap testing using a liquid medium to capture epithelial cells
the fulgurated tissue with a sponge or curette is an effective for analysis (34). Patients with suspicious cytology can be
adjunct. Cryotherapy is similar to electrocautery in that the referred for surgical evaluation. High-resolution anoscopy
condyloma and underlying tissue are destroyed with liquid can be used to inspect for areas of high-grade dysplasia. The
nitrogen. Application causes tissue damage by formation best management of high-grade anal dysplasia, which is the
of ice crystals, leading to disruption of cell membranes and precursor to invasive squamous cell carcinoma, remains
cell death. Treatment success rates from randomized tri- unclear. Patients with high-grade intraanal dysplasia who
als range from 44% to 75% (28). Laser therapy is another undergo ablation have recurrence rates in the range of 50%.
destructive technique useful in the elimination of condy- This number is higher in HIV-positive patients; however,
loma. Clearance and recurrence rates after CO2 laser treat- there is a very low risk of progression to anal cancer (35).
ment vary widely (29). Care must be taken to protect care There are currently two HPV vaccines (bivalent and
providers from aerosolized viral particles that can cause quadrivalent) on the market, which offer protection against
respiratory papilloma (30). high-risk oncogenic HPV types. Both are effective at pre-
There are a number of topical therapies available to vention of HPV-associated dysplasia, but treatment of
treat anal condyloma. These therapies are most effective in anogenital warts with any HPV vaccine formulation is
patients with a low number of discrete warts. Trichloroacetic not recommended. Either vaccine should be given prior to
acid is a caustic agent used to chemically burn the wart. As becoming sexually active, and both are approved for both
Anorectal melanoma / Clinical presentation 151
boys and girls aged 9–26 years (36). The quadrivalent vac- esophagogastroduodenoscopy, and computed tomography
cine has been shown to be effective at reducing the rates of of the chest/abdomen/pelvis is prudent.
anal condyloma development and high-grade anal dyspla-
sia, and it also may decrease the risk of anal cancer (37,38). TREATMENT
HPV vaccination of MSM is likely to be a cost-effective
intervention for the prevention of genital warts and anal Previous management of perianal Paget disease included
cancer (39). wide local excision combined with perianal mapping.
Mapping involved sequential punch biopsies in a circular
pattern from the anus while maintaining location within
four quadrants; if a biopsy in the quadrant demonstrated
PERIANAL PAGET DISEASE disease, the resection margins were enlarged to accom-
modate the affected skin. There is debate between wide
Perianal Paget disease (cutaneous adenocarcinoma in situ) local excision and perianal mapping with wide local exci-
was first reported in literature in the late 1800s. Due to sion. Perez et al. found a high rate of local recurrence
its rarity, current recommendations for management are despite utilizing punch biopsies in 1 cm intervals beyond
limited. Despite the infrequent occurrence of this disease, the visualized tumor followed by a wide local excision.
Paget disease and other malignancies warrant a strong This led the authors to question the necessity of mapping
clinical suspicion when dealing with persistent and unusual procedures (41). Isik et al. described local excisions for
perianal lesions. The perineum is the second most com- perianal Paget disease as 0.5–1 cm macroscopic margins;
mon location for extramammary Paget disease. Primary wide local excision was described as a macroscopic mar-
disease is described as the presence of intraepithelial Paget gin of more than 1 cm. They found no significant differ-
cells without an associated malignancy. Secondary perianal ence in the survival of patients with local excision and
Paget disease has an associated malignancy from either epi- wide local excision for primary perianal Paget disease.
dermoid extension of an adenocarcinoma of the anus and Additionally, there were comparable survival rates in
rectum or a synchronous cancer of the bladder, thyroid, patients with primary perianal Paget disease despite posi-
endometrium, and breast (40,41). Associations with mela- tive surgical margins (40).
nomas and acute leukemias also are described (40,41). Local excision with sphincter preservation should be
the goal with noninvasive disease. Patients with invasive
CLINICAL PRESENTATION perianal Paget disease extending from an associated can-
cer proximal to the dentate line should be considered for
Patients typically present with nonspecific anorectal prob- abdominal perineal resection (APR).
lems of pain, pruritus, and perianal bleeding or with a Other modalities used to treat perianal Paget disease
palpable mass. Physical exam findings can range from an include topical imiquimod, photodynamic therapy, radia-
erythematous, scaly plaque to a hypopigmented or hyper- tion, or chemoradiation (40–42). The paucity of Paget dis-
pigmented lesion. Often, the lesion is associated with peri- ease cases makes the studies of adjuvant therapies difficult,
neal fungal infections, psoriasis, and pruritus ani, which and there is not enough data to support their use.
can lead to a delay in diagnosis. A report from the Mayo Follow-up in those with invasive disease should include
Clinic found a median delay in diagnosis of 2 years from the annual imaging of the chest and abdomen as well as an ano-
initial onset of symptoms (42). Patients seen in the colorec- rectal and inguinal lymph node exam with proctoscopy/
tal clinic with longstanding complaints and perianal disease flexible sigmoidoscopy. Those with noninvasive disease
should undergo a punch biopsy to rule out the diagnosis of should have a visual inspection and endoscopy yearly (41).
perianal Paget disease.
ETIOLOGY AND DIAGNOSIS

ANORECTAL MELANOMA
Apocrine glands, hair follicles, and sebaceous glands of the
skin are the sites of origin of extramammary Paget disease CLINICAL PRESENTATION
(40,41). A punch biopsy in the office can confirm the diag-
nosis of a suspicious lesion. While perianal Paget disease Anal melanomas typically present with features similar to
can have a similar gross appearance to Bowen disease, his- benign anorectal disease: bleeding, mass, pruritus, pain,
tologic slides of perianal Paget disease show signet ring cells and tenesmus (43,44). Due to the association with other
that demonstrate vacuolated and pale cytoplasm and acid- anorectal conditions, diagnosis can often be delayed. The
Schiff positive mucin stain (41). average duration of time from symptoms to diagnosis is
Preoperative workup for perianal Paget disease should 4–6 months, and a benign anorectal condition was initially
first focus on ruling out potential associated malignan- misdiagnosed in 46 of 79 patients in one case series (43,45).
cies. There are no guideline recommendations due to the Further confusing the diagnosis is the fact that a large num-
rarity of this disease; however, workup with colonoscopy, ber of reported anal melanomas are amelanotic (43).
ETIOLOGY AND DIAGNOSIS can be made by the colorectal surgeon” (50). HIV screen-
ing in high-risk patients is prudent and can make the initial
Anorectal melanoma comprises only 2% of all melanomas HIV diagnosis in an otherwise asymptomatic patient.
and 4% of all anal cancers (43,44). It is most commonly seen Perianal ulcers have a multifactorial etiology, and treat-
in patients in their seventh decade of life. Patients seen in ment involves identification of the causative agent and
clinic with prolonged symptoms should have a punch biopsy appropriate medical management. Etiologies of anal ulcer
performed in order to rule out the disease. Hillenbrand et al. in the HIV patient include herpes virus, syphilis, cytomeg-
found no difference in patient survival of pigmented versus alovirus, and Cryptococcus (51,52). Surgical management
nonpigmented lesions and highlighted the importance of is reserved for chronic, nonhealing ulcers despite medi-
keeping melanoma in the differential when no pigment is cal management of the underlying disease and includes
present (46). local debridement and unroofing of ulcerative cavities.
Unfortunately, a large proportion of patients present Complications of surgery include prolonged drainage and
with nodal and/or metastatic disease (43,44). Therefore, poor wound healing. Incontinence and superinfections are
appropriate staging after diagnosis should include an ingui- risks that need to be discussed preoperatively, but these
nal node exam, colonoscopy, and computed tomography of risks are unlikely.
the head, chest, abdomen, and pelvis (43). With the advent of highly active antiretroviral therapy
(HAART), patients with well-controlled HIV develop
TREATMENT the same anorectal disorders seen in non-HIV-infected
patients. Abramowitz et al. screened 473 HIV patients on
There are limited data to support a consensus statement on HAART and found condyloma, hemorrhoids, and anal fis-
the management of nonmetastatic anal melanoma. Initially, sures to be the most common anorectal disorders in the
anal melanomas were treated with an APR. While an APR HIV population followed by dermatologic disorders, ano-
does give better local control, there is no clear improvement rectal infections, and incontinence (53). Their screening
in patient survival (44,47). Multiple case series comparing included a visual inspection of the perineum, digital rectal
APR to wide local excision have shown no difference in sur- exam, and anoscopy, which revealed a 44% prevalence of
vival, which has led to newer recommendations of wide local anal condyloma in the 473 patients screened between 2003
excision for initial treatment. Disease-free survival has been and 2004. Of those infected with HIV, men who had sex
shown to be largely determined by metastatic spread and with men were found to have the highest incidence of anal
not local control; therefore, APR is reserved for large or cir- condylomas. Interestingly, they found no increased risk of
cumferential tumors with concern for eventual obstruction hemorrhoidal disease or anal fissures in patents with HIV
(43). Inguinal lymph node dissection has been performed disease in relation to sexual practices (53).
with positive disease, but this is a morbid surgery with lim- Due to a depressed immune system, HIV-positive indi-
ited data on patient outcomes. Radiation for anal melanoma viduals are at increased risk of wound complications follow-
has been described with equivocal results to an APR, leaving surgery. Of those affected, more severe HIV disease leads
ing this as a potential option (48). Treatment with systemic to higher morbidity and mortality from minor surgical pro-
chemotherapy has not been shown to increase survival cedures including hemorrhoidectomy, lateral internal anal
(43,49). Five-year survival remains low with reports ranging sphincterotomy, and transrectal biopsies as well as major
from 6% to 20% (43,44,47,49). A review of the Surveillance, surgical cases. Due to the high complication rates, surgical
Epidemiology, and End Results (SEER) Program database treatment of benign anorectal diseases should be approached
by Kiran et al. demonstrated that stage of the anal mela- carefully, and previous recommendations called for a CD4
noma is the best predictor of prognosis and that patients count higher than 200 (54). With improved drug regimens
with local and regional disease had equivalent outcomes to treat HIV (HAART), Barrett et al. described their experi-
despite local excision or APR (49). ence in operating on the HIV population from 1989 to 1996
As anal melanoma remains a lethal disease regardless and reported on 260 patients receiving treatment for their
of therapy, treatment plans should be individualized with perianal disease (50). They described anorectal and abdom-
each patient with a clear understanding of patient goals and inal surgery in an HIV patient population with a median
potential complications. CD4 count of 175 and with 34% meeting criteria for the
Centers for Disease Control and Prevention’s definition of
acquired immune deficiency syndrome (AIDS). Their study
showed minimal complications from major and minor sur-
HUMAN IMMUNODEFICIENCY VIRUS gery with 2% preoperative complications. The main com-
plication was slow wound healing, and they showed 44%
Patients infected with HIV may have no outward symptoms of patients were completely resolved at 13 weeks. Surgical
save for chronic perianal complaints; therefore, “patients management was not limited in patients with CD4 counts
not known to be HIV-positive but who have multiple peri- less than 200, despite concerns for poor wound healing,
anal disorders, should be questioned regarding their risk of and their patients experienced significant symptom relief.
HIV infection, because the initial diagnosis of this disease Their conclusion was that aggressive therapy was warranted
References 153
regardless of the high recurrence of pathology and pro- Symptoms can include pruritus ani, bloody discharge, and
longed wound healing. pain. Disseminated gonorrhea occurs if the disease is not
Human immunodeficiency disease complicates benign treated; pericarditis, meningitis, and arthritis are manifes-
anorectal disease. While complication rates following sur- tations of disseminated disease. A thick, purulent discharge
gery are low in the HAART era, discussions with patients can be expressed from the anal crypts and is highly suspi-
regarding surgical risks and prolonged wound healing need cious for gonococcal proctitis. This discharge should be col-
to be emphasized. There is a high rate of recurrence with anal lected on Thayer-Martin plates for identification via culture.
condyloma in the HIV population, and continued surveil- Management includes systemic antibiotics with amoxicillin,
lance is recommended after treatment to evaluate both for fluoroquinolones, or ceftriaxone. Current treatment of gon-
recurrent disease and for squamous cell carcinoma (50,53). orrhea also includes treatment of a presumed Chlamydia
infection with doxycycline and azithromycin.
Another common sexually transmitted disease is syphi-
lis, caused by the spirochete, Treponema pallidum. Anorectal
COMMON ANORECTAL SEXUALLY disease presents much like other sites of inoculation: a chan-
TRANSMITTED DISEASES cre represents the first stage of the disease. These ulcerative
lesions may be associated with pain and inguinal adenopa-
Anorectal sexually transmitted diseases (STDs) typically thy. Rectal symptoms may include discharge or bleeding.
present with perianal skin changes, pruritus, pain, and If untreated, the first stage of syphilis resolves within 2–4
bleeding. However, many also can be asymptomatic. A weeks with subsequent progression to secondary syphilis.
study of men having sex with men found that 85% of rectal A macular rash on the torso and extremities denotes sec-
infections were asymptomatic, supporting the need for rou- ondary syphilis. Condyloma may be present during this
tine screening (55). Coinfections also can be present at time time as well as mucosal ulcerations. Without treatment, this
of STD diagnosis and must be taken into consideration and condition will spontaneously resolve within a few weeks.
tested (56). One needs to have a high index of suspicion and Tertiary syphilis with its neurologic and vascular sequelae
ensure testing in concerning patients or in those who are will eventually develop if left untreated. Serologic testing
not improving with conservative management. with Venereal Disease Research Laboratory (VDRL) and
Herpes simplex virus (HSV) is transmitted via direct rapid plasma reagin (RPR) will provide the diagnosis. The
skin contact and results in small, painful vesicles about treatment of choice remains penicillin G and doxycycline.
the perianal skin. Lesions typically last for 2 weeks and
remain contagious even in the asymptomatic stage. Vesicles REFERENCES
can become secondarily infected and are noted to have
erythematous edges. Proctitis can occur and is diagnosed 1. Nelson RL et al. Dis Colon Rectum. 1995;38(4):341–4.
with endoscopic evaluation demonstrating an inflamed 2. Hanno R, Murphy P. Dermatol Clin. 1987;5(4):811–6.
and friable mucosa. Swabs taken from the ulcerations are 3. Mazier WP. Surg Clin North Am. 1994;74(6):1277–92.
sent for viral culture and polymerase chain reaction (PCR). 4. Zuccati G et al. Dermatol Ther. 2005;18(4):355–62.
Treatment involves medical management (valacyclovir) and 5. Metcalf A. Postgrad Med. 1995;98(5):81–4, 87–9,
local debridement for superimposed infections. Additional 92–4.
management includes local analgesic creams for symptom- 6. Silvestri DL, Barmettler S. Dermatitis. 2011;22(1):
atic relief and good hygiene to prevent secondary infections 50–5.
of the affected area. Patients must be counseled that viral 7. Abu-Asi MJ et al. Contact Dermatitis. 2016;74(5):
shedding and transference can occur at any stage in the dis- 298–300.
ease progression, even when the patient is asymptomatic. 8. Siddiqi S et al. Ann R Coll Surg Engl. 2008;90(6):
Chlamydia trachomatis infections can lead to proctitis, 457–63.
with symptoms of rectal urgency, bleeding, and pain. If the 9. Kahlke V et al. Colorectal Dis. 2013;15(5):602–7.
infection progresses proximally, bloody diarrhea can occur. 10. Kranke B et al. Wien Klin Wochenschr. 2006;118(3–4):
Endoscopic evaluation demonstrates diffuse inflamma- 90–4.
tion and ulcerations. PCR and cultures reveal the diagno- 11. Markell KW, Billingham RP. Surg Clin North Am.
sis. Treatment includes antibiotics such as doxycycline and 2010;90(1):125–35, Table of Contents.
azithromycin. 12. Dodi G et al. Br J Surg. 1985;72(12):967–9.
Neisseria gonorrhea is a gram-negative diplococcus that 13. Marchesa P et al. Dis Colon Rectum. 1997;40(11):
infects the mucous membranes via direct contact. This 1286–93.
infection can lead to proctitis, urethritis, cervicitis, phar- 14. Halverson AL. Semin Colon Rectal Surg. 2003;14(4):
yngitis, and conjunctivitis. In men, transmission occurs via 222–5.
anal receptive intercourse. Women may become infected 15. Bowyer A, McColl I. Proc R Soc Med. 1970;63(Suppl):
by similar means or from autoinoculation secondary to 96–8.
a vaginal infection. After an incubation period ranging 16. Murie JA et al. Br J Surg. 1981;68(4):247–9.
from 3 days to 2 weeks, proctitis or cryptitis may occur. 17. Farouk R et al. Br J Surg. 1994;81(4):603–6.
18. Daniel GL et al. Dis Colon Rectum. 1994;37(7):670–4. 38. Swedish KA, Goldstone SE. PLOS ONE. 2014;9(4):
19. Friend WG. Dis Colon Rectum. 1977;20(1):40–2. e93393.
20. Samalavicius NE et al. Tech Coloproctol. 2012;16(4): 39. Kim JJ. Lancet Infect Dis. 2010;10(12):845–52.
295–9. 40. Isik O et al. Int J Colorectal Dis. 2016;31(1):29–34.
21. Suys E. J Am Acad Dermatol. 2012;66(2):327–8. 41. Perez DR et al. Dis Colon Rectum. 2014;57(6):
22. Sauder DN et al. Sex Transm Dis. 2003;30(2):124–8. 747–51.
23. Nadal SR et al. Int J Colorectal Dis. 2010;25(5):663–4. 42. Padrnos L et al. Rare Tumors. 2016;8(4):6804.
24. Silvera RJ et al. Dis Colon Rectum. 2014;57(6):752–61. 43. Meguerditchian AN et al. Dis Colon Rectum. 2011;
25. Giuliano AR et al. Lancet. 2011;377(9769):932–40. 54(5):638–44.
26. Wexner SD. Dis Colon Rectum. 1990;33(12):1048–62. 44. Homsi J, Garrett C. Dis Colon Rectum. 2007;50(7):
27. Gollock JM et al. Br J Vener Dis. 1982;58(6):400–1. 1004–10.
28. Lacey CJ et al. J Eur Acad Dermatol Venereol. 2013; 45. Zhang S et al. J Cancer Res Clin Oncol. 2010;136(9):
27(3):e263–70. 1401–5.
29. Carrozza PM et al. Dermatology. 2002;205(3):255–9. 46. Hillenbrand A et al. Colorectal Dis. 2008;10(6):612–5.
30. Ilmarinen T et al. Eur Arch Otorhinolaryngol. 2012; 47. Matsuda A et al. Ann Surg. 2015;261(4):670–7.
269(11):2367–71. 48. Ballo MT et al. J Clin Oncol. 2002;20(23):4555–8.
31. Grillo-Ardila CF et al. Cochrane Database 49. Kiran RP et al. Dis Colon Rectum. 2010;53(4):402–8.
Syst Rev. 2013; Issue 2. Art. No.: CD010389. 50. Barrett WL et al. Dis Colon Rectum. 1998;41(5):606–
DOI:10.1002/14651858.CD010389 11; discussion 611–2.
32. Chu QD et al. Dis Colon Rectum. 1994;37(9):950–7. 51. Cohen SM et al. Int J Colorectal Dis. 1994;9(4):
33. Long CA et al. Am Surg. 2013;79(6):2E228–30. 169–73.
34. Palefsky JM et al. J Acquir Immune Defic Syndr Hum 52. Viamonte M et al. Dis Colon Rectum. 1993;36(9):
Retrovirol. 1998;17(4):320–6. 801–5.
35. Goldstone SE et al. Dis Colon Rectum. 2014;57(3): 53. Abramowitz L et al. Dis Colon Rectum. 2009;52(6):
316–23. 1130–6.
36. Workowski KA et al. MMWR Recomm Rep. 2010; 54. SDW. Perspect Colon Rectal Surg. 1989;2:19–54.
59(RR-12):1–110. 55. Kent CK et al. Clin Infect Dis. 2005;41(1):67–74.
37. Palefsky JM et al. N Engl J Med. 2011;365(17): 56. Assi R et al. World J Gastroenterol. 2014;20(41):
1576–85. 15262–8.
17
Outcomes and quality
SHAUN R. BROWN AND KURT G. DAVIS
44,000–98,000 deaths occur annually in the United States

CHALLENGING CASE due to medical errors. A second report published in 2001,
titled “Crossing the Quality Chasm: A New Health System
A 56-year-old man living in a rural community pres- for the 21st Century,” attempted to provide guidance on
ents to his primary care physician for what he believes how to achieve the goals of improving the quality of health
are symptomatic hemorrhoids. On physical exam he care delivered (2). As a consequence of these reports, aware-
is found to have a low rectal mass. Further workup, ness and attention increased, resulting in an explosion of
to include colonoscopy with biopsies, confirms the research aimed at improving quality and outcome mea-
diagnosis of rectal cancer. The patient’s primary care sures. In this chapter, we explore the concept of quality, and
physician refers the patient to the local oncologist the various methods of quality improvement with special
who orders a staging workup. The patient has no emphasis on quality improvement related to the specialty of
evidence of metastatic disease on preoperative com- colon and rectal surgery.
puted tomography (CT) imaging, and pelvic magnetic
resonance imaging (MRI) demonstrates a T2N1MO
lesion, 5 cm from the anal verge. The patient’s com-
munity has a regional hospital with a three-person HISTORY OF QUALITY ASSESSMENT
general surgery group. After consultation with the AND IMPROVEMENT
local surgeon, the patient is offered an abdominoper-
ineal resection to treat his rectal cancer. The patient
is unsure of this recommendation and seeks a second —where all the women are strong, all the men
opinion. are good-looking, and all the children are above
average …
Garrison Keillor
CASE MANAGEMENT
Like the children in the fictitious hometown of Garrison
The man is seen by a colorectal surgeon who reviews Keillor, surgeons certainly all consider themselves above
the patient’s workup and presents the patient to a average. A certain amount of hubris is required to per-
Multidisciplinary Tumor Board. The consensus was for form complex operations on patients with multiple, often
the patient to receive long-course chemoradiother- complex medical comorbidities and with significant conse-
apy followed by a sphincter saving resection. quences for complications that can be labeled as a mistake
whether or not an error actually occurred. It is inherent in
what we, as surgeons, do daily both in and out of the operat-
ing room to have the training, experience, and confidence
INTRODUCTION to perform technically challenging operations and make
difficult decisions for our patients. Despite the difficulty
The landmark paper published by the Institute of Medicine in defining, measuring, and codifying medical and surgi-
in 2000 titled, “To Err Is Human, Building a Safer Health cal quality, surgeons take great pride and are committed to
System,” shined a spotlight on the epidemic of medical mis- improving the care that is delivered to their patients.
takes, bringing quality and safety concerns to the forefront The search for quality has likely always been an aim for
for many American hospitals (1). The report estimated that physicians. Hippocrates is attributed with saying, “Make a
155
156 Outcomes and quality
habit of two things: to help; or at least to do no harm (3).” Donabedian in 1966 (6). In this model, information regard-
This is nothing else if not a call for physicians to pursue qual- ing quality of medical care can be taken from three separate
ity. That is what surgeons are attempting to do—to help the categories: structure, process, and outcomes. The structure
patient. Yet all surgeons know that harm does sometimes is devoted to the context in which medical care is delivered,
befall our patients. How do we look at our own results—or the process is the actual medical care that is delivered, and
those of our colleagues or peers and determine if quality the outcomes are the effects that are actually seen by the
surgery is being practiced? patients.
A surgeon, Ernest Amory Codman was one of the pio-
neers of medical quality. Although also an innovative tech- STRUCTURE
nical surgeon whose eponymous exercises and tumor are
still in use, he is most widely hailed today as the forefather The structure as it relates to the Donabedian model is the
of outcomes and quality measurement. While it seems self- setting where the medical care is delivered. This includes
evident, he was the first physician to look at results from the the physical building, the equipment, as well as the human
patient’s perspective, using the end result. resources including the qualifications and the training of
Codman used end-result cards to keep track of informa- these individuals.
tion on every patient that he operated on and sought to track The structure is relatively easy to observe and measure.
the long-term outcomes from his surgeries. He sought to Examples of structure as it relates to colon and rectal sur-
determine if the treatment that was administered had been gery could include how many surgeons at an institution
successful, and if it had not, then the reason for the failure have additional training in colon and rectal surgery or have
was sought. He felt that this should be practiced by all hos- attained certification by the American Board of Colon and
pitals and surgeons, and he aimed to make this informa- Rectal Surgeons (ABCRS). In addition, whether a facility
tion available to the public, so that it could be used to assist employs a coordinator for the cancer program or a multi-
them in their decisions on where to receive medical care. disciplinary approach is used for all patients with defeca-
He felt that by tracking outcomes we, as physicians, could tory disorders, the majority of the early focus on quality in
learn from our patients and advance the field of medicine. colorectal surgery revolved around structural measures. For
His ideas were not well received in his time, yet he leaves colorectal cancer surgery, hospital and surgeon volume are
behind a broad legacy (4). associated with a reduction in operative morbidity and an
One facet that he left is the practice of regular morbid- improvement in 5-year survival (7,8). Additionally, there is
ity and mortality conferences. He sought to determine if a similar relationship between complications and volume in
there was an unsuccessful outcome, “was it the fault of the complex inflammatory bowel disease surgery (9).
surgeon, the disease, or the patient,” and could we prevent While the structure is relatively easy to measure, it is
similar failures in the future (5)? He was indeed prescient more difficult to extrapolate these measurements or attri-
in that before the explosion of information and the devel- bute any outcome measures based on the structure. Do
opment of the Internet, he knew that in order for hospitals improved outcomes result from a program coordinator,
to implement these tasks, that they would be “difficult, more training, a multidisciplinary approach, or the pres-
time-consuming, and troublesome, and would lead … to ence of a water feature in the lobby of the hospital? These
much onerous committee work” by members of the staff (5). are all elements of the structure of health care, but extrapo-
Another legacy of Ernest Codman remains within the Joint lating these to quality measurements can be arbitrary.
Commission. He developed and then chaired the Committee
on the Standardization of Hospitals that became the Joint PROCESS
Commission on Accreditation of Hospital Organizations,
now simply the Joint Commission, which continues to be a The process in the Donabedian model is the combination of
continuous aspect of modern medical care. all of the actions that comprise the medical care that is deliv-
It is obvious that the ideas of Codman were revolution- ered to the patient. It is the collection of everything that is
ary, but it remains to be seen if adopting his ideas, which is done in the administration of health care. This includes any
certainly done today, has actually led to improved quality preventative services, diagnostic tests, or treatments that
or better outcomes. In addition, while Codman focused on are rendered. Process measures focus on the details of the
outcomes, is outcomes research the only marker of quality, health care delivered—prophylactic antibiotics, preopera-
or is it the only marker that should be studied? tive bowel preparation, and deep vein thrombosis prophy-
laxis—and are relatively easy to monitor for compliance.
Unfortunately, excellent compliance with surgical process
measures has largely failed to demonstrate an improve-
THE DONABEDIAN MODEL OF ment in outcomes. An example of this is the Surgical Care
QUALITY OF CARE Improvement Project (SCIP). The SCIP initiative was for-
malized in 2006 and met with widespread endorsement,
One of the more common models of quality medical and it resulted in mandatory implementation. While there
care in use today was developed by the physician Avedis was significant improvement in the surgical care process
Quality and outcome measures specific to colorectal surgery 157
(compliance with preoperative antibiotics), this failed to and the nature of complications. The results demonstrated
translate into an improvement in outcomes (10,11). that overall compliance was high, and surgeons are willing
Process measures are generally evaluated by examin- to report their outcomes, and generally work together on a
ing the medical records or by directly interviewing health quality improvement project under peer review protection
professionals or the patients. Some examples of process as (14). The New England Colorectal Cancer Quality Project
it relates to colon and rectal surgery are the percentage of confirmed these findings, on a broader geographical scale,
patients who undergo screening colonoscopy, the percent- demonstrating a willingness of surgeons to participate and
age of colectomies that are performed laparoscopically, or report their data to a multi-institutional quality database
the understanding of postoperative instructions in patients (15). The authors concluded that surgeon data entry can
undergoing surgical hemorrhoidectomy. Again, while it is supplement the data entered by physician surrogates and
possible to examine numerous aspects of the process of med- can extend the achievements in quality seen in NSQIP (15).
ical care, it remains arbitrary to attribute these to quality. It should be emphasized that the Donabedian method
was developed to assess the quality that is in clinical prac-
OUTCOMES tice, but it suffers from the lack of a definition of quality,
and all three of the domains have both advantages and dis-
The outcomes, as in the time of Codman, are the effects advantages that require researchers to draw conclusions
of the process of health-care delivery on the patients and regarding causality. Donabedian points out that while at
patient populations. Encompassed within this are not only one time, the idea of measuring quality was ridiculed, now
direct outcomes, but also patient satisfaction. So outcome perhaps the pendulum has swung too much in the oppo-
measures of colon and rectal surgery will include not only site direction. Those who wish to measure medical quality,
anastomotic leak rate or permanent ostomy rates but also especially those who are not clinicians, “demand measures
patient quality of life measurements. that are easy, precise, and complete—as if a sack of potatoes
Outcome measures garner a significant amount of atten- was being weighed” (16). This is clearly not realistic, and it
tion from researchers and hospital administration in regard is imperative that surgeons remain at the forefront of how
to quality. The majority of data collected in regard to out- quality is defined and measured.
come measures is collected from administrative data and
clinical registries (12). Administrative data are collected ret-
rospectively and derived from hospital billing information,
whereas clinical registries collect clinical variables with
QUALITY AND OUTCOME MEASURES
high fidelity utilizing dedicated personnel to extract and
SPECIFIC TO COLORECTAL SURGERY
audit the data from patient charts. An excellent example
of a surgical registry is the American College of Surgeons The practice of colorectal surgery, while considered a sur-
National Surgical Quality Improvement Program (NSQIP). gical subspecialty, encompasses a broad range of pathol-
The ACS NSQIP database includes numerous data points, ogy, from infectious processes to cancer. Several quality
including demographic information, intraoperative data, improvement measures specific to colorectal surgery have
and postoperative complications, to name a few. In 2015 previously been addressed in this chapter. However, two
alone, the NSQIP Participant Use Data File (PUF) included areas of interest in regard to both quality and outcomes that
885,502 cases from 603 hospitals (13). Colorectal surgery is have garnered significant attention are the management of
one of the key areas of focus of the NSQIP database, with rectal cancer and the performance of colonoscopy.
a separate targeted database for colorectal surgery. Again Rectal cancer surgery is more technically difficult com-
in 2015, the targeted colectomy PUF included 31,307 cases, pared to colon cancer surgery, due to the constrained
from 239 hospitals (13). This high-volume data collected working space of the pelvis and the resulting challenge of
from a diverse collection of hospitals across the United achieving adequate resection margins with close proximity
States provides significant power for research related to out- to vital structures. In the United States, the majority of rec-
come measures. One of the pitfalls identified with clinical tal cancer surgeries are performed by general surgeons who
registries is that physician-surrogates who rely on the medi- may or may not have had additional training or possess an
cal records typically enter the data. This may limit the qual- interest in colorectal surgery. There is significant room for
ity of the data entered, omitting critical information known improvement in the outcomes of rectal cancer surgery, as
only to the surgeon. In an attempt to overcome these short- evident by the variations in quality of resection, such as pos-
comings, the Vermont Colorectal Cancer Project was estab- itive distal or circumferential radial margins, local recur-
lished. Published in 2002, the purpose of this study was to rence, use of neoadjuvant/adjuvant chemoradiotherapy,
assess the feasibility of performing a quality study of the and permanent stoma creation (7,17,18). The fundamentals
surgical management of colorectal cancer in Vermont using of rectal cancer management have evolved over the past
a surgeon-initiated, prospective database (14). The motiva- 30 years with the introduction of total mesorectal excision
tion for this study was, in the authors’ opinion, that the sur- (TME), chemoradiotherapy (CRT), and multidisciplinary
geon is in the best position to provide reliable and accurate team (MDT) approach to care (19). Throughout this evo-
data regarding tumor characteristics, surgical technique, lution, five main principles have been identified, and when
Table 17.1 Five principles of rectal cancer management minimize complications, and finally appropriate postpro-
cedure recommendations (24). The appropriate goal for an
Principles of rectal cancer management
individual metric such as withdrawal time remains contro-
1 Total mesorectal excision and associated surgical versial. While it is easy to measure, the relationship of with-
principles drawal time to other metrics such as ADR or the incidence
2 Pathologic assessment of surgical quality and of interval cancers is problematic. Will the identification of
accurate staging 2–3 mm adenomas in a patient who will undergo regular
3 Preoperative imaging to identify those patients surveillance lead to improved care or just add cost to the
at risk for local recurrence procedure? As in other areas of medicine, improving some
4 Implementation of updated neoadjuvant/ of these process measures has yet to translate into improved
adjuvant chemotherapy and radiotherapy outcomes.
5 Multidisciplinary team (MDT) approach to every A quality indicator is often reported as a ratio between
individual patient the incidence of correct performance and the opportunity
for correct performance (25). As previously stated, qual-
ity indicators can be divided into three categories: struc-
combined, have led to a significant reduction in the rates tural measures, process measures, and outcome measures.
of local recurrence, increased disease-free and overall sur- Specific to colonoscopy, structural measures assess the
vival, and a reduction in permanent stoma rates (19). These health-care structure, process measures assess the perfor-
five principles are depicted in Table 17.1. mance of the procedure (ADR, cecal intubation rate, and
Implementation of these principles requires a hospital withdrawal time), and outcome measures assess the results
system, with specialized surgeons, pathologists, oncologists, of care (prevention of cancer through screening and reduc-
radiation oncologists, radiologists, and associated support tion in colonoscopic complications) (26). The American
staff available to deliver higher-quality care. These five prin- Society of Gastroenterology has defined specific quality
ciples related to quality are the basis for the formation of indicators and performance targets based on preprocedure,
Rectal Cancer Centers of Excellence (ReCCoEs). Several intraprocedure, and postprocedure metrics (Table 17.2)
European countries have established ReCCoEs, which has (26). For the first time, in 2015, the ASGE task force rec-
resulted in a significant improvement in short- and long- ommended three priority indicators that every colonoscopy
term outcomes in rectal cancer (20,21). practice should track. The priority indicators are as follows:
The Rectal Cancer Coordinating Committee (RCCC)
of the American Society of Colorectal Surgery (ASCRS) 1. Frequency with which adenomas are detected in asymp-
was established to coalesce knowledge and activity regard- tomatic average-risk individuals
ing ongoing North American rectal cancer initiatives (22). 2. Frequency with which colonoscopies follow recom-
A primary goal of the RCCC is to develop a surgical skills mended postpolypectomy and postcancer resection
education module for TME. This committee, along with the surveillance intervals and 10-year intervals between
OSTRiCH Consortium (Optimizing the Surgical Treatment screening colonoscopies in average-risk patients who
of Rectal Cancer) and the American College of Surgeons have negative examination results and adequate bowel
Commission on Cancer, is working to define and establish cleansing
Centers of Excellence for rectal cancer care. 3. Frequency with which visualization of the cecum by
The OSTRiCH Consortium is a group of health-care notation of landmarks and photo documentation of
institutions that have come together with the ultimate goal landmarks is documented in every procedure
of providing access to high-quality rectal cancer care for all
Americans (23). The Consortium plans to achieve this goal The advantage of these priority indicators is that they are
through the establishment of new Rectal Cancer Centers all actionable metrics that readily measure quality and can
throughout the United States with a highly trained multi- be acted upon accordingly (26).
disciplinary team focused on providing care based on the
five core principles of evidence-based rectal cancer care.
PUTTING IT TOGETHER
COLONOSCOPY The focus and resources allocated to researching and imple-
menting quality and outcome measures have expanded
There are approximately 3.3 million screening colonos- exponentially since the start of the twenty-first century.
copies performed in the United States annually. Optimal The terms “quality,” “value,” and “outcomes” have become
effectiveness and quality of colonoscopy depend on sev- part of the common vernacular permeating throughout our
eral factors, including preprocedure bowel preparation, health-care systems, yet they remain poorly defined. This
meticulous inspection and appropriate withdrawal times, evolution has contributed to the expansion of sophisticated
adenoma detection rate (ADR), technical expertise to data systems and the creation of task forces dedicated to
Putting it together 159
Table 17.2 Summary of proposed quality indicators for colonoscopy
Grade of Measure Performance

Quality indicator recommendation type target (%)
Preprocedure
1. Frequency with which colonoscopy is performed for an 1C+ Process >80
indication that is included in a published standard list of
appropriate indications, and the indication is documented
2. Frequency with which informed consent is obtained, 1C Process >98
including specific discussions of risks associated with
colonoscopy, and fully documented
3. Frequency with which colonoscopies follow recommended 1A Process ≥90
postpolypectomy and postcancer resection surveillance
intervals and 10-year intervals between screening
colonoscopies in average-risk patients who have negative
examination results and adequate bowel cleansing
(priority indicator)
4. Frequency with which ulcerative colitis and Crohn colitis 2C Process ≥90
surveillance is recommended within proper intervals
Intraprocedure
5. Frequency with which the procedure note documents the 3 Process >98
quality of preparation
6. Frequency with which bowel preparation is adequate to 3 Process ≥85 of
allow the use of recommended surveillance or screening outpatient
intervals examinations
7. Frequency with which visualization of the cecum by notation 1C Process –
of landmarks and photodocumentation of landmarks is
documented in every procedure (priority indicator)
Cecal intubation rate with photography (all examinations) – – ≥90
Cecal intubation rate with photography (screening) – – ≥95
8. Frequency with which adenomas are detected in 1C Outcome –
asymptomatic average-risk individuals (screening)
(priority indicator)
Adenoma detection rate for male/female population – – ≥25
Adenoma detection rate for male patients – – ≥30
Adenoma detection rate for female patients – – ≥20
9a. Frequency with which withdrawal time is measured 2C Process >98
9b. Average withdrawal time in negative-result screening 2C Process ≥6 minutes
colonoscopies
10. Frequency with which biopsy specimens are obtained when 2C Process >98
colonoscopy is performed for an indication of chronic
diarrhea
11. Frequency of recommended tissue sampling when 1C Process >98
colonoscopy is performed for surveillance in ulcerative
colitis and Crohn colitis
12. Frequency with which endoscopic removal of pedunculated 3 Outcome >98
polyps and sessile polyps <2 cm is attempted before
surgical referral
Postprocedure
13. Incidence of perforation by procedure type (all indications 1C Outcome –
versus colorectal cancer screening/polyp surveillance) and
postpolypectomy bleeding
(Continued)
Table 17.2 (Continued) Summary of proposed quality indicators for colonoscopy
Grade of Measure Performance

Quality indicator recommendation type target (%)
Incidence of perforation—all examinations – – <1:500
Incidence of perforation—screening – – <1:1,000
Incidence of postpolypectomy bleeding – – <1
14. Frequency with which postpolypectomy bleeding is 1C Outcome ≥90
managed without surgery
15. Frequency with which appropriate recommendation for 1A Process ≥90
timing of repeat colonoscopy is documented and provided
to the patient after histologic findings are reviewed
improving both quality and outcomes. There is no ques- 12. Lawson EH et al. Ann Surg 2012;256:973–81.
tion that colorectal surgeons are committed to improving 13. American College of Surgeons National Surgical
the health care delivered to our patients, and that we must Quality Improvement Project: ACS NSQIP. (2017).
maintain a voice in the constant search for quality. https://www.facs.org/quality-programs/acs-nsqip.
Accessed July 23, 2018.
REFERENCES 14. Hyman N, Labow SB. Arch Surg 2002;137:413–6.
15. Hyman NH et al. J Am Coll Surg 2006;202:36–44.
1. Medicine IO. To Err Is Human: Building a Safer 16. Donabedian A. JAMA 1988;260:1743–8.
Health System. Washington, DC: National Academy 17. Wexner SD, Rotholtz NA. Dis Colon Rectum 2000;
Press, 2000. https://doi.org/10.17226/9728 43:1606–27.
2. Richardson WC et al. nationalacademies.org, 2001. 18. Archampong D et al. Cochrane Database Syst Rev
3. Grammaticos PC, Diamantis A. Hell J Nucl Med 2008; 2012;(3):CD005391.
11:2–4. 19. Abbas MA et al. Dis Colon Rectum 2014.
4. Brand RA. Biographical sketch: Ernest Amory 20. Dahlberg M et al. Br J Surg 1999.
Codman, MD (1869–1940). Clin Orthop Relat Res 21. Khani MH, Smedh K. Colorectal Dis 2010;12:874–879.
2013;471(6):1775–7. 22. Rectal Cancer Coordinating Committee. May 2017.
5. Codman EA. Boston, MA: Thomas Todd Co.; 1918. https://www.fascrs.org/rectal-cancer-coordinating-
6. Donabedian A. The Milbank memorial fund quarterly, committee
1966. 23. OSTRiCh consortium: Optimizing the Surgical
7. Gietelink L et al. Ann Surg 2016;263:745–50. Treatment of Rectal Cancer. May 2017. http://www.
8. Liu CJ et al. Cancer 2015;121:2782–90. ostrichconsortium.org/about.htm#.WvG75dMvwWo
9. Kennedy ED et al. Dis Colon Rectum 2006;49:958–65. 24. Rex D et al. Am J Gastroenterol 2015.
10. Stulberg JJ et al. JAMA 2010;303:2479–85. 25. Harewood GC et al. Gastrointest Endosc 2003;58:76–9.
11. Munday GS et al. Am J Surg 2014;l208:835–40. 26. Rex D et al. Am J Gastroenterol 2015;110:72–90.
18
Hemorrhoidal surgery
JEFFERY MINO AND MASSARAT ZUTSHI
the often feared red hot poker (3,4). Although few people
CHALLENGING CASE have died of hemorrhoidal disease, some patients wish they
had, particularly after therapy, and this fact led to the beati-
A 38-year-old male presents to the emergency fication of St. Fiachre, the patron saint of gardeners and
department with complaints of severe and sudden hemorrhoidal sufferers (5). This chapter guides the practi-
onset of perianal pain. He has a history of internal and tioner to a more humane approach to hemorrhoidal disease
external hemorrhoids, which have progressed over with an emphasis on short- and long-term outcomes.
time, and had a hemorrhoidal banding 2 years ago.
On physical exam, you notice a black mass protruding
from the anus which is acutely painful to manipulation ANATOMY AND PHYSIOLOGY
and another smaller, firm nodule in the perianal area
that is exquisitely tender to palpation.
The understanding of the anatomy of hemorrhoidal disease
has not changed substantially since 1975 when Thomson
published his master’s thesis based on anatomic and radio-
CASE MANAGEMENT logic studies and first used the term “vascular cushions” (6).
The hemorrhoidal cushions are located in the upper anal
The patient’s findings are suggestive of a strangulated
canal and exist within the submucosal layer, which clusters
internal hemorrhoid with an associated thrombosed
into columns in the anal canal rather than forming a con-
external component. The patient should be evaluated
tinuous circumferential ring. These cushions are a normal
urgently, admitted, started on broad-spectrum antibi-
component of anorectal anatomy and contribute up to 20%–
otics and IV fluids, as well as IV analgesia. After appro-
30% of the resting tone of the anus, as well as allowing the
priate preoperative bloodwork, the patient should be
anal canal to dilate during defecation without tearing. They
taken expeditiously to the operating room for excision
are composed of sinusoidal blood plexuses, smooth muscle
of the strangulated internal hemorrhoid and throm-
from the internal sphincter, and conjoined longitudinal
bosed external components, lest pelvic sepsis occur.
muscle, connective tissue, and elastic tissue (Figure 18.1).
Because they lack a defined muscular wall, they are predis-
posed to bleeding, and may prolapse as the elastic connec-
tive tissue holding them in suspension wears down from age
and trauma during defecation.
INTRODUCTION While they have classically been described as forming
three pillars in the right anterior, right posterior, and left
Hemorrhoids are a normal part of the anal canal and are lateral positions within the anal ring, in truth there are
symptomatic when they protrude into the anal canal caus- often secondary hemorrhoidal complexes, which can lead
ing bleeding with bowel movements or prolapse. Few to pathologic hemorrhoidal disease at any position. Blood
diseases are more chronicled in human history than symp- supply is derived from the paired vessels of the superior rec-
tomatic hemorrhoidal disease (1,2). Citations of hemor- tal arteries as branches of the inferior mesenteric artery, the
rhoidal disease have been noted in historic texts dating middle rectal arteries as branches of the internal iliac arter-
back to Babylonian, Egyptian, Greek, and Hebrew cultures ies, and the inferior rectal arteries as branches of the inter-
(1,2). A multitude of treatment regimens have been offered nal pudendal arteries (themselves branches of the internal
including anal dilation, various topical liniments, and even iliac arteries). Venous drainage corresponds to the arterial
161
162 Hemorrhoidal surgery
within these cushions, producing bleeding with or without

prolapse of the hemorrhoidal tissue (7). Dietary patterns
including a low-fiber diet, behavioral factors including
Vein excessive straining or prolonged sitting on a commode,
chronic constipation, and sphincter dysfunction can all
Internal
result in symptomatic hemorrhoids. The prevailing expla-
hemorrhoid
nation is that during straining, the hemorrhoids become
engorged due to Valsalva pressure impairing venous return.
Dentate Over time, as this process continues, the connective tissue
line
Muscularis
that supports the muscular submucosa weakens, allowing
mucosa the hemorrhoidal tissue to elongate, ultimately resulting in
Sinusoid prolapse. Once the tissue prolapses far enough into or past
Vascular the internal sphincter, its contracted resting tone further
space prevents blood return leading to a vicious cycle of further
Arteriole engorgement. The stretch of this tissue also results in thin-
ning of the vascular walls, leading to transudative bleeding
and/or rupture of the vessels, resulting in the classic symp-
toms associated with the disease. Haas et al. noted that the
External
hemorrhoid connective tissue can be shown to deteriorate by the third
decade of life on microscopic examination (8). Additionally,
Figure 18.1 Hemorrhoidal anatomy. neovascularization appears to play a role in enlarging hem-
orrhoids. Several studies have found evidence that terminal
branches of the superior hemorrhoidal artery were larger
supply, with the superior rectal veins draining the region
with greater flow in patients with symptomatic hemorrhoids
above the dentate line into the portal system, while the
(9,10), greater microvascular density, higher markers of pro-
inferior veins drain into the systemic system below the den-
liferation, and increased angiogenesis-related proteins such
tate line. Consequently, hemorrhoids are divided into two
as vascular endothelial growth factor (11).
groups, internal and external, separated by their relation to
the dentate line. External hemorrhoids are located distally
and covered by the anoderm, a modified squamous epi-
thelium. Internal hemorrhoids are located proximally and CLINICAL PRESENTATION
covered by columnar or transitional epithelium. This dis-
tinction has clinical importance, as the relative lack of sen-
When patients present with any of a varied number of ano-
sation in the internal hemorrhoids allows for significantly
rectal pathologies, their presenting complaint is often, “I have
more invasive and thus definitive management options
hemorrhoids.” The clinician should be wary of taking this
compared to external hemorrhoids.
statement at face value. A study by Ganchrow et al. reported
Internal hemorrhoids are commonly classified into one
on a series of 500 patients with anorectal complaints they
of four grades based on the level of prolapse. Grade I inter-
associated with hemorrhoids, while only 35% of patients
nal hemorrhoids protrude into the lumen, grade II protrude
were found to have any hemorrhoidal disease at all (12). The
spontaneously with bowel movements and regress spon-
most common complaints include bleeding and protrusion
taneously once straining ceases, grade III protrude spon-
for internal hemorrhoids and pain and itching for external.
taneously with bowel movements and must be manually
Pain is an unusual presentation for internal hemorrhoids
reduced, and grade IV are permanently prolapsed and irre-
due to their relative lack of sensory nerves corresponding
ducible. Mixed hemorrhoids include a component of both
to their location above the dentate line. Exceptions to this
internal and external cushions. While the grade of internal
are thrombosed or strangulated internal hemorrhoids.
hemorrhoids correlates well with appropriate medical and
Barring these, the clinician should thoroughly investigate
operative management, ultimately the therapeutic decisions
other causes of the problem. Similarly, hemorrhoidal bleed-
should be based on symptoms and overall patient health
ing is typically fresh and thus bright red in nature. Dark
rather than grade alone. The presence or absence of an
blood or melena should raise suspicion of another etiol-
external component must be factored in as well (7).
ogy. The bleeding from hemorrhoids is most often seen on
tissue paper after wiping or streaked in the toilet bowel.
Occasionally larger volumes of blood can occur as hemor-
ETIOLOGY rhoids enlarge, particularly in advanced stages when a por-
tion of the complex is fixed externally, allowing the blood
There are a number of potential clinical causes of symptom- to drip or spurt into the commode. Frequency will often
atic hemorrhoidal disease. Ultimately, they converge in the increase over time as the hemorrhoids enlarge through
physiologic condition of disruption of the connective tissue their stages. Nonetheless, bleeding to a sufficient extent to
Excisional hemorrhoidectomy / Open/milligan-morgan 163
cause clinical anemia is extremely rare, estimated at 0.5 per this, a decision must be made regarding further evaluation.
100,000 per year (13). Thus, the index of suspicion should be This should be done based on the age of the patient, family
high for another source of blood loss in patients who present history, characteristics of bleeding, etc. In general, anyone
with anemia. over 50 years should undergo full colonoscopy even with
Prolapse of the internal hemorrhoid is also quite com- “classic” findings, while a young patient with no family his-
mon, hence the creation of the grading system. While the tory likely needs no further testing. In a large series of clas-
vast majority may reduce either spontaneously or with min- sic “outlet” bleeding, colonoscopy revealed adenomas in less
imal manipulation, grade IV internal hemorrhoids are pro- than 2% and no cancers in patients less than 50 years. When
lapsed all the time and cannot be manually reduced. These considering all age groups, 6.7% of patients had a significant
are at high risk of strangulation and definitive management lesion that would have otherwise been missed (15).
should be offered expeditiously.
External hemorrhoids have a rich vascularization and
innervation. Itching due to leakage of minute amounts of
rectal content via capillary action and pain likely due to MEDICAL AND NONSURGICAL
acute and chronic inflammation of the tissues as they chafe MANAGEMENT
against one other during movement are the hallmarks of
these entities, and bleeding is common as well from much Excisional hemorrhoidectomy is becoming less common as
the same etiology as internal hemorrhoids, thus leading we learn more about the disease and better nonoperative
to overlapping symptoms between the two. In addition, options have emerged. Dietary and lifestyle modifications,
because these enlarge in size due to engorgement with including increasing dietary fiber, increasing water intake,
blood, it is often difficult to tell on symptoms alone without stool softeners, and limiting straining and the duration of
clinical examination between an external hemorrhoid and time on the commode are all first-line interventions for
a prolapsed internal one. A cause of severe pain is the clas- symptomatic hemorrhoids. Sclerotherapy, infrared coagu-
sic thrombosed external hemorrhoid, which typically leads lation, and rubber band ligation are the most common
to a sudden onset, constant and unrelenting pain without nonsurgical modalities utilized for control of grade II hem-
palliating factors. If brought to attention before the clot has orrhoids. These procedures are covered in Chapter 19.
matured, it may be extracted and the hemorrhoid excised,
typically within 72 hours. However, after this it is often bet-
ter to treat conservatively, and allow the thrombosis to reab-
sorb over the next several weeks. EXCISIONAL HEMORRHOIDECTOMY
Approximately 5%–10% of patients will require surgical

management of their hemorrhoids (16). Patients with low-
CLINICAL EVALUATION grade hemorrhoids that are nonetheless refractory to non-
surgical therapies, high-grade symptomatic hemorrhoids,
Prior to inspection of the anus, a focused physical examina- and those with complications such as thrombosis or stran-
tion should be performed, paying attention to the abdomen, gulation are candidates for surgical treatment. Excisional
groins, and perineal area (14). Patients should be examined hemorrhoidectomy remains the gold standard treatment
first in the supine position and then either in the left lat- for grades II–IV or complicated hemorrhoids; however,
eral or prone jackknife position. Careful examination of techniques are continually developing to find an alterna-
the anus should then commence, with detailed explanation tive approach with less postoperative pain that nonethe-
by the clinician for everything that they are doing. Visual less maintains the low rate of recurrence and complications
inspection should commence with the spreading of the found with these procedures. Options for surgical resection
buttocks to evaluate for external hemorrhoids, and other include the Milligan-Morgan open hemorrhoidectomy, the
conditions such as skin tags, condylomata, anal and peri- Ferguson closed hemorrhoidectomy, stapled procedure for
anal neoplasms, fissures, fistulae, and stigmata of Crohn prolapsed hemorrhoids (PPH), Doppler-guided hemor-
disease. The patient should be asked to Valsalva, to inspect rhoidal arterial ligation with or without mucopexy, and his-
for engorgement of external hemorrhoids and also prolapse torically, the Whitehead hemorrhoidectomy.
of internal ones. A digital rectal exam should then be per-
formed with focus on any internal masses, tenderness at any OPEN/MILLIGAN-MORGAN
specific point along the sphincter, the robustness of sphinc-
ter tone, and in a male, the contour and size of the prostate. The Milligan-Morgan open hemorrhoidectomy is popu-
Anoscopy should follow, paying specific attention to the left lar in Europe and has stood the test of time and rigorous
lateral, right anterior, and right posterior regions of the anal documentation. First described in 1937, it involves resec-
canal for documentation of size, friability, presence of bleed- tion of the internal and external hemorrhoids and liga-
ing or inflammation, ease of prolapse of any hemorrhoidal tion of the arterial blood supply, leaving the excision sites
tissue, and reducibility of the prolapsed tissue. Following open to granulate in by secondary intention. The procedure
begins with digital rectal examination and then insertion hemorrhoid surgery performed in the United States. Nearly
of a Hill-Ferguson retractor through the anus. Lidocaine identical to the open technique, the Ferguson procedure
with epinephrine 1:200,000 is injected around the anus and differs in that the suture used to ligate the vascular pedicle
within the hemorrhoids themselves to decrease bleeding is then used to close the defect longitudinally in a running
during the case. The external and internal hemorrhoidal fashion (Figure 18.3). This reduces postoperative pain and
components are grasped with Allis clamps and retracted discomfort as the underlying tissue is no longer exposed to
caudally. After the patient is adequately sedated, an ellip- the fecal stream. It is important in this procedure to take care
tical or rhomboid incision is created, encompassing both to line up the anal verge precisely during reapproximation
components, beginning with the external, and the dissec- in order to avoid formation of an ectropion. Occasionally,
tion is carried out. The mucosa and submucosa are elevated it is necessary to undermine flaps of anoderm and perianal
and then lifted off of the internal anal sphincter with either skin to allow excision of intermediate hemorrhoidal tissue,
blunt or sharp dissection. The dissection should continue while preserving the bridges of anoderm between pedicles.
proximally until the feeding artery is reached. At this point, This technical adjustment will avoid postoperative stricture.
the base of the hemorrhoid should be clamped and the hem-
orrhoid excised. The vessel is then suture ligated. If multiple WHITEHEAD
hemorrhoids are to be excised, intervening skin bridges
must be left in place in order to minimize the risk of an anal The Whitehead procedure is a technique rarely practiced
stricture. The open wounds typically heal in approximately today due to its potential for significant adverse outcomes.
4–8 weeks, leading to significant postoperative pain and Developed in the 1890s, it begins with a circumferential
discomfort. The procedure is summarized in Figure 18.2. incision around the sphincter complex. The entirety of the
hemorrhoid-bearing mucosa and submucosa is dissected
CLOSED/FERGUSON from the internal sphincter and removed as specimen.
Hemostasis is achieved with suture ligation or cautery, and
Due to the long recovery time of the Milligan-Morgan the rectal mucosa is then mobilized and anastomosed down
technique, the closed hemorrhoidectomy has enjoyed to the distal cut edge of the anoderm at the dentate line
widespread adoption since its introduction in the 1950s by (Figure 18.4). This results in removal of all hemorrhoidal
Dr. Lynn Ferguson, and it is currently the most common tissue. Typically, this procedure was performed only in
Figure 18.2 Open/Milligan-Morgan procedure.

Excisional hemorrhoidectomy / Stapled hemorrhoidectomy or procedure for prolapsed hemorrhoids 165
Figure 18.3 Closed/Ferguson.
Figure 18.4 Whitehead.
patients with multiple, severe grade IV hemorrhoids, where utilized for the treatment of grade III and some grade IV
they cannot be reduced, and it is not possible to distinguish hemorrhoids and involves a 33 mm circular stapler with
and separate the three piles. If the procedure is performed resection of a cuff of tissue above the hemorrhoids resulting
incorrectly, complications are numerous and affect quality in retraction and relocation of the tissue more proximally.
of life significantly. Ectropion can develop if the mucous- The procedure starts with the insertion of an anoscope into
producing rectal mucosa is sutured down to the anal verge, the anal canal with reduction of the prolapsing hemorrhoids.
rather than the dendate line, resulting in constant drain- Next a circumferential purse-string suture through the sub-
age, known as the Whitehead deformity. Anal stenosis is mucosa is applied 4 cm proximal to the dentate line above
another potential complication of the Whitehead radical the apices of the hemorrhoids. The circular stapler is then
hemorrhoidectomy, as the rectal mucosa can be difficult to inserted with the nondetachable anvil fully extended, and
mobilize and is thus sutured with tension, resulting in poor this is placed above the purse-string. This is then tightened
healing, dehiscence, and subsequent stricture formation. around the shaft of the anvil, which is then retracted into
Last, when stricture does not occur, incontinence may be the body of the stapler while holding gentle traction on the
significant, as the hemorrhoidal tissues account for as much purse-string externally. As the stapler is closed, the prolaps-
as 20% of the resting tone of the anus and are especially ing hemorrhoidal tissue is brought within the circle (Figure
important in the continence of flatus. As a result of these 18.5). Once tightened, the vagina should be inspected in
potential complications, the Whitehead procedure is rarely females to ensure that the vaginal wall has not been caught
performed. within the anvil and the stapler. The stapler is then fired,
excising a cuff of rectal mucosa and submucosa. This pulls
STAPLED HEMORRHOIDECTOMY the residual hemorrhoidal tissue up into the rectum while
OR PROCEDURE FOR PROLAPSED often cutting off the vascular pedicle of the hemorrhoids
HEMORRHOIDS leading to atrophy of any remaining tissue. The staple line
should be inspected for hemostasis as several small sites of
The stapled hemorrhoidopexy is a procedure first described bleeding are common. These should be controlled with fig-
in 1998 (17) as an alternative to the standard excisional ure-of-eight absorbable sutures.
hemorrhoidectomy. While not truly a primarily resectional Multiple randomized controlled trials have been con-
procedure, it is also referred to as PPH. This technique is ducted comparing stapled PPH to traditional excisional
Figure 18.5 Stapled hemorrhoidectomy or procedure for prolapsed hemorrhoids (PPH).
hemorrhoidectomy. Overall findings include decreased caught within the staple line. Chronic debilitating rectal
operative times, less pain and analgesic use, earlier return pain is a particularly difficult to resolve complication that
to work, and similar symptom control (18–21). However, as has been reported. Because the recurrence rate is higher
more long-term data are accumulated, it appears clear that than with traditional excisional procedures, and because
the stapled hemorrhoidopexy is a less durable procedure of the severity of these rare complications, the stapled pro-
with increased recurrence after 1 year. A Cochrane analy- cedure for prolapse using a hemorrhoidopexy has begun to
sis of 25 randomized controlled trials found a recurrence fall out of favor.
incidence of 5.7% (for PPH) versus 1% (for excisional hem-
orrhoidectomy) at 1 or more years (21). In addition to this, DOPPLER-GUIDED HEMORRHOIDAL
unique complications are attributable to this procedure, ARTERIAL LIGATION
which while rare, can be devastating. These include rectal
perforation or leak if the purse-string suture is taken full Transanal hemorrhoidal ligation using Doppler assistance
thickness, pelvic sepsis, incontinence if the purse-string is is a relatively new technique first described by Morinaga in
placed too low resulting in resection of the dentate line and/ 1995 (22). This was further refined by Dal Monte with the
or part of the sphincter complex, rectal obstruction, and inclusion of hemorrhoidopexy. It is an alternative to tradi-
anovaginal or anoprostatic fistulae if extraneous tissues are tional hemorrhoidectomy in that it involves no excisional
Comparison of techniques and results 167
Step 1 Step 2 Step 3
Figure 18.6 Doppler-guided hemorrhoidal arterial ligation.
component. It is based on the premise that the prolapsing reduction in postoperative pain and operating time using
tissue is a secondary result of its vascular inflow, and dealing the LigaSure; however, there are inconsistent results regard-
with the latter will result in resolution of the former with- ing short-term benefits of the Harmonic Scalpel for postop-
out need for resection. The preparation for the procedure erative pain. A multicenter, prospective, randomized study
is the same as for excisional hemorrhoidectomy, namely by Altomare et al. showed significantly less pain 12 hours
the use of preoperative enemas to clear the rectal vault. A after defecation, lower analgesic requirements, and faster
specially designed anoscope that incorporates a Doppler return to work and normal activity, with no difference in
transducer is inserted into the anus, and the Doppler is early or late complications. However, this benefit was not
used to identify the hemorrhoidal vessels 2–3 cm above significant after 2 weeks (35). Regardless, both energy
the dentate line. Once identified, the vascular pedicles are devices are safe and effective; however, the added cost, con-
suture ligated with absorbable sutures leading to cessation flicting short-term outcomes, and lack of long-term follow-
of flow into the hemorrhoidal plexuses (Figure 18.6). The up preclude recommendations for their routine use purely
anoscope is moved around circumferentially, and all of the from a cost standpoint. Nonetheless, they may have benefit
vessels leading to prolapsing tissue are similarly identified in reducing short-term pain.
and ligated. Once done, the redundant, prolapsing hemor-
rhoids are then relocated proximally and pexied using a 2–0
Vicryl running suture starting in the distal anal canal and
ending at least 5 mm above the dentate line (23), thereby COMPARISON OF TECHNIQUES
repositing the hemorrhoid back into the anal canal. This is AND RESULTS
an effective procedure for grades II and III hemorrhoids,
and while it may be safely used in grade IV, it appears to As one might imagine with the number of different pro-
have a higher rate of recurrence. Postoperative complica- cedures that are practiced for the treatment of symptom-
tions and pain appear to be lower, as would be expected with atic hemorrhoids, debate remains vigorous as to the best
a nonexcisional procedure. Bleeding requiring intervention approach. Outcomes including postoperative pain, bleeding,
is approximately 11% (24), and urinary retention, hema- time to complete healing, wound infection, and anal steno-
toma, and thrombosis have been described in the literature, sis are the common results for which various procedures
although with low overall incidences. Prolapse recurrence have been compared. Perhaps the most well-studied com-
rates are higher than either excisional hemorrhoidectomy parison is between the open and closed excisional hemor-
or PPH at 10.3% at 1 year (25). rhoidectomy techniques. Multiple studies have found that
the Ferguson closed technique is superior to the Milligan-
Morgan procedure in terms of time to postoperative healing,
wound infection, and bleeding. This is intuitive, as healing
USE OF ALTERNATIVE ENERGY by secondary intention will always take longer than primary
SOURCES closure, and bleeding is to be expected with passage of stool
across an open wound. However, there appears to be no sig-
Due to the extreme amount of pain classically associated nificant reduction in postoperative pain, and operative times
with traditional hemorrhoidectomy procedures, alternative are understandably longer (37). Rehman et al. in fact noted
approaches and devices have been investigated, including an increase in postoperative pain in the closed group com-
the Harmonic Scalpel and LigaSure. These devices confine pared with an open technique, while Arroyo et al. (2004)
energy to minimal collateral damage and limited tissue noted the opposite, with increased pain in the open hemor-
charring, as thermal spread is confined within 2 millime- rhoidectomy group, but the difference was only significant
ters of surrounding tissue (26). There have been multiple during bowel movements (38,39). Regardless, nearly all stud-
randomized controlled trials published in order to assess ies agree that long-term results are similar with either proce-
the efficacy of these (26–36). Most of these demonstrate a dure, with recurrence rates of approximately 1%.
Overall findings for stapled hemorrhoidopexy include To avoid postoperative constipation, patients should be
multiple improved outcomes (39–42). Randomized con- encouraged to consume adequate fiber and enough polyeth-
trolled trials on stapled hemorrhoidopexy versus conven- ylene glycol to keep their stool soft. If an impaction occurs,
tional excisional hemorrhoidectomy showed significantly disimpaction may require a general anesthetic.
lower operative times, less pain at first defecation, lower
mean pain scores, less analgesia consumption, and an ear-
lier return to work (20,40,41).
Complications of stapled hemorrhoidectomy are similar SPECIAL SITUATIONS
to those of closed hemorrhoidectomy, in addition to rare but
potentially catastrophic complications including anasto- THROMBOSED EXTERNAL
motic leak with pelvic sepsis, anovaginal fistula, and Fournier HEMORRHOIDS
gangrene (42,43). While PPH causes less postoperative pain
compared with conventional excisional hemorrhoidectomy, Acute thrombosis of an external hemorrhoid may be an
some patients will experience chronic pain after the proce- exquisitely painful occurrence. While a precipitating event
dure. Unrelenting pain of unknown etiology after stapled may be present, often these will occur at random, for no
hemorrhoidopexy is known as PPH syndrome (44); 15.1% apparent reason that the patient can recall. Known risk
of surgeons surveyed report experience with patients having factors include recent constipation and traumatic vagi-
unrelenting pain lasting for months. This has been postu- nal delivery (52,53). The patient will typically describe the
lated to be due to fibrosis around the staples or direct trauma acute onset of constant, sharp pain in the anal region, often
to the pudendal or sacral nerve spindles. Even removal of accompanied by a tender, palpable nodule in the perianal
the staples does not always result in relief. In addition, recur- region. The pain typically will increase over time through
rence rates are higher with stapled hemorrhoidopexy at 5.7% days 3–4, after which it will gradually subside as the inflam-
versus 1% for excisional hemorrhoidectomy (21) for grades II mation decreases. In some situations, the thrombosis will
and III disease and a surprisingly high rate of recurrence up result in pressure necrosis of the overlying skin, resulting in
to 50% for grade IV hemorrhoids (45,46). spontaneous evacuation of the clot and immediate improve-
Doppler-guided hemorrhoidal artery ligation is effec- ment of symptoms.
tive for grade II hemorrhoids, with a recurrence rate of Conservative treatments include stool softeners, increased
5.3%–6.7% at less than 1 year follow-up and a recurrence fiber, increased fluid intake, Sitz baths, and nonsteroidal
of 12% at greater than 1 year (47). Complications are few, anti-inflammatory drugs (54–56). Topical calcium chan-
with improved postoperative pain and return to work, nel blockers have shown improvement of pain compared
and the procedure may be repeated without difficulty. to traditional topical lidocaine (57). Nonetheless, none of
However, when used for grade III hemorrhoids, recurrence these treatments have been shown to shorten the time to
rates are higher, up to 18%–31% (48,49). In a randomized symptom resolution or reduce the frequency of recurrence
controlled trial comparing ligation versus stapled hemor- compared to surgical excision. This is a simple bedside pro-
rhoidopexy, ligation was found to have a shorter operating cedure within the purview of an office-based or emergency
time, lower mean pain scores, less postoperative discom- practitioner and offers low recurrence and complication
fort, and less postoperative complications (49). Nonetheless, rates with high levels of patient acceptance and satisfac-
this improved pain profile appears to come at the cost of tion (58). It can typically be done under local anesthesia. An
increased recurrence. Ultimately, excisional hemorrhoidec- elliptical incision is made overlying the thrombosis, and the
tomy remains the gold standard in terms of durability of clot is enucleated. (Figure 18.7) Bleeding is usually minimal
repair. but can be addressed with pressure, silver nitrate, or suture
ligation. The skin may then either be left open to close by
secondary intention, or reapproximated with absorb-
able suture (59). A retrospective study by Greenspon et al.
POSTOPERATIVE PAIN AND BOWEL showed that surgical excision resulted in faster symptom
MANAGEMENT resolution (3.9 versus 24 days) and lower recurrence rates
(6.3% versus 25.4%) (60).
Traditionally, hemorrhoid surgery has been very painful,
and the postoperative pain was managed with opioids. The STRANGULATED HEMORRHOIDS
pain and the medication used to treat it often resulted in
constipation or impaction, which added to patients’ mor- Strangulated hemorrhoids are internal hemorrhoids that
bidity. Modern multimodality pain management has sig- have become irreducibly incarcerated with compromised
nificantly improved the patient’s experience (50,51). The blood flow due to the constricting action of the anal sphinc-
editors currently use preoperative IV ibuprofen and acet- ter. While internal hemorrhoids are classically nonpainful,
aminophen and intraoperative liposomal bupivacaine. strangulated ones will often present with an acute episode
Additional discussion is provided in Chapters 3, 9, and 10. of pain. Urinary retention may also be present. Edema is
Special situations / Crohn disease 169
Figure 18.7 Excision of thrombosed external hemorrhoid.
often severe in these situations and is both a result of and technique. Treatment is almost universally nonsurgical.
contributing factor to further strangulation. This is con- Banding should be avoided as it may result in profuse bleed-
sidered a surgical emergency, as these hemorrhoids may ing at home after tissue necrosis. Sclerotherapy or photoco-
progress to necrosis, gangrene, pelvic sepsis, and ultimately agulation are acceptable options for the treatment of bleeding
death. As such, a careful examination must be carried out. varices, and a transjugular intrahepatic portosystemic shunt
If the hemorrhoids are incarcerated but show no signs of (TIPS) procedure, surgical shunt, or liver transplant should
necrosis, a manual reduction may be attempted (with the be considered to address the underlying etiology.
aid of an anal block of lidocaine with epinephrine). Often
this can be augmented by the application of sugar over the PREGNANCY
hemorrhoids to draw out the edema or injection with hyal-
uronidase. Topical nitrates may also be used as an adjunct Postpartum hemorrhoids that are refractory to con-
to relax the sphincter muscle. If successful, the patient may servative measures may require surgical management.
be observed for signs of systemic sepsis. However, if necrosis Hemorrhoidectomy in this setting is safe, has a low preva-
or other concerning signs of strangulation are present, the lence of complications, and in many cases will minimize
patient should be taken expeditiously to the operating room recovery time. Proper patient positioning (usually left lateral
for an excisional hemorrhoidectomy. Sims) and good anesthetic techniques are important. As in
other urgent hemorrhoidectomies, preservation of as much
PORTAL HYPERTENSION anoderm as possible is also critical.
Unlike hemorrhoids, varices result from portal venous CROHN DISEASE

hypertension. These arise from the anastomoses between
the portal and systemic venous circulation via the inferior Unfortunately, patients with Crohn disease are just as if not
and middle rectal veins. Differentiation from hemorrhoids more susceptible to hemorrhoids as the general population,
is essential because excision of varices may result in venous due to their increased frequency of bowel movements and
bleeding that may be difficult or impossible to control. diarrhea. Patient selection is critical in this group if surgical
Anorectal varices are extremely common in patients with excision is considered. A history of perianal Crohn disease
portal hypertension, found in up to 78% of individuals, is a relative contraindication due to the prolonged wound
although the vast majority are asymptomatic (61). A history healing and increased risk of wound complications. Surgery
of anal bleeding in a cirrhotic patient should arouse suspi- should be reserved for patients with no history of perianal
cion. Varices may be present in the rectum, anal canal, or disease or in patients who have well-controlled disease off
anal verge (62). Doppler ultrasonography of the anorectum steroids. Resection in patients with active disease may result
can confirm the diagnosis. Active bleeding from varices in proctectomy for surgical complications not manageable
will usually require oversewing with a continuous suture with conservative means (63).
IMMUNOCOMPROMISED STATE 28. Armstrong DN et al. Dis Colon Rectum. 2001;44(4):

558–64.
Any invasive intervention should generally be avoided 29. Tan JJ, Seow-Choen F. Dis Colon Rectum. 2001;44(5):
if possible in patients with immunocompromised states. 677–9.
Similar to Crohn patients, sepsis and prolonged wound 30. Bessa SS. Dis Colon Rectum. 2008;51(6):940–4.
healing are the major concerns. Thus, conservative manage- 31. Wang JY et al. World J Surg. 2006;30(3):462–6.
ment is the mainstay of treatment. Injection sclerotherapy 32. Chung YC, Wu HJ. Dis Colon Rectum. 2003;46(1):
appears to be a safer alternative to band ligation or surgi- 87–92.
cal excision (64,65); however, if a more invasive approach is 33. Franklin EJ et al. Dis Colon Rectum. 2003;46(10):
required, antibiotic prophylaxis should be given due to the 1380–3.
increased risk of bacteremia. 34. Jayne DG et al. Br J Sur. 2002;89(4):428–32.
35. Altomare DF et al. Dis Colon Rectum. 2008;51(5):
REFERENCES 514–9.
36. Palazzo FF et al. Br J Sur. 2002;89(2):154–7.
1. Holley CJ. South Med J. 1946;39:536–41. 37. Arbman G et al. Dis Colon Rectum. 2000;43(1):31–4.
2. Madoff R. Presented at the Midwest Society of Colon 38. Rahman ASMT RA, Biswas SK. Faridpur Med Coll J.
and Rectal Surgeons’ Meeting Brechenridge, CO. 2012;7:37–41.
1991. 39. Arroyo A et al. Int J Colorectal Dis. 2004;19(4):370–3.
3. Dirckx JH. Am J Dermatopathol. 1985;7(4):341–6. 40. Huang WS et al. Int J Colorectal Dis. 2007;22(8):
4. Maimonides M. Treatise on Hemorrhoids. 955–61.
Philadelphia, PA: JB Lippincott. 1969. 41. Ammaturo C et al. Il Giornale di chirurgia. 2012;33(10):
5. Rachochot J et al. Am J Proctol. 1971;22:175–9. 346–51.
6. Thomson WH. Br J Surg. 1975;62(7):542–52. 42. Molloy RG, Kingsmore D. Lancet. 2000;355(9206):810.
7. Morgado PJ et al. Dis Colon Rectum. 1988;31(6): 43. Wong LY et al. Dis Colon Rectum. 2003;46(1):116–7.
474–80. 44. Yeo D, Tan KY. World J Gastroenterol. 2014;20(45):
8. Haas PA et al. Dis Colon Rectum. 1984;27(7):442–50. 16976–83.
9. Aigner F et al. J Gastrointest Surg. 2006;10(7):1044–50. 45. Ortiz H et al. Br J Surg. 2002;89(11):1376–81.
10. Aigner F et al. Int J Colorectal Dis. 2009;24(1): 46. Ortiz H et al. Dis Colon Rectum. 2005;48(4):809–15.
105–13. 47. Avital S et al. Tech Coloproctol. 2012;16(4):291–4.
11. Han W et al. Zhonghua Wei Chang Wai Ke Za Zhi. 48. Avital S et al. Tech Coloproctol. 2012;16(1):61–5.
2005;8(1):56–9. 49. Avital S et al. Tech Coloproctol. 2011;15(3):267–71.
12. Ganchrow MI et al. Dis Colon Rectum. 1971;14(2): 50. Haas E et al. Am Surg. 2012;78(5):574–81.
128–33. 51. Beck DE et al. Ochsner J. 2015;15(4):408–12.
13. Kluiber RM, Wolff BG. Dis Colon Rectum. 1994;37(10): 52. Oh C. Mount Sinai J Med. 1989;56(1):30–2.
1006–7. 53. Abramowitz L et al. Dis Colon Rectum. 2002;45(5):
14. Steele SR et al. The ASCRS Textbook of Colon and 650–5.
Rectal Surgery 3rd Ed. Springer 2016. 54. Stites T, Lund DP. Semin Pediatr Surg. 2007;16(1):
15. Monson JR et al. Dis Colon Rectum. 2013;56(5): 71–8.
535–50. 55. Alonso-Coello P et al. Am J Gastroenterol. 2006;
16. Bleday R et al. Dis Colon Rectum. 1992;35(5):477–81. 101(1):181–8.
17. Longo A. Proceedings of the 6th World Congress of 56. Mounsey AL, Henry SL. J Fam Pract. 2009;58(9):492–3.
Endoscopic Surgery, Rome, Italy. 1998:777–84. 57. Perrotti P et al. Dis Colon Rectum. 2001;44(3):405–9.
18. Shalaby R, Desoky A. Br J Surg. 2001;88(8):1049–53. 58. Jongen J et al. Dis Colon Rectum. 2003;46(9)
19. Pavlidis T et al. Int J Colorectal Dis. 2002;17(1):50–3. :1226–31.
20. Senagore AJ et al. Dis Colon Rectum. 59. Grosz CR. Dis Colon Rectum. 1990;33(3):249–50.
2004;47(11):1824–36. 60. Greenspon J et al. Dis Colon Rectum. 2004;47(9):
21. Tjandra JJ, Chan MK. Dis Colon Rectum. 2007;50(6): 1493–8.
878–92. 61. Chawla Y, Dilawari JB. Gut. 1991;32(3):309–11.
22. Morinaga K et al. Am J Gastroenterol. 1995;90(4): 62. Corman M. Complications of hemorrhoidal and
610–3. fissure surgery. In Ferrari BT, Ray JE, Gathright JB
23. Infantino A. Colorectal Dis. 2010;12(12):1274. (eds). Complications of Colon and Rectal Surgery.
24. Greenberg R et al. Dis Colon Rectum. 2006;49(4): Philadelphia, PA: WB Saunders, 1985:91–100.
485–9. 63. Jeffery PJ et al. Lancet. 1977;1(8021):1084–5.
25. LaBella GD et al. Tech Coloproctol. 2015;19(3):153–7. 64. Buchmann P, Seefeld U. Int J Colorectal Dis. 1989;
26. Kennedy JS et al. Surg Endosc. 1998;12(6):876–8. 4(1):57–8.
27. Khan S et al. Dis Colon Rectum. 2001;44(6):845–9. 65. Scaglia M et al. Dis Colon Rectum. 2001;44(3):401–4.
19
Nonoperative therapy for hemorrhoidal disease
JOSEPH C. ADONGAY AND SCOTT A. BRILL
CHALLENGING CASE to that of nonanticoagulated patients. An option used

by the editors is to do RBL with the patient anticoagu-
A 72-year-old woman taking Coumadin for atrial fibril- lated, with the understanding that the risks of bleed-
lation is referred for treatment of bleeding hemor- ing are less than the risks associated with stopping the
rhoids and resulting anemia with a hemoglobin of anticoagulation. Finally, some would argue that the
9.2 g/dL. She reports daily painless bright red bleed- safest approach in such a patient is operative hemor-
ing and prolapse usually requiring manual reduction rhoidectomy, as it can be performed with the shortest
with bowel movements. Recent colonoscopy reports suspension of anticoagulation.
nonbleeding diverticulosis of the sigmoid colon.
Anoscopy reveals enlarged, friable, and prolapsing
internal hemorrhoid in each of the three major posi-
tions. The patient reports soft, formed stools daily, INTRODUCTION
and has been on a high-fiber diet for years as recom-
mended for her diverticular disease. It is customary to begin discussions of hemorrhoid manage-
ment by providing compelling data regarding the number of
Americans who suffer from this entity or the money spent
CASE MANAGEMENT annually on over-the-counter and prescription remedies of
questionable efficacy. Perhaps a more meaningful contem-
Given that this patient has already achieved optimal porary figure is that the term “hemorrhoids” was Google
stool consistency, efforts to increase fiber consump- searched up to 120,000 times per week in 2013, and was the
tion are unlikely to improve her symptoms. While top trending U.S. health-care issue in the Google Zeitgeist
management of anticoagulation complicates rubber 2012 data review. Hemorrhoidal disease remains as trouble-
band ligation (RBL), treatment with sclerotherapy some in our digital age as when Hippocrates wrote of it in his
or infrared coagulation (despite their safer bleeding Treatises of 460 bc (1), and patients will go to great lengths
profiles) is unlikely to succeed at treating prolapsing to self-diagnose and treat rather than seek help from a
grade III hemorrhoids. Options for managing antico- knowledgeable physician for their embarrassing symptoms.
agulation for RBL include (1) performing the proce- Although surgical intervention is for some the most success-
dure fully anticoagulated, (2) discontinuing Coumadin ful and durable intervention, excellent symptom control can
for 1 week prior with continued suspension for up to 2 be achieved for most with nonoperative management.
weeks postprocedure, or (3) using regimens designed
to minimize the period of thromboembolic risk while
off therapy by beginning Coumadin suspension on the
day of RBL and resuming regular dosing on post-RBL
ANATOMY AND PATHOPHYSIOLOGY
day 7. This schedule exploits the drug’s therapeutic
nadir to coincide with the period of greatest risk of The understanding of hemorrhoids has not changed sig-
hemorrhage (days 5–10), and in one published retro- nificantly since Thomson published his seminal paper in
spective study achieved a bleeding risk comparable 1975. He described hemorrhoids as specialized arterio-
venous sinusoids between the terminal branches of the
171
172 Nonoperative therapy for hemorrhoidal disease
hemorrhoidal arteries and veins forming cushions of con-

nective tissue in the subepithelial space of the anal canal. TREATMENT
Age and trauma may compromise the connective tissue
matrix that supports these cushions leading to their pro- CONSERVATIVE MANAGEMENT
lapse. As the sinusoids lack a defined muscular wall, pro-
lapse and trauma predispose to characteristic hemorrhoidal Constipation and straining may result in congestion,
bleeding (2). prolapse, and trauma to the hemorrhoids. Optimization
of stool consistency should therefore be an early goal of
conservative management. The average American con-
sumes only 10–15 g of the 20–30 g that are recommended
CLINICAL EVALUATION daily (9). Dietary consumption can be improved with
patient education materials outlining fiber content of
As patients and their referring providers will attribute common fiber-rich foods. Efficacy of fiber consumption
anorectal symptoms of all sorts to “hemorrhoids,” a care- results from bulking and softening the stool by the pas-
ful history and physical examination must be obtained sive absorption of water. Psyllium is a water-soluble natu-
to distinguish hemorrhoid disease from other anorectal ral source of pure fiber from the husks of psyllium plant
pathology. seeds that is sold in powder form.
The most frequent patient complaint is painless bright Patients who are unable to tolerate psyllium because of
red bleeding that may be noted in scant amounts on the increased flatulence or bloating from its fermentation may
toilet tissue, dripping into the commode, or more sig- prefer insoluble fibers such as FiberCon (polycarbophil) or
nificantly as pulsatile bleeding from the anus. Anemia Benefiber (wheat dextrin—United States, inulin—Canada)
attributable solely to hemorrhoidal bleeding is reportedly that pass through the gastrointestinal tract largely unfer-
rare (0.5 patients/100,000 population) (3–5) but has been mented. Gradual increases in fiber consumption should
encountered not infrequently in the authors’ practices. be prescribed with adequate fluid (six to eight glasses of a
Prolapsing hemorrhoids may cause irritation often noncaffeinated beverage daily) to avoid constipation (9).
described by patients as “burning,” as well as pruritus, the Behavioral modification should minimize stooling times to
sensation of incomplete fecal evacuation, and minor mucus 3–5 minutes with minimal straining. A Cochrane review
seepage resulting in hygiene-related concerns. Patients of seven investigations including 378 patients randomized
can reliably describe the occurrence of prolapse, and if to treatment with fiber supplement versus controls demon-
this reduces spontaneously (grade II), or requires manual strated a statistically significant benefit in the reduction of
reduction (grade III) (Table 19.1). Complaints of more symptomatic prolapse (relative risk [RR] = 0.53) and bleed-
intense pain should guide the clinician toward other likely ing (RR = 0.50) (10,11).
diagnoses such as thrombosis, anal fissure, or abscess. In Little evidence exists supporting the use of stool soften-
order to avoid unnecessary surgical treatment of hemor- ers and laxatives in the treatment of symptomatic hemor-
rhoids, it is important to correctly diagnose burning and rhoids, but such adjuncts may be useful to patients who are
itching secondary to prolapse as opposed to those symp- chronically constipated and for whom bulking agents and
toms being caused by perianal skin irritation from con- increased fluid alone are insufficient.
tact irritants and/or excoriation from overzealous hygiene Warm sitz baths have been theorized to be beneficial
practices. in the treatment of symptomatic hemorrhoids due to an
Physical examination should include visual inspec- observed decrease in anal canal pressure. Their use, how-
tion of the anus, digital rectal examination, and anoscopy. ever, has not been subjected to a randomized trial (12,13).
Concomitant or alternative diagnoses such as fissure, fis- Their use is typically most beneficial in the acute setting.
tula, abscess, or evidence of Crohn disease will frequently Numerous topical agents are available for the treatment of
be discovered (6,7). For patients who present with hema- hemorrhoids that combine varying concentrations of local
tochezia, there should exist a low threshold for evaluation anesthetics (pramoxine, lidocaine, and dibucaine), anti-
with colonoscopy (8). inflammatories (hydrocortisone and 5-ASA), local vaso-
constrictors (phenylephrine), and astringents (witch hazel).
As some prescription preparations can be quite costly for
Table 19.1 Grading of symptomatic internal hemorrhoids
patients, thoughtfulness on the prescriber’s part is required
Grade Physical characteristics given the limited proven efficacy of these compounds. Only
5-ASA has been subjected to a randomized clinical trial,
I Prominent vasculature; bleeding without
which did show a decrease in hemorrhoid-associated pain
prolapse
and bleeding (14). The others do seem to offer relief from
II Prolapse with Valsalva; reduces spontaneously
acute symptoms and have favorable safety profiles with local
III Prolapse with Valsalva; requires manual
allergic reaction being the most common adverse effect.
reduction
Topical steroids should not be advocated for long-term use
IV Prolapsed hemorrhoid that does not reduce due to a potential to cause chronic perianal dermatitis (15).
Treatment / Rubber band ligation 173
Flavonoids are a group of plant metabolites found in a Takano and colleagues published a review of sclero-
variety of fruits and vegetables. They are thought to provide therapy using aluminum potassium sulfate and tannic
numerous health benefits through their antioxidant effects. acid (ALTA), a recently developed agent in Japan. Treating
Their proposed utility in hemorrhoid treatment includes internal hemorrhoids grades II through IV, they reported
modulation of venous tone and capillary resistance with similar relief of symptoms compared to those treated with
resulting improvement of lymphatic drainage, venous aci- hemorrhoidectomy at 28 days postintervention. Recurrence
dosis, and inflammation of the microcirculation (16,17). rate after 1 year was 16% versus 2% in the excisional hemor-
Several studies have assessed oral micronized purified rhoidectomy group (30,31). At present, this formulation is
flavonoid fraction (MPFF) in treating hemorrhoid symp- not approved for use in the United States.
toms. Two placebo-controlled trials showed symptom-
atic improvement with flavonoid dosing, but results were RUBBER BAND LIGATION
inconsistent when combined with fiber. Ho and colleagues
reported that MPFF with fiber led to faster relief of bleeding Ligation of the hemorrhoid mass has been the basis of
than either fiber and RBL or fiber alone (18). In contrast, hemorrhoidal treatments for centuries (1). The modern
Thanapongsathorn compared fiber with and without MPFF application of this principle utilizes a small rubber band to
in a double-blind trial and found similar improvement in obliterate the hemorrhoid’s feeding vessels leading to ulcer-
both groups at 14 days (19). Diosmiplex consists of diosmin, ation, sloughing, and fixation of the ulcer base to the under-
synthetically produced from the citrus flavonoid hesperi- lying sphincter muscle. This was described by Blaisdell in
din, coupled with an alka4 complex. It has been commer- 1958 and popularized by Barron in 1963, who initially used
cially available as Vasculera in the Analpram Advanced kit bands cut from pieces of a urinary catheter fired from a
(Ferndale Laboratories, Inc., Ferndale, Michigan) as 630 mg modified umbilical ligator (32,33). The technique’s simplic-
tablets with a recommended dose of one tablet three times ity, effectiveness, and overall safety have made it the most
daily for 4 days, followed by one twice daily for 9 days. It widely used technique in the United States for treating first-,
has a U.S. Food and Drug Administration GRAS (generally second-, and third-degree hemorrhoids (21).
recognized as safe) designation for medical foods. After obtaining informed consent, the patient is posi-
tioned either on a proctoscopy table or in the left lateral
SCLEROTHERAPY Sims’ position. Anoscopy allows for the hemorrhoid to be
identified. Two types of ligators can be used: the McGown
Injection sclerotherapy, first described by John Morgan of ligator draws the hemorrhoid bundle into its barrel using
Dublin in 1869 using iron persulfate, can be effective in the gentle pneumatic suction, whereas the Barron or McGivney
treatment of symptomatic grade I and II internal hemor- ligator requires an atraumatic grasper for this (Figures 19.1
rhoids (20). Currently used sclerosants include 5% phenol and 19.2). Both devices use a trigger release to deploy a small
in oil, 5% quinine and urea, and 1%–3% sodium tetradecyl rubber band at the base of the hemorrhoid. Typically, two
sulfate. Volumes of 1–3 mL of agent are injected using a bands are preloaded onto the apparatus to prevent slippage
25-gauge spinal needle into the submucosa at the apex of or premature band rupture. As the McGown suction ligator
each hemorrhoid. Treatment aims to obliterate the vascular can be used without an assistant, it has become the authors’
component of the hemorrhoid leading to fibrosis and fixa- and editors’ preference. With increasing prevalence of
tion with prevention of further prolapse. Repetitive sclero-
therapy may lead to scarring and stricture and is therefore
not recommended (21).
Complications stem from injection into unintended (a)
spaces. Bleeding can be exacerbated by superficial injection
causing mucosal slough. Significant pain, infection, and
abscess can occur with injection too deeply into or through
the underlying muscle (22). Impotence, prostatic abscess, (b)
and urinary retention may result from deep anterior injec-
tion in the male patient (23,24). There have been numerous
case reports of life-threatening retroperitoneal sepsis and
necrotizing fasciitis after injection sclerotherapy (25–27).
Mann and colleagues in 1988 surveyed 100 patients (c)
after sclerotherapy for first-degree hemorrhoids and found
62% of patients had no bleeding at 24 hours, but only 41%
remained symptom free after 28 days. Nonetheless, 88%
felt their symptoms had improved with treatment (28). The
same year, Senapati and Nicholls found no significant dif- Figure 19.1 (a) McGown suction rubber band ligator;
ference in bleeding in a randomized controlled trial com- (b) Barron-McGivney rubber band ligator; (c) Infrared
paring sclerotherapy to bulk laxatives (29). coagulator.
(a) apparatus. Although they comment on their need to excise

thrombosed external hemorrhoids in 2.8% of patients
treated thusly, they conclude that this approach is a safe,
cost-effective, and well-tolerated method of treating hemor-
rhoids and excluding more serious colonic pathology at one
setting (38). The editors’ practice is to band the two largest
bundles at the first treatment.
Complications are infrequent following RBL (<2%) and
include vasovagal symptoms, pain, bleeding, external hem-
orrhoidal thrombosis, and rare cases of life-threatening pel-
vis sepsis (39,40). Allowing the patient to rest and recline
(b)
for several minutes with a cold forehead compress is usually
all that is required for vasovagal response. For mild pain
(c) immediately following band placement, injection of local
anesthetic into the base of the banded tissue is often effec-
tive, as are the use of mild analgesics and warm sitz baths.
More severe pain in the immediate postprocedure setting is
best managed by removing the band with a hooked cutting
probe after the site has been locally anesthetized.
A review of 10 papers describing 17 patients for whom
RBL resulted in pelvic sepsis was made by McCloud and col-
Figure 19.2 Banding an internal hemorrhoid. The internal
leagues in 2006 (40). First reported in 1980, the complication
hemorrhoid is teased into the barrel of the ligating gun
with (a) a suction (McGown) ligator or (b) a McGivney liga- is believed to be secondary to necrosis resulting in infection
tor. (c) The apex of the banded hemorrhoid is well above of the adjacent soft tissues (41,42). Although the absolute risk
the dentate line to minimize pain. seems very small in immune-competent patients, common
presenting factors that should alert the surgeon to this dev-
astating possibility are worsening perineal or pelvic pain,
patient latex allergy, the authors and editors have switched urinary retention, and fever. These findings mandate urgent
to commercially available latex-free bands (George Percy examination, often under anesthesia, computed tomogra-
McGown, Brooklyn, New York) without any noticeable phy scan, empiric broad-spectrum antibiotics, and drainage
increase in premature rupture or band slippage. of any identified abscess with debridement of necrotic tissue
Proper placement of the bands is crucial to the proce- and possible fecal diversion. It has been the authors’ prac-
dure’s success, and for this reason the authors have not tice to warn and reinforce this possibility to patients with
endorsed the blind deployment of bands using disposable written instructions upon leaving the office following each
banding systems marketed for this sort of use. Inclusion session of RBL.
of somatically innervated anoderm will result in excruci- Bleeding is a far more commonly described complication,
ating pain. To prevent this, the bands should be applied at and although most episodes are insignificant, cases result-
the apex of the hemorrhoid bundle, well proximal to the ing in hemodynamic instability and transfusion require-
visualized dentate line. We favor the use of a lighted beveled ments do occur and are more common in patients receiving
anoscope to optimize visualization. If the patient expresses platelet inhibitors or systemic anticoagulation. Many sur-
discomfort when either the grasper or suction is applied, geons have therefore avoided RBL in these patients, or at
the application should be aborted and a more proximal site least insisted that such medications be suspended to reduce
selected. Some have suggested injection of local anesthetic this risk. Few data exist regarding safe strategies for RBL
either into the banded pedicle itself or the upper anal canal in such patients. Part of the difficulty lies in the unpre-
(5 mm above dentate line, 1.5 mL in four quadrants) (34,35). dictable time frame in which the banded hemorrhoid will
The authors have found that this is seldom necessary. slough and put the patient at greatest risk for hemorrhage.
Disagreement exists as to how many hemorrhoids should Although this usually occurs within 5 to 10 days, delayed
be banded per session. Barron advocated ligation of one bleeding is indeed encountered (43,44).
hemorrhoid at a time, with future sessions at 3-week inter- Some have recommended offering this procedure only
vals so as to minimize patient discomfort and to allow reas- after antiplatelets have been held for 7 days, followed by
sessment as to the need for additional bandings (33). Others another 7–10 days thereafter before restarting (35,45).
have described no significant increase in patient discomfort If only one hemorrhoid is banded per session, this will
with multiple bandings in a single session (36,37). Davis result in a significant number of days off therapy and an
and colleagues reported their experience with combined increased risk of thromboembolic events. Nelson and col-
colonoscopy under moderate sedation followed by imme- leagues described their preference of minimizing patients’
diate three-quadrant hemorrhoid ligation using a TriView time off these medications by offering RBL on the day of
(tri-slotted) anoscope and a Short-Shot multifire banding initial consultation, followed by immediate suspension of
Comparison of techniques 175
antiplatelet medications (aspirin or clopidogrel) for 10 days, Complications after IRC are infrequent. As with RBL,
and Coumadin for 7 days. This approach allows for recovery pain occurs if the energy is applied too near the somatically
of platelet and coagulation capacity by the time the patient innervated anoderm; the discomfort, however, is typically of
is theoretically at greatest risk of bleeding (5–10 days), and a shorter duration and lesser severity than misplaced hem-
minimizes the total number of days they are at increased orrhoidal bands. For this reason, IRC can be used to treat
risk of thromboembolic events while these medications are small distal internal hemorrhoids that are not suitable for
suspended. They described 605 bands placed on 364 patients RBL. Excessive application can result in bleeding. Rarely,
while on either anticoagulation or antiplatelet therapy, with ulceration can progress to fissure formation (39). IRC has
a risk of significant bleeding of less than 1%. If clopidogrel been shown to be effective for the treatment of hemorrhoidal
was not included, this risk dropped to less than 0.5%. They bleeding, but less so for alleviating significant prolapse (48).
point out that these rates of bleeding do not differ signifi- It is our practice to use IRC as a preferred alternative to
cantly from what has been reported in the literature for sclerotherapy for hemorrhoids too small or distal for RBL.
patients not taking antiplatelet or anticoagulation agents,
but that patients taking clopidogrel showed a trend toward
greater bleeding rates (46,47). This thoughtful approach
appears to safely allow these medically challenging patients COMPARISON OF TECHNIQUES
the benefits of RBL over less successful (sclerotherapy) or
more invasive (hemorrhoidectomy) alternatives. In comparing RBL to sclerotherapy, RBL is associated
with a more frequent occurrence of postprocedure pain
INFRARED PHOTOCOAGULATION from 30 minutes up to 72 hours. RBL is also associated
with more episodes of bleeding in the immediate 24 hours
Infrared photocoagulation (IRC) uses infrared radiation postprocedure period (49). In a meta-analysis compiled in
focused by a photoconductor to create protein coagula- 1995, MacRae and associates showed that patients treated
tion at the base of the pedicle, causing scarring and fibrosis with RBL were less likely to need additional treatment
(Figures 19.1 and 19.3). Alteration of the optical wavelength versus patients treated with sclerotherapy (p = 0.031) or
of the coagulator and contact time controls the depth of IRC (p = 0.0014) (50). However, a more recent random-
tissue penetration to approximately 3 mm (24). A slot- ized trial showed the number of sessions needed to reach
ted anoscope is used to visualize the base of the targeted symptomatic relief was equivalent between RBL and IRC
hemorrhoid. Three or four pulses of 1–1.5 seconds each are (1.6 SD 0.9) (48). In the case of third-degree hemorrhoids,
made around each hemorrhoid apex, producing a 3–4 mm2 patients treated with RBL were more consistently symptom
area of coagulation, which appears immediately as a whit- free versus those treated with IRC, with no difference noted
ish circular eschar. Proponents advocate treating one to between these therapies for treatment of first- and second-
three hemorrhoids per session. Treatment of additional degree hemorrhoids (48,51).
hemorrhoids can be offered at 3- to 4-week intervals (21,39).
However, governmental reimbursement currently applies
a 90-day global to IRC. Additional treatments during this SUMMARY
period will not be reimbursed.
The great majority of patients with hemorrhoidal
disease may be treated nonoperatively, beginning
with dietary modification intended to optimize
stool consistency and ease of defecation, selective
use of over-the-counter and prescription strength
preparations for alleviation of mild acute symp-
toms, and escalation to office-based interventions to
treat chronic symptoms such as bleeding and pro-
lapse. Although treatment with sclerotherapy and
infrared coagulation shows results comparable to
treatment with RBL for stages I and II internal hem-
orrhoids, patients with more significant prolapse
(grade III) achieve improved and more durable
Sites of results with RBL. Management of platelet inhibi-
coagulation
tors and anticoagulation around these procedures
requires some thoughtfulness and should be tai-
lored to the procedure’s bleeding risk (greatest with
Figure 19.3 The infrared photocoagulator creates a small RBL), and the patient’s unique thromboembolic risk
thermal injury. Thus, several applications are required for while these agents are suspended. Although quite
each hemorrhoidal column.
21. Beck DE. Hemorrhoidal disease. In Beck DE, Wexner

rare, the presenting symptoms and signs of pelvic SD. (eds). Fundamentals of Anorectal Surgery, 2nd
sepsis following these procedures must be conveyed Edition. Bailliere Tindell, 1998, pp. 237–53.
to the patient, and clinicians, once alerted to this 22. Sim AJ et al. Surg Gynecol Obstet. 1983;157:534–6.
possibility, must evaluate and intervene rapidly in 23. Bullock N. BMJ. 1997;314:419.
order to salvage patients from this potentially fatal 24. Hardy A et al. Dig Surg. 2005;22:26–33.
complication. Last, although some have claimed that 25. Ribbans WJ, Radcliffe AG. Dis Colon Rectum. 1985;
nearly all hemorrhoidal disease can be treated non- 28:188–9.
operatively, good clinical judgment can often pre- 26. Kaman L et al. Dis Colon Rectum. 1999;42:419–20.
dict which patients will be better served by operative 27. Barwell J et al. Dis Colon and Rectum. 1999;42(3):
management. Such patients, as well as those whose 421–3.
symptoms are not well controlled after an appro- 28. Mann CV et al. J R Soc Med. 1988;81:146–8.
priate trial of nonoperative intervention, should 29. Senapati A, Nicholls RJ. Int J Colorectal Dis. 1998;3:
be offered excisional hemorrhoidectomy without 124–6.
f urther delay. 30. Takano M et al. Int J Colorectal Dis. 2006;21:44–51.
31. Takano M et al. 2010, May 15. Proceedings of the
Annual Meeting of 2010. The American Society of
REFERENCES Colon and Rectal Surgeons (ASCRS), Minneapolis,
MN, 52–4.
1. Parks AG. Guys Hosp Rep. 1955;104:135–56. 32. Blaisdell PC. Surg Gynecol Obstet. 1958;106:485–8.
2. Thomson WHF. Br J Surg. 1975;62:542–52. 33. Barron J. Dis Colon Rectum. 1963;6:109–13.
3. Nakama H et al. Am J Med. 1997;102:551–4. 34. Law WL, Chu KW. Dis Colon Rectum. 1999;42:363–6.
4. Korkis AM, McDougall CJ. Dig Dis Sci. 1995;40: 35. Gordon PH, Nivatongs S. Principles and Practice of
1520–3. Surgery for the Colon, Rectum and Anus, 3rd Edition.
5. Kluiber RM, Wolff BG. Dis Colon Rectum. 1994;37: New York, NY: Informa Healthcare, 2007, p. 148.
1006–7. 36. Khubchandani IT. Dis Colon Rectum. 1983;26:705–8.
6. Madoff RD, Fleshman JW. Gastroenterology. 2004; 37. Lee HH et al. Dis Colon Rectum. 1994;37:37–41.
126:1463–73. 38. Davis KG et al. Dis Colon Rectum. 2007;50:1445.
7. Fazio VW, Tjandra JJ. Adv Surg. 1995;29:59–78. 39. Larach S et al. Nonoperative treatment of hemor-
8. Rivadeneira DE et al. Dis Colon Rectum. 2011;54: rhoidal disease. In Hicks TC, Beck DE, Opelka FG,
1059–64. Timmcke AE. (eds). Complications of Colon and
9. Villalba H, Abbas M. Perm J. 2007;11(2):74–6. Rectal Surgery. Baltimore, MD: Williams and Wilkens,
10. Alonso-Coello P et al. Cochrane Database Syst Rev. 1997, pp. 173–80.
2005;(4):CD004649. 40. McCloud JM et al. Colorectal Dis. 2006;8:748–55.
11. Alonso-Coello P et al. Am J Gastroenterol. 2006;101 41. O’Hara VS. Dis Colon Rectum. 1980;23:570–1.
:181–8. 42. Russell TR, Donohue JH. Dis Colon Rectum. 1985;28:
12. Dodi G et al. Dis Colon Rectum. 1986;29:248–51. 291–3.
13. Tejirian T, Abbas M. Dis Colon Rectum. 2005;48: 43. Corman ML. Colon and Rectal Surgery, 6th Edition.
2336–40. Lippincott Williams and Wilkins, Philadelphia, PN,
14. Gionchetti P et al. Can J Gastroenterol. 1992;6(1): 2013, p. 288.
18–20. 44. Patel S et al. World J Clin Cases. 2014;2(4):86–9.
15. Johanson JF. J Gastrointest Surg. 2002;6:290–4. 45. Beattie GC et al. Ulster Med J. 2004;73:139–41.
16. Desnoyers P. In: European Symposium on Venous 46. Bat L et al. Dis Colon Rectum. 1993;36:287–90.
Disease and its Treatment. Rome, Italy, April 1978. 47. Iyer VS et al. Dis Colon Rectum. 2004;47:1367–70.
Gazette Medicale de France, 1987:17–21. 48. Poen AC et al. Eur J Gastroenterol Hepatol. 2000;
17. Klemn J. Gazette Medicale de France. 1976;83: 12:535–9.
3158–65. 49. Marques CF et al. Tech Coloproctol. 2006;10:312–7.
18. Ho YH et al. Dis Colon Rectum. 2000;43:66–9. 50. MacRae HM, McLeod RS. Dis Colon Rectum. 1995;
19. Thanapongsathorn W, Bajrabukka T. Dis Colon 38:687–94.
Rectum. 1992;35:1085–8. 51. Linares Santiago E et al. Rev Esp Enferm Dig. 2001;
20. Morgan J. Medical Press and Circular. 1869;29–30. 93:238–47.
20
Surgery and nonoperative therapy of perirectal
abscess and anal fistulas
MOHAMMED IYOOB MOHAMMED ILYAS AND CRAIG A. REICKERT
CHALLENGING CASE has decided not to pursue further surgery for the fis-
tula and remain diverted via a colostomy with draining
A 40-year-old male with poorly controlled HIV setons in place. Anorectal abscesses and fistulas can
infection presents with incision and drainage of a be incredibly frustrating, both for the patient and the
perirectal abscess 2 cm right posterolateral to the managing physician. Meaningful outcomes data with
anal verge. He is treated with a short course of oral large, prospective randomized trials regarding the
antibiotics with resolution of the acute event. Several management of these entities are extremely limited.
months later, he presents with purulent discharge This chapter addresses the surgical as well as non-
from the drainage site as well as a second area in the operative management of these common problems,
posterior midline, 4 cm from the anal verge. Exam focusing on means of improving clinical outcomes.
under anesthesia demonstrates a single primary
fistula opening in the anal canal, in the posterior
midline 3 cm proximal to the dentate line, which com- INTRODUCTION
municate with both secondary openings. Draining
setons are placed, and biopsies from the fistula tracts
show no evidence of Crohn disease or malignancy. A good understanding of the anatomy of the various peri-
Six weeks later, he presents in septic shock due to anal spaces in relation to the sphincters and the pelvic floor
worsening perineal infection. is essential in the clinical management of perianal abscess
and fistulas. It also requires understanding the local ana-
tomical factors along with the etiology, systemic, and local
patient factors, and the various treatment options available.
CASE MANAGEMENT
The patient is initially managed with additional drain-

age and a diverting colostomy. After recovering,
ANATOMICAL CONSIDERATIONS
he is initiated on highly active antiretroviral therapy
(HAART) with an excellent response in his CD4 count, Successful management of perianal abscesses and fistulas
and his viral load becomes undetectable. After mag- requires a good understanding of the anatomy of the anal
netic resonance imaging (MRI) of the pelvis demon- canal, perianal region, and etiopathogenesis of abscesses
strates no further infection in the pelvis, an anal fistula and fistulas. The anal canal anatomically extends from the
plug is placed to attempt to close the fistula; within a anal verge to the anorectal junction (cephalad extent of the
week, the plug has dislodged. An attempt at closing pelvic diaphragm), measures about 2–4 cm, and is divided
the fistula with an endorectal advancement flap sev- by the dentate line into the proximal endothelium lined
eral weeks later also fails. Further biopsies again show (columns of Morgagni) part and distal part lined with squa-
no evidence of Crohn disease or malignancy, and a mous epithelium. Between the columns of Morgagni are
second attempt at endorectal advancement flap clo- unevenly distributed anal crypts, the ducts of which extend
sure performed 12 weeks later also fails. The patient through the internal sphincter into the intersphincteric
space or into the external sphincter (2–4).
177
178 Surgery and nonoperative therapy of perirectal abscess and anal fistulas
There are six potential spaces around the anorectum (4) by the external anal sphincter and levator muscle and the
that can become sites of infection (Figure 20.1). The subcu- lateral wall formed by the obturator internus muscle. The
taneous space is filled with fat and fibers of the corrugator two ischiorectal spaces communicate posteriorly behind
cutis ani and is in communication with the central space the anus through the retrosphincteric space. The pelvirectal
and laterally with the ischiorectal space. The subcutaneous spaces are found between the levators and the pelvic perito-
space is also considered to be part of the ischioanal space. neum. The submucosal space lies between the anal mucosa
The central space is considered the main perianal space and the internal sphincter muscle (4–9).
which is in communication with others. It surrounds the The cryptoglandular theory initially proposed by
anal canal and is bounded above by the termination of lon- Eisenhammer and Parks has been generally espoused in
gitudinal muscles and below by the muscular loop of the the etiopathogenesis of perianal abscess. Stasis within the
external sphincter complex. The intersphincteric spaces ducts of these glands leads to obstruction and formation of
are four upward longitudinal extensions from the central abscess formation in a majority (90%) of these cases (10).
space below into and between the longitudinal intersphinc- Chronic anal fissure has also been reported to be involved
teric muscles spaces. From lateral to medial, the first and in the etiopathogenesis of anal fistula. The remaining 10%
the third of the four spaces open into the ischioanal space, of the abscesses are secondary to other predisposing factors
the second space opens into the pelvirectal space, and the like inflammatory bowel disease, trauma, malignancy, radi-
medial most space communicates with the submucosal ation, tuberculosis, actinomycosis, and lymphogranuloma
space. Knowledge of these spaces, particularly of the cen- venereum. Perianal abscesses and fistulas are considered
tral and intersphincteric spaces, helps in understanding the two phases of the same disease process (11). The first phase
progression of perianal abscesses when not drained. The involves obstruction of the mucous anal gland leading to
ischioanal or ischiorectal is a pyramidal space posterior stasis and infection. Most glands are confined to the sub-
to the urogenital diaphragm, with its medial wall formed mucosa and when obstructed and infected will discharge
into the anal canal with spontaneous healing. Nearly half of
(a)
these glands do have extensions traversing the internal anal
sphincter and may not discharge easily in the anal canal sec-
ondary to tonic contraction of the internal anal sphincter.
Supralevator
Pus may then follow along with the path of least resistance
between the internal and external anal sphincter or through
the external sphincter into the ischioanal fossa (12).
A fistula is defined as an abnormal communication
between two epithelial lined surfaces with 90% of cases
associated with cryptoglandular etiology. Trauma, postop-
Intersphincteric erative states, inflammatory bowel disease, anal fissure, and
tuberculosis accounts for the rest. Anal fistulas are charac-
terized as simple and complex. Simple fistulas are those that
Perianal
are superficial or low transsphincteric, have a single exter-
Ischioanal Submucosal nal opening, and have no associated abscess (Figure 20.2).
Complex fistulas include those that are high intersphinc-
(b) teric, high transsphincteric, extra- and suprasphincteric fis-
tulas, or have multiple external openings or are associated
with abscesses, rectovaginal fistulas, or anorectal strictures.
Fistulas involving more than 30% of the external sphinc-
ter or anteriorly located in females and fistulas associated
Retrorectal
with local irradiation or Crohn disease are also considered
as complex fistulas.
Supralevator The true incidence of the anorectal abscess is difficult to
Deep
assess, as many such abscesses drain spontaneously or are
postanal drained in the physician’s office or in the emergency room.
The incidence of anorectal abscess in the United States falls
between 68,000 and 96,000 per annum (1).
Intersphincteric fistulas account for nearly one-third
Superficial of the cases and are known to branch out in the inter-
postanal sphincteric plane only without involving the external
anal sphincter. Transsphincteric fistulas account for up to
two-thirds of the fistulas with variable involvement of the
Figure 20.1 Anorectal spaces. (a) Coronal section. external sphincter muscle. Transsphincteric fistulas may
(b) Sagittal section. be associated with a blind tract leading to the ischiorectal
Management of abscesses 179
(a) (b) (c) (d)
Figure 20.2 Classification of fistula-in-ano. (a) Intersphincteric. (b) Transsphincteric. (c) Suprasphincteric. (d) Extrasphincteric.
fossa or through the levator muscle into the pelvic cavity.

Perforation of the blind tract during evaluation may lead
to an extrasphincteric fistula. Suprasphincteric fistulas
are uncommon and have tracts extending from above the
puborectalis, through the levator plate, and into the ischio-
rectal fossa and skin. Suprasphincteric fistulas can be asso- Curved tracts Exc
ciated with an ischiorectal abscess. Extrasphincteric fistulas
ep
tion
are outside of the external sphincter complex and are less Transverse line
common and extend from the perianal skin through the 3 cm
ischiorectal fossa and levator muscles into the rectum. Straight tracts
DIAGNOSIS OF ABSCESSES
Physical examination (in the prone position preferably) is

the most important diagnostic study when an anorectal
abscess is suspected. Severe tenderness or fluctuance on the
digital rectal exam are suggestive of intersphincteric abscess.
Examination under anesthesia should be considered in
cases of ambiguous clinical exam. Imaging is useful when
Figure 20.3 Goodsall’s rule.
a diagnosis could not be confirmed on physical exam or
when supralevator abscesses are suspected. Triple contrast
computed tomography imaging of the pelvis with 2.5 mm have a colonoscopy performed. MRI is helpful in delineat-
cuts is more helpful in an acute setting than magnetic reso- ing the anatomy of the fistula in relation to the sphincter
nance imaging (MRI) of the pelvis. Imaging is also useful complex and demonstrating pelvic sepsis, if any. It is opera-
to evaluate recurrent or incompletely drained abscesses or tor independent and does not need any instrumentation of
to evaluate complex abscesses. Endoanal ultrasound and the anus. It is indicated when a complex fistula is suspected
transperineal sonography are other imaging modalities with multiple external openings, in those where an internal
used to evaluate perianal and perirectal abscesses. opening could not be identified, or in recurrences following
Fistulas resulting from the cryptoglandular disease a fistulotomy when cure is expected.
often have a history of prior anorectal abscesses. Associated Endoanal ultrasound with hydrogen peroxide injection
symptoms include bleeding secondary to granulation tissue into the external opening can identify primary and sec-
from healing and irritation from the discharge. Pain is usu- ondary tracks. Endoanal ultrasound is more helpful with
ally associated with chronic inflammation, but severe pain diagnosing transsphincteric fistula compared with inter- or
is associated more with Crohn disease and malignancy. suprasphincteric fistulas.
Clinical evaluation of patients with perianal fistulas should
include the history of prior anal procedures, history of
incontinence, history of inflammatory bowel disease, radia-
tion to the pelvis, and obstetric history. Physical examina-
MANAGEMENT OF ABSCESSES
tion in the office is more helpful with delineating superficial
and transsphincteric fistulas and is expectedly less accurate Adequate drainage of abscesses remains the cornerstone
with supralevator fistulas and intersphincteric fistulas. of management of perianal and perirectal abscesses.
Goodsall’s rule has been estimated to have a positive Antibiotics are indicated when there is associated cellulitis
predictive value in less than two-thirds of the cases and is or when patients are immunosuppressed but are not ade-
more accurate in locating posterior compared with anterior quate treatment without incision and drainage and do not
fistulas (Figure 20.3). Patients with abdominal symptoms or have any protective effect on subsequent fistula formation.
findings suggestive of noncryptoglandular etiology should Amoxicillin-clavulanic acid combination has been shown
to offer the antibacterial spectrum necessary to cover the

gram-negative enteric bacteria and gram-positive epider-
mal bacteria typically found in the perianal region (13).
Escherichia coli and Klebsiella pneumoniae are the most Posterior drainage
common antibiotics isolated from perianal abscesses in Counter
External sphincter
patients without and with diabetes, respectively (14). The drainage
Internal sphincter
Counter
specificity of the microbiologic information from the ano- drainage
rectal abscesses has been shown to be low and is not a reli-
able predictor of fistula formation; thus, routine cultures
are not performed during drainage of perianal abscesses.
Studies also have shown no differences in progression to
fistula formation with or without the use of antibiotics (13).
Dentate line
DRAINAGE OF ABSCESSES Figure 20.4 Hanley procedure.
The extent of the incision and drainage depends on the type (a)
of abscess. Superficial perianal and ischiorectal abscesses
can be drained in the surgeon’s office or in the emergency
room, with or without conscious sedation. Understanding
the concept of the skeletal muscle rule helps deciding
“inward” versus “outward” drainage of perianal/perirectal
abscesses. As a rule, submucosal abscesses, intersphincteric
abscesses, and supralevator abscesses from an intersphinc-
teric fistula or from the pelvic disease are drained “inward”
into the rectum. Supralevator abscess from an upwardly
extending transsphincteric fistula, ischioanal abscess from
a transsphincteric fistula, or ischioanal abscess from a
supralevator abscess caused by an intersphincteric fistula
penetrating the levator plate are drained “outward.” The
rule of thumb is that if the sepsis does not pass through the
skeletal muscle (external anal sphincter and levator ani), the
abscess should be drained inward, whereas if it traverses the
muscle it should be drained outward (15). MRI can direct (b)
the surgeon to choose the correct direction of the drainage
and is indicated in all supralevator abscesses. If the source of
the abscess is intraabdominal, transrectal drainage is indi-
cated. Also, abdominal drainage can be performed based on
ease of access and directionality of the abscess cavity. Also,
percutaneous drainage may prevent the creation of the fis-
tulous track through the levator plate and is more successful
than transrectal drainage.
Horseshoe abscesses or bilateral abscesses arise from
the deep postanal space and require operative drainage
for source control and to delineate the abscess. Hanley or
modified Hanley procedure consists of open drainage of
Figure 20.5 Catheter drainage of abscess.
the postanal space through the anococcygeal ligament, pos-
terior midline incision of the internal sphincter, and open
drainage of the bilateral ischiorectal fossae. Modifications (Figure 20.5a,b). The catheter can be secured to the skin to
to this drainage include elliptical incisions on both ischio- ensure adequate drainage of the abscess cavity until reso-
rectal space and internal sphincterotomy and drainage of lution of the abscess, and could also be used to perform a
deep space with or without placement of seton in the mid- drain study to decipher the anatomy.
line (16) (Figure 20.4). Drainage with primary fistulotomy in the acute setting
Drainage catheters help minimize the perianal inci- is controversial. Historically it is performed along with the
sions with adequate drainage of large abscess cavities abscess drainage to obtain source control and increase the
Management of fistula 181
healing rate without the need for the subsequent procedure. wound discharge (26–28). Incontinence rates are higher with
A meta-analysis of six randomized control studies showed fistulotomy when it is associated with complex fistulas, female
that fistula surgery during the initial incision and drainage gender, prior fistula surgeries, and surgeon experience.
of the abscesses reduced significantly the recurrence, per- Setons have been in use since ancient times, when a caus-
sistent abscess/fistula, or repeat surgery. Incontinence at 1 tic chemical seton from a plant extract known as kshara-
year following drainage with fistula surgery was not statis- sutra was used to obliterate the tract. A variety of other
tically different (pooled relative risk [RR] 3.06, 95% confi- materials have been used since, including different types of
dence interval [CI] 0.7–13.45) (17). But, when the amount sutures, stainless steel wires, Pezzer catheters, self-locking
of sphincter involvement is confounded by the acute cables, Penrose drains, vessel loops, and rubber bands (29–
inflammation, placement of seton may be indicated to pre- 31). Grooved Lockhart-Mummery probes are used to probe
vent unintended consequences of the excessive division of fistula track and help with seton insertion. Fistula probes
sphincter and its associated consequences (18,19). are modified with an eye near the tip of the probe through
Identification of the crypt of origin during the initial which the seton may be passed and withdrawn through the
incision and drainage of the abscess could be performed by tract. Other modifications used for fistula probe insertion
manual pressure over the abscess cavity, by careful inser- include using a railroad technique with the use of a plastic
tion of the probe into the suspected duct by direct visual- infusion line or use of a plastic infusion line and olive-tip
ization, or injection of hydrogen peroxide via a catheter malleable metaguide for seton placement (32–34). Setons
inserted into the external opening. If the offending duct are used when fistulotomy is not possible, as with higher fis-
is identified, the incidence of subsequent fistula could be tulas, or not advisable, as with anterior fistulas in women or
lower than 30% as reported, confirming the role of primary in Crohn disease. The use of a seton helps stimulate fibrosis
fistulotomy during initial incision and drainage procedure around the track to make it obvious and helps identify the
(17,20). Microbiological origin of the abscess has not been track for sphincter-sparing procedures like placement of a
shown to be helpful in the management of anorectal abscess fistula plug, fibrin glue, or ligation of the intersphincteric
of cryptoglandular origin (21–23). tract (LIFT).
Postoperative management should include local wound Seton could be used as a single-stage cutting seton,
care with sitz baths, and packing of the abscess cavity two-stage seton fistulotomy, or draining seton followed by
should be avoided to facilitate drainage of the abscess cav- a definitive procedure. The cutting seton is used purely to
ity. Antibiotics are indicated when cellulitis persists or in divide the muscle, while the staged technique is advocated
patients who are immunosuppressed. when the muscle is too thick or to avoid the pain on the
patient that a cutting seton would inflict while cutting
through the anoderm. Cutting setons require maintaining
tension by the use of leg strap and tourniquet, use of a syn-
MANAGEMENT OF FISTULA thetic cable tie, use of a hangman’s tie using a polypropylene
or nylon suture, or use of rubber band ligator to manage
The challenges associated with anal fistula management the tension. Use of cutting setons has a reported recurrence
have been touted as the chief reason for the opening of rate with variation based on the preservation or partial divi-
St. Mark’s Hospital in London, England. The goals of fistula sion of the internal anal sphincter. When the internal anal
management include elimination of sepsis, closure of the sphincter is preserved, the recurrence rate was 3%, and the
fistula tract, maintenance of continence, and prevention of incontinence rate was 5.6% versus 5% and 25.2% when the
recurrence (24). internal anal sphincter partially divided. A “snug” seton
Examination under anesthesia with the use of adjuncts technique was also described where the seton is tied snugly
(fistula probe or methylene blue or hydrogen peroxide around the sphincter muscles with minimal tension. A
injection) helps identify the internal opening and assess the “two-thread” technique involves passing two No. 0 silk ties
sphincter involvement. The treatment modality depends on around the fistula track, tying one of the threads, and leav-
the amount of sphincter involvement. ing the second untied initially and tying after 1 month when
The simple and most distal intersphincteric fistula can be the first thread is removed. Another technique described is
managed with laying open the fistula tract to the anoderm. the use of a drainage seton along with a primary cutting
Multiple studies since 1987 have shown fistulotomy to be seton. The use of multiple setons simultaneously has also
effective with recurrence rates ranging from 3% to 7% (21). been discussed with five 1–0 silk sutures, and at weekly
Recurrence rates are lower (4%) for intersphincteric fistulas intervals each suture is tightened progressively after tak-
and higher for transsphincteric (7%) and suprasphincteric ing out the previously tightened suture, which becomes lax
fistulas (33%) (25). Fistulotomy wounds typically heal by 4–6 (24,35). Multiple setons could also be used as modified cut-
weeks, and marsupialization of the wound edges leads to ting setons along with partial fistulotomy (staged seton),
less bleeding and a decrease in the size of the wound, with- wherein the fistula tract is laid open to the dentate line. Use
out increasing postoperative pain and sepsis. It has also been of the seton as a long term-draining seton for low trans-
shown to result in faster healing with a shorter duration of sphincteric and intersphincteric fistulas has been shown
to be a good alternative to primary surgical treatment with closure rate of 50% with one attempt, increasing to 72%
similar healing rates and recurrence rates compared with with two or three attempts, and 26% failure rate even after
primarily surgically treated patients, but incontinence has multiple injections of fibrin (40). Since then multiple studies
been shown to be rare (36). The major disadvantage is that have reported conflicting success rates ranging from 10% to
the treatment takes a longer period compared with primary 78% (41–50). A figure-of-eight absorbable suture to prevent
surgery (37). When draining setons are used, the second- early expulsion of the glue into the rectum was placed by
stage surgery when performed is usually 6–8 weeks after, some investigators but did not show a significant difference
to allow for adequate healing of the cephalad portion of the in the success rate in a prospective controlled study.
sphincter. Fibrin glue instillation is contraindicated in the pres-
ence of active inflammation or sepsis or in the presence
of advanced HIV infection with low CD4 count and high
viral load, as these are known to reduce the success rate.
DRAINAGE TUBES IN FISTULAS Inadequate tract preparation with persistent tract epithe-
lialization and granulation tissue lining the tract, leading
A tube loop seton drainage has been described with the use to its chronicity, are all attributed to failure of fistula clo-
of a 16–18Ch Nelaton catheter for multiple recurring high- sure associated with fibrin instillation. Fibrin clot extrusion
spreading extrasphincteric fistulas. It helps with irrigating during coughing or straining or due to liquid consistency
the cavity and bidirectional drainage from the space. Use of of the glue are other possible reasons for failure. The disad-
an 8 mm Penrose drain with multiple holes with a Penrose vantage attributed to fibrin glue instillation is the theoreti-
drain fixed to the distal one-third point of the punctured cal risk associated with transmission of infectious agents to
Penrose drain has also been described for draining abscess humans, including hepatitis B or HIV from human pooled
cavities associated with high intersphincteric or trans- blood or exposure to spongiform encephalopathy from
sphincteric fistulas. Similarly, Malecot catheters have been bovine aprotinin (43).
used to drain the abscess cavity as well.
There are conflicting data on the predisposition of cer-
tain perianal abscesses to fistula formation. Earlier reports
were suggestive of perianal abscesses more predisposed
FISTULA PLUG
to fistula formation than horseshoe abscess. Other stud-
ies showed eightfold higher fistula formation with ischio- Experience with the fibrin glue instillation leads to the
rectal abscess and threefold increase with intersphincteric development of a material that could be secured to the fis-
abscesses compared with perianal abscesses (38). tula tract and improve the fistula closure rate. Fistula plus
is used to obliterate the track and secure it with sutures to
the internal opening. Surgisis Anal Plug (Cook Surgical,
Belington, Indiana) is the first-generation bioabsorbable,
FIBRIN GLUE xenograft made from porcine small intestinal submucosa.
Fistula plug, like fibrin glue instillation, is performed when
Fibrin glue is a mixture of fibrinogen and thrombin and has local sepsis is controlled and the tract is relatively dry. A
been in use for over three decades in the management of draining seton is typically needed for approximately 8
anal fistulas. The fibrin sealant (Tissucol; Baxter Laboratory, weeks prior to fistula plug application. During the proce-
Maurepas, France) contains Aprotinin as a fibrin stabilizer dure, the seton is removed, and the tract is cleaned with
along with coagulation factor XIII and calcium chloride. hydrogen peroxide and debrided with fistula brush. The
It is injected using an appropriate-sized catheter from the plug is then pulled through the internal opening until mild
external opening of the fistula, which is withdrawn slowly resistance is met and then secured in four quadrants at the
to leave a bleb of glue at the external opening. The patient internal opening with an absorbable suture. The excess plug
is kept immobile on the operating table for a few minutes is trimmed at the external opening. The external opening is
to ensure the glue solidifies. When injected into the fistula also widened to ensure adequate drainage. Strenuous activ-
tract, it is expected to seal the tract, enhance wound heal- ity could still extrude the plug, and patients are advised to
ing, promote hemostasis and angiogenesis, and serve as a avoid such activities for 2 weeks postoperatively.
matrix for fibroblast ingrowth and formation of a collagen Newer prefabricated cone-shaped devices such as Surgisis
network over a period of 1–2 weeks. Fibrin installation is Anal Fistula Plug (AFP) (Cook Surgical, Bloomington,
carried out after control of local sepsis and ensuring that the Indiana) have been introduced since, which can be easily
tract is relatively dry (39). secured into the fistula track and are available now. Like fis-
Fibrin glue instillation is easy to apply, associated with tula glue instillation, the results vary widely and are reported
minimal discomfort, minimal postoperative analgesia, and to be between 14% and 87% (24,51–54). Review of 20 stud-
early return to normal activity. Hjortrup et al. reported their ies (2 abstracts and 18 original articles) with 530 patients
initial modest success with fibrin instillation with a fistula showed a fistula closure rate at 54% and plug extrusion rate
Ligation of internal fistula tract 183
at 8.7%. This review also showed no differences in the fis- (a)

tula closure rate between Crohn and non-Crohn patients.
An absence of the preoperative seton, aggressive curettage
of the track during the procedure leading to widening of
the lumen, and inadequate fixation are all factors leading
to extrusion of the plug. Inappropriate selection of patients
also leads to higher failure rates, such as patients with mul-
tiple tracks.
Delayed absorbable synthetic plug (GORE BIO-A) was
purported as an improvement with an expanded disc at one
end and several tails. The disc is anchored to the internal
sphincter and buried under a mucosal flap. The length of
the fistula plug is adjusted to fit the length of the fistula track
with studies reporting use with fistula track ranging from
1.5 to 8 cm (55,56). These delayed absorbable flaps undergo
hydrolytic and enzymatic degradation, and studies have (b)
reported that polymer scaffolding is replaced by a layer of
newer tissue within 7 months without generating a chronic
inflammatory response (57,58). The rate of fistula healing
ranges between 15.8% and 72.7%. There has been some
disagreement over the use of a draining seton before inser-
tion of the GORE BIO-A plug, with Stamos et al. reporting
higher plug implantation and healing rate without the use of
a prior seton (59,60).
LIGATION OF INTERNAL FISTULA

TRACT
The ligation of internal fistula tract (LIFT) procedure Figure 20.6 Ligation of intersphincteric Tract (LIFT).
(a) Incision in the intersphincteric groove to expose fistula
includes identifying the fistula tract in the intersphincteric
tract containing a flexible probe. (b) Tract is ligated. (With
plane, ensuring secure ligation of the tract, curetting the kind permission from Springer Science+Business Media:
external portion of the tract, and performing partial fistu- Fundamentals of Anorectal Surgery, Anorectal Abscess
lotomy or drainage along with laying open the distal part of and Fistula-in-ano, 2018, Vogel JD, Vasilevsky CA.)
the tract (Figure 20.6). Any defect in the external sphincter
muscle is repaired with sutures. This procedure was first
described by Rojanasakul in 2007 with healing rates of up requiring more extensive dissection in the intersphincteric
to 94% (61,62). Although similar results are not reproduced plane. The principle behind BIOLIFT appears to have been
by many, the LIFT technique continues to remain popular. extrapolated from studies using biological graft in rec-
A meta-analysis of 18 studies with 592 patients reported a tovaginal fistula repairs (66). Studies show the BIOLIFT
mean healing rate of 74.6% during a mean follow-up period procedure to have a fistula closure rate of 63%–94%, with
of 42.3 weeks with a mean healing time of 5.5 weeks. No poorer outcomes with anterior fistulas. A human acellular
incontinence was reported from the procedure (63). Similar dermal matrix has been used as an adjunct with LIFT, called
results were reported from a recent meta-analysis of 24 orig- BIOLIFT plug, and has been shown to be successful in 90%
inal articles including 1,100 patients, which showed a simi- of cases in a study involving 21 patients (67,68). The meta-
lar result with a success rate of the repair at 76.4%, with no analysis by Vergara et al. did not show enough evidence
reported incontinence. Failure of the LIFT procedure pres- that variants in the surgical technique of the LIFT proce-
ents as discharge from the intersphincteric incision and has dure showed better outcomes (63). But, the meta-analysis by
the anatomy of simple fistulas and could be managed with Hong et al. showed the LIFT success rate to increase from
fistulotomy or local wound care (64,65). 74.6% to 83.5% when all the combined LIFT procedures
Modifications to improve and reinforce LIFT repair fur- (LIFT, BIOLIFT, LIFT plus coring, and LIFT plus fistula
ther, by insertion of a bioprosthetic graft in the intersphinc- plug) are included in the analysis (69).
teric plane to act as a physical barrier, is called the BIOLIFT Currently, a randomized, multicenter prospective trial
procedure. The bioprosthetic graft ideally must overlap (NCT01478139) is in progress comparing Bio-LIFT with
the closure of the fistula tract by 1–2 cm in all directions, LIFT alone (70).
The advantages of ERAF include avoiding division of

STEM CELLS the sphincter with less risk for incontinence, maintaining
the anal contour, avoiding keyhole deformity, and promot-
Autologous expanded adipose tissue derived stem cells ing quicker healing. Additional procedures can be incor-
(eADSC) has been used in the management of fistula. These porated into the operation, like sphincteroplasty without
cells are obtained easily with liposuction and have been pro- the need for fecal diversion. Failure of the repair does not
posed in suppressing inflammation and promote healing usually lead to worsening of symptoms, as disruption of
(71). After identifying the internal opening and curettage the internal sphincter at the level of the anorectal junction
of the tract, the internal opening of the tract is closed, and is offset by more rigidity at the same level due to the scar
eADSC suspension is injected at the level of the intersphinc- tissue. Advancement flap techniques are not recommended
teric tract and into the walls of the tract. Early results from when associated with anorectal sepsis, active proctitis,
the use of eADSC have been modest at best, with 71.5% ini- malignant or radiation-related fistulas, stricture of the ano-
tial healing and 33% at 3 years (72–74). An autologous bone rectum, severe sphincter defect, or severe perianal scarring
marrow–derived mesenchymal stromal cells (MSC) solu- from previous fistula surgery (76,77). Smokers are associ-
tion has also been evaluated in a smaller trial with intracta- ated with lower healing rates (60%) compared to nonsmok-
ble Crohn perianal fistulas by injecting into the lumen and ers (79%) (78).
the walls of the fistula tract. It has shown promising results The healing rates after ERAF were reported to be between
with 70% fistula healing at 1 year (75). 60% and 93% with a recurrence rate of 19.7% over a median
follow-up period of 42 months (range 24–65 months)
(76,79,80). The median time to relapse was 5 months (range
1–11.7 months), with no recurrences seen after 1 year of the
ADVANCEMENT FLAP PROCEDURES repair. For lesions below the dentate line—fistulotomy or
dermal advancement flap was preferred to avoid the cre-
Advancement flap procedures are indicated in more complex ation of mucosal ectropion (81). Mitalas et al. reported the
fistulas that are not suitable for fistulotomy without compro- healing rate after first ERAF to be 67%, and patients with a
mising continence. They are also indicated in cases with rec- failed previous repair had a 69% healing rate after the second
tourethral or rectovaginal fistulas. The mechanics of the anal repair, thus resulting in an overall success rate of 90% (82).
fistula with higher pressure on the luminal side of the anus ERAF when combined with fibrin glue injection had worse
help with the technique of flap closure as it brings a layer of outcomes, while a combination of ERAF with platelet-rich
healthy tissue over the internal opening (76). plasma was shown to have better outcomes (83,84). Full-
The flaps are based from the rectum or perianal skin. The thickness rectal flaps were shown to have better recurrence
endorectal advancement flap is based from the rectum and rates compared to partial-thickness flaps without higher
the anocutaneous and transanal sleeve advancement flap incontinence rates in a retrospective review involving 54
and based from the perianal skin. patients (85). Smoking and obesity negatively impacted the
outcomes after ERAF (86).
ENDORECTAL ADVANCEMENT FLAP
ANO-CUTANEOUS ADVANCEMENT FLAP
Endorectal advancement flap (ERAF) is considered as the
gold standard sphincter preserving operation and usu- An ano-cutaneous advancement flap is an alternative to
ally performed as a second-stage procedure after a seton a rectal advancement flap. It includes a V-shaped incision
fistulotomy for high transsphincteric or suprasphincteric made in the perianal skin with the base of the incision
fistulas. The essential principles behind ERAF includes including the internal opening and the lateral edge includ-
debridement or excision of the fistula tract, and mobiliza- ing the external opening. The flap is raised by mobilizing
tion of a vascularized, tension-free mucosal flap to cover the along the underlying fat to allow advancement without ten-
area over the internal opening. After curetting the tract, the sion (Figure 20.8). The track is cored out, including both
defect in the internal anal sphincter is closed with absorb- internal and external openings. The flap is sutured to the
able sutures. The broad-based flap with adequate blood sup- anal mucosa above the internal opening with one side of the
ply is developed along the level of the submucosal plane and flap left open to ensure adequate drainage (76).
includes a few superficial fibers of the internal anal sphinc- The transanal sleeve advancement flap procedure takes
ter. The base of the flap should be approximately twice the the concept of flap advancement further by mobilizing the
width of its length to ensure absence of tension at the suture circumference of the anal canal and is useful in patients
line. Adequate hemostasis and a tension-free suture line with severe complex fistulas associated with Crohn dis-
are essential to ensure success of the flap repair. Although ease (87). It includes a near circumferential incision at or
a majority of them use a U-shaped flap, a semicircular flap below the dentate line to create a sleeve of full thickness
avoids ischemia at the corners. It is recommended to excise of the bowel wall, which was mobilized to allow the flap
a small rim of the anoderm below the internal opening to to be advanced into the supralevator space without ten-
create a neodentate line (Figure 20.7). sion. The distal suture line is kept below the level of the
Video-assisted anal fistula management 185
(a) (b)
(c) (d)
Figure 20.7 Endorectal advancement flap. (a) Transphincteric fistula-in-ano. (b) Enlargement of external opening. (c) Flap
of muscle and muscle is mobilized. (d) Flap is advanced, the distal tip is removed, and the flap is sutured in place. (From
Vasilevsky CA. Fistula-in-ano and abscess. In Beck DE, Wexner SD. (eds). Fundamentals of Anorectal Surgery, 2nd edition,
1998, Figure 10–15, p.176, with permission.)
FISTULOTOMY WITH SPHINCTER

RECONSTRUCTION
This is an effective management strategy when complex

anal fistulas are associated with incontinence or have higher
risk for incontinence, as those with a history of multiple or
complicated childbirths, those who present with preexist-
ing sphincter defect, or those women who have anterior fis-
tulas. The first German guideline for the treatment of anal
fistulas considered “fistula excision with reconstruction”
as a therapeutic option (92). These radical procedures are
associated with higher healing rates ranging from 83.3%
to 97.4% but are associated with higher incontinence rates
ranging between 3.7% and 21.4% (80,93–99). Although the
Figure 20.8 Anocutaneous flap. results from U.S. centers are reassuring, most available data
are from European centers (80).
internal opening with absorbable sutures. The advance-

ment flap procedures are safe and effective in around
VIDEO-ASSISTED ANAL FISTULA
70% patients with minimal or no disturbance of conti-
nence. Studies show the healing rates of the fistula range
MANAGEMENT
from 64% to 95% with a recurrence rate ranging from 0%
to 36% during a follow-up period ranging from 1 to 31 Video-assisted anal fistula treatment (VAAFT) uses a fis-
months (88–91). tuloscope manufactured by Karl Storz GmbH (Tuttlingen,
Germany), obturator, unipolar electrode, endobrush, and 13. Sözener U et al. Dis Colon Rectum 2011;54(8):923–9.
cyanoacrylate glue. The procedure involves diagnostic fis- 14. Liu CK et al. J Microbiol Immunol Infect 2011;44(3):
tuloscopy under irrigation, fulguration of the fistula tract, 204–8.
closing of the internal opening with a mucocutaneous flap, 15. Zinicola R, Cracco N. Colorectal Dis 2014;16(7):562.
or using a stapler followed by reinforcement of the repair 16. Browder LK et al. Tech Coloproctol 2009;13(4):301–6.
with cyanoacrylate glue. Cyanoacrylate glue reinforcement 17. Malik AI et al. Cochrane Database Syst Rev 2010;7:
has not been uniformly accepted as part of VAAFT as it CD006827.
could contribute to impaired drainage of the tract and delay 18. Ramanujam PS et al. Surg Gynecol Obstet 1983;157(5):
healing. Appropriate widening of the external opening is 419–22.
necessary with daily irrigation to ensure adequate drain- 19. Cariati A. Updates Surg 2013;65(3):201–5.
age of the debrided tract. VAAFT is also helpful to identify 20. Quah HM et al. Int J Colorectal Dis 2006;21(6):602–9.
the secondary tracts, which are one of the main reasons for 21. Cox SW et al. Am Surg 1997;63:686–9.
failure of anal fistula surgery. The early results of VAAFT 22. Chrabot CM et al. Dis Colon Rectum 1983;26:105–8.
repair are promising with primary healing in 73.5% with 23. Lunniss PJ, Phillips RK. Br J Surg 1994;81:368–9.
maintenance of healing in up to 87.1% of patients who 24. Davis BR, Kasten KR. Anorectal Abscess and Fistula.
showed healing over a short period (100–102). ASCRS Textbook of Colon and Rectal Surgery,
3rd Edition. New York, NY: Springer, 2016.
25. Garcia-Aguilar J et al. Dis Colon Rectum 1996;39(7):
723–9.
FILAC 26. Pescatori M et al. Colorectal Dis 2006;8(1):11–4.
27. Ho YH et al. Br J Surg 1998;85(1):105–7.
Fistula layer closure is a novel sphincter saving procedure 28. Jain BK et al. J Korean Soc Coloproctol 2012;28(2):
where the primary closure of the track is achieved using 78–82.
laser energy from a diode laser. After identifying the track 29. Williams JG et al. Br J Surg 1991;78:1159–61.
and curettage, a 14 Fr catheter was inserted into the fistula 30. Awad ML et al. Colorectal Dis 2009;11:524–6.
track using a guidewire. A 400 µm radial-emitting dispos- 31. Takesue Y et al. J Gastroenterol 2002;37(11):912–5.
able laser fiber was inserted through the 14 Fr catheter and 32. Seow-Choen F. Colorectal Dis 2003;5:373.
delivers laser energy (1470 nm wavelength) homogeneously 33. Gurer A et al. American J Surg 2007;193:794–6.
at 360° while the fiber is withdrawn from the internal open- 34. Subhas G et al. Dig Surg 2012;29(4):292–300.
ing of the fistula track outward by 1 mm/s. The radial pen- 35. García Olmo D et al. Br J Surg 1994;81:136–7.
etration depth of the energy has been shown to be 2–3 mm 36. Durgun V et al. Dig Surg 2002;19:56–8.
beyond the fistula track. Studies show a successful clo- 37. Lentner A, Wienert V. Dis Colon Rectum 1996;39:
sure rate of 71.4% over a 20-month follow-up period, but 1097–101.
reported a 22.9% incidence of significant postoperative pain 38. Sözener U et al. Dis Colon Rectum 2011;54(8):923–9.
and anismus. Oztürk and colleagues reported an 82% heal- 39. de Parades V et al. Colorectal Dis 2010;12(5):459–63.
ing rate at 12 months of follow-up after the procedure. The 40. Hjortrup A et al. Dis Colon Rectum 1991;34:752–4.
FiLaC procedure remains investigational; additional data 41. Patrlj L et al. Dig Surg 2000;17:77–80.
must be made available before further recommendations 42. Cintron JR et al. Dis Colon Rectum 2000;43:944–9.
can be made (80,103,104). discussion 949–50.
43. Lindsey I et al. Dis Colon Rectum 2002;45:1608–15.
REFERENCES 44. Zmora O et al. Dis Colon Rectum 2003;46:584–9.
45. Buchanan GN et al. Dis Colon Rectum 2003;46:
1. Abcarian H. Clin Colon Rectal Surg 2011;24(1):14–21. 1167–74.
2. Eglitis J. Ohio J Sci 1961;61(2):65–79. 46. Maralcan G et al. Surg Today 2006;36:166–70.
3. Seow-Choen F, Ho JM. Dis Colon Rectum 1994;37(12): 47. Johnson EK et al. Dis Colon Rectum 2006;49:371–6.
1215–8. 48. Grimaud JC et al. Gut 2006;55(Suppl. V):A40.
4. Shafik A. Invest Urol 1976;13:424. 49. Tyler KM et al. Dis Colon Rectum 2007;50:1535–9.
5. Milligan ETC et al. Lancet 1937;2:1119. 50. Singer M et al. Dis Colon Rectum 2005;48:799–808.
6. Courtney H. Am J Surg 1950;79:155. 51. Ommer A et al. Ger Med Sci 2012;10:Doc13.
7. Milligan ETC. Proc R Soc Med 1943;36:365. 52. Heydari A et al. Dis Colon Rectum 2013;56(6):774–9.
8. Haagensen CD et al. The Lymphatics in Cancer. 53. Buchberg B et al. Am Surg 2010;76(10):1150–3.
Philadelphia, PA: Saunders, 1972. 54. de la Portilla F et al. Dis Colon Rectum 2011;54(11):
9. Parks AG. Br Med J 1961;1:463. 1419–22.
10. Seow-Choen F, Ho JM. Dis Colon Rectum 1994;37: 55. Ratto C et al. Colorectal Dis 2012;14(5):e264–9.
1215–8. 56. Narang SK et al. Colorectal Dis 2016;18(1):37–44.
11. Gosselink MP et al. Colorectal Dis 2015;17(12):1041–3. 57. Limura E et al. World J Gastroenterol 2015;21:12–20.
12. Mitalas LE et al. Tech Coloproctol 2012;16:113–7. 58. Katz AR et al. Surg Gynecol Obstet 1985;161:213–22.
References 187
59. Stamos MJ et al. Dis Colon Rectum 2015;58:344–51. 82. Mitalas LE et al. Dis Colon Rectum 2007;50(10):
60. de la Portilla F. Colorectal Dis 2013;15:628–9. 1508–11.
61. Rojanasakul A et al. J Med Assoc Thai 2007;90(3):581–6. 83. Jacob TJ et al. Cochrane Database of Syst Rev 2010;
62. Rojanasakul A. Tech Coloproctol 2009;13(3):237–40. 5:CD006319.
63. Vergara-Fernandez O, Espino-Urbina LA. World J 84. Gottgens KW et al. Dis Colon Rectum 2014;57(2):
Gastroenterol 2013;19(40):6805–13. 223–7.
64. Tan KK et al. Dis Colon Rectum 2011;54(11):1368–72. 85. Dubsky PC et al. Dis Colon Rectum 2008;51(6):852–7.
65. van Onkelen RS et al. Colorectal Dis 2013;15(5):587–91. 86. Schwandner O. BMC Gastroenterol 2011;11:61.
66. Ellis CN. Dis Colon Rectum 2010;53:1361–4. 87. Marchesa P et al. Br J Surg 1998;85:1695–8.
67. Chew M-H et al. Int J Colorectal Dis 2013;28:1489–96. 88. Jun SH, Choi GS. Br J Surg 1999;86:490–2.
68. Tan K-K, Lee PJ. ANZ J Surg 2014;84:280–3. 89. Amin SN et al. Dis Colon Rectum 2003;46:540–3.
69. Hong KD et al. Tech Coloproctol 2014;18(8):685–91. 90. Del Pino A et al. Dis Colon Rectum 1996;39:224–6.
70. https://clinicaltrials.gov/ct2/show/NCT01478139. 91. Nelson RL et al. Dis Colon Rectum 2000;43:681–4.
71. Scoglio D et al. Clin Colon Rectal Surg 2014;27(4): 92. Ommer A et al. Dtsch Arztebl Int. 2011;108:707–13.
172–81. 93. Parkash S et al. Aust N Z J Surg 1985;55:23–7.
72. Mizuno H et al. Plast Reconstr Surg 2002;109(1): 94. Christiansen J, Rønholt C. Int J Colorectal Dis 1995;
199–209. discussion 210–1. 10:207–9.
73. Garcia-Olmo D et al. Dis Colon Rectum 2009;52(1): 95. Roig JV et al. Colorectal Dis 2010;12:145–52.
79–86. 96. Kraemer M, Picke D. Coloproctology 2011;33:104–8.
74. Guadalajara H et al. Int J Colorectal Dis 2012;27(5): 97. Arroyo A et al. Ann Surg 2012;255:935–9.
595–600. 98. Perez F et al. Am J Surg 2006;192:34–40.
75. Ciccocioppo R et al. Gut 2011;60(6):788–98. 99. Ratto C et al. Dis Colon Rectum 2013;56:226–33.
76. Williams JG et al. Colorectal Dis 2007;9(Suppl 4):18–50. 100. Meinero P, Mori L. Tech Coloproctol 2011;15:417–22.
77. Ozuner G et al. Dis Colon Rectum 1996;39:10–4. 101. Wałęga P et al. Pol Przegl Chir 2014;86:7–10.
78. Zimmerman DD et al. Br J Surg 2003;90:351–4. 102. Seow-En I et al. Tech Coloproctol 2016;20(6):389–93.
79. Abbas MA et al. Am Surgeon 2008;74:921–4. 103. Giamundo P et al. Colorectal Dis 2014;16:110–5.
80. Santoro GA, Abbas MA. Complex Anorectal Fistulas. 104. Oztürk E, Gülcü B. Dis Colon Rectum 2014;57:360–4.
ASCRS Textbook of Colon and Rectal Surgery, 3rd 105. Vogel JD, Vasilevsky CA. Anorectal Abscess and
Edition. Cham, Heidelerg, New York, Dordrecht, Fistula-in-ano. Beck DE, Wexner SD, Steele SR. (eds).
London: Springer, 2016. Fundamentals of Anorectal Surgery, 3rd edition. NY:
81. Christante DH, Thorsen AJ. The Management of Springer, 2018.
Anorectal Abscess and Fistula. Current Surgical 106. Vasilevsky CA. Fistula-in-ano and abscess. In Beck
Therapy, 12th Edition. Philadelphia, PA: Elsevier DE, Wexner SD. (eds). Fundamentals of Anorectal
Health Sciences, 2016. Surgery, 2nd edition. 1998. Figure 10–15, p.176.
21
Surgery and nonoperative therapy
of anal fissure
MARY T. O’DONNELL AND CARY B. AARONS
CHALLENGING CASE INTRODUCTION
A 28-year-old female presents to your clinic with rec- Anal fissure is a common condition in which the epithe-
tal bleeding and extreme anal pain with bowel move- lium is torn at the anal verge or distal anal canal. While true
ments that lasts for hours afterward. These symptoms population incidence is unknown, estimates suggest 10% of
have been present for 6 months since the birth of referrals or up to 250 visits per year of a colorectal surgery
her first child. She reports intermittent constipation, clinic visits are for anal fissures (2). Patients present with
and the pain is partially relieved with hydrocortisone symptoms of mild bleeding and sharp, tearing pain with
creams and sitz baths. During physical exam, gentle defecation that can last for hours afterward. Precipitating
exposure of the anal verge reveals an anteriorly events include constipation and passage of hard stool that
located fissure. Given the chronicity of her symptoms, causes trauma to the anal canal, though diarrhea has also
she desires surgery. What is your next step? been reported prior to symptoms. A history consistent with
an anal fissure and visualization on external exam are usu-
ally sufficient for diagnosis. Spreading the buttocks with
CASE MANAGEMENT opposing traction of the thumbs at the anal verge usually
reveals the fissure defect. Digital rectal exam and anos-
Despite the duration of symptoms and the patient’s copy are often not well tolerated given patient discomfort;
request, treatment in this young postpartum female however, they should be performed if there is concern
should begin with fiber supplementation, increased for the accuracy of the diagnosis. In these cases, an exam
fluid intake, and lifestyle modifications. In a young under anesthesia with possible biopsy may aid in diagnosis
female who may experience sphincter, perineal, or of abnormally located fissures or other anorectal etiologies
pudendal nerve trauma from childbirth in the future, it for pain.
is especially important to pursue conservative measures
as a first-line treatment due to their proven efficacy
without impairment in fecal continence. Specifically, CLASSIFICATION
postpartum fissures are unique in that they are associ-
ated with decreased pressures in the sphincter com-
plex, rather than hypertonia associated with the typical Anal fissures are classified by chronicity and etiology. A fis-
anal fissure. Treatment should not be focused on sure that has been present for more than 4 weeks can begin
medication or surgery to relax the sphincter, but rather to demonstrate secondary findings of chronic inflamma-
aid in healing the wound. Greater than 90% of these tion, including a sentinel pile (external skin tag) and/or a
will heal with topical anesthetic and improved bowel hypertrophied anal papilla proximal to the fissure (Figure
habits, with the remaining patients needing procedures 21.1). In some cases, the fibers of the internal sphincter
such as anorectal advancement flaps (1). If surgical are visible through the fissure opening. Once a fissure has
intervention is planned, preoperative anal manometry been present for over 6 weeks, usually having been refrac-
or ultrasound should be considered. tory to conservative lifestyle modifications, it is considered
chronic. The etiologies of anal fissures are vast. While the
188
Nonoperative management / Nitrates 189
Posterior
1. Hypertrophied
anal papilla Acute and
chronic anal fissure
2. Fissure with Crohn’s disease

Crohn’s disease Ulcerative colitis
heaped up Ulcerative colitis
scarred edges Syphilis
Syphilis Tuberculosis
Tuberculosis Leukemia
Leukemia Cancer HIV
Cancer HIV Acute
3. Sentinel skin tag
and
chronic
Figure 21.1 Chronic anal fissure. (Reprinted from the anal fissure
American Society of Colon and Rectal Surgeons website.
Image provided by Robin Noel, Graphic Artist: University
of Pennsylvania.)
Anterior
typical acute anal fissure in a healthy person can be caused Figure 21.2 Location and etiologies of anal fissure at the
by the local trauma of a hard stool, fissures can be associated anal canal.
with underlying inflammatory bowel disease, cancer, sexu-
ally transmitted infections, and even childbirth.
The location or multiplicity of an anal fissure can help trial, fissure recurrence was 16%, 60%, and 68% in patients
determine both the etiology and chronicity of a fissure. Most receiving 15 g unprocessed bran, 7.5 g unprocessed bran,
commonly, the acute tear leads to anal pain that precipitates and placebo (4). Studies have shown that half of acute fis-
patient aversion to defecation and increased sphincter tone. sures resolve with sitz baths, psyllium, topical anesthetics,
The problem is exacerbated by the hypertonic sphincter at or anti-inflammatory ointments, making this the current
rest, which in turn causes decreased perfusion of the ano- recommendation in the American Society of Colon and
derm. The posterior midline, which has the lowest perfusion Rectal Surgeons (ASCRS) Clinical Practice Guideline for
of all quadrants of the anoderm, is predisposed to mild isch- initial management of acute fissure (5). Unfortunately, up
emia with resultant poor healing, especially in the setting to 50% of these fissures will recur within 5 years, and a
of increased anal pressure (3). For this reason, up to 90% 2012 Cochrane Review suggests no difference between lido-
of fissures can be found at the posterior midline. Fissures caine, bran, and hydrocortisone and placebo (6). There are
have been reported anteriorly in up to 25% of women and few, if any, side effects to these interventions, and therefore,
8% of men, with 3% being found simultaneously in ante- little risk exists in recommending these simple therapies to
rior and posterior positions. Atypical presentations, such patients as initial treatment for acute anal fissure. If diet and
as lateral, painless, or multiple fissures, should raise clini- lifestyle modifications have failed for a period of 6–8 weeks,
cal suspicion for Crohn disease, ulcerative colitis, leuke- the fissure is classified as chronic and may only improve
mia, trauma, tuberculosis, sexually transmitted diseases, with topical medications or surgery.
HIV/AIDS, or anal carcinoma (Figure 21.2). Women with
reduced anal canal pressures more often demonstrate ante- NITRATES
riorly located fissures, which should prompt consideration
for anal manometry or endorectal ultrasound prior to sur- The smooth muscle tone of the internal anal sphincter is
gical interventions. partially affected by the nonadrenergic, noncholinergic
neurotransmitter nitric oxide, which is released with the
application of nitroglycerin cream. A compounded diluted
mixture of 0.2%–0.4% glyceryl trinitrate (GTN) or nitro-
NONOPERATIVE MANAGEMENT glycerin cream is applied to the anus two to three times daily
for 4–8 weeks to relax the internal sphincter and allow the
DIET AND LIFESTYLE MODIFICATION fissure to heal. Glyceryl trinitrate has been associated with
an overall healing rate of 48.9% and an overall improve-
Once the diagnosis of an anal fissure is confirmed, the ment in healing rate of 13.5% when compared to placebo
initial and most conservative steps in management are to or lidocaine alone in patients with chronic anal fissures (6).
improve bowel consistency and control symptoms. Fiber GTN applications in different dosages (from 0.05% to 0.4%)
supplementation with adequate water intake improves as well as methods of administration (intraanal injection
consistency and bulk of stool leading to less straining and or distant dermal patch application) have been investigated
trauma to the area. In a double-blind placebo-controlled with similar results.
190 Surgery and nonoperative therapy of anal fissure
When compared with lateral internal sphincterotomy that botulinum toxin was slightly superior to placebo. In a
(LIS), nitrates generally demonstrate significantly lower randomized controlled trial of lateral internal sphincterot-
healing rates after 6–8 weeks of therapy. Since the previ- omy versus botulinum injection, LIS demonstrated superior
ous version of this volume, which referenced six studies outcomes (17). This study also demonstrated the concerning
demonstrating large differences in healing rates between finding of recurrence of 50% of fissures that had previously
LIS and nitrate therapy, only one additional study directly healed 1 year after botulinum treatment. The paralytic effect
comparing the two therapies has been reported. These are of the toxin subsides after 3–5 months, which may contrib-
summarized in Table 21.1. There are a number of other ute to fissure recurrence rates. The temporary effect may be
studies that compare these treatments, but their data used to the clinician’s advantage since any side effects will
groups involve crossover treatments and are not discussed also be temporary. Because healing rates of up to 43% have
in this chapter. While side effects of nitrates are minimal, been demonstrated (18) and some equivalence in topical
headaches are reported in up to 30% of patients using medications has been suggested, it can be recommended for
nitrates, which often results in discontinuation of therapy use in the treatment of chronic anal fissure. Due to the tem-
and poor compliance. Therefore, the available data sup- porary and few side effects, clinicians may prefer botulinum
port that topical nitrates can be used to treat anal fissures toxin to sphincterotomy in patients over 50 years old due to
with low risk, but with the expectation of lowered success the risk of incontinence.
rates in more chronic fissures and an associated risk of
headaches.
SURGICAL MANAGEMENT
CALCIUM-CHANNEL BLOCKERS
Calcium-channel blockers, which include diltiazem and ANAL DILATION
nifedipine, work similarly to relax the smooth muscle of the
Dilation of the anal sphincters was thought to decrease the
hypertonic internal sphincter and allow a fissure to heal.
maximum resting anal pressure and therefore aid in healing
These medications can be used orally or topically with simi-
of fissure due to hypertonic anal sphincters. Historically,
lar success. Topical diltiazem 2% must be prescribed from a
manual digital dilation or graduated serial dilations with
compounding pharmacy and applied three times daily over
anal dilators were often used, but complications of inconti-
8 weeks. Numerous studies have demonstrated the success
nence from diffuse sphincter damage have largely led to the
of calcium channel blockers, though the heterogeneity of
abandonment of this method as it has been shown to have
dosing, small study number, and short follow-up time have
worse healing rates with higher rates of incontinence (5).
limited the applicability of their conclusions. Healing of fis-
However, balloon dilation was evaluated in comparison to
sures with diltiazem appears to be similar to that of GTN
LIS in a prospective randomized trial and found to demon-
(over 85%) when treated over 8 weeks with good follow-
strate similar fissure healing rates (19). After 2 years, there
up (15). Unfortunately, recurrence of these fi ssures within
was less incontinence in the balloon dilation group than in
6 months is common. Calcium-channel blockers have dem-
the LIS group (0% versus 16%). Therefore, balloon dilation
onstrated equivalent healing rates to topical nitrates, but
by a physician, not manual dilation, can be considered in
with fewer side effects, such as headaches. Therefore, these
the treatment of fissure; however, more research must be
medications are recommended more commonly as first-line
done before it could become a recommended treatment
medical therapy than nitrates. However, like other medical
option for chronic anal fissures.
therapies with low-risk profiles, their long-term success rates
are lower than sphincterotomy. One study demonstrated a
20% difference in anal fissure healing rates between 0.3% LATERAL INTERNAL SPHINCTEROTOMY
nifedipine plus lidocaine ointment with gentle dilation and
The surgical procedure that has shown the most efficacy
LIS (16).
in the resolution of a chronic anal fissure is the LIS. This
involves the partial division of the internal sphincter
BOTULINUM TOXIN to counteract the hypertonicity that is the source of the
typical anal fissure. Not surprisingly, incontinence of gas
The exotoxin of the bacterium Clostridium botulinum can be or liquid stool is an associated side effect in up to 10% of
injected locally to cause sympathetic blockade of the inter- patients (20). The sphincterotomy is best made laterally on
nal anal sphincter. Temporary paralysis of the internal anal the internal sphincter so as to avoid risk of incontinence
sphincter can allow an acute anal fissure to heal. Botulinum in women with thinned anterior sphincter or the dreaded
injection of 20 units bilaterally has demonstrated superior “keyhole” deformity posteriorly caused by division of the
effectiveness to nitroglycerin ointment. Healing rates of internal sphincter through the posterior fissure. It has been
anal fissures with botulinum injection are between 37% and shown in a number of studies that the extent of the sphinc-
43%, though the published data demonstrate a variety of terotomy need not be the entire length of the muscle to the
botulinum doses in use. A Cochrane Review (6) suggested dentate line. No significant difference in healing rates was
Table 21.1 Randomized controlled trials of nitrates versus surgical sphincterotomy for the treatment of chronic anal fissure
Number Treatment
of Treatment groups (% length Overall side HA IC flatus Follow-up Recurrence
Author/year patients ointment) (weeks) Fissure healing (%) effects (%) (%) (%) (months) (%)
Oettle (7) 24 NTG/LIS TID 4 83 versus 100 (p = NS) NR NR NR 1 NR
Richard (8) 82 0.25/0.5 GTN/LIS TID 6 30 versus 90 (p = 0.0) 84 versus 29 (p < 0.001) 21 None 6 38 versus 3
Evans (9) 60 0.2 GTN/LIS TID 8 61 versus 97 (p < 0.001) NR 33 7.4 5 45 versus 4
Libertiny (10) 70 0.2 GTN/LIS TID 8 54 versus 100 (p < 0.02) NR 20 2.9 24 16 versus 2.9
Parellada (11) 54 0.2 IDN/LIS TID 6 67 versus 96 (p < 0.001) 30 versus 44 (p = NR) NR 44@5 wk 24 13 versus 0
15@24 wk
Mishra (12) 40 0.2 GTN/LIS BID 6 90 versus 85 (p = 0.347) 40 versus 70 (p = NR) 15 15 4 NR
Brown (13) 51 0.25/0.5 NTG/LIS TID 6 59 versus 100 (p < 0.01) NR NR 67 79.4 37 versus 25%
Source: Adapted from Whitlow CB et al. Improved Outcomes in Colon and Rectal Surgery. Boca Raton, FL: CRC Press, 2009, p. 203 (14).
Note: BID, twice daily; GTN, glyceryl trinitrate; HA, headache; IC, incontinence; IDN, isosorbide dinitrate; LIS, lateral internal sphincterotomy; NR, not reported; NS, not significant; NTG,
nitroglycerin; TID, three times a day.
Surgical management / Lateral internal sphincterotomy 191
found between patients who underwent sphincterotomy ADVANCEMENT FLAPS

the length of dentate line versus sphincterotomy to the apex
of the fissure (21). Current recommendations have moder- In traditional fissures associated with hypertonicity, no
ate-quality evidence to demonstrate that this tailored LIS significant difference has been found between healing
yields equivalent healing rates with traditional LIS to the rates in patients who underwent advancement flaps versus
dentate line, and further investigation is needed. In con- sphincterotomy. These flaps can be combined with LIS or
clusion, traditional LIS remains the standard of care until botulinum toxin injection for decreased postoperative pain
more data are obtained. and good primary fissure healing. Combining anocutane-
The sphincterotomy procedure can be performed as ous flap with injection of botulinum toxin has shown heal-
a “closed” or “open” technique, and numerous studies ing rates from 86.7% to 92% at 2 years with minimal fecal
have shown equivalency of these methods. In the operat- incontinence (28,29). Similarly, when combined with LIS,
ing room, an exam under anesthesia is performed and the advancement flaps can lead to less postoperative pain, faster
sphincter complex digitally palpated for the intersphinc- healing, and fewer soiling episodes (30).
teric groove. In the “open” lateral internal sphincterotomy, Anocutaneous flaps are considered a safe alternative to
an anal retractor of choice (Pratt bivalve or Hill-Ferguson LIS due to good fissure healing rates (81%–100%) and lower
of appropriate size) is used to place the sphincter complex rates of minor fecal incontinence (0%–6%) (31–33). If the
on stretch, and a radial incision is made over the inter- underlying fissure pathology is hypotonicity rather than
sphincteric groove. The internal sphincter is then elevated hypertonicity, an advancement flap may be ideal. Tissue
under direct visualization and divided. The anoderm is then healing can be aided by advancement flap without the need
closed with an absorbable suture or left open. for a procedure to decrease anal tone.
The “closed” LIS is performed by similarly palpating the Different techniques have been described for anorectal
sphincter anatomy, inserting an 11-blade knife through the advancement flaps, all of which are created with the prin-
anoderm laterally in the intersphincteric groove. The blade ciple of transferring healthy tissue coverage for the chronic
is then rotated 90°, and the internal sphincter is then cut fissure wound. The simplest of these is the Y-V anoplasty.
medially toward a finger in the anus. A Y-shaped incision is made to debride the fissure bed, and
Success rates of LIS have been consistently described near then the V flap is advanced into the anal canal and sutured in
90% in multiple studies and Cochrane reviews and is the place. The wide base of the V allowed for a well-vascularized
treatment of choice for chronic or medically refractory anal flap (Figure 21.3). Alternatively, the V-Y anoplasty involves
fissures (22–26). The decision to proceed with LIS should be a triangular incision with the apex on the perianal skin and
made judiciously in patients at high risk for incontinence, the base made just past the fissure. After fissurectomy, the
such as women, people older than 50 years, those with a V is mobilized from the subcutaneous tissue and advanced
history of previous anal surgery, or patients with sphincter into the anal canal with closure of the skin behind the V
injury after childbirth. aiding in advancement of the flap (Figure 21.4). Less com-
Last, recurrent anal fissure after LIS can be safely treated mon techniques include the “house” advancement pedicle
with repeat LIS on the contralateral side. This has been flap (Figure 21.5) and the diamond flap (Figure 21.6), which
evaluated in a study of 55 people with 98% healing rate can be used if more circumferential coverage is needed
demonstrated at over a year after surgery (27). along the anal canal. The editors prefer the House flap.
(a) (b) (c)
Figure 21.3 Y-V Anoplasty. (Reprinted with permission from Whitlow CB et al. Improved Outcomes in Colon and Rectal
Surgery. Boca Raton, FL: CRC Press, 2009, pp. 199–214.)
Special situations / Crohn disease 193
(a) (b)
Figure 21.4 V-Y Anoplasty. (Reprinted with permission from Whitlow CB et al. Improved Outcomes in Colon and Rectal
(a) (b) (c)
Figure 21.5 House advancement flap. (Reprinted with permission from Whitlow CB et al. Improved Outcomes in Colon
and Rectal Surgery. Boca Raton, FL: CRC Press, 2009, pp. 199–214.)
(a) (b) (c)
Figure 21.6 Diamond flap. (Reprinted with permission from Whitlow CB et al. Improved Outcomes in Colon and Rectal
patients are usually a manifestation of active inflammation

SPECIAL SITUATIONS and should be suspected if fissures are atypical in location
or appearance, or are multiple in number. Fissures in Crohn
CROHN DISEASE disease tend to be deeper in nature and can be associated
with fistulas, which if left unmanaged can lead to significant
Anorectal surgery for anal fissures should be avoided in sphincter deformity. Therefore, efforts should be directed
patients with active Crohn disease in favor of medical toward continence preservation with medical management
management of the underlying disease. Fissures in these of the underlying Crohn disease.
SEXUALLY TRANSMITTED DISEASES has consistently demonstrated high rates of healing and is
considered the treatment of choice in patients with chronic
Sexually transmitted diseases such as syphilis and HIV have anal fissure. Last, repeat contralateral LIS can be performed
a known association with anal fissures. HIV-related anorec- in patients with recurrent anal fissure who continue to dem-
tal disease can manifest as ulcers or fissures in atypical loca- onstrate hypertonic sphincter.
tions. These fissures are unique in that instead of the typical
sharp pain on defecation, patients often have painless or per-
sistent “gnawing” pain (34). These are more often associated REFERENCES
with poor sphincter tone. Examination under anesthesia 1. Corby H et al. Br J Surg. 1997;84:86–8.
with viral culture or biopsy should be performed to confirm 2. Beck DE et al. The ASCRS Textbook of Colon and
acute infections and to rule out similarly presenting pathol- Rectal Surgery. New York: Springer, 2011.
ogy such as HIV-related anal ulcer. Appropriate antibiotics 3. Schouten WR et al. Dis Colon Rectum. 1994;37:664–9.
should be used to treat syphilis or other sexually transmit- 4. Jensen SL. Br Med J (Clin Res Ed). 1986;292:1167–9.
ted diseases. HIV-related fissure treatment should be aimed 5. Stewart DB et al. Dis Colon Rectum. 2017;60:7–14.
at optimizing treatment with antiretroviral therapy. 6. Nelson RL et al. Cochrane Database Syst Rev. 2012;
Issue 2: Art. no. CD003431. DOI: 10.1002/14651858.
FISSURES WITHOUT HYPERTONICITY CD003431.pub3.
7. Oettlé GJ. Dis Colon Rectum. 1997;40:1318–20.
Acute low-pressure anal fissures are most commonly seen
8. Richard CS et al. Dis Colon Rectum. 2000;43:1048–57;
in postpartum patients, and these should be managed con-
discussion 1057.
servatively with medications and lifestyle modification.
9. Evans J et al. Dis Colon Rectum. 2001;44:93–7.
Because fissures in the presence of sphincter hypotonicity
10. Libertiny G et al. Eur J Surg. 2002;168:418–21.
are not improved by sphincter relaxants or sphincterotomy,
11. Parellada C. Dis Colon Rectum. 2004;47:437–43.
treatment measures should be directed toward healing of
12. Mishra R et al. ANZ J Surg. 2005;75:1032–5.
the fissure wound.
13. Brown CJ et al. Dis Colon Rectum. 2007;50:442–8.
Chronic fissures in postpartum patients can be managed
14. Whitlow CB et al. Improved Outcomes in Colon and
with fissurectomy and advancement flap with near perfect
Rectal Surgery. Boca Raton, FL: CRC Press, 2009.
results (35). Similarly, other fissures in the setting of hypo-
15. Bielecki K, Kolodziejczak M. Colorectal Dis. 2003;
tonicity should be considered for anorectal advancement
5:256–7.
flap as therapy in the chronic setting.
16. de Rosa M et al. Updates Surg. 2013;65:197–200.
17. Arroyo A et al. Am J Surg. 2005;189:429–34.
18. Samim M et al. Ann Surg. 2012;255:18–22.
19. Renzi A et al. Dis Colon Rectum. 2008;51:121–7.
CONCLUSION 20. Nelson R. Cochrane Database Syst Rev. 2005;
Issue 2: Art. no. CD002199. DOI: 10.1002/14651858.
Acute anal fissure in the otherwise healthy individual with CD002199.pub2.
resultant hypertonic sphincter can be treated with lifestyle 21. Elsebae MM. World J Surg. 2007;31:2052–7.
modifications, topical relaxants, and close clinical follow- 22. Saad AM, Omer A. East Afr Med J. 1992;69:613–5.
up to determine resolution of symptoms. Chronic anal 23. Olsen J et al. Int J Colorectal Dis. 1987;2:155–7.
fissures in the setting of a hypertonic sphincter, includ- 24. Weaver RM et al. Dis Colon Rectum. 1987;30:420–3.
ing anal fissure refractory to medical management, should 25. Ram E et al. Tech Coloproctol. 2007. https://doi.org/
be considered for further therapies, including botulinum 10.1007/s10151-007-0373-7
injection, lateral internal sphincterotomy, or combination 26. Nelson RL et al. Cochrane Database Syst Rev.
of botulinum injection/LIS with anorectal advancement 2011; Issue 11: Art. No.: CD002199. DOI: 10.1002/
flap. Fissures due to inflammatory bowel disease, sexually 14651858.CD002199.pub4.
transmitted diseases, or HIV should be managed by con- 27. Leong AF, Seow-Choen F. Dis Colon Rectum. 1995;
firming diagnosis with examination under anesthesia and 38:69–71.
biopsy or culture, and medical management of the underly- 28. Halahakoon VC, Pitt JP. Int J Colorectal Dis 2014;
ing disease process. In patients with anal fissure with sus- 29(9):1175–7.
pected hypotonic sphincter mechanism or who are at high 29. Patti R et al. Updates Surg. 2012;64:101–6.
risk for incontinence (multiparous female or previous ano- 30. Theodoropoulos GE et al. Am Surg. 2015;81:133–42.
rectal surgery), anal manometry or endorectal ultrasound 31. Gupta P. ANZ J Surg. 2006;76:718–21.
should be considered. Patients with chronic anal fissure 32. Kennedy ML et al. Dis Colon Rectum. 1999;42:1000–6.
with reduced resting anal pressures should be conserva- 33. Giordano P et al. World J Surg. 2009;33:1058–63.
tively managed, then considered for fissurectomy and ano- 34. Viamonte M et al. Dis Colon Rectum. 1993;36:801–5.
rectal advancement flap. Lateral internal sphincterotomy 35. Patti R et al. Colorectal Dis. 2010;12:1127–30.
22
Surgery for pilonidal disease and
hidradenitis suppurativa
JOHN D. HUNTER AND LEANDER M. GRIMM JR.
the gluteal cleft from riding long distances over rough terrain
PILONIDAL DISEASE in jeeps under hot and sweaty conditions (8).
INTRODUCTION ETIOLOGY
Pilonidal disease is a chronic inflammatory condition of Pilonidal disease was initially thought to be congenital in
the skin and associated hair follicles typically found in the origin. Failed involution of neural tube structures left rem-
sacrococcygeal region. Clinically, pilonidal disease can nants that eventually became pilonidal cysts and sinuses (9).
present on a wide spectrum, including asymptomatic inci- In 1946 Patey challenged this theory, suggesting that pilo-
dentally found pits, acute or recurrent abscess, chronically nidal disease was acquired from hair piercing the skin of
draining sinus, or a large nonhealing wound from a previ- the gluteal cleft, leading to local inflammation and eventual
ous attempted repair. The environment of the natal cleft can cysts and sinuses. Karydakis studied over 6,000 cases and
be moist, hirsute, unhygienic, and subject to a wide range of surmised three main factors leading to pilonidal disease: (1)
forces throughout the day. These characteristics can lead to a an “invader,” that is, loose hair applying a (2) force causing
unique challenge for the surgeon treating pilonidal disease. hair insertion, composed of factors including depth, nar-
The true incidence of pilonidal disease is unclear. It has rowness, and friction of the gluteal cleft, and (3) vulnerabil-
been reported anywhere from 26 per 100,000 to 4.6% (1,2). ity of the skin and soft tissues. Loose hairs from the head,
Typical age at presentation is in the late teens to early 30s. back, and natal cleft accumulate in the natal cleft. Friction
It is rarely found in patients older than 45 years. Classically from walking drives the hairs into the skin creating a for-
it is found in young adult Caucasian males. However, recent eign-body reaction. This eventually creates a “pit” opening
military data reported a similar incidence rate among males through which more hairs are pulled and eventually form
and females (1.9 and 1.7 per 1,000 p-years) (3). a chronic cavity (10). Bascom also suggests that native hair
follicles of the natal cleft become infected with keratin,
BACKGROUND which leads to infection and abscess formation. Friction
forces then “suck” hair into the cavity, further seeding the
The first description of pilonidal disease is credited to Herbert inflammatory cycle (11). (Figure 22.1) The acquired theory
Mayo in 1833. In his chapter on fistula disease, he describes a is further supported by reports of pilonidal disease in the
young woman with a sinus opening an inch behind the anus interdigital regions of the hands of barbers, pet groomers,
and “upon examining it with a probe, I found that instead sheep shearers, and cow milkers (12–15).
of running towards the rectum, it extended upwards for a Risk factors for the development of pilonidal disease
length of five inches between the skin and os sacrum” (4). include hirsutism, deep natal cleft, obesity, poor hygiene,
In 1844, Anderson gave a description of hair extracted from prolonged sitting, excessive sweating, and family history.
an ulcer (5). Hodges later coined the term “pilonidal” in A prospective case control study by Harlak in 2010 stud-
1880, derived from the Latin words for “hair” and “nest” (6). ied 587 patients with pilonidal disease and compared them
Pilonidal disease became much more prevalent in World to 2,780 healthy controls. The four most predictive factors
War II, during which 80,000 soldiers were hospitalized with were (from strongest to weakest) body hair stiffness, num-
the disease for an average of 55 days each (7). Buie coined the ber of baths per week, hours spent seated per day, and body
term “jeep disease,” attributing pilonidal disease to trauma to mass index. They showed that hirsute individuals who sit
195
196 Surgery for pilonidal disease and hidradenitis suppurativa
Figure 22.1 Pilonidal cyst opened after excision showing

hair inside cavity.
for >6 hours per day and bathe two or less times per week
have a 219-fold increased risk of pilonidal disease. This fits
with the higher incidence of the disease seen in military
populations (16).
Figure 22.2 Acute pilonidal abscess. Note midline
PRESENTATION AND DIAGNOSIS openin g with abscess slightly to the right of midline.
With the wide spectrum of pilonidal disease, the surgeon
can see a variety of clinical presentations. Some patients are
referred for an incidental finding of midline pits and are
completely asymptomatic. Others may have had an acute
abscess already drained and are referred for further man-
agement. Then there are patients with persistent disease
that can range from minimally symptomatic to debilitating
wounds that have failed multiple operations.
The diagnosis of pilonidal disease is made by history and
physical. Patients often complain of a vague pain over the
sacrum that is often associated with intermittent clear or
bloody drainage. The drainage is sometimes confused with
rectal bleeding. An acute abscess may have accompanying
fevers or purulent drainage. (Figure 22.2) Classic physi-
cal examination findings are one or more midline “pits”
3–5 cm posterior to the anus (Figure 22.3). These are often
associated with an acute abscess or chronic draining sinus
off midline. Probing the sinus with a fine clamp will reveal
numerous tufts of hair. Differential diagnosis includes
hidradenitis, anorectal abscess, and anorectal fistula, and
digital rectal examination and anoscopy (or proctoscopy)
should be performed to exclude these diagnoses. Figure 22.3 Chronic pilonidal disease showing
midline pit.
MANAGEMENT OF ACUTE DISEASE
decubitus or prone positioning allows proper exposure. The
An acute abscess secondary to pilonidal disease is managed area of fluctuance is typically 1–2 cm off midline; an ellipti-
like any other abscess. Incision and drainage can be car- cal incision is made over this area, and the cavity is debrided
ried out at the bedside under local anesthetic or in the oper- of all purulence and hair. Thorough irrigation, hemosta-
ating room under general or regional anesthesia. Lateral sis, and packing complete the procedure. Postoperative
Pilonidal disease / Nonoperative and minimally invasive management 197
antibiotic use is typically not needed unless significant cel- pilonidal protocol over a period of 17 years to the popu-
lulitis is encountered. lation of Tripler Army Medical Center in Hawaii. During
The importance of the off-midline incision was reported the 17-year study period of 150 pilonidal admissions, there
in a 2011 study, where 96 patients with acute presentation of were only 23 excisional procedures performed. Admissions
pilonidal abscess were treated with midline versus off-mid- were reduced by 78% between the first 4 years and last
line incisions. This retrospective review found the patients 4 years of the study (19).
receiving an off-midline incision healed about 3 weeks Laser depilation has been used with some success in the
faster (17). treatment of recurrent pilonidal disease. It offers the advan-
Incision and drainage can at times be definitive. In a tage of longer-term hair control. However, the hair loss is
series of 73 patients who underwent incision and drainage not permanent, and treatments can be painful and often
for acute abscess in the setting of first-time pilonidal pre- require local anesthetic. Odili et al. reported 14 patients
sentation, 58% had healed wounds at 10 weeks. Of those with recurrent pilonidal wounds after multiple surgical
patients, 21% had recurrence during the mean follow-up of procedures (average three) who underwent laser depilation.
60 months. Those patients with fewer pits and tracts had a Eight patients were healed by 1 year and 10 by the second
better chance of complete cure (18). year. The remaining four healed after a second treatment.
They postulated that without hair present, secondary heal-
MANAGEMENT OF CHRONIC DISEASE ing proceeded rapidly (20).
Razor and laser depilation have also been shown to have
In the setting of chronic disease, it is crucial for the sur- adverse effects on the recurrence rate of pilonidal disease.
geon to tailor treatment to the individual patient. Patient Peterson and colleagues reviewed 504 patients who were
expectations should be managed early and often, stressing treated surgically, 113 of whom shaved the area postopera-
the possibility of extended periods of wound care that may tively. They observed a 30.1% recurrence rate in those who
be borne out of failed smaller procedures or more aggressive shaved versus 19.7% in those who did not (21). Another
procedures. study looked at laser depilation prior to undergoing flap
reconstruction for pilonidal disease. They found a 20%
NONOPERATIVE AND MINIMALLY recurrence rate in laser depilation group versus 4% in the
INVASIVE MANAGEMENT control group (22).
Patients who present with asymptomatic or minimally Fibrin

symptomatic disease can be safely observed. Nonoperative
management should focus on good hygiene with atten- Fibrin glue injections have been used to treat chronically
tion directed toward keeping the natal cleft clean and dry. draining pilonidal sinuses. The idea is that forming a fibrin
Patients whose job requires them to sit for prolonged peri- clot within the sinus would obliterate dead space and pro-
ods should be encouraged to briefly stand and walk around mote healing through clotting cascade signaling. A similar
multiple times a day. technique has been applied to anal fistulas with variable
success. Lund and Leveson treated six patients with chronic
Depilation draining pilonidal sinuses with fibrin glue. Half were recur-
rent patients, while the others were primary presentations.
Given that hair is the inciting factor in pilonidal disease, Under general anesthesia, they curetted the pits and then
removing hair from the natal cleft would intuitively lessen injected 2 mL of fibrin. They reported minimal postopera-
the progression and potentially provide an environment to tive discomfort, 2–3 day return to work, and one recurrence
heal some pilonidal disease. Weekly shaving can improve 4 months after treatment. There were no other recurrences
the environment of the natal cleft and aid in healing. at 18 months (23). In a similar study in 2014, Isik et al.
A pilot study in the mid-1970s compared conservative reported on 40 patients treated with fibrin glue injection.
management of pilonidal disease in 101 soldiers at a single With a patient selection that excluded more than one sinus
Army training hospital to the previous 2 years of exci- tract, no previous sinus infection, no previous treatment,
sional therapy. They applied conservative measures such and no history of recurrence, they achieved a 10% recur-
as hygiene education, meticulous hair shaving, simple lat- rence rate at 1 year (24).
eral incision and drainage for acute abscess, and avoidance A variation on direct injection, simple excision fol-
of certain exercises that put stress on the sacrococcygeal lowed by fibrin injection was performed on 25 patients by
region (sit-ups and leg lifts). Hair shaving consisted of a Saleem et al. in 2004. Again, these were first-time present-
5 cm strip shaved from the anus to the presacrum every ers with noninfected tracts numbering from 1 to 3. Simple
week until healing occurred. Over the 1-year period, the excision of the entire pit/tract complex was performed
conservative group required no operations, compared to down to the postsacral fascia with minimal skin excision,
the 240 operations during the 2 years prior. Number of followed by injection of the cavity with fibrin glue. With a
occupied bed days decreased to 83 from 4,760 during the mean follow-up on 10.8 months, they achieved a 96% heal-
2 years prior. A follow-up study applied the conservative ing rate (25).
Phenol be taken to keep as much of the wound off midline as pos-

sible. Wound edges are then excised to prevent early closure.
Sclerotherapy with phenol has been used to treat nonrecur- Once hemostasis is obtained, the wound is packed and cov-
rent pilonidal disease. The technique consists of first remov- ered. Patients are discharged home the same day with ade-
ing the hairs from the deep cavity through the pits. Phenol quate oral analgesia. Wound care consists of daily packing
is then placed through the pits into the cavity, either by and showering, with the help of a family member or home
injection or in crystal form, and left in for 2 minutes. After, health provider. Close follow-up in the immediate postop-
the phenol is removed through expression. Necrotic debris erative phase allows for adequate monitoring and as-needed
is then removed 24 hours later. At follow-up, if there is con- debridement and separation of wound edges.
tinued leakage, the process is repeated. Kepenecki and colleagues reported on their experience
A group of 76 patients were treated with phenol sclero- with this technique in 2009. In this report, 297 patients
therapy in a 2012 study. The overall success rate was 67%. presenting with acute, chronic, or recurrent disease were
Return to work was immediate for all patients. Patients subjected to the unroofing technique; 84.5% were done
with a previous abscess drainage or more than three sinus under local anesthesia. Return to work was 3.2 days, and
openings, or those who required more than one treatment average healing time was 5.4 weeks. During their aver-
were at higher risk of failure (26). A more recent prospective age follow-up time of 54 months, they reported only a 2%
randomized control trial (RCT) compared phenol injection recurrence rate (28).
to excision and healing by secondary intent in 140 primary Marsupialization involves suturing the skin edges to the
and recurrent pilonidal patients (27). Wound healing time, base of the unroofed wound with absorbable suture. The
operative time, time to normal activity, and pain scale all goal is to create a smaller shallow wound that heals faster
heavily favored the phenol group. At 3 weeks, 89% of the and is easier for the patient to manage (Figure 22.4). A 2007
patients in the phenol group were healed; at 6 weeks all were study examined 26 patients who underwent wide local
healed. At 40 months of follow-up, there was no significant excision and 42 who underwent unroofing and marsupial-
difference in recurrence rates (18.6% versus 12.9%). ization. Mean time to healing (21 versus 6 weeks), postop-
erative complications, and reoperative rate (35% versus 2%)
SURGICAL TREATMENT all favored the unroofing and marsupialization group (29).
A more recent study examined 39 patients who underwent
Recurrent, chronic, and nonhealing pilonidal disease either wide local excision or unroofing and marsupializa-
requires a more aggressive surgical approach for cure. tion. Their evidence also favored unroofing and marsupi-
Complete excision of the disease leaves a large open wound alization, demonstrating a faster healing time (5.9 versus
that is difficult to manage given the location. Thus, many 23 weeks), reduced recurrence (4% versus 31%), and no need
different approaches have been developed in an attempt to for reoperation (0% versus 25%) (30).
find the optimal operation for pilonidal disease. Of course, Wide local excision involves removing the entire pilo-
none of them are appropriate in every patient, and ulti- nidal cyst/sinus complex. An elliptical incision is made to
mately selection is based on patient parameters and surgeon include all the sinus tracts and is carried down to the sacro-
comfort. The most common techniques are presented here. coccygeal fascia to ensure the entire cyst cavity is removed.
In general, these procedures are performed in the prone The resulting wound can be either closed or left open to heal
jackknife position with the buttocks taped apart. Standard by secondary intention. Closing the wound makes postop-
preoperative antibiotics are given. The simpler procedures erative care simpler, but this can lead to wound breakdown
can be done as an outpatient, whereas the more complex or infection. Leaving the wound open lessens the chance of
flap procedures will usually require a short hospital stay for infection but leaves the patient with difficult and cumber-
bedrest, pain control, and wound monitoring. some wound care that can often take months to heal. The
use of negative pressure wound therapy to augment healing
MINIMAL/MODERATE DISEASE of large pilonidal wounds has been investigated. A recent
study treated 24 patients with wound vac therapy after wide
Unroofing and wide local excision local excision for 2 weeks postoperatively and then switched
to regular dressing changes. When compared with the con-
Unroofing consists of laying open all sinus tracts and cysts trol group, the wound vac group had significantly smaller
that make up the pilonidal complex in a manner similar to wounds at 2 weeks. Complete healing was achieved at 84
fistulotomy for simple anal fistulas. This technique, along days versus 93 days in the control group (31).
with wide local excision, is among the earlier and most Several RCTs have examined primary closure versus
common treatments for pilonidal disease. leaving the wound open to heal secondarily. A recently
Once the patient is positioned and prepped, the pits are updated Cochrane Review of these trials showed there was
probed to identify sinus tracts. The sinus tracts are then no difference in wound infection between the two, and heal-
opened along the length of the probe. All extensions and ing rates were faster in the closure group. The recurrence
tracts of the sinus are opened and then debris, hair, and rate was 35% less in the open group, but when the midline
granulation tissue are removed with curette. Care should closures were removed from analysis, the recurrence rates
Pilonidal disease / Minimal/moderate disease 199
(a) (b)
(c) (d)
Figure 22.4 Marsupialization the diseased tissue is excised with electocautery (a), and the cavity is debrided (b). The
edges of the wound are then sutured down to the base of the wound using absorbable suture (c), resulting in a small
open wound (d).
were no different. Thus, primary off mid-line closure is rec- patients demonstrated a median healing time of 5 weeks
ommended in wide local excision. The risk of infection and and a 5% recurrence rate at 2 years (34). They later reported
recurrence is similar to open healing, and the patient gains long-term data on 257 patients. At 3.6 years, the overall
the benefit of faster healing with less wound care and scar- recurrence was 7%, and median time to work was 7 days.
ring (32). They also emphasized the majority of the procedures (93%)
were able to be performed under local anesthesia (35).
Limited excision Mohamed and colleagues prospectively compared this
limited excision technique to wide local excision left open
Limited sinus excision was developed to minimize morbid- and wide local excision with primary closure. They found
ity in the treatment of limited pilonidal disease, defined the limited excision group to have shorter operative time,
as four or less pits and no active infection. Oncel and col- shorter hospital stay, and less postoperative pain. Healing
leagues reported a technique that excises the individual pits time was significantly shorter for the closed incision and
in an upside-down funnel-shaped manner. Methylene blue limited incision groups. With a minimum follow-up of 15
is injected into the pits to aid in identification of the tract months, there was no difference in recurrence rates (36).
and sinus. They stressed that only the minimum amount of
tissue is removed, and there is no need to dissect all the way Bascom procedure
to the postsacral fascia. If two sinuses are connected, the
fistula tract and overlying skin are also excised. The wounds In 1980, Bascom described a minimally invasive technique
are left open to heal by secondary intention (33). They later for pilonidal disease that focused on removing the midline
reported on 62 consecutive patients with limited pilonidal pits rather than large amounts of tissue. He based this off the
disease treated with the technique. Mean return to work idea that the hair follicle itself was the source of the inflam-
was 2 days; mean healing was 43 days. At 1-year follow-up, mation that leads to the development of the sinus and cyst.
they had only one recurrence. The abscess cavity is incised off midline and debrided. The
Soll and colleagues described a similar technique for cavity walls are not excised. The associated fistula tracts are
limited excision. Along with patients having more than four dissected out and excised. The midline pits are then indi-
pits, they also excluded those with a distance of more than vidually excised (sometimes referred to as “pit picking”) and
8 cm between pits. Their initial report of 93 consecutive closed with absorbable suture. Punch biopsy devices can be
scarring. These patients will require excision of diseased

(a) Vertical (b)
incision
Open tissue and subsequent closure of the defect with various tis-
wound sue transfer techniques. Here, we discuss the more common
Excision advanced techniques that can be employed by the colorectal
of midline
pits
or general surgeon. More advanced tissue transfers should
involve consultation with a plastic surgeon.
Cyst
cavity Anus and
Karydakis flap
midline
In 1973 Karydakis described his operation as one “which
places resistant skin at the depth of the intergluteal fold”
Figure 22.5 Bascom operation. (a) A vertical incision (41). With the patient prone and the buttocks taped apart,
is made overlying the cyst, 1cm away from the gluteal an off-midline elliptical incision is made in a vertical orien-
cleft. The cyst cavity any communicating fistula tracts are tation centered about the bulk of the disease. This is carried
debrided. The midline pits are excised, with the wounds down to the sacrococcygeal fascia, and all the disease tissue
communicating to the cavity. (b) The midline wounds are is removed en bloc. A skin flap is then raised across midline,
closed primarily with absorbable suture and the vertical
such that when the tape is released the wound edges come
wound is packed lightly and left to heal by secondary
intention.
together allowing for a closure under minimal tension. The
result is an off-midline incision and a flatter gluteal cleft less
prone to recurrent pilonidal disease (Figure 22.6). Given
employed for this portion. The lateral wound is left open or the large potential space created, drainage can be employed
closed partially to heal by secondary intention (37) (Figure to help prevent fluid collections. A 2005 study showed that
22.5). His initial experience with 149 patients showed an routine drainage decreased the risk of significant seroma
average healing time of 3 weeks, time to work of 1 day, and a formation in Karydakis flaps by 24%, avoiding painful post-
16% minor complication/recurrence rate at 3.5 years (38). In operative office drainage procedures (42).
2000, Senapati and colleagues performed the Bascom tech- Karydakis published his initial experience with 6,545
nique on 218 patients: 68% of the patients had more than cases and reported a <1% recurrence rate (10). Kitchen
two midline pits and/or a lateral discharging sinus. They later published on a series of 141 patients with an 18-month
achieved an initial success rate of 90% (39). Others have follow-up and demonstrated a 96% success rate. Findings
shown that for simple pilonidal disease, the Bascom proce- were that 23% of the patients presented with recurrence
dure can be performed with minimal pain, early return to from previously treated disease and were all completely
work, and an acceptable success rate (40). cured with the Karydakis flap (43). Other studies have
continued to show the Karydakis flap to be a durable pro-
ADVANCED DISEASE cedure for the treatment of pilonidal disease. A prospec-
tive randomized trial in 2015 compared 161 Karydakis flap
A small but significant portion of pilonidal patients will go patients to 160 undergoing excision alone with secondary
on to develop chronic nonhealing wounds. These may be the healing. The recurrence rate in the flap group was 1.2%
result of wide local excision left to heal secondarily, break- versus 7.5% in the excision alone, suggesting an impor-
down of a closure, recurrence of disease, or unacceptable tant role in the flattening of the gluteal cleft to prevent
(a) (b) (c) (d) (e)
Figure 22.6 Karydakis flap. (a) Schematic of the operative field depicting a pilonidal cyst slightly to the left of midline
with 2 midline pits. (b) An elliptical incision is made encompassing the cyst and pits. The excision is carried down to the
sacrococcygeal fascia. (c) The medial edge of the wound is raised as a flap crossing the midline. (d) The tape retracting the
buttocks is released so that the wound edges are able to be approximated without tension. (e) Final result showing a verti-
cal incision closed primarily away from the midline, resulting in a flattened gluteal cleft.
Pilonidal disease / Conclusion 201
recurrence (44). A 2011 prospective randomized study of Advanced flaps

269 patients compared the Karydakis flap to the Limberg
flap (discussed in the following). The Karydakis group had More advanced flaps include rotational and myocutaneous
a lower complication rate, lower postoperative pain score, flaps. The rhomboid flap, and its variants the Limberg and
and better perceived cosmetic result. The recurrence rate Dufourmental flaps, are commonly used rotational flaps.
was lower (3% versus 9%), but this was not statistically sig- The cyst and midline pits are excised down to the sacrococ-
nificant (45). cygeal fascia via a rhomboid-shaped skin incision with its
long axis oriented vertically. A triangular incision with the
Bascom cleft lift same acute angle and side-length as the rhomboid is made
based on the lateral extent of the initial incision and rotated
In 2002 Bascom offered a modification to the Karydakis in to fill the defect (Figure 22.8). A randomized trial compar-
flap to use for patients with severe recurrent disease that has ing the Limberg flap to primary closure showed the Limberg
become known as the “cleft lift” or “Bascom II.” The pro- flap patients to have shorter hospital stays, less time off work,
cedure begins with preoperative marking—the buttocks fewer postoperative complications, and a lower recurrence
are pressed together while standing and the line of contact rate. Equivalent outcomes have been reported between the
is marked. In the operating room, the ellipse of skin to be Limberg and Karydakis flaps (48). One advantage the Limberg
removed is marked to include the midline pits and previ- flap has over the Karydakis flap is the ability to excise very
ous scar but does not extend past the preoperatively marked lateral disease without the risk of tension in the closure (49).
line. The ellipse is excised into the cyst cavity, unroofing it. Other flaps used to treat extensive pilonidal disease
The cyst wall is scrubbed with gauze to promote granulation include the V-Y flap and Z-plasty, along with other rota-
and incised in various locations to release tension, while tional flaps. Myocutaneous flaps based off the gluteus maxi-
secondary openings are enlarged to promote drainage. mus can be used to fill large chronic nonhealing defects.
A flap is then raised across midline to the preoperatively The choice of which approach to use will depend on patient
marked line. When the tape is released, the edges come factors and surgeon comfort. When planning complex pilo-
together and an off-midline wound is closed with minimal nidal repair with large flap coverage, consultation with a
tension (Figure 22.7). A drain is left in place. In his initial plastic surgeon is recommended. Poor flap selection in an
series of 31 patients with severe recurrent pilonidal dis- initial operation can limit options for the patient should the
ease, all of them healed completely, and none had recurred patient require additional procedures.
at 20 months (11). Bascom later reported on 69 patients, all
of whom had undergone at least two prior surgeries, who CONCLUSION
were treated with the cleft lift procedure. All but three were
healed with a single operation. The others required one to Pilonidal disease can present in a variety of ways. The patient
two repeat procedures (46). Others have reproduced similar who presents with a straightforward case of pilonidal dis-
results, with or without a drain, reporting very little if any ease can evolve into a complex problem with significant
recurrence (47). morbidity. It is imperative for the surgeon to understand
(a) (b)
Figure 22.7 U-flap technique as described by Bascom for non-healing wounds. (a) One side of would is excised and the other
side is elevated as a flap. (b) Deep wound is closed and remaining flap is advanced to allow closure of incision off midline.
(a) (b)
D D
B 1
Pits
1 C
C
Cyst E B 2 E
2 A
A
Midline
Anus
Figure 22.8 Rhomboid flap. (a) A rhomboid incision encompassing the pilonidal cyst and midline pits is marked on the
skin, along with a lateral extension. The rhomboid is composed of 2 120° angles and 2 60° angles. Line BC is drawn at a
90° angle to Line CD. Line AB is drawn vertically down. All lines should be of equal length. The cavity is excised down to
fascia and debrided. The flap is raised and mobilized to cover the defect. (b) The flap is rotated into the defect so that
Point 2 meets Point E, Point 1 meets Point D, and Point A meets Point C. This results in a primarily closed wound and
flattened gluteal cleft.
the disease process, variety of treatments, as well as unique studies, but rather by noting that the most common areas of
individual patient factors when evaluating pilonidal disease. disease distribution in HS correlated with areas where these
Appropriate selection of treatment early on will ultimately sweat glands are primarily found (51,52).
lead to better outcomes for the patient (Figure 22.9). More modern histopathologic studies have argued quite
vociferously and convincingly that HS is actually a disease
process that begins with occlusion of hair follicles, similar
PERIANAL HIDRADENITIS to the pathogenesis of the common acne vulgaris, and that
SUPPURATIVA apocrine sweat glands are only involved secondarily, if at all,
in cases of widespread follicular obstruction and infection.
BACKGROUND AND PATHOPHYSIOLOGY Whereas acne vulgaris is more frequently seen in areas with
a predominance of sebaceous sweat glands such as the face
Historically, hidradenitis suppurativa (HS) has long been and upper back/chest, HS tends to most commonly appear in
defined as a chronic, waxing and waning disease process the opposite or inverse anatomical areas, primarily the axilla,
involving inflammation and infection of the apocrine sweat groin, and perianal regions, but also any other intertriginous
glands, which predominate in the axillary, inguinal, and region, such as the inframammary folds, scrotum, labia, and
perianal regions. This disease process was first described in perineum (52). Thus, HS is now more accurately referred to
the literature by the French physician Velpeau in 1832 (50). as acne inversa, a term first used in this manner by Plewig
It was not until 1864 that another French physician, Vernuil, and Stegar in 1989 (53). Despite this more accurate name,
associated the disease process with sweat glands in the skin, the term hidradenitis suppurativa has been maintained and
thus coining the term hidradenitis suppurativa; however, is still used most frequently for this disease, especially in the
Vernuil made this association not through histopathological surgical literature.
Chronic pilonidal disease
Limited disease Complicated disease Large

wound
Karydakis flap
Rhomboid flap
Midline excision ± closure Cleft lift (Bascom II)
+ contraindications to surgery or marsupialization
– Hygiene V-Y flap Myocutaneous flaps
Sinus excision
– Depilation Z-plasty Vacuum sponge
Bascom operation
Figure 22.9 Algorithm for management of pilonidal disease.

Perianal hidradenitis suppurativa / Diagnosis and natural history 203
A recent study by Danby et al. further characterized the differences in typical phenotypic expression between the
process of follicular obstruction by identifying a defect in groups. LC1 patients (48%) primarily express lesions in the
the basement membrane at the sebofollicular junction of axilla and breast, while LC2 patients (26%) display a follicu-
the folliculopilosebaceous unit as the initiating event, which lar pattern more widely distributed on the body, as well as a
leads to bacterial colonization and subsequent inflamma- concurrent history of pilonidal disease and a family history
tion and infection (54,55). When there are multiple adja- of HS. Most relevant to this chapter, LC3 patients (26%) are
cent areas of follicular obstruction, these areas may rupture most likely to have gluteal disease, have less severe disease
and even connect via sinuses and fistulas as they become overall, and are less often obese (59).
infected. This can lead to wide areas of purulent drainage
and discomfort. Some mild cases may respond at least ini- DIAGNOSIS AND NATURAL HISTORY
tially to a course of antibiotics or simple drainage or unroof-
ing of the tracts, but in severe cases, the recurrent nature of The prevalence of HS has frequently been reported to be
the disease can often lead to areas of active infection coalesc- around 1% (60,61); however, more recent and robust epi-
ing with widespread areas of fibrosis and scar, rendering a demiologic studies in the United States show that it is sig-
regional catacomb of disease in a patient who is not only in nificantly rarer than previously thought, as low as 0.053%
a constant state of discomfort but also often severely socially (62,63). A genetic component is frequently seen, as 34%–40%
isolated due to the malodor and dyshygienic appearance of of patients with HS have a family history of the disease, and
those suffering from chronic severe disease (56). many potential gene groups have been identified (59,64–66).
HS is rare in prepubertal children, with a peak onset in one’s
CLASSIFICATION early 20s and a decline in disease after 50 (58,59). It affects
women more than men at a ratio of 3:1, but a higher propor-
Most commonly, HS is classified by degree of severity tion of males have perianal disease (58,59).
according to the Hurley staging system (57). Hurley stage I The diagnosis of HS is generally a clinical one, where
disease is characterized by simple abscess formation, either patients present with the classic findings as described by the
single or multiple, but without sinus tracts or scarring. Hurley classification system. In cases of perianal HS, it is
Stage II disease is characterized by recurrent abscesses with important to examine the other common areas affected by
the formation of sinus tracts and scarring, but these occur HS, such as the axilla and inguinal regions, to aid in obtain-
as either single lesions or multiple lesions that are widely ing the correct diagnosis. Mild perianal HS can sometimes
separated. Hurley stage III (Figure 22.10) represents the be confused with an anal fistula of cryptoglandular origin;
most severe stage of HS, defined by diffuse or nearly diffuse however, as opposed to a classic anal fistula, HS will at most
involvement of the entire affected region with numerous only involve the very distal anal verge and not the dentate
interconnected, honeycombing sinus tracts and abscesses line. Smoking (82%) and obesity are the two most common
throughout (55–58). Hurley III patients are those most conditions associated with both the diagnosis and increas-
likely to ultimately require radical surgical excision. ing severity of HS (55,56,58–63,67), although perianal dis-
Recently, a group from France further stratified Hurley ease is often seen in the nonobese and even underweight
II/III patients into three latent classes (LC1-3) based on (59). Depilation, sweating, tight-fitting clothes, deodorant
use, friction, androgen excess, and even pilonidal disease
have been associated with HS (55,56). Patients with meta-
bolic syndrome have also been shown to have an increased
prevalence of HS (68,69), but this did not correlate with
increased HS disease severity (69).
HS has also long seen an increased prevalence in patients
with inflammatory bowel disease (IBD). A 2009 Dutch
survey of patients with IBD reported that 17% of patients
with Crohn disease and 14% with ulcerative colitis had
coexisting HS (70). Similarly, a 2016 retrospective review
from the Mayo Clinic concluded that patients with IBD
were nine times more likely to suffer from HS than the gen-
eral population (71). Although quite rare, transition into
invasive squamous cell carcinoma (SCCA) within areas of
long-standing and chronically inflamed HS has also been
described (72). Due to the extremely scarred, inflamed set-
ting in these instances, SCCA is often either a delayed diag-
nosis or incidental finding after radical excision, and several
Figure 22.10 Chronically recurrent Hurley stage III HS in cases of death from metastatic disease have been reported;
the setting of multiple incision and drainage and local therefore, a high index of suspicion is necessary in individu-
excision procedures. als with a long-standing history of severe disease (73).
NONEXCISIONAL THERAPY in itself is quite straightforward, in that it requires complete

excision of all the gross disease with a margin of normal-
In cases of milder disease, lifestyle modifications have a role appearing tissue. The excision is carried at least into healthy
in helping prevent disease progression or flares. In fact, edu- subcutaneous fat, but in cases of severe disease, excision all
cating patients to keep the affected areas as clean and dry the way down to the fascia of the gluteus muscle and expo-
as possible, promoting perianal hygiene, avoiding excessive sure of the external anal sphincter may be necessary. Failure
sweating or friction, and not wearing tight-fitting clothes to completely excise the affected areas will ensure recur-
should be an important conversation with every patient rence, though the risk of recurrence is still quite high with
encountered with HS. Antibiotics, both topical and sys- any modality. A retrospective review from 1998 reported
temic, have long been used in the treatment of acute flares, a recurrence rate of 100% for incision and drainage proce-
especially in conjunction with simple unroofing or inci- dures, 43% for limited excision, and only 27% for radical
sion and drainage of abscesses (73,74). A frequently cited excision (p < 0.05) (82).
1983 RCT showed that topical clindamycin for 3 months The more difficult decision is wound management after
was superior to placebo in cases of mild disease, with bet- excision, as simple primary closure is rarely technically
ter resolution of flares and fewer recurrences (75). In 1998, possible in Hurley II/III patients. When primary closure is
another RCT showed no advantage of systemic tetracycline attempted, recurrence rates are high, as the surgical mar-
over topical clindamycin (76). gin is often compromised to allow primary closure. Many,
As antibiotics cannot be used indefinitely for long-term perhaps most, surgeons still opt to allow the wounds to heal
control, a myriad of more aggressive attempts at control by secondary intention, with quite acceptable long-term
via systemic therapy have been described (55,56,58,77). outcomes that rarely lead to any functional disability in
Antiandrogen therapy in females (78), immunosuppres- the perianal region. The biggest drawback is wound healing
sants such as cyclosporine (79), and retinoids (80,81) have time, which usually takes 12–14 weeks (83,84).
all been used with varying degrees of success at inducing Immediate or even delayed split-thickness skin grafting
remission, decreasing flares, and improving quality of life, alone or in combination with myocutaneous advancement
especially in mild to moderate HS. flaps if enough healthy adjacent tissue remains has been
With the fairly common coexistence of IBD and HS, it is associated with decreased wound healing time and less scar
not surprising that biologic agents, primarily tumor necro- contracture; however, at least partial graft loss from con-
sis factor-α (TNF-α) inhibitors, have been studied exten- tamination and shearing forces is common (Figure 22.11).
sively in the treatment of HS. A recent Dutch systematic Others have at times employed rectal tubes and vacuum-
review evaluated all papers available in the Medline and assisted closure devices to help speed healing and decrease
Embase databases regarding the systemic treatment of HS contamination, with mixed results. It is likely that all options
with immunosuppressants and retinoids (81). Included in are important to have in one’s armamentarium to treat each
the review were 87 papers representing 518 patients. Patients patient on a case-by-case basis. An able and willing plastic
treated with infliximab (89%) and adalimumab (79%) had surgery colleague is also helpful to tailor and employ the
by far the highest response rate, while etanercept showed optimum approach for each individual patient. Only rarely
less of a response (56%). All three biologics included one is fecal diversion necessary (82–85).
manuscript of level A evidence and more than five of level A recent case series from Japan describes a two-stage
B in their analyses. However, most patients had recurrence approach to wound closure using artificial dermis. At the
of active disease within a couple of months of discontinuing initial wide excision, attempts to preserve the deep subcu-
therapy, and to date only adalimumab has been approved taneous fat are made to allow for a more suitable graft bed,
for use in patients with moderate to severe HS. and then the wound is immediately grafted with artificial
There are also numerous described techniques of local dermis. Two weeks later, the artificial dermis is covered
invasive therapy for HS short of radical excision. In addition with split-thickness skin graft after re-excising any areas of
to unroofing and local excision techniques, both YAG and persistent HS and lightly curetting the surface. Their results
CO2 laser therapies have been employed, and even radiation show a very high graft take rate with very low recurrence.
therapy in some instances. However, all of these options Although they do not report wound healing time, they cite
have fairly high recurrence rates, especially if used in severe the minimal to no hypertrophic scarring or wound con-
disease, and can still result in the significant scarring that tracture seen with this technique as the primary advantage,
contributes to much of the decreased quality of life seen in leaving the grafted areas flexible and supple long term (86).
recurrent disease (55,58). This approach is similar to approaches commonly employed
at our institution and many others in the management of
EXCISIONAL THERAPY severe burn injuries.
A newer even less invasive excisional technique described
Despite these recent attempts at improved medical control by the Dutch group of Blok et al. in 2014 is the skin-tissue-
with aggressive systemic therapy, radical wide surgical exci- sparing excision with electrosurgical peeling (STEEP) pro-
sion remains the best chance for cure or at least significant cedure. This procedure focuses on excising all the affected
disease burden control in severe perianal HS. The excision areas via tangential unroofing incisions that then focus
References 205
(a) (b) (c)
(d) (e) (f )
Figure 22.11 (a) Severe Hurley III perianal HS affecting the left > right. (b) Status/post wide radical excision with exposed
external sphincter. (c) Immediate reconstruction via a right-sided V-Y advancement flap and split-thickness skin grafting. (d)
Early loss of large portions of the skin graft due to fecal contamination and poor patient mobility. (e) Excellent reepithelializa-
tion from areas of viable graft at 6 weeks. (f) Complete healing without disease recurrence at 16 weeks.
on minimal excision of only the inflamed sinus tracts, rather than one of apocrine sweat gland origin. Regardless
attempting to leave the epithelialized sinus floor and non- of the inciting event, HS is a chronic disease that can lead to
affected subcutaneous fat in situ to minimize wound size severe disability and social withdrawal. While topical and
and shorten healing time (87). Their preliminary retrospec- systemic options may be successful at controlling milder
tive data of 363 operations in 133 patients with Hurley II/III disease, and while simple incision and drainage or unroof-
HS shows a 29% recurrence rate but high satisfaction scores ing procedures may temporize severe areas of suppura-
(8/10). The most frequent complication was hypergranula- tion, wide radical excision currently remains the mainstay
tion tissue in 7% (88). Certainly more prospective studies of treatment, especially in severe disease. Radical excision
are needed before this becomes more widely adopted. allows the best chance of significant long-term disease
Despite the postexcision wound management technique control and quality of life improvement, regardless of the
employed, the large majority of patients report a signifi- postexcision wound management decision, a decision that
cantly improved quality of life after excision as well as long- is best made on an individualized basis in multimodal fash-
term satisfaction with their decision to have undergone ion. Fecal diversion is rarely required.
excision. A large quality of life survey from the Mayo Clinic
was recently published in which all their patients with HS of REFERENCES
any region who underwent operative treatment from 1976
to 2014 were surveyed. Of the 113 respondents, 83% were 1. Søndenaa K et al. Int J Colorectal Dis. 1995;10(1):39–42.
satisfied with their surgical result, and 96% were glad that 2. Aysan E et al. Surg Today. 2013 Nov;43(11):1286–9.
they underwent surgery. Only a small minority of patients 3. Armed Forces Health Surveillance Center (AFHSC).
underwent flap of skin graft wound closure. The means MSMR. 2013;20(12):8–11.
quality of life for all respondents increased significantly 4. Mayo OH, Observations on Injuries and Diseases
from 5/10 to 8.4/10 after surgery. Additionally, those with of the Rectum. London, UK: Burgess and Hill, 1833,
isolated gluteal and perianal disease reported the greatest pp. 115–6.
increase in quality of life after surgery (89). 5. Anderson AW. Boston Med Surg J. 1847;36:74.
6. Hodges RM. Boston Med Surg J. 1880;103:485–6.
7. Abramson DJ. Mil Med. 1978;143:753–7.
8. Buie LA. South Med J. 1944;37:103–9.
CONCLUSION 9. da Silva JH. Dis Colon Rectum. 2000;43(8):1146–56.
10. Karydakis GE. Aust N Z J Surg. 1992;62(5):385–9.
In summary, hidradenitis suppurativa, or acne inversa, 11. Bascom J, Bascom T. Arch Surg. 2002;137(10):1146–50.
appears to be a disease instigated by follicular occlusion 12. Ballas K et al. J Hand Surg Br. 2006;31(3):290–1.
13. Patel MR et al. J Hand Surg Am. 1990 Jul;15(4):652–5. 54. Danby FW et al. Br J Dermatol. 2013;168:1034–9.
14. Phillips PJ. Med J Aust. 1966;2(24):1152–3. 55. Gill L et al. F1000 Prime Reports 2014;6:112.
15. Mohanna PN et al. Br J Plast Surg. 2001;54(2):176–8. 56. Alikhan A et al. J Am Acad Dermatol. 2009;60(4):
16. Harlak A et al. Clinics (Sao Paulo). 2010;65(2):125–31. 539–61.
17. Webb PM, Wysocki AP. Tech Coloproctol. 2011;15(2): 57. Hurley HJ. Axillary hyperhidrosis, apocrine bromhi-
179–83. drosis, hidradenitis suppurativa, and familial benign
18. Jensen SL, Harling H. Br J Surg. 1988;75(1):60–1. pemphigus: surgical approach. In: Roenigk RK,
19. Armstrong JH, Barcia PJ. Arch Surg. 1994 Sep;129(9): Roenigk HH Jr (eds). Dermatologic Surgery:
914–7; discussion 917–9. Principles and Practice. 2nd edition. New York, NY:
20. Odili J, Gault D. Ann R Coll Surg Engl. 2002;84(1): Marcel Dekker, 1996, pp. 623–45.
29–32. 58. Jemec G. N Engl J Med. 2012;366(2):158–64.
21. Petersen S et al. Dis Colon Rectum. 2009;52(1):131–4. 59. Canoui-Poitrine F et al. J Invest Dermatol. 2013;133:
22. Demircan F et al. Int J Clin Exp Med. 2015;8(2): 1506–11.
2929–33. 60. Revuz JE et al. J Am Acad Dermatol. 2008;59:
23. Lund JN, Leveson SH. Dis Colon Rectum. 2005;48(5): 596–601.
1094–6. 61. Jemec G et al. J Am Acad Dermatol. 1996;35:191–4.
24. Isik A et al. Int J Clin Exp Med. 2014;7(4):1047–51. 62. Cosmatos I et al. J Am Acad Dermatol. 2013;68(3):
25. Seleem MI, Al-Hashemy AM. Colorectal Dis. 2005; 412–9.
7(4):319–22. 63. McMillan K. Am J Epidemiol. 2014;179(12):1477–83.
26. Dag A et al. Surgery. 2012;151(1):113–7. 64. Fitzsimmons JS et al. J Med Genet. 1984;21:281–5.
27. Calikoglu I et al. Dis Colon Rectum. 2017;60(2):161–9. 65. Fitzsimmons JS, Guilbert PR. J Med Genet. 1985;
28. Kepenekci I et al. World J Surg. 2010;34(1):153–7. 22:367–73.
29. Tejirian T et al. Am Surg. 2007;73(10):1075–8. 66. Melnik BC, Plewig G. Exp Dermatol. 2013;22(3):
30. Rouch JD et al. JAMA Surg. 2016;151(9):877–9. 172–7.
31. Biter LU et al. Dis Colon Rectum. 2014;57(12):1406–11. 67. Sartorious K et al. Br J Dermatol 2009;161:831–9.
32. Al-Khamis A et al. Cochrane Database Syst Rev. 2010 68. Sabat R et al. PLOS ONE. 2012;7:e31810.
Jan 20; Issue 1: Art. no. CD006213. 69. Gold DA et al. J Am Acad Dermatol. 2014;70:
33. Oncel M et al. Tech Coloproctol. 2002;6(3):165–9. 699–703.
34. Soll C et al. Int J Colorectal Dis. 2008;23(2):177–80. 70. van der Zee HH et al. Br J Dermatol. 2010;162:195–7.
35. Soll C et al. Surgery. 2011;150(5):996–1001. 71. Yadav S et al. Clin Gastroenterol Hepatol. 2016;14(1):
36. Mohamed HA et al. Surgeon. 2005;3(2):73–7. 65–70.
37. Bascom J. Surgery. 1980;87(5):567–72. 72. Losanoff JE. Am Surg. 2011;77(11):1449–53.
38. Bascom J. Dis Colon Rectum. 1983;26(12):800–7. 73. Shah N. Am Fam Physician. 2005;72(8):1547–52.
39. Senapati A et al. Br J Surg. 2000;87(8):1067–70. 74. van der Zee HH et al. J Am Acad Dermatol. 2010;63(3):
40. Colov EP, Bertelsen CA. Dan Med Bull. 2011;58(12):348. 475–80.
41. Karydakis GE. Lancet. 1973;2(7843):1414–5. 75. Clemmensen OJ. Int J Dermatol. 1983;22:325–8.
42. Gurer A et al. Dis Colon Rectum. 2005;48(9):1797–9. 76. Jemec GB, Wendelboe P. J Am Acad Dermatol. 1998;
43. Kitchen PR. Br J Surg. 1996;83(10):1452–5. 39(6):971–4.
44. Keshvari A et al. J Surg Res. 2015;198(1):260–6. 77. Nazary M et al. Eur J Pharmacol. 2011;672:1–8.
45. Ates M et al. Am J Surg. 2011;202(5):568–73. 78. Kraft JN. Searles, GE. J Cutan Med Surg. 2007;11:
46. Bascom J, Bascom T. Am J Surg. 2007;193(5):606–9. 125–31.
47. Theodoropoulos GE et al. Dis Colon Rectum. 2003; 79. Buckley DA, Rogers S. J R Soc Med. 1995;88:289–90.
46(9):1286–91. 80. Boer J, Nazary M. Br J Dermatol. 2011;164(1):170–5.
48. Khan PS et al. Indian J Surg. 2013;75(3):192–4. 81. Blok JL et al. Br J Dermatol. 2013;168:243–52.
49. Can MF et al. Am J Surg. 2010;200(3):318–27. 82. Ritz J et al. Int J Colorect Dis. 1998;13:164–8.
50. Velpeau A. Dictionnaire de Medicine, un Repertoire 83. Balik E et al. World J Surg. 2009;33(3):481–7.
General des Sciences Medicales sous la Rapport. 84. Wollina U et al. Int J Dermatol. 2017;56:109–15.
Therique et Pratique. Paris: Bechet Jeune, 1839. 85. Menderes A et al. Int J Med Sci. 2010;7(4):240–7.
51. Vernuil A. Arch Gen Med Paris. 1864;114:537–57. 86. Yamashita Y et al. Dermatol Surg. 2014;40(2):110–5.
52. Sellheyer K, Krahl D. Int J Dermatol. 2005;44:535–40. 87. Blok JL et al. J Eur Acad Dermatol Venereol. 2015;
53. Plewig G, Steger M. Acne inversa (alias acne triad, 29(2):379–82.
acne tetrad or hidradenitis suppurativa). In Marks R, 88. Blok JL et al. J Eur Acad Dermatol Venereol. 2015;
Plewig G (eds). Acne and Related Disorders. London, 29(8):1590–7.
UK: Martin Dunitz, 1989, pp. 345–57. 89. Kohorst JJ et al. Dermatol Surg. 2017;43(1):125–33.
23
Surgical treatment of fecal incontinence
NICOLE M. SAUR AND JOSHUA I.S. BLEIER
CHALLENGING CASE 1 CASE MANAGEMENT 2
A 30-year-old F presents to your office with complaints A complete history is otherwise negative. Physical
of incontinence to solid and liquid stool and gas. She examination reveals weak resting and squeeze pres-
is G2P2 (gravida 2, para 2) and had vaginal deliver- sures. A full colonoscopy was negative for masses
ies with need for perineal laceration repair after both or mucosal abnormalities. Anorectal manometry
deliveries. Her last child was delivered 1 year ago. She revealed decreased resting and squeeze pressures
reports worsening passage of both solid and liquid circumferentially. Endoanal ultrasound did not reveal
stool, often without her knowledge. The incontinence a sphincter defect. Pudendal nerve terminal motor
is affecting her lifestyle, and she is wearing a pad. latencies are prolonged bilaterally. He is counseled
and started on a trial of fiber supplementation and
biofeedback. He has a modest improvement in his
symptoms. After further discussion, he is offered
CASE MANAGEMENT 1
sacral nerve stimulation.
Fecal incontinence (FI) and its implications can
The remainder of her history is negative. Physical
have a major impact on a patient’s quality of life.
examination reveals a thin perineal body, decreased
Whether it is soiling, inadvertent passage of fla-
resting and squeeze pressures, and an anterior
tus, or the leakage of stool, the symptoms can be
sphincter defect. Flexible sigmoidoscopy did
debilitating. Therefore, the surgeons treating this
not reveal any masses or mucosal abnormalities.
condition should be well versed in the medical and
Anorectal manometry shows decreased average rest-
surgical treatment options to offer a patient-centered
ing and squeeze pressures (worst anteriorly), and an
approach.
endoanal ultrasound revealed an anterior sphincter
defect and a perineal body measurement of 6 mm.
After discussion with the patient, the decision is
made to proceed with overlapping sphincterotomy.
EPIDEMIOLOGY
CHALLENGING CASE 2 The true incidence of FI has largely been unknown; how-
ever, a recent systematic review by Sharma et al. included 30
A 65-year-old male presents with incontinence to studies and showed a rate of FI ranging from 1.4% to 19.5%.
liquid and solid stool. He reports that he has had They found that if incontinence to flatus was included in
incontinence for “years” but describes that it has the definition, the prevalence rates were highest (15%–17%).
progressed from incontinence to only liquid stool and In addition, the prevalence of FI to liquid or solid stool at
gas to incontinence to solid stool. He has a history least once a month was found to be 8.3%–8.4% for face-to-
of an excisional hemorrhoidectomy 20 years ago. He face interviews or phone interviews and as high as 12.4%
also has a diagnosis of irritable bowel syndrome. He for surveys conducted by mail (1). Groups of individuals at
states that the incontinence is affecting his quality of high risk for incontinence include the elderly, the mentally
life, and he is having one to three accidents a week. ill, institutionalized patients, those with neurologic disor-
ders, and parous women. Macmillan et al. (2) had reported
207
208 Surgical treatment of fecal incontinence
similar results in an older systematic review. The estimated function: the worse the function, the higher is the score.
prevalence of FI (including flatus incontinence) varied from Summary scores are considered more accurate in quan-
2% to 24%, and the estimated prevalence of FI (excluding tifying the patient’s symptoms, comparing patients, and
flatus incontinence) varied from 0.4% to 18% in that study. gauging treatment response. These scales also include items
Data have been published recently on the socioeconomic such as urgency, cleaning difficulties, the use of pads, and
impact of FI. The average yearly cost for FI was $4,110 per lifestyle alterations. Numerous summary scales have been
patient. Direct medical costs accounted for $2,353 yearly. designed, such as those according to Rockwood, Wexner,
Multivariate regression showed that more severe FI was Pescatori, and Vaizey. The assignment of values to types
associated with higher costs (3). In addition, we are likely and frequencies of incontinence varies between scales. The
not fully treating the entire appropriate patient population, frequently cited and validated Wexner/Cleveland Clinic
as two-thirds of patients with FI symptoms do not seek Florida Incontinence Score is outlined in Table 23.1 (10).
care, and of those who do, more than half are seen only with More in-depth scoring systems have been adopted to incor-
their primary care physicians-clinicians who may not have porate quality of life measures (FI severity index [FISI], FI
a complete understanding of the workup and treatment Quality of Life Scale [FIQLS]) (11,12).
a lgorithms (4–7). Most discussions of etiology of anal incontinence have
been based on the assumption that women, particularly
women younger than 65 years of age, are more at risk for
FI than men. Obstetric injury to the pudendal nerve or
ETIOLOGY AND SCORING sphincter muscles is described as the primary risk factor,
irritable bowel syndrome as a secondary factor (a disease
thought to be more prevalent in women; secondary to
Fecal continence requires coordination of learned and reflex
urgency and frequency of bowel movements), and other eti-
responses to colonic and rectal stimuli. Normal individual
ologies such as diabetes were listed as a less common third
variation in bowel habits makes the definition of abnor-
cause. Yet, population-based surveys of anal incontinence
mal defecation and incontinence difficult. Normal physi-
prevalence, including that by Nelson et al., demonstrate
ologic continence depends on a number of general (mental
FI in men at higher than expected rates (63% women, 37%
function), colonic (colonic stool transit, stool volume, and
men). Therefore, etiologies other than childbirth must be
stool consistency), and anorectal (rectal distensibility,
evaluated (13).
anal sphincter function, anorectal sensation, and anorec-
The true percentage of incontinence attributable to each
tal reflexes) variables (8). Definitions, although imprecise,
of the possible causes is unknown. However, surgical and
have been utilized including complete or full incontinence
obstetric injuries are the most common. In addition, nerve
relating to complete loss of control of solid feces and partial
damage secondary to diabetes has been reported, and spinal
incontinence involving inadvertent soiling or leakage of liq-
uid stool or inadvertent passage of flatus. Patients may have cord injuries also account for cases of FI (14–16). For pur-
difficulty qualitatively or quantitatively describing partial poses of discussion and classification, we break down the
incontinence. causes of FI into those involving (1) the anal sphincter, (2)
In an effort to more accurately classify the severity of the rectum, (3) the colon/stool consistency, and (4) the cen-
symptoms, Browning and Parks proposed the following tral nervous system.
criteria: category A, those patients with normal continence
who are continent of solid stool, liquid stool, and flatus; B, THE ANAL SPHINCTER
those continent of solid and usually liquid stool but not
flatus; C, acceptable continence of solid stool but no control Obstetric injury
over liquid stool or flatus; and D, continued fecal leakage
of solid or liquid stool (9). In addition, numerous sever- Obstetric injury is the most common risk factor for FI in
ity scores exist and are simple to use to reflect sphincter women following childbirth. In 1993 Sultan et al. published
Table 23.1 Wexner/Cleveland clinic florida fecal incontinence scale
Never Rarely Sometimes Usually Always

Incontinence to solid stool 0 1 2 3 4
Incontinence to liquid stool 0 1 2 3 4
Incontinence to gas 0 1 2 3 4
Alteration of lifestyle 0 1 2 3 4
Wears a pad 0 1 2 3 4
Source: Adapted from Jorge JM, Wexner SD. Dis Colon Rectum. 1993;36(1):77–97.
Note: Rarely: <1/month; sometimes: <1/week, ≥1/month; usually: <1/day, ≥1/week, always: ≥1/day. Score of 0 corresponds to perfect
continence, score of 20 to complete incontinence.
Etiology and scoring / The anal sphincter 209
their well-known article evaluating anal sphincter inju- Surgical injury

ries during vaginal delivery and described the results of
an endosonographic study of 79 primiparous women. Surgical injuries are possible following surgery of the anal
Endoanal ultrasound was performed 6 weeks before and 6 canal including lateral internal sphincterotomy (LIS), fis-
months after routine vaginal delivery. After vaginal deliv- tula surgery, anal dilatation, and hemorrhoidectomy. Injury
ery, sphincter defects were detected in 35% of the patients. is most common to the internal sphincter, but external
A similar evaluation was performed in 23 primiparous sphincter injury can occur as well. Khubchandani and
women who had undergone a cesarean section, and none of Reed evaluated the effect on continence with lateral internal
these women demonstrated a sphincter defect after delivery sphincterotomy, a procedure in which the internal sphinc-
(17). Eason et al. (18) studied 949 women 3 months after ter is purposely divided. They found that lack of control of
delivery, and 3.1% reported incontinence to solid or liquid flatus was the most common complaint (35%), followed by
stool, while 25.5% reported gas incontinence. FI was more fecal soiling (22%) and accidental bowel movements (5%)
frequent in women who delivered vaginally with third- or (25). A significantly higher proportion of patients who
fourth-degree perineal lacerations than in those without reported accidental bowel movements were older than 40
(7.8% versus 2.9%). Forceps delivery (relative risk 1.45) and years. Most recently, Liang and Church reported on a pro-
sphincter tears (relative risk 2.09) were independent risk spective series of 57 patients undergoing LIS for chronic
factors for all types of incontinence. Anal sphincter injury fissure. Only two (4%) reported any changes in continence,
was most strongly associated with first vaginal births (rela- and overall satisfaction in this cohort was 9.7 ± 0.9 out of 10
tive risk 39.2) but was also independently associated with patients (p < 0.001) (26). The differences in reported incon-
median episiotomy (relative risk 9.6), forceps delivery (rela- tinence rates between studies is likely secondary to differ-
tive risk 12.3), and vacuum-assisted delivery (relative risk ences in the length of internal sphincter divided. In the
7.4). Birth weight (relative risk for birth weight 4000 g or later study, the internal sphincter was only divided for the
more: 1.4) or length of the second stage of labor (relative length of the fissure and not the entire length of the sphinc-
risk for second stage 1.5 hours or longer compared with ter, which likely resulted in less incontinence to liquid stool
less than 0.5 hours: 1.2) were not strongly associated with and gas.
sphincter injuries. Fistula surgery is the anorectal procedure most com-
Seventy percent of women with postpartum sphincter monly followed by postoperative incontinence. Gross
defects do not demonstrate FI in the postpartum period incontinence of feces generally may be avoided if the ano-
(19). The question is whether women with an occult rectal ring is preserved. However, minor abnormalities in
sphincter defect are at increased risk for FI over time. continence may result if even a small amount of sphinc-
According to Rieger and Wattchow, many women remain ter muscle is divided. This complication can be reduced
asymptomatic because the number of occult sphincter by avoiding division of the sphincter muscles. This goal
defects is far greater than the documented prevalence of is accomplished by placing a seton followed by utilization
FI in the community (20). However, we must keep in mind of the ligation of the intersphincteric fistula tract (LIFT),
that FI is largely underreported in community observa- or the fistula plug, or endorectal advancement flap for
tional studies. Oberwalder et al. examined elderly females transsphincteric fistulae. However, although the endorec-
with late-onset FI, all of whom had a history of vaginal tal advancement flap repair is thought to heal the fistula
delivery. They observed sphincter defects in more than without dividing sphincters or compromising continence,
70% of the patients (21). Despite these findings, it is still impairment of continence has been reported with an inci-
not possible to determine the exact risk for asymptom- dence between 8% and 35% (27–30). This could be sec-
atic women with a sphincter defect to develop FI later in ondary to inclusion of the internal sphincter in the flap
life. More studies including those longitudinally follow- mobilization. In addition, it has been suggested that anal
ing these patients are necessary to determine the true stretch caused by the Parks’ retractor is a contributing fac-
incidence. tor to sphincter injury and incontinence (31).
It has been theorized that elective cesarean section at Currently, after surgery for hemorrhoids, incontinence
term before the onset of labor protects the anal sphincters is a rare complication. However, inadvertent injury of the
and prevents FI (22,23). However, although cesarean section sphincter mass (e.g., in a blind-clamping technique causing
performed during labor protects the anal sphincters, it does the internal sphincter to be grasped by a clamp) may lead to
not necessarily prevent FI. This finding suggests a signifi- incontinence. In addition, minor alterations in continence
cant component of pudendal nerve injury during labor. A can be attributed to removal of the hemorrhoidal cushions
Cochrane Review evaluated almost 32,000 women, more that are known to contribute to continence via anal canal
than 6,000 of whom underwent delivery by C-section, and occlusion at rest (32).
of the 21 studies reviewed, only 1 revealed a difference in
preservation of anal continence between vaginal and cesar- Miscellaneous
ean deliveries. The authors concluded that preservation of
anal continence should not be used as a reason for selecting Congenital malformations such as imperforate anus can
C-section for delivery (24). disrupt normal anal sphincter structure and function.
Rectal prolapse, too, is frequently associated with FI because of anorectal disorders such as hemorrhoids and prolapse.
the rectum is intermittently or chronically distal to the Incontinence also must be distinguished from diarrhea
sphincter mechanism. Persistence of FI after repair of rec- with urgency, in which the patient’s diet or underlying
tal prolapse occurs in up to 50% of patients (33), which has bowel condition lead to frequent passage of liquid stool
been attributed, at least in part, to nerve injury (34). FI can accompanied by a sense of urgency. In such cases, simple
also occur with trauma to the anus associated with pelvic dietary change or the addition of medications (antispasmot-
trauma. Inflammatory conditions (inflammatory bowel ics, fiber, antidiarrheals, etc.) may be all that is necessary.
disease [IBD], infectious proctitis, sexually transmitted When these underlying diseases are treated, persistent urge
diseases, etc.) can result in FI as can carcinoma of the anal incontinence has been reported to be a marker of external
canal. anal sphincter dysfunction (36).
Female patients should be asked about number of child-
RECTUM births and type of delivery as well as a history of instru-
ment-assisted vaginal deliveries. A history of anal sphincter
The rectum serves as a reservoir for the stool bolus. If the trauma such as episiotomies, perineal tears, and prior ano-
reservoir function is impaired, most commonly in the rectal procedures should be obtained as well as a history
form of decreased wall compliance, this can result in FI. of the patient’s continence in the postpartum period. Also
Mechanisms for rectal reservoir function impairment notable are a history of associated conditions such as uri-
include rectal masses, radiation proctitis, infectious or nary incontinence, prolapsing tissue, history of associated
inflammatory proctitis, history of proctectomy, pelvic nerve fistulae, diabetes mellitus, medications, radiation treat-
damage, and rectal prolapse, as mentioned. ment, and previous colon, anorectal, or rectal operations.
A history of conditions that can lead to diarrhea such as
COLON/STOOL CONSISTENCY IBD, IBS, and infectious colitis as well as food and bever-
age history are important to ascertain especially because,
IBD and irritable bowel syndrome (IBS) can be associated for example, some beverages such as coffee or beer can lead
with changes in stool transit and consistency, which lead to to frequent loose bowel movements. Inquiring about associ-
FI. Liquid stool is more likely to be associated with FI as is ated motor or sensory symptoms may point to a neurologic
stool with increased transit. In addition, constipation can be lesion (37). A clue to the severity of the problem is to deter-
associated with overflow FI. mine the frequency of the incontinence and the necessity to
wear a protective pad.
CENTRAL NERVOUS SYSTEM Grading and scoring the severity of the FI can be under-
taken at the time of the history taking. It is especially use-
Patients with history of stroke or dementia are more likely ful to identify the severity of FI as well as the impact of the
to have FI and lack of awareness of need for bowel move- problem and past treatments on the patient’s quality of life.
ments. In cases of myelomeningocele, the nerve supply, both Specific scoring scales and quality of life scales are discussed
sensory and motor, is disturbed in a variety of ways, lead- in more detail in the section “Etiology and Scoring.”
ing to various forms of incontinence. Any form of trauma,
neoplasm, vascular accident, infection, or demyelinating Physical examination
disease to the central nervous system or spinal cord can
interfere with normal sensation or motor function, leading When beginning the physical examination, undergarments
to incontinence. In addition, diabetic patients with auto- or pads should be inspected for staining by stool, mucus,
nomic neuropathy may have impaired reflex relaxation of or pus. In addition, the perineum must be inspected and
the internal sphincter (35). the perineal body measured. A decreased length of the
perineum is frequently associated with a defect of the exter-
nal anal sphincter after sphincter injury. By simple retrac-
tion of the gluteal muscles, the large patulous anus, for
DIAGNOSIS example, that occurs with rectal prolapse can be recognized
easily and large prolapsing hemorrhoids or evidence of pru-
HISTORY AND PHYSICAL ritus may suggest seepage and soiling. Scars from previous
anorectal operations or episiotomies may also be identified.
History With straining, perineal descent or mucosal or full-thick-
ness rectal prolapse may become obvious. Examination in
A proper history can delineate the severity of the patient’s the sitting position or aided by Valsalva may be necessary to
symptoms and the likely etiology. In addition, treatment demonstrate prolapse.
recommendations are based on the cause and characteris- Digital rectal examination can reveal the strength of the
tics of the incontinence with assessment of the sphincter sphincters (resting tone and augmentation on squeeze) or
status. True incontinence should be distinguished from large sphincter defects. However, the assessment of anal
seepage and soiling, which may be associated with a variety tone is, at best, a very indistinct barometer of sphincter
Diagnosis / Testing 211
function. The assessment of the strength of voluntary best tool for the assessment of sphincter defects associated
sphincter contraction is subjective. Contraction of the with FI. In the last decade, the use of three-dimensional
puborectalis at the tip of the finger versus contraction of the (3D) techniques has increased, and studies have validated
external sphincter over the midportion of the finger may that compared to two-dimensional (2D) ultrasound, 3D
be distinguished and the anorectal angle can be assessed. ultrasound has an improved concordance with sphincter
Anoscopic and proctosigmoidoscopic examinations may findings at the time of operation (44).
demonstrate inflammatory or neoplastic processes contrib-
uting to the patient’s FI. Anorectal manometry
TESTING Resting and squeeze pressures, measures of rectal compli-

ance, and the rectoanal inhibitory reflex (RAIR) are obtained
Multiple tests are available for the evaluation of FI, but the during a manometry study. The absence of the RAIR raises
usefulness of each test depends on the cause of incontinence the suspicion for Hirschsprung disease. However, in addi-
and the likely proposed treatment. Practically speaking, tion, the RAIR may be absent after low anterior resection.
many investigations do not influence the choice of initial Basal pressure is thought to represent the activity of the
treatment. If conservative measures fail, further investiga- internal sphincter, although the spontaneous activity of
tion, particularly with endorectal ultrasound or manometry the external sphincter also affects maximal basal pressure.
may be useful. Squeeze pressure is secondary to the voluntary function of
the external sphincter and the pelvic floor muscles. If both
Ultrasonography basal and squeeze pressures are low, patients are likely to be
totally incontinent. If only the voluntary function (squeeze
Endoanal ultrasound is easily available and accurate. It has pressure) is low, the patients are likely partially incontinent
been shown to be superior to electromyographic mapping, (45). External sphincter function is critical for achieving
anorectal manometry, and physical examination (sensitivity continence of solid stool (46).
for sphincter defect detection 100% versus 89% versus 67% Penninckx et al. (47) studied the relationship between
versus 56%) (38). Based on these and other studies, endo- incontinence symptoms and the results of manometric
anal ultrasound is considered to be the gold standard diag- data. Values >40 mmHg for maximum basal pressure
nostic tool for the assessment of sphincter injury. However, and >92 mmHg for squeeze pressure could identify conti-
interpretation of ultrasound images of the external anal nent patients with 96% accuracy and incontinent patients
sphincter and associated defects can be subjective, operator with 88% accuracy. Unfortunately, there is a 10% overlap
dependent, and difficult to differentiate from normal ana- between the manometric values obtained from incontinent
tomical variations. Differentiation of normal variants from and continent patients (46). Following childbirth, pudendal
sphincter defects can be difficult, especially in the upper nerve damage increases the risk of FI in women with anal
part of the anal canal in female patients, which is due to sphincter injury, but manometric findings are not able to
asymmetry of the external anal sphincter at that level (39). differentiate these patients from patients without pudendal
In 75% of asymptomatic nulliparous women, Bollard et al. nerve injuries and, therefore, the manometric readings do
found a natural gap in the anterior external anal sphincter, not correlate to the likelihood of incontinence (48). In addi-
just below the level of the puborectalis. According to these tion, manometric values do not correlate with the severity of
authors, the gap explains the difficulties of interpretation of incontinence, and they do not predict the success of opera-
postpartum ultrasounds (40). Sentovich et al. evaluated the tive intervention (49). Finally, normal manometric readings
accuracy and reliability of endoanal ultrasound for anterior do not fully exclude incontinence (45).
sphincter defects (41). In incontinent, parous women, the
sphincter defects detected by ultrasound were confirmed Defecography
at the time of operation in 100% of the cases. However, in
continent, nulliparous women, the two ultrasonographers The anorectal angle is more obtuse (increased) in patients
identified sphincter defects in 55% and 75%, respectively, with incontinence (49). Voiding defecography or balloon
which may be partially explained by the difficulty in delin- proctography can demonstrate this increased angle. This
eating the anterior sphincter anatomy. To reduce the rate examination will not likely add information about the etiol-
of false positives, some advocate measuring perineal body ogy or treatment of fecal incontinence but has been shown
thickness. Zetterstrom et al. reported that the perineal to identify an occult internal intussusception.
body thickness was 6 ± 2 mm in patients with an anterior
sphincter defect and 12 ± 3 mm without (42). Pudendal nerve terminal motor latency
Endoanal ultrasound is associated with substantial
interobserver variability in assessment of the thickness Although the severity of nerve injury does not appear to
of the sphincters; however, the qualitative assessment of influence the severity of incontinence, it may affect the out-
sphincter defects is similar among observers (43). Despite come of sphincter repair (50,51). Assessment of the puden-
several disadvantages, endoanal ultrasound is, to date, the dal nerve terminal motor latency (PNTML) provides a
useful tool in defining pathology of the pudendal nerves. addition, patients should be given a trial of conservative
Roig et al. (52) found pudendal neuropathy in 70% of their symptomatic management prior to proceeding with surgi-
patients with FI (59% in patients with a sphincter defect, 94% cal intervention. An algorithm is proposed in Figure 23.1.
in patients without a sphincter defect). Therefore, puden-
dal neuropathy is a likely etiologic or, at least, contributing MEDICAL THERAPY
factor in FI. Laurberg et al. demonstrated that pudendal
neuropathy affects surgical treatment, and in their series, Initial treatment of FI should be conservative. Dietary
sphincter repair was successful in 80% of patients without changes, including fiber supplementation, and perineal
neuropathy and in only 10% of patients with neuropathy exercises are often recommended for patients with FI, but
(51). However, the relationship between pudendal nerve generally are useful in patients with incontinence to liquid
injury and surgical outcome is not universally accepted. and not solid stool. Rosen et al. (57) reviewed the various
Rasmussen et al., Chen et al., and Young et al. were unable antidiarrheal agents that can be used in patients with incon-
to identify a relationship between pudendal neuropathy and tinence to liquid stool. Substances such as kaolin, activated
a poor outcome after sphincteroplasty (53–55). charcoal, pectin, and bulk-forming agents such as fiber act
on the intestinal contents to solidify them. Agents such as
Magnetic resonance imaging bismuth salts and astringents such as aluminum hydrox-
ide may produce a barrier between intestinal contents and
Magnetic resonance imaging is comparable to endoanal the intestinal wall, which can be particularly useful in the
ultrasound in reliability of identification of sphincter presence of inflammation. Anticholinergic agents such as
defects, and its use is based on surgeon and institution atropine act to inhibit intestinal secretion and gut motil-
preference. Magnetic resonance imaging has been shown ity, but at therapeutic doses these drugs may produce trou-
to be more accurate in identifying sphincter atrophy than bling side effects. The opium derivatives such as tincture of
ultrasound, a finding that may correlate with the outcome opium, paregoric, and codeine act directly on the smooth
of sphincter repair (56). muscle of the intestinal wall, but the risk of addiction limits
long-term use. One of the most frequently used antidiar-
rheals is loperamide (Imodium), which inhibits intestinal
motility by directly affecting the circular and longitudinal
TREATMENT muscles of the intestinal wall. It both solidifies the stool
and increases rectal compliance, both of which decrease
Underlying disorders (IBD, rectal prolapse, carcinoma, etc.) urgency. In addition, it has also been shown to increase rest-
should be optimally managed prior to treatment of FI. In ing anal pressures (58) and, therefore, improve continence
Treatment of
underlying
etiology and
comorbidities
Sphincteroplasty
Sphincter Fail,
Especially no Yes
injury early post- prior
partum SNS
Detailed
history Stool bulking
Consider EAUS:
and +/– Fail Sphincter
physical biofeedback SNS
defect
Remote/ present
no
sphincter
Consider Multiple
injury
failed
procedures,
No significant
Consider comorbidities
Minimally
invasive,
experimental
therapies Artificial Ostomy
sphincter
Figure 23.1 Proposed algorithm for the treatment of FI. (SNS: sacral nerve stimulation, EAUS: Endoanal ultrasound.)
Treatment / Biofeedback 213
after restorative proctocolectomy (59). For patients with diabetic patients exhibit multiple abnormalities of anorectal
certain neurologic conditions, the regular administration sensory and motor functions. Pharmacologic treatment and
of enemas may achieve a level of social continence by emp- dietary interventions to decrease diarrhea, as well as bio-
tying the rectum and left colon of stool. feedback to improve rectal sensory thresholds and striated
Amitriptyline, a tricyclic antidepressant agent with muscle responsiveness, may be successful especially in this
anticholinergic and serotoninergic properties, has been patient population.
used empirically in the treatment of idiopathic FI. Santoro Several literature reviews have been performed to deter-
et al. (60) conducted an open study to test the response to mine the efficacy of biofeedback in the management of FI.
amitriptyline 20 mg daily for 4 weeks in 18 patients with Norton et al. conducted a Cochrane Review of controlled
idiopathic FI. Amitriptyline improved incontinence scores studies of biofeedback and sphincter exercises for FI.
(scale 1–18; median pretreatment score 6 versus median Twenty-one studies met the inclusion criteria of being a ran-
posttreatment score 3, p < 0.001). Treatment also reduced domized or quasi-randomized trial, which included 1,525
the number of bowel movements per day and improved participants. The authors concluded that they were not able
symptoms in 89% of patients with FI. The major effects of to definitively evaluate biofeedback with the number and
amitriptyline are a decrease in the amplitude and frequency quality of studies available. They did find that biofeedback is
of rectal motor complexes and an increase in colonic transit enhanced with electrical stimulation in addition to the exer-
time, which lead to the formation of a more firm stool that cises (64). Heymen et al. searched the Medline database for
is passed less frequently. In combination, these results may papers published between 1973 and 1999 including the terms
be the source of the improvement in continence in patients “biofeedback” and “FI,” which included 35 studies. Only
treated with amitriptyline. six studies used a parallel treatment design, and just three
Stool bulking with fiber plus or minus a low-dose antidi- of those randomized subjects to treatment groups. A meta-
arrheal is used routinely as first-line treatment for incon- analysis comparing the treatment outcome of studies using
tinence to predominately liquid stool in our practice. This coordination training (i.e., coordinating pelvic floor muscle
combination has a low risk of severe side effects and can be contraction with the sensation of rectal filling) to studies
efficacious. In addition, if a patient has a history of chole- using strength training (i.e., pelvic floor muscle contraction
cystectomy and has FI to liquid stool, a bile acid sequestrant alone) failed to show any advantage for one treatment strat-
is utilized. egy over another. The mean success rate was 67% and 70%,
respectively. Despite these positive results, the authors state
BIOFEEDBACK that the conclusions of the reviewed studies are limited by the
quality of the studies available (65). The largest randomized
Engel et al. (61) first described biofeedback training for FI, controlled trial has been conducted by Norton et al. They
which involves screening patients for incontinence, partic- randomly assigned 171 patients with FI into four treatment
ularly to liquid stool, and selecting motivated, cooperative arms: (1) standard care; (2) standard care plus instruction
patients for a three-phase instruction of voluntary control in sphincter exercises; (3) same as (2) plus computer-assisted
mechanisms. There are at least three components to biofeed- biofeedback involving coordination techniques; (4) same
back treatment: (1) exercise of the external sphincter muscle, as (3) plus daily use of an electromyography (EMG) home
(2) training in the discrimination of rectal sensations, and trainer device. About half of the participants in all four
(3) training synchrony of the internal and external sphinc- groups who completed their treatment protocol showed
ter responses during rectal distention (62). All or some of improvement, which was maintained at 1-year follow-up.
these components may be effective for some patients. These data indicate that improvement can be realized with-
Biofeedback involves placement of a balloon in the rec- out sphincter exercises and without biofeedback. Patient and
tum and connection of pressure transducers to a graph therapist interaction and the development of better coping
to give the patient a visual feedback corresponding to his strategies seem to be important factors associated with suc-
or her sphincter responses to command. Initially, large cess (66). Another randomized controlled study, conducted
amounts of air are injected into the rectal balloon; gradu- by Solomon et al., revealed that instrument-guided biofeed-
ally the volume of distention is reduced until the patient back offers no advantage over simple pelvic floor retraining
can sense small distention and contract the external anal with digital guidance alone (67).
sphincter. Subsequently, visual feedback is eliminated, and The mechanisms for the effect of biofeedback are not
the patient is tested to respond to rectal sensations alone. clear. It has been suggested that biofeedback is beneficial
Training occurs weekly generally for 4–8 weeks and is sup- by improving the contraction of the external anal sphincter
plemented by at-home sphincter exercises. The goals of this and the pelvic floor muscles due to strength training. Initial
training are to increase the strength of external sphincter attempts to demonstrate objective manometric changes
contraction and detect and respond to small volumes of rec- secondary to biofeedback have proved difficult. Fynes et al.
tal distention. conducted a randomized controlled trial to compare the
One major disadvantage is the time required for ther- effects of biofeedback alone with those of biofeedback com-
apy as each session takes at least 2 hours and involves a bined with electrical stimulation. The manometric param-
significant amount of equipment. Wald (63) reported that eters did not change after the biofeedback alone, whereas
anal resting and squeeze pressures increased after com- It has been suggested that biofeedback is also beneficial
bined biofeedback and electrical stimulation (68). Beddy as an adjuvant therapy following anal sphincter repair.
et al. observed a significant improvement in anal resting Davis et al. performed a randomized controlled trial where
pressure, duration of the squeeze, and amplitude of the 38 patients were randomly assigned to sphincter repair or
squeeze after EMG-guided biofeedback, but no improve- sphincter repair plus biofeedback. Shortly after surgery,
ment in the squeeze pressure (69). Biofeedback might also there was no difference in functional outcome between the
work by enhancing the ability to perceive and respond to two groups. More studies are warranted to elucidate the role
rectal distensions, known as sensory training. Chiarioni of adjuvant biofeedback (78).
et al. reported that sensory retraining is the key to biofeed-
back treatment of FI. Although they observed an increase
of maximum squeeze pressure and squeeze duration after
biofeedback, the sphincter strength alone did not separate OPERATIVE TECHNIQUES
responders from nonresponders. However, responders had
lower thresholds for first sensation (70). Critics of this treat- Once conservative management has failed, operative inter-
ment modality argue that the improvement is a result of vention is indicated. All operative techniques except sacral
the supportive interaction between the physiotherapist and nerve stimulation involve preparing the patient by evacu-
the patient, resulting in decreased anxiety and increased ation of the large bowel with laxatives and enemas or oral
confidence. Despite many unanswered questions, it seems lavage solutions. At the time of operation, an indwelling
obvious that biofeedback is beneficial for more than half of urethral catheter is placed and maintained until decreased
the patients with FI, at least in the short term. pain permits voluntary voiding. In all cases, perioperative
Another question that remains is whether the outcome broad-spectrum antibiotics are administered as per the
of biofeedback can be predicted. Manometric parameters, institution’s Surgical Care Improvement Project guidelines.
except for increased cross-sectional asymmetry, do not pre-
dict response to biofeedback therapy (71). Another study OVERLAPPING SPHINCTEROPLASTY
revealed that incomplete anal relaxation during straining
adversely affects the outcome of biofeedback (72). The long- External anal sphincter defects, most frequently in the ante-
term results after biofeedback have also been questioned. rior sphincter, are the principal cause of FI secondary to
Most studies offer a follow-up of less than 2 years. Enck obstetric trauma. These anatomic defects can be treated by
et al. posted a questionnaire to patients who were treated an anterior anal sphincter repair, which is the classic surgi-
by biofeedback 5–6 years earlier. The same questionnaire cal procedure for an early (4 months to several years) post-
was also sent to patients who had not entered the treatment partum sphincter injury. Most surgeons use an overlapping
program. In both groups 78% of the patients experienced technique to repair the divided external anal sphincter;
episodes of incontinence. However, the severity of incon- however, a primary or end-to-end repair may be employed
tinence was significantly less in the treatment group. Five very early (less than 3 months) after sphincter injury, as
to 6 years after the treatment, the severity of incontinence there is no significant scar, and muscle tissue is healthy.
was similar to that reported immediately after therapy (73). The technique of sphincteroplasty, as applied by Fang
However, two other studies revealed deterioration over time et al. (79) and Parks and McPartlin (80), provides good to
(74,75). Ryn et al. reported an overall success rate of 60% excellent results in most patients who have adequate residual
immediately after the treatment. This dropped to 41% after a muscle mass. The operation is performed with the patient
median follow-up of 44 months (76). Based on this deterio- in the prone jackknife position, with the buttocks elevated
ration over time, it has been suggested that it could be useful over a Kraske roll. Anesthesia may be either regional or gen-
to reinitiate biofeedback training. Pager et al. were not able eral, but local anesthesia is also utilized.
to demonstrate this worsening with time. At a median of The first step is the mobilization of the anoderm from the
42 months after completion of the training program, 75% underlying sphincter mechanism and scar tissue via a cur-
of their patients still perceived a symptomatic improve- vilinear incision that parallels the outer edge of the external
ment, and 83% reported improved quality of life. They sphincter. The incision should extend no more than 180°,
also observed that patients continued to improve during depending on the amount of scar tissue present (Figure
the years following the training, possibly due to the strong 23.2a). Further incision risks injury to the pudendal nerves
emphasis placed on them to continue the exercises on their laterally. Cephalad mobilization should extend approxi-
own (77). Because biofeedback treatment is multimodal, mately to the distal edge of the anorectal ring. The entire
more studies are needed to establish selection criteria, to sphincter mechanism is then dissected widely from its bed,
compare different biofeedback techniques, and to establish and care is taken to preserve the branches of the pudendal
valid endpoints and follow-up. Although biofeedback is nerves as they enter into the muscle posterolaterally. Wide
time consuming and labor intensive, it is noninvasive and dissection permits approximation without tension. In one
safe. Based on the reported outcomes, it can be considered approach, the entire sphincter mechanism is sectioned
as initial treatment in patients with FI, especially predomi- transversely through the middle of the scar tissue, with pres-
nantly to liquid stool and with intermittent incontinence. ervation of the scar for suture placement. The muscle ends
Operative techniques / Overlapping sphincteroplasty 215
(a) are overlapped to decrease the size of the anal aperture until
it fits snugly over the index finger. Multiple mattress sutures
are carefully placed to maintain the desired aperture. The
suture used is generally a 2–0 synthetic slowly absorbable
suture. The sphincter should lay easily overlapped, because
the separation of the ends of the sphincter is a sign of inad-
equate mobilization of the muscle from its bed and will pre-
dispose to separation at the suture line. When all sutures
have been placed, they are pulled tight, and the orifice is
checked again to ensure proper placement of the sutures
and properly sized aperture and the sutures are then tied.
A second approach involves preservation of midline scar,
and imbrication of the tissue, and then the performance of
overlapping repair.
If the perineal body is thin, attempts should be made
to bulk it, which is performed by an anterior levatorplasty.
(b) Tissues from each side of the perineum (transverse perinei
muscles and/or scar tissue) are approximated in the midline.
This reconstruction supports the anovaginal septum and
effectively separates the anal orifice from the introitus. The
anoderm is sutured carefully over the sphincter with inter-
rupted or running absorbable sutures. The horseshoe-shaped
defect outside the muscle is partially closed, and the remain-
der is packed open with gauze or over a drain (Figure 23.2c).
Postoperative management varies based on surgeon
preference. The recent trend has been toward early feeding.
Although there is concern for subsequent need for laxa-
tives, we administer opiates to decrease the pain and fre-
quency of bowel movements. Mahony et al. (81) conducted
a randomized trial designed to compare a laxative regimen
with a constipating regimen in early postoperative manage-
ment after primary obstetric and anal sphincter repair. A
(c)
total of 105 females were randomized after primary repair
of a third-degree tear to receive lactulose (laxative group,
56) or codeine phosphate (constipated group, 49) for 3 days
postoperatively. The first postoperative bowel movement
occurred at a median of 4 days in the constipated group
and 2 days in the laxative group. Patients in the constipated
group had a significantly more painful first evacuation
compared with the laxative group. Continence scores, anal
manometry, and endoanal ultrasound findings were similar
in the two groups at 3 months postoperatively. Sitz baths are
given two to three times a day for comfort and to wash away
secretions. Some surgeons are concerned about skin mac-
eration and prefer to irrigate the wound with warm saline
solution or even diluted hydrogen peroxide (dilute 1:4) to
provide both comfort and cleanliness. With the introduc-
tion of food, fiber supplementation is started to eliminate
Figure 23.2 Overlapping sphincteroplasty. (a) The any straining at defecation and disruption of the suture
sphincter muscle is dissected free on both sides with repair. Performing a diverting stoma is not required. An
care to preserve the scar tissue attached to the muscle to alternative strategy is to teach the patient to irrigate his or
improve the strength of the repair. (b) The muscle is over-
her rectum by sliding a catheter posteriorly along the natal
lapped and sutured in place with mattress sutures. (c) The
wound is closed with a drain or packing in the center to
cleft and into the anal canal, away from the surgical site.
facilitate drainage. (With permission from Gurland B, Hull This may avoid issues of delayed constipation or impaction
T. Wexner SD, Fleshman JD (eds). Master Techniques in upon discharge.
Surgery. Colon and Rectal Surgery: Anorectal Operations. Numerous reports have been published on the short-term
Philadelphia, PA: Wolters Kluwer, 2012.) outcomes of sphincter repair, and overall, initial success is
positive, with approximately 60% of patients achieving sig- were identical in both groups. Resting pressures and maxi-
nificant benefit. However long-term success has not been as mal squeeze pressures were no different, and subjective suc-
durable. In one of the larger studies published, Karoui eval- cess scores were also no different (91).
uated 86 patients undergoing sphincteroplasty. At 3 months Briel et al. conducted a prospective study looking at
postoperatively, 30% of patients were totally continent, and whether bulk overlapping repair was superior to separate
an additional 33% were incontinent only to gas. However, at internal and external sphincter repair. In this study, 31
40 months follow-up, <30% of patients remained fully con- patients underwent separate internal and external sphinc-
tinent and >70% were incontinent to either gas or feces (82). ter repair, and 24 patients underwent standard overlapping
Malouf et al. evaluated 55 consecutive patients undergoing repair. There was no statistically significant difference in
overlapping sphincteroplasty as a result of obstetric injury. these two groups (92).
At 15 months postoperatively, 42/55 patients were conti- Hasegawa et al. (93) conducted a randomized trial to
nent to both solid and liquid stool. At 5 years follow-up, no assess whether fecal diversion would improve primary
patient was fully continent to both solid and liquid stool, wound healing and functional outcome after sphincter
which shows the deterioration of continence over time (83). repair. Patients were randomly assigned to a defunction-
Similarly, Halverson reported on a series of 71 consecutive ing stoma (n = 13) or no stoma (n = 14). The incontinence
patients undergoing sphincteroplasty assessed at a median score improved significantly in both groups (stoma 13.5–
of 69 months after surgery. Forty-nine (69%) were available 7.8; no stoma 14–9.6). There was no significant difference
for follow-up, and of these patients, four patients had under- in the functional outcome or the number of complications
gone fecal diversion and 54% were incontinent to liquid or of sphincter repair between the groups. However, stoma-
solid stool, while only six patients (14%) remained fully con- related complications occurred in 7 of 13 patients (parasto-
tinent (84). mal hernia, 2; prolapsed stoma, 1; incisional hernia at the
Attempts at determining predictive factors for success stoma site requiring repair, 5; and wound infection at the
have evaluated age, pudendal nerve injury, and type of closure site, 1). They concluded fecal diversion in sphincter
repair. Despite the wealth of literature, there are no clear repair is unnecessary because it gives no benefit for wound
answers. Age of the patient at the time of sphincter repair or functional outcomes and itself is a source of morbidity.
has been assessed by several investigators. Simmang et al. Despite mixed data, sphincteroplasty should remain
evaluated 14 patients with ages ranging from 55 to 81 a viable approach to address continence in patients with
years, where almost all of the patients reported improve- sphincter injuries, especially in the first years follow-
ment in symptoms and half reported complete continence. ing injury (Figure 23.1). In our practice, this procedure is
In this admittedly small series, advanced age did not seem reserved for young women with sphincter defects with the
to predict failure (85). Rasmussen assessed postoperative knowledge that they may need another operation in the
continence in 24 women under the age of 40 years and in future when the repair loses its durability.
14 women older than 40 years and found a significant dif-
ference in postoperative continence in the older cohort. This OTHER SPHINCTEROPLASTIES
was hypothesized to be, at least partially, related to weak-
ening of the pelvic floor (86). However, patient perception Postanal repair
may also play a role. Young evaluated 57 women undergoing
overlapping sphincter repair and found that 78% of patients Prior to the introduction of endoanal ultrasound, most
younger than 40 deemed the repair a success compared to cases of FI were classified as “idiopathic” or neurogenic. For
93% of patients in the older group, while formal inconti- the treatment of patients presenting with this type of incon-
nence scores improved equally in both groups (54). tinence, Parks devised the postanal repair, and he believed
Pudendal nerve injury as evidenced by prolonged that this procedure worked by restoring the anorectal angle
PNTML has often been cited as one of the etiologies of post- and increasing the length of the anal canal. Several stud-
obstetric injury incontinence. Numerous publications have ies, however, have revealed that a postanal repair does not
attempted to assess this, and data are still divided. This is result in a significant change of the anorectal angle (94–98).
supported by studies by Barisic, Londono-Schimmer, and However, several studies have demonstrated an increase in
Giliand (87–89) comparing groups with and without puden- the length of the anal canal after successful postanal repair
dal neuropathy showing a significant difference in inconti- (9,94,99). Conflicting data have been reported regarding
nence scoring after sphincteroplasty between the groups; the impact of postanal repair on anal pressure. Some have
however, this is refuted by multiple studies, including found that resting and squeeze anal pressure increase after
those by Bravo-Gutierrez, Halverson, Malouf, and Koroui successful postanal repair (95–97,100,101). However, oth-
(82–84,90). Tjandra attempted to evaluate whether repair ers have shown that postanal repair does not affect anal
technique, overlapping versus end-to-end repair, was asso- pressure (102,103). Due to the lack of consistent changes in
ciated with different functional outcomes. In this study, 23 anatomy and physiology, it is unclear why postanal repair is
patients with anterior defects underwent sphincter repair, effective in some patients, but it might be due to lengthen-
12 randomized to end-to-end repair and 11 to overlapping ing and narrowing of the anal canal. Van Tets and Kuijpers
repair. At a median follow-up of 18 months, Wexner scores suggested that this procedure might improve continence by
Operative techniques / Sacral neuromodulation 217
a placebo effect and not by enhanced muscle function (104). in 2011 for the indication of FI in patients who have failed
For the reasons above and the lack of durability of the repair best conservative therapy.
when it does improve symptoms, the procedure has fallen To provide nerve stimulation, a quadripolar lead elec-
out of favor. trode is placed transcutaneously and passed via the sacral
foramen to follow the path of the S3 nerve root (Figure 23.3).
GRACILIS MUSCLE TRANSPOSITION An initial test phase with a temporary external pacer is
(GRACILOPLASTY)
(a)
Several advanced techniques have been described using
Approx
pedicled muscle flaps to replace a damaged or nonfunc- 1.5 cm
tional sphincter. In general, due to technical difficulty, high Marked location-S3 60°
rates of morbidity, and the success of newer, less invasive angle
techniques, graciloplasty is rarely performed and may only
be accessible in select situations and at specialized centers
where expertise and experience may maximize results.
This technique was first described more than a century
ago by Chetwood and renewed as a technique for use in
pediatric FI by Pickrell (105). More recently, it was cham-
pioned by Wexner as a salvage treatment after catastrophic
sphincter damage in otherwise healthy patients. In this
technique, the gracilis muscle is harvested as a pedicled flap.
The muscle is tunneled around the sphincter complex and
Points of
sutured in place. Initially described as only a muscle trans- insertion
position, the wrapped muscle functions as a biologic cer-
clage, akin to the Thiersch procedure. In addition, patients (b) (External)
learn techniques to voluntarily contract this muscle to aug- rubber ground pad
ment control, at the expense of altered gait (106).
To additionally augment the procedure, electrostimula-
tion is employed in order to convert the more easily fatigued
fast-twitch skeletal muscle fibers to slow twitch fibers via
neuromodulation resulting in relatively tonic contraction. Temporary
lead
This was dubbed a “dynamic graciloplasty.” Isolated series
produced favorable results; however, due to surgical diffi-
culty and high complication rates, it was rarely performed.
Wexner, reporting on one of the largest series, showed that
at 2 years >60% of patients had a significant improvement
in quality of life and incontinence scores (107). Long-term
Test
success was less favorable as Thornton and others showed stimulator
a significant decrease in success at 5 years, with only 16%
maintaining continence, and overall complication rates
>70%. Therefore, this procedure has also largely been
abandoned.
SACRAL NEUROMODULATION
Sacral nerve stimulation (SNS) (Interstim, Medtronic,
Minneapolis, Minnesota) has emerged as the most prom-
ising modality in the treatment of medically refractory FI. Figure 23.3 SNS. (a) The needle is passed at the level
The treatment was originally studied and developed for the of the S3 nerve root at a 30° angle, and the electrode is
treatment of urinary incontinence, but its efficacy for the passed through the needle. The electrode is stimulated to
treatment of FI quickly became apparent due to the inci- confirm position along the S3 nerve root. (b) A temporary
dence of mixed urinary incontinence and FI. Clinically, simulator is attached to the electrode for the trial period.
If the patient responds to treatment, a permanent simula-
it seems to be more efficacious for the treatment of FI
tor is placed in a subcutaneous pocket. (With permis-
(108). Following the 2010 American Society of Colon and sion from Matzel KE. In: Wexner SD, Fleshman JD (eds).
Rectal Surgeons presentation and subsequent publications Master Techniques in Surgery. Colon and Rectal Surgery:
(109,110) based on multi-institutional trial data, SNS was Anorectal Operations. Philadelphia, PA: Wolters Kluwer,
approved by the U.S. Food and Drug Administration (FDA) 2012.)
utilized to test efficacy before the implantable stimulator 5-year follow-up for the Australian experience in 53 patients
is placed. The minimum definition of success of the trial and reported that mean Wexner scores improved from a
period (1–3 weeks) is a 50% reduction in the number of baseline of 11.5 to 8 (116). Michelsen published the Danish
episodes of fecal incontinence. If needed, two separate test experience with a 6-year follow-up of 126 patients and dem-
phases can be used. If preferred, an office-based temporary onstrated mean Wexner score improvement from a baseline
unipolar nontined lead can be placed using either anatomic of 20 to 7 (117). The European SNS outcomes study group
landmarks or with fluoroscopic guidance rather than utiliz- reported on 7-year outcomes in a multinational study incor-
ing placement in the operating room. However, as this is a porating 10 European centers and 407 patients with a mean
nontined lead, this test phase lasts for a maximum of 3–7 follow-up of 84 months (118). Side-by-side comparison of
days because of the high likelihood of lead dislodgment. multiple incontinence scoring parameters including num-
This test is best suited for patients with frequent FI episodes, ber of incontinent episodes, Wexner score, and St. Mark’s
often one or more per day. score all showed dramatic and significant improvements
Stimulator programming is based on best motor persisting to 7 years of follow-up. George (119) published a
responses obtained (therefore, if the procedure is performed 10-year follow-up study of 25 patients. Ninety-two percent
in the operating room, only light sedation is performed). A of patients still had a >50% improvement, and full conti-
successful lead placement causes bellowing of the levators nence was maintained in almost 50%. In an attempt to
and toe turning of the great toe, which indicate specific S3 define predictive variables for success, Brouwer and Duthie
stimulation. If the test period reveals a >50% reduction in evaluated a cohort of patients with 4 years of follow-up
the number of FI episodes, the stimulator is placed in a sub- looking at variables including sphincter defect, neuropathy,
cutaneous pocket just inferior to the posterior superior iliac and prior sphincter repair. The therapy was overwhelmingly
spine. At typical settings, the current stimulator model has positive in all groups, regardless of these variables present
a battery life of approximately 5 years (111). (120). As a result of the overwhelming success of SNS, the
The seminal prospective trial validating SNS was con- effective indications of pelvic floor testing for these patients
ducted by Tjandra et al., published in 2008, and compared have decreased. Since no predictors of success are based on
SNS to best medical therapy. The study included patients manometric values, and since the technique works irrespec-
with FI varying etiologies and with sphincter defects of up tive of a sphincter injury, preoperative pelvic floor testing
to 120°. There were 120 patients in the initial cohort, 60 in is of negligible use and should not be considered routine,
the control group and 60 in the SNS group. Remarkably, or even necessary at all in patients deemed appropriate
90% of patients reported success with initial testing and for SNS.
moved on to implantation. Of this group, almost 50%
achieved perfect continence, compared to no improvement POSTERIOR TIBIAL NERVE STIMULATION
in the control group (112). The FDA qualifying trial echoed
Tjandra’s results, with an 87% success rate and an >40% Posterior tibial nerve stimulation (PTNS) was first used
rate of patients achieving perfect continence. Interim fol- for the treatment of urinary incontinence in 1983 by
low-up at 3 years was conducted by Mellgren and found that Nakamura (121). As with SNS, it was serendipitously found
success was sustained: 83% of patients still reported over- to have efficacy for the use of FI as well. Either transcu-
all success with a mean decrease in the number of FI epi- taneous or percutaneous electrodes are applied over the
sodes from a baseline of 9.4–1.7 (109,113). Forty percent of posterior tibial nerve and stimulation is typically per-
this cohort still reported perfect continence. Complications formed twice daily, for 20-minute sessions over a 3-month
in this trial were minimal, the most frequent complication period. An early series was published by Shafik in which
being implant site pain in 28%, with a 10% infection rate 32 patients with medically refractory FI were treated with
noted. Hull reported on 5-year follow-up in this cohort in percutaneous stimulation, and significant improvement in
2013. Impressively, 89% of patients still reported success FI was achieved in >78% of the patients (122). Leroi, using a
with therapy, and 36% still reported maintenance of perfect modified transcutaneous technique, performed a random-
continence. Rates of complications remained similar to the ized prospective trial on 144 patients. Almost all patients
earlier studies. They demonstrated that over 5 years, 24.4% in the treatment group showed improvement in FI scores
of patients underwent at least one revision or replacement compared to only 27% in the sham group, though this
of the stimulator (114). It should be noted that in this trial, did not reach statistical significance (123). A more recent
patients with sphincter defects of up to 60° were included, systematic review identified nearly 300 patients treated
and this had no effect on overall success. Longer-term with tibial nerve stimulation. Success, defined by at least
data for this therapy have been published by numerous a 50% improvement in incontinence scores, was achieved
groups, primarily from Europe. In the Italian SNS registry, by between 63% and 82% of patients (124). Thin et al. per-
Altomare published 5-year follow-up on 52 patients. Mean formed a randomized clinical trial comparing tibial nerve
Wexner incontinence scores decreased from a baseline of stimulation to SNS. Although SNS showed greater suc-
15 to 5. Seventy-four percent of patients had at least a 50% cess than tibial nerve stimulation, both treatments showed
improvement in the number of FI episodes with full conti- clinical efficacy (125). Knowles reported on the largest
nence maintained in 20% of patients (115). Lim published randomized sham-controlled trial including 227 patients.
Operative techniques / Artiﬁcial bowel sphincter 219
Compared to 31% in the sham group, 38% of patients in the little positive data reflecting any durable success with this
treatment group achieved a >50% reduction in the num- modality. Guerra followed a cohort of 19 patients with a
ber of weekly incontinent episodes (126). Edenfield recently mean follow-up of 7 years who underwent treatment with
published a systematic review inclusive of 15 studies. The Durasphere, PTQ, or Solesta. Patients underwent clinical
group was composed of 745 patients, and although the assessment, anal manometry, and ultrasound evaluation.
majority of the studies were of poor quality, both percuta- In this group, the vast majority of implants were no longer
neous and transcutaneous approaches showed significant detectable or clinically effective (132).
improvement over controls (127). Bulking agents have largely been abandoned by colorec-
PTNS remains an interesting potential option for moti- tal surgeons for the treatment of FI. A niche may still exist
vated patients who may be otherwise ineligible for standard in patients with seepage and soilage, especially after anorec-
sacral neuromodulation; however, it is not currently FDA tal surgery.
approved for this indication.
ARTIFICIAL BOWEL SPHINCTER
INJECTABLES
The artificial bowel sphincter (ABS), developed by American
A recent modality used to treat FI is injection of bulking Medical Systems (Minnetonka, Minnesota), utilizes an
agents, which are synthetic or biomaterials injected into the inflatable cuff that is tunneled around the native sphincter
tissue around the anal canal in order to bulk the area and complex. The cuff is controlled by a fluid-filled pressure-
cause a relative physical obstruction. Numerous materials regulated balloon that is implanted anterior to the bladder
have been tested including autologous fat, collagen, as well and controlled manually by an actuator implanted in the
as slowly absorbable biomaterials including hydrogel cross- scrotum or labia majora. First reported on by Christiansen
linked with polyacrylamide synthetic calcium hydroxyapatite and Sparso (133), 12 patients with anal incontinence due
ceramic microspheres, silicone biospheres (PTQ, Cogentix to neurologic disease or failure of previous incontinence
Medical Incorporated, Minnetonka, Minnesota), carbon- surgery underwent implantation of an artificial bowel
coated beads (Durasphere EXP, Coloplast Corporation, sphincter. The system used was a modification of the AMS
Minneapolis, Minnesota), and nonanimal stabilized dex- 800 artificial urinary sphincter. In two patients, infection
tranomer in hyaluronic acid (NASHA Dx – Solesta, Salix necessitated removal of the system, and in four patients
Pharmaceuticals, Raleigh, North Carolina) (111). eight revisional procedures had to be performed because of
The majority of published data surrounds the use of mechanical failure. Erosion through the anal canal did not
NASHA Dx, under the trade name Solesta. The seminal occur. Among 10 patients with the system in place for more
study, a randomized double-blind trial, compared a treat- than 6 months, the result was considered excellent in five,
ment arm with a sham saline injection arm. In this study, with only occasional leakage of flatus; good in three, with
52% of patients in the treatment group experienced a 50% occasional leaked liquid feces and flatus; and acceptable in
or more reduction in the number of incontinent episodes; two, in whom the cuff obstructed defecation. The authors
however, in the sham group, 31% also achieved this end- concluded that implantation of an artificial bowel sphinc-
point (128). Some significant adverse events were reported ter is a valid alternative to permanent colostomy in patients
including rectal and prostatic abscesses. Maeda performed with anal incontinence due to neurologic disorders and
a recent systematic review looking at all trials of injectable in patients in whom other types of incontinence surgery
bulking agents. Not surprisingly, as the majority of these have failed. Since the initial report, more data cast a ques-
were industry funded, many were found to be at high risk tion of device safety. Mundy published a systematic review
for bias. Only the Solesta trial showed a statistically signifi- in 2004 (134) showing encouraging initial functional suc-
cant improvement in continence. However, one of the tri- cess in two-thirds of patients. However, complication rates
als comparing silicone biospheres (PTQ) to carbon-coated were unacceptably high, including infection rates >20%
beads (Durasphere) showed some short-term advantages as well as mechanical failures resulting in explantation in
(129). Some of the key questions that have yet to be well more than half of patients. Darnis published an even more
answered include optimum dose and delivery method. In ominous series showing a >75% complication rate (135).
Maeda’s review, ultrasound-guided delivery was found to be Wong published a more balanced series of 52 patients with
superior to manually guided injection. A study of Solesta a greater than 5-year follow-up with a revision rate of 50%,
conducted by La Torre (130), looking at longer-term results, but explantation in only 27%, In this series, more than two-
found some durable efficacy at 24 and 36 months with just thirds of patients who retained their implants at 5 years had
over half of patients still maintaining a >50% reduction in a significant improvement in FI scores and quality of life
fecal incontinent episodes. scores (136). In order to better determine predictors of suc-
In order to identify predictors for failure, Hussain con- cess, Wexner followed a cohort of 51 patients over a 9-year
ducted a systematic review of all injectable materials and period. In this study, there was a 41% rate of infection, 18/23
found that the only significant predictors for failure were of which were early postop. Multivariate analysis showed
the use of local anesthetic for injection as well as the failure that prior perianal infection and time between implant and
to use laxatives in the postoperative period (131). There is first bowel movement were predictive of infection (137).
Overall, despite success in a highly select population, the that the needle electrodes start to penetrate the tissue 1 cm
overwhelming technical and infectious complications have distal to the dentate line. Additional lesions are created up to
resulted in the device no longer being implanted except at 1.5 cm above the dentate line in all four quadrants. Mucosal
select centers. temperature is cooled by surface irrigation. In this way,
thermal lesions are created in the muscle below the mucosa,
MAGNETIC ANAL SPHINCTER (FENIX) while preserving the mucosal integrity. In contrast to the
belief that RF therapy would cause scarring and tightening
With SNS successfully treating the majority of patients, of the anal canal, essentially causing mild obstruction akin
the group of patients who still fail has become smaller and to cerclage, new or histologic assessment of RF-treated tis-
more difficult to treat. Not surprisingly, new techniques sue reveals that nonablative RF energy causes morphologic
have taken inspiration from past failure. The magnetic anal changes in damaged sphincter muscle to become more his-
sphincter is a modern update of the cerclage technique tologically normal (141). Efron published some of the earli-
popularized by Thiersch called the FENIX (Torax Medical est data in a multicenter trial involving 50 patients. In this
Inc., Shoreview, Minnesota). This technique involves the cohort, mean Wexner scores improved from a baseline of
implantation of a string of titanium beads with magnetic 14.5 to 11.1 at 6 months. All quality of life parameters were
cores linked together by wire. As in the Thiersch procedure, also improved, and only minor complications were noted
this wire is tunneled around the native sphincter complex (142). Five-year data were reported by Takahashi showing
and secured in place. At rest, the magnetic cores are drawn that mean Wexner scores remain significantly improved
together, providing an occlusive force. The technology was from baseline of 14 to 8 with nearly 85% of patients showing
first used (and abandoned) for esophageal closure to pre- a >50% improvement (143). Ruiz reported on more modest
vent reflux. Force generated in the rectum during Valsalva is results at 2-year follow-up showing mild improvement in
enough to overcome the magnetic forces, and allow passage incontinence scores from 15.6 to 12.9 (144). Currently, few
of stool (111). centers are employing this modality, in favor of the more
Lehur first published a feasibility study in 2010, report- reliable results achieved with sacral neuromodulation.
ing on 14 patients. Complications were reported in 7/14
patients including one erosion into the anal canal. Of those ANTEGRADE CONTINENCE ENEMA
who maintained their implants, 5 patients at 6-month (ACE PROCEDURE)
follow-up had a >90% reduction in the number of FI epi-
sodes, and a significant improvement in the Wexner score The ACE procedure was first described by Malone in 1990
(17.8 to 7.8) (138). Barussaud published a prospective study (145). This treatment was developed for patients with dis-
on 23 patients implanted with the magnetic anal sphincter abling colonic motility disorders and difficulty with evacu-
with an 18-month follow-up. They reported that median ation of solid stool. In this description, the appendix is
incontinence scores decreased from 15.2 to 6.9 at 6 months. reversed and tunneled into the wall of the cecum, creating
With follow-up as long as 36 months, they reported that FI a one-way valve. The proximal end is brought through the
scores remained low, with a Wexner score of 5.3. Only two abdominal wall as a small stoma in the right lower quadrant,
patients required explantation due to infection (139). This flush with the skin. In order to eliminate, the appendicos-
device has been FDA approved under the Humanitarian tomy is intubated, and the colon is flushed in an antegrade
Device Exception. There is an ongoing SaFaRI trial com- manner with fluid in order to clear the colon of stool. This
paring the FENIX to SNS for FI (140). has been employed primarily in the pediatric population,
afflicted with congenital motility disorders, but is also rarely
RADIOFREQUENCY THERAPY used in adults. Several modifications of this procedure have
been described in order to simplify the technical rigors.
Radiofrequency (RF) therapy is a technique that was origi- Several reports have shown good results (146,147). Worsoe
nally described for use to treat gastroesophageal reflux dis- reported long-term results on a series of 80 patients with a
ease and was termed the Stretta technique. The adaptation mean follow-up of 75 months. In this cohort, there was an
for use in the anal sphincter is termed the Secca procedure overall success rate of 74%, with very positive subjective
(Curon Medical Incorporated, Fremont, California). This results (148). ACE can be considered in select patients as an
technique uses submucosally applied RF energy to induce option prior to colostomy.
tissue remodeling. The Secca procedure can be performed
on an ambulatory basis using conscious sedation and local COLOSTOMY
anesthesia. The patient is positioned in the prone jack-
knife position. A special RF energy device that utilizes an For patients with anal incontinence so severe that they are
anoscopic barrel with four nickel-titanium curved needle disabled by their symptoms and unable to maintain con-
electrodes is used. Within the tip and at the base of each tinence by either conservative or surgical modalities, con-
electrode, thermocouples are present to monitor tissue and struction of a colostomy or ileostomy may become necessary.
mucosal temperature during RF delivery. The instrument is Despite the assumed negative associations with a stoma, the
introduced into the anal canal under direct visualization, so patient’s quality of life may be significantly enhanced, and
References 221
self-imposed isolation corrected. A colostomy is especially REFERENCES

appropriate in patients with profound incontinence related
to anorectal disease or tumor, and in whom other surgi- 1. Sharma A et al. Br J Surg 2016;103:1589–97.
cal approaches are contraindicated, or in situations where 2. Macmillan AK et al. Dis Colon Rectum 2004;47(8):
reconstruction is impossible. Norton et al. (149) reviewed 1341–9.
the formation of a permanent stoma as a last resort when 3. Xu X et al. Dis Colon Rectum 2012;55(5):586–98.
all other interventions for FI have failed. Questionnaires 4. Brown HW et al. Female Pelvic Med Reconstr Surg
were sent asking about the stoma, previous incontinence, 2013;19(2):66–71.
anxiety, depression, and quality of life. Sixty-nine replies 5. Damon H et al. Dis Colon Rectum 2002;45(11):1445–50;
were received from 11 males and 58 females with a median discussion 1450–1.
age of 64 years and a median of 59 months since the opera- 6. Perry S et al. Gut 2002;50(4):480–4.
tion. Rating their ability to live with their stoma on a scale 7. Brown HW et al. Int J Clin Pract 2012;66(11):1109–16.
of 0–10, the median response was 8. The majority (83%) 8. Madoff RD et al. N Engl J Med 1992;326(15):1002–7.
felt that the stoma restricted their life “a little” or “not at 9. Browning GG, Parks AG. Br J Surg 1983;70(2):101–4.
all,” a significant improvement from perceived restriction 10. Jorge JM, Wexner SD. Dis Colon Rectum 1993;36(1):
from former incontinence. Satisfaction with the stoma was 77–97.
a median score of 9. Eighty-four percent would “probably” 11. Rockwood TH et al. Dis Colon Rectum 1999;42(12):
or “definitely” choose to have the stoma again. They con- 1525–32.
cluded the majority of previously incontinent people felt 12. Rockwood TH. Gastroenterology 2004;126(1 Suppl 1):
positively about the stoma and the difference it had made to S106–13.
their life. However, a few had not adapted and disliked the 13. Nelson R et al. JAMA 1995;274(7):559–61.
stoma intensely. 14. Cerulli MA et al. Gastroenterology 1979;76(4):742–6.
15. Ctercteko GC et al. Aust N Z J Surg 1988;58(9):
ONGOING INVESTIGATIONS 703–10.
16. Keighley MR, Fielding JW. Br J Surg 1983;70(8):
Other modalities are currently being tested in order to 463–8.
increase our armamentarium for the treatment of this 17. Sultan AH et al. N Engl J Med 1993;329(26):1905–11.
debilitating issue. Current clinical trials include the TOPAS 18. Eason E et al. CMAJ 2002;166(3):326–30.
device (American Medical Systems), which is a polypro- 19. Oberwalder M et al. Br J Surg 2003;90(11):1333–7.
pylene mesh sling implanted to reinforce the puborectal 20. Rieger N, Wattchow D. Aust N Z J Surg 1999;69(3):
angle, as well as the LIBERATE trial (Pelvilon, Inc.), which 172–7.
employs an inflatable vaginal insert used to occlude the anal 21. Oberwalder M et al. Arch Surg 2004;139(4):429–32.
canal. In addition, the use of an occlusive anal plug has been 22. Faridi A et al. J Perinat Med 2002;30(5):379–87.
employed as a minimally invasive option. Examples include 23. Fynes M et al. Obstet Gynecol 1998;92(4 Pt 1):496–500.
the Peristeen anal plug (Coloplast UK), and the Procon-2 24. Nelson RL et al. Cochrane Database Syst Rev 2010;
device. (2):CD006756.
25. Khubchandani IT, Reed JF. Br J Surg 1989;76(5):
431–4.
26. Liang J, Church JM. Am J Surg 2015;210(4):715–9.
CONCLUSIONS 27. Aguilar PS et al. Dis Colon Rectum 1985;28(7):496–8.
28. Golub RW et al. J Gastrointest Surg 1997;1(5):487–91.
Modern surgery is experiencing an increase in the number 29. Ortiz H, Marzo J. Br J Surg 2000;87(12):1680–3.
of modalities available to treat fecal incontinence. Especially 30. Schouten WR et al. Dis Colon Rectum 1999;42(11):
regarding the success of SNS, there has been a profound par- 1419–22; discussion 1422–3.
adigm shift in the more conservative, yet equally effective 31. van Tets WF et al. Dis Colon Rectum 1997;40(9):
management of all but the most extreme cases of FI. Despite 1042–5.
the wide range of surgical options, most patients with FI 32. Stelzner F. Prog Pediatr Surg 1976;9:1–6.
can be treated very successfully with appropriate conserva- 33. Parks AG. Proc R Soc Med 1975;68(11):681–90.
tive management and the use of stool bulking agents and 34. Suilleabhain CB et al. Dis Colon Rectum 2001;44(5):
antidiarrheal medications. For those who failed conserva- 666–71.
tive treatment, there are a wide variety of surgical options. 35. Schiller LR et al. N Engl J Med 1982;307(27):1666–71.
The classical historical gold standard of sphincter repair 36. Gee AS, Durdey P. Br J Surg 1995;82(9):1179–82.
has been largely supplanted with much less invasive and 37. Henry MM. Gastroenterol Clin North Am 1987;16(1):
much more effective techniques, with SNS leading the way. 35–45.
Nevertheless, a thorough understanding of all of the medi- 38. Sultan AH et al. Br J Surg 1993;80(4):508–11.
cal and surgical options for management of FI is required 39. Bharucha AE. Gastroenterology 2004;126(1 Suppl 1):
of all surgeons who may evaluate and treat these patients. S90–8.
40. Bollard RC et al. Dis Colon Rectum 2002;45(2):171–5. 79. Fang DT et al. Dis Colon Rectum 1984;27(11):720–2.
41. Sentovich SM et al. Dis Colon Rectum 1998;41(8): 80. Parks AG, McPartlin JF. Proc R Soc Med 1971;64(12):
1000–4. 1187–9.
42. Zetterstrom JP et al. Dis Colon Rectum 1998;41(6): 81. Mahony R et al. Dis Colon Rectum 2004;47(1):12–7.
705–13. 82. Karoui S et al. Dis Colon Rectum 2000;43(6):813–20.
43. Gold DM et al. Br J Surg 1999;86(3):371–5. 83. Malouf AJ et al. Lancet 2000;355(9200):260–5.
44. Xue Y et al. Zhonghua Wei Chang Wai Ke Za Zhi 84. Halverson AL, Hull TL. Dis Colon Rectum 2002;45(3):
2014;17(12):1187–9. 345–8.
45. Hiltunen KM. Dis Colon Rectum 1985;28(12):925–8. 85. Simmang C et al. Dis Colon Rectum 1994;37(11):1065–9.
46. Read NW et al. Br J Surg 1984;71(1):39–42. 86. Rasmussen OO et al. Dis Colon Rectum 1999;42(2):
47. Penninckx F et al. Acta Gastroenterol Belg 1995; 193–5.
58(1):51–9. 87. Gilliland R et al. Dis Colon Rectum 1998;41(12):1516–22.
48. Tetzschner T et al. Acta Obstet Gynecol Scand 1995; 88. Londono-Schimmer EE et al. Int J Colorectal Dis
74(6):434–40. 1994;9(2):110–3.
49. Bartolo DC et al. Br J Surg 1983;70(11):664–7. 89. Barisic GI et al. Int J Colorectal Dis 2006;21(1):52–6.
50. Baig MK, Wexner SD. Br J Surg 2000;87(10):1316–30. 90. Bravo Gutierrez A et al. Dis Colon Rectum 2004;47(5):
51. Laurberg S et al. Br J Surg 1988;75(8):786–8. 727–31; discussion 731–2.
52. Roig JV et al. Dis Colon Rectum 1995;38(9):952–8. 91. Tjandra JJ et al. Dis Colon Rectum 2003;46(7):937–42;
53. Rasmussen OO et al. Int J Colorectal Dis 1990;5(3): discussion 942–3.
135–41. 92. Briel JW et al. Dis Colon Rectum 1998;41(2):209–14.
54. Young CJ et al. Dis Colon Rectum 1998;41(3):344–9. 93. Hasegawa H et al. Dis Colon Rectum 2000;43(7):
55. Chen AS et al. Dis Colon Rectum 1998;41(8):1005–9. 961–4; discussion 964–5.
56. Briel JW et al. Int J Colorectal Dis 2000;15(2):87–90. 94. Womack NR et al. Br J Surg 1988;75(1):48–52.
57. Rosen L et al. Am Fam Physician 1986;33(3):129–37. 95. Miller R et al. Br J Surg 1988;75(2):101–5.
58. Bannister JJ et al. Br J Surg 1989;76(6):617–21. 96. Orrom WJ et al. Dis Colon Rectum 1991;34(4):
59. Hallgren T et al. Dig Dis Sci 1994;39(12):2612–8. 305–10.
60. Santoro GA et al. Dis Colon Rectum 2000;43(12): 97. Athanasiadis S et al. Langenbecks Arch Chir 1995;
1676–81; discussion 1681–2. 380(1):22–30.
61. Engel BT et al. N Engl J Med 1974;290(12):646–9. 98. Healy JC et al. Dis Colon Rectum 2002;45(12):
62. Loening-Baucke V. Dig Dis 1990;8(2):112–24. 1629–34.
63. Wald A. Eur J Gastroenterol Hepatol 1995;7(8):737–9. 99. Setti Carraro P et al. Br J Surg 1994;81(1):140–4.
64. Norton C, Cody JD. Cochrane Database Syst Rev 100. Browning GG, Motson RW. Ann Surg 1984;199(3):
2012;(7):CD002111. 351–7.
65. Heymen S et al. Dis Colon Rectum 2001;44(5): 101. Jameson JS et al. Dis Colon Rectum 1994;37(4):369–72.
728–36. 102. Keighley MR. Int J Colorectal Dis 1987;2(4):236–9.
66. Norton C et al. Gastroenterology 2003;125(5):1320–9. 103. Laurberg S et al. Br J Surg 1990;77(5):519–22.
67. Solomon MJ et al. Dis Colon Rectum 2003;46(6): 104. van Tets WF, Kuijpers JH. Dis Colon Rectum 1998;
703–10. 41(3):365–9.
68. Fynes MM et al. Dis Colon Rectum 1999;42(6):753–8; 105. Pickrell K et al. Am J Surg 1955;90(5):721–6.
discussion 758–61. 106. Wexner SD et al. Dis Colon Rectum 1996;39(9):
69. Beddy P et al. J Gastrointest Surg 2004;8(1):64–72; 957–64.
discussion 71–2. 107. Matzel KE et al. Dis Colon Rectum 2001;44(10):
70. Chiarioni G et al. Am J Gastroenterol 2002;97(1): 1427–35.
109–17. 108. Chodez M et al. Tech Coloproctol 2014;18(12):
71. Sangwan YP et al. Dis Colon Rectum 1995;38(10): 1147–51.
1021–5. 109. Wexner SD et al. Ann Surg 2010;251(3):441–9.
72. Fernandez-Fraga X et al. Dis Colon Rectum 2003; 110. Wexner SD et al. J Gastrointest Surg 2010;14(7):
46(9):1218–25. 1081–9.
73. Enck P et al. Dis Colon Rectum 1994;37(10):997–1001. 111. Wexner SD, Bleier J. Expert Rev Gastroenterol
74. Guillemot F et al. Dis Colon Rectum 1995;38(4): Hepatol. 2015;9:1577–89.
393–7. 112. Tjandra JJ et al. Dis Colon Rectum 2008;51(5):
75. Glia A et al. Dis Colon Rectum 1998;41(3):359–64. 494–502.
76. Ryn AK et al. Dis Colon Rectum 2000;43(9):1262–6. 113. Mellgren A et al. Dis Colon Rectum 2011;54(9):
77. Pager CK et al. Dis Colon Rectum 2002;45(8): 1065–75.
997–1003. 114. Hull T et al. Dis Colon Rectum 2013;56(2):234–45.
78. Davis KJ et al. Aliment Pharmacol Ther 2004;20(5): 115. Altomare DF et al. Dis Colon Rectum 2009;52(1):11–7.
539–49. 116. Lim JT et al. Dis Colon Rectum 2011;54(8):969–74.
References 223
117. Michelsen HB et al. Dis Colon Rectum 2010;53(4): 133. Christiansen J, Sparso B. Ann Surg 1992;215(4):383–6.
414–21. 134. Mundy L et al. Br J Surg 2004;91(6):665–72.
118. Altomare DF et al. Br J Surg 2015;102(4):407–15. 135. Darnis B et al. Dis Colon Rectum 2013;56(4):505–10.
119. George AT et al. Dis Colon Rectum 2012;55(3): 136. Wong MT et al. Ann Surg 2011;254(6):951–6.
302–6. 137. Wexner SD et al. Dis Colon Rectum 2009;52(9):
120. Brouwer R, Duthie G. Dis Colon Rectum 2010;53(3): 1550–7.
273–8. 138. Lehur PA et al. Dis Colon Rectum 2010;53(12):
121. Nakamura M et al. Hinyokika Kiyo 1983;29(9):1053–9. 1604–10.
122. Shafik A et al. Eur Surg Res 2003;35(2):103–7. 139. Barussaud ML et al. Colorectal Dis 2013;15(12):
123. Leroi AM et al. Am J Gastroenterol 2012;107(12): 1499–1503.
1888–96. 140. Williams A et al. Int J Colorectal Dis 2016;31(2):
124. Thomas GP et al. Colorectal Dis 2013;15(5):519–26. 465–72.
125. Thin NN et al. Br J Surg 2015;102(4):349–58. 141. Herman RM et al. Colorectal Dis 2015;17(5):433–40.
126. Knowles CH et al. Lancet 2015 Oct 24;386(10004): 142. Efron JE. Surg Technol Int 2004;13:107–10.
1640–8. 143. Takahashi-Monroy T et al. Dis Colon Rectum 2008;
127. Edenfield AL et al. Obstet Gynecol Surv 2015;70(5): 51(3):355–9.
329–41. 144. Ruiz D et al. Dis Colon Rectum 2010;53(7):1041–6.
128. Graf W et al. Lancet 2011;377(9770):997–1003. 145. Malone PS et al. Lancet 1990;336(8725):1217–8.
129. Maeda Y et al. Cochrane Database Syst Rev 2013;2: 146. Lawal TA et al. J Laparoendosc Adv Surg Tech A
CD007959. 2011;21(5):455–9.
130. LaTorre F, de la Portilla F. Colorectal Dis 2013;15(5): 147. Ellison JS et al. J Urol 2013;190(4 Suppl):1529–33.
569–74. 148. Worsoe J et al. Dis Colon Rectum 2008;51(10):
131. Hussain ZI et al. Br J Surg 2011;98(11):1526–36. 1523–8.
132. Guerra F et al. Tech Coloproctol 2015;19(1):23–7. 149. Norton C et al. Dis Colon Rectum 2005;48(5):1062–9.
24
Surgery for rectal prolapse
STEVEN R. HUNT
the anal canal. While it is often clinically obvious, sev-

CHALLENGING CASE eral other anorectal disorders can imitate the condition.
Circumferential prolapsed internal hemorrhoids, when
A 75-year-old woman with a history of chronic large, are frequently diagnosed as prolapse, and prolapse
obstructive pulmonary disease (COPD) and conges- is often diagnosed as hemorrhoids. Rectal polyps or can-
tive heart failure (CHF) presents with full-thickness cer can protrude through the anus and mimic prolapse. It
rectal prolapse. Her colonoscopy was normal. is important to differentiate between true procidentia and
mucosal prolapse, as the entities may have similar present-
ing symptoms. Patients with mucosal prolapse frequently
CASE MANAGEMENT have a history of prior anorectal procedures or trauma, and
the prolapse is often asymmetric. Solitary rectal ulcers and
The patient is offered a perineal proctosigmoidec- colitis cystica profunda can present with symptoms simi-
tomy (Altmeier procedure). lar to rectal prolapse. These disorders are associated with
internal intussusception of the rectum but may coexist in
patients with rectal prolapse. Both solitary rectal ulcer and
colitis cystica profunda are hypothesized to result from
INTRODUCTION
repeated mucosal trauma and ischemia at the lead point of
the prolapse. Significant rectal bleeding is relatively rare in
Rectal prolapse (rectal procidentia) is defined as the full- patients with procidentia, although it is a common presen-
thickness intussusception of the rectum through the anal tation in solitary rectal ulcer syndrome and colitis cystica
canal. The annual incidence of rectal prolapse is estimated profunda.
to be 2.5 per 100,000 population (1). The disorder tends to Internal rectal intussusception without prolapse may
affect elderly women, psychiatric patients, and patients with be a predecessor to rectal prolapse, although this associa
neurologic disorders. Presenting symptoms are usually tion has not been proven (2). Rectal intussusception is
referable to the prolapse itself. Additional presenting com- frequently identified during defecography performed to
plaints include constipation, straining, incontinence, and evaluate obstructed defecation or constipation. It is also
mucous soilage of the undergarments. a common finding in asymptomatic patients (3). While it
Surgery remains the only definitive therapy for rectal is clear that surgery is the mainstay of treatment for com-
prolapse. Over 100 operations have been described for the plete rectal prolapse, the indications for surgical inter-
treatment of procidentia. Generally, these procedures can vention in cases of internal intussusception are less clear.
be divided into peritoneal and abdominal approaches. The Some authors advocate surgical intervention in cases of
optimal procedure for each patient should be determined by symptomatic intussusception, while others are more cau-
presenting symptoms and patient comorbid disease. tious in their approach (4,5). In the author’s section, the
initial approach to patients with rectal intussusception is
dietary modification and pelvic floor retraining through
CLASSIFICATION biofeedback. Surgery is generally reserved for patients with
complications of internal prolapse (solitary rectal ulcers
In the strictest sense, rectal prolapse refers only to full- and colitis cystica profunda) who have failed conservative
thickness, circumferential protrusion of the rectum beyond therapy.
224
Operative repairs / Perineal repairs 225
that demonstrates a nonrelaxing puborectalis should prompt

PATIENT EVALUATION AND initiation of biofeedback therapy as an adjunct to surgery.
INVESTIGATIONS Postoperative continence can also be predicted on the
basis of a prolonged pudendal nerve terminal motor latency
As rectal prolapse is a benign disease, surgery need only be (PNTML) and poor resting sphincter tone (7,8). We do not
considered if the symptoms are debilitating. Frequency of routinely obtain physiologic studies in the evaluation of
the prolapse and initiating factors (defecation, straining, or procidentia, as they are expensive and a prolonged PNTML
standing) should be documented. The presence of severe is not a contraindication for surgery.
constipation or symptoms of obstructed defecation should
be noted, as these patients may require further evaluation. ADDITIONAL STUDIES
Fecal incontinence occurs in 60%–80% of patients, and a
When the patient is unable to reproduce the rectal prolapse
frank discussion should ensue regarding expected surgi-
in the office, defecography may be used to evaluate for inter-
cal outcomes with regard to continence. Most large series
nal prolapse or other defecatory pathology. In patients with
show improvement in fecal incontinence in greater than
severe constipation and prolapse, a colonic transit study
40% of procidentia patients after surgery, regardless of the
may be obtained. Concentration of the markers in the left
approach (6). Continence may continue to improve over the
and sigmoid colon on day 5, in the setting of severe consti-
first 6–12 months postoperatively.
pation, is an indication for a resection rectopexy.
A detailed surgical history should be obtained, with spe-
cial attention to anorectal and pelvic operations, as this may
influence the ultimate surgical approach. In patients with
recurrent rectal prolapse, operative notes from the prior OPERATIVE REPAIRS
procedures should be obtained and scrutinized. Female
patients may have a history of bladder or uterine prolapse, Although the modern operative procedures for rectal pro-
requiring consultation with a urogynecologist and a com- lapse are not particularly morbid, the patients are frequently
bined approach. All patients with prolapse should have a elderly, and morbidity is not trivial. Some series report mor-
recent colonoscopy to rule out any mucosal lesions. talities as high as 7%. These patients often have significant
comorbid conditions, and the operative approach (peri-
PHYSICAL EXAM neal, open, or laparoscopic) should take these factors into
account. The choice of procedure is frequently dictated by
The diagnosis of procidentia is made by demonstra- surgeon preference and experience; however, a one-size-fits-
tion of the prolapse in the surgeon’s office. Patients with all approach may not be suitable for all patients.
advanced prolapse may be able to produce the prolapse In addition to the morbidity of the procedure, evaluation
on the examination table with minimal straining. If the of the various surgical approaches to rectal prolapse must
patient cannot prolapse on the examination table, he or take efficacy and functional outcomes into account. Some
she should be examined after straining on the toilet. Once techniques have excellent results in terms of recurrence but
the prolapse has been achieved, the examiner should first can predispose the patients to constipation or evacuatory
differentiate between full-thickness prolapse and hemor- difficulties, trading one problem for another.
rhoidal or mucosal prolapse. Full-thickness prolapse is
characterized by concentric mucosal rings, as opposed to PERINEAL REPAIRS
the radially oriented sulci seen with mucosal and hemor-
rhoidal prolapse. The preponderance of the historical literature suggests
The digital rectal exam should exclude other anorectal that the abdominal approach to rectal prolapse is superior
pathology, and the sphincter tone and the squeeze pressure to the perineal approach in terms of recurrence rates. While
should be evaluated. Female patients should be evaluated most single institution studies report better outcomes for
for the presence of an enterocele or rectocele. Rigid proctos- abdominal procedures, this difference is not demonstrated
copy should be performed in the office to rule out any rectal in meta-analysis (9,10).
tumors, and to evaluate for solitary ulcer and colitis cystica The major advantage of the perineal approach is the
profunda. ability to conduct the operation under spinal or even local,
Generally, a physical exam and demonstration of the anesthetic. The avoidance of general anesthesia and an
complete prolapse in the office are sufficient evaluation abdominal dissection makes this the preferred approach for
prior to surgery. Additional studies are sometimes required patients with significant comorbidities.
in certain cases.
Perineal proctosigmoidectomy
ANAL PHYSIOLOGY (Altmeier procedure)
Patients with chronic severe straining at stool should be The technique of perineal proctosigmoidectomy involves
evaluated with anal physiology testing. Electromyography mobilization and resection of the prolapsed rectosigmoid
226 Surgery for rectal prolapse
colon via a perineal approach. Patients should have a com- peritoneal cavity facilitates delivery of the prolapsed rectum
plete mechanical bowel preparation. The prone-jackknife or and division of the mesorectum. The mesorectum is then
left lateral position is preferred over lithotomy, as it allows divided and ligated with ligatures, or alternatively, a vessel-
easy access to the operative field for the surgeon and assis- sealing device may be used. Division of the mesorectum
tant. While general anesthetic provides more comfort for should be continued, advancing proximally on the bowel
the patient, it is often necessary to use local or spinal anes- until tension is encountered (Figure 24.1).
thesia in frail patients. The buttocks should be taped apart, Once the redundant rectosigmoid has been mobilized,
and a Lonestar retractor is used to efface the anus and pro- the anterior peritoneum should be repaired, including sero-
vide optimal exposure. muscular bites of the anterior bowel wall, with a running
The procedure is begun by recreating the prolapse. Once absorbable suture to obliterate the pouch. A levatorplasty
the bowel has been completely prolapsed, a circumferen- should be considered if a defect is present in the pelvic
tial incision is made in the rectum approximately 1.5–2 cm floor. If the levator muscles can be identified without exten-
proximal to the dentate line. Using the electrocautery, this sive dissection, plication should be performed anteriorly
incision should be continued until the full thickness of the and posteriorly. The redundant bowel is then divided and
rectal wall has been incised circumferentially. The incised a hand-sewn anastomosis is fashioned using interrupted
rectum is then everted and pulled downward. The vaginal absorbable sutures. Alternatively, the anastomosis may be
wall is frequently adherent to the prolapsed segment and created using an EEA stapler with acceptable results (11,12).
should be dissected away from the rectum to avoid the dev- Generally, patients have minimal narcotic requirements
astating complication of a postoperative colovaginal fistula. postoperatively and ileus is exceedingly rare. Patients should
The peritoneal cavity is then entered by incising the perito- be ambulated and their diet is advanced on postoperative
neum of the pouch of Douglas anteriorly. Entrance into the day 1. Constipating regimens have no proven beneficial
(a) (b) (c)
(d) (e)
Figure 24.1 Perineal rectosigmoidectomy. (a,b) Incision of rectal wall. (c) Division of vessel adjacent to bowel wall.
(d) Mesenteric vessels ligated. Stay sutures previously placed in distal edge of outer cylinder are placed in cut edge of
inner cylinder. (e) Anastomosis of distal aspect of remaining colon to the short rectal stump.
Operative repairs / Perineal repairs 227
results. It is the author’s practice to discharge patients after in the prone-jackknife or left lateral position with efface-
the first bowel movement, but in some centers, the Altmeier ment of the anus. Again, local or spinal anesthesia may be
procedure is performed on an outpatient basis (11). used for infirm patients. The rectal prolapse is delivered,
In experienced hands, the Altmeier procedure has excel- and the submucosal plane is infiltrated with local anesthetic
lent results, rivaling the abdominal procedures for recur- containing epinephrine. A circumferential mucosal incision
rence rates. Several recent large series report recurrence is made 1 cm proximal to the dentate line. The submucosal
rates ranging from 6% to 16% (7,9,11). Both incontinence plane is identified, and downward traction is applied to the
and constipation are also significantly improved after peri- mucosal tube. Dissection is carried out within this plane to
neal proctectomy (7,9,13). Some authors describe significant the apex of the prolapsed segment of rectum. At this point,
improvement in recurrence rates if a levatorplasty is per- the exposed muscularis propria is plicated with multiple
formed (14). bites in four quadrants using an absorbable monofilament
Fortunately, major morbidity and mortality for this pro- or braided suture. The redundant mucosa is then excised
cedure are rare. The anastomotic leak rates are reportedly and the plication sutures are tied. The mucosal edges are
1%–2%, with significant bleeding occurring in a similar then reapproximated using interrupted absorbable sutures
percentage of patients (7,9,14). (Figure 24.2).
The recurrence rate in most recent large series ranges
Delorme procedure from 13% to 27% (15–17). The morbidity and mortality rates
are similar to those of the Altmeier repair. Improvement is
Delorme procedure offers another alternative to the reported in both continence and constipation in most series
Altmeier repair. The technique involves a submucosal resec- where these functional outcomes were evaluated (16–18).
tion of the prolapsed rectum, with plication of the muscula- Given the uniformly inferior results of Delorme proce-
ris propria. The submucosal nature of the dissection in this dure relative to the Altmeier repair, it is the author’s feel-
procedure does not allow for a concomitant levatorplasty. ing that this approach should not be used as a first-line
As with the Altmeier procedure, mechanical bowel prep- perineal procedure. Many advocate this procedure for the
aration should be performed and the procedure conducted treatment of mucosal prolapse; however, other, less involved
(a) (b) (c)
(d) (e) (f )
Figure 24.2 Delorme procedure. (a) Subcutaneous infiltration of dilute epinephrine solution. (b) Circumferential mucosal
incision. (c) Dissection of mucosa off muscular layer. (d) Plicating stitch approximating cut edge of mucosa, muscular wall,
and mucosa just proximal to dentate line. (e) Plicating stitch tied. (f) Completed anastomosis.
techniques exist for this disorder. Elastic rubber band liga-

(a)
tion is frequently adequate for modest mucosal prolapse.
The circular stapler technique used in the treatment of hem-
orrhoids is a second appealing option for more advanced
mucosal prolapse.
Anal encirclement (Thiersch repair)

(b)
Anal encirclement has almost reached the status of histori-
cal interest, as it has been replaced by other procedures with
more favorable results. The procedure can be performed in
a short period of time with only local anesthetic. The origi-
nal repair described by Thiersch used a silver wire to encir-
cle the anal sphincter complex. The wire encirclement has
fallen out of favor as the wire can break or erode through
the sphincters and anoderm. Marlex or Mersilene mesh are (c)
the preferred alternatives to wire, as they are softer and less
prone to breakage or erosion.
The operation can be performed in the prone-jackknife
or lithotomy position. After meticulous antiseptic prepa-
ration, small posterior and anterior incisions are made
1 cm outside the anal verge. A curved clamp is then tun-
neled through the ischiorectal fossa from the anterior inci-
sion to the posterior incision, and one end of the mesh is
then pulled through the tunnel. This is duplicated on the
opposite side, and the other end of the mesh is delivered.
The redundant mesh is pulled through and the prosthetic
is tightened around an 18F Hegar dilator. The mesh is then Figure 24.3 Anal encirclement (Thiersch). (a) Lateral
overlapped anteriorly and sewn to itself with a nonabsorb- incisions with prosthetic mesh tunneled around the
able suture. The small incisions are then closed with absorb- anus. (b) Mesh completely encircling the anal opening.
able subcuticular sutures, and the wounds are sealed with (c) Completed anal encirclement procedure.
Dermabond, to prevent subsequent soilage of the wounds
(Figure 24.3). be said for constipation, as rectopexy alone tends to worsen
Anal encirclement procedures do not repair the prolapse constipation. In cases of severe constipation preoperatively,
but merely prevent external prolapse. Infectious complica- a sigmoidectomy may be combined with rectal fixation.
tions are common with the synthetic mesh, occurring in up The repairs discussed in the following all involve com-
to 33% of patients (19). Postoperatively, these patients fre- plete rectal mobilization. In all cases, the rectal mobiliza-
quently experience tenesmus and difficulty with evacuation tion should be carried out in the avascular plane outside the
(20). This procedure should be reserved for patients who have mesorectal fascia. The peritoneum at the sacral promontory
significant contraindications to more formal repairs. One is incised, and the plane posterior to the superior rectal
relative indication for this repair is the patient with signifi- artery is identified. Great care should be taken to prevent
cant hepatic ascites (not amenable to transjugular intrahe- injury to the hypogastric plexus, and the ureters should be
patic portosystemic shunt) and debilitating rectal prolapse. identified and avoided. When rectal fixation sutures are
placed, the position of the ureters should be reconfirmed to
OPEN ABDOMINAL REPAIRS prevent inclusion in the suture. These approaches are not
immune to the usual pitfalls of open laparotomy, with com-
A prerequisite to the open approach is the patient’s ability plications including small bowel obstruction, prolonged
to tolerate a general anesthetic and laparotomy. A variety of ileus, and wound complications.
abdominal repairs are described in the literature, but only There is some controversy regarding the extent of rectal
a few have withstood the test of time. The common theme mobilization. While some authors advocate division of the
among these time-tested procedures is complete rectal lateral rectal ligaments to improve recurrence rates, there
mobilization and fixation of the rectum to the sacrum. It is are some reports of worsening constipation if the lateral
suggested that the fibrosis resulting from the rectal mobili- ligaments are divided (22). In a small randomized prospec-
zation is responsible for the long-term fixation of the rectum tive study comparing rectal mobilization with and without
and avoidance of recurrence (21). division of the lateral ligaments, Mollen et al. reported no
All of the large series involving abdominal procedures difference between the two groups with regard to constipa-
show improvement in fecal continence. The same cannot tion scores or to total colonic transit time. Anterior rectal
Operative repairs / Open abdominal repairs 229
mobilization is recommended with all of these procedures, enthusiasm for this procedure has waned because of reports
but this is generally a minimal dissection as these patients of mesh erosion into the rectum, late colovaginal fistulas,
tend to have a deep pouch of Douglas. stenosis, and significant constipation following the proce-
dure (23). In light of these complications and the success of
Mesh sling repair (Ripstein procedure) other alternative therapies, the Ripstein procedure’s role in
the modern treatment of rectal prolapse should be limited.
The Ripstein procedure involves the posterior mobilization
of the rectum down to the pelvic floor followed by fixa- Ventral mesh rectopexy
tion of the rectum to the sacrum using a mesh sling. Before
the advent of the laparoscopic approach, this procedure was A variation of an anterior mesh rectopexy is to use an ante-
one of the most commonly employed abdominal techniques riorly placed vertical strip attached posteriorly, secured to
for rectal prolapse. one side of the sacral promontory. This approach is known
Patients should undergo a complete mechanical bowel as a ventral mesh rectopexy (25). This differs substantially
preparation, and the operation is performed in the lithot- from a traditional Ripstein in that the posterior dissection
omy position. A complete rectal mobilization is carried is typically limited to exposure of the sacral promontory.
down to the pelvic floor. A 3–4 cm wide piece of polytetra- With this technique, the right aspect of the distal sigmoid
fluoroethylene (PTFE) or polypropylene mesh is then fixed mesentery and upper rectum are mobilized by incising
to the sacrum approximately 1 cm to the right of the mid- the right side of the peritoneal reflection at the level of the
line using several nonabsorbable sutures. Traction is then sacral promontory and then sweeping the mesentery off of
applied to the rectum in a cephalad direction, and the mesh the retroperitoneum in order to expose a site of fixation of
is fixed at multiple points to the anterior rectum by sero- the mesh to the sacral promontory. The left side of the peri-
muscular bites of nonabsorbable suture. The mesh is then toneum is left intact, while the right peritoneum is incised
secured to the left side of the sacrum approximately 1 cm in a curvilinear fashion, over the lateral ligaments and into
off the midline, taking care to ensure that the mesh does not the anterior peritoneal reflection. The anterior dissection
constrict the rectum (Figure 24.4). is undertaken to the level of the mid-vagina or seminal
The results of the Ripstein repair are excellent in terms vesicles. The mesh is fashioned to the appropriate width
of recurrence, with recurrence rates of 0%–7% reported and length and is then secured to the anterior rectum with
in large series (6,23,24). In spite of these enviable results, interrupted suture. The mesh then courses along the right
(a) (b)
(c) (d)
Figure 24.4 Mesh rectopexy (Ripstein): (a) posterior fixation of sling on one side, (b) sling brought anteriorly around mobi-
lized rectum, (c) sling fixed posteriorly on the opposite side, and (d) sagittal view of the completed rectopexy.
repair. Some surgeons prefer the robotic approach for ease

of deep pelvic dissection and knot tying.
Results of mesh rectopexy

In its various iterations, mesh rectopexy has performed well
in terms of low recurrence rates, but occasionally high mor-
bidity rates have led to evolution of the procedure over time
(6,28–32). With the original Ripstein repair, recurrence rates
ranged from 4% to 10%, but the morbidity was high, with
complications including mesh erosion, large bowel obstruc-
tion, fecal impaction, and rectovaginal fistula in up to 50%
of patients (30,33,34). Secondary to this relatively high com-
plication rate, Ripstein modified the procedure to involve
posterior fixation of the mesh with anterolateral fixation to
the rectum/mesorectum with improved results (35). Ventral
mesh rectopexy has demonstrated excellent results in terms
of low rates of major morbidity and mortality with recurrence
rates of 0%–6% and no higher than 8.2% in long-term follow-
up (36–43). A multi-institutional review of 2,203 patients
undergoing laparoscopic ventral rectopexy with either syn-
thetic mesh (80.1%) or biologic grafts (19.9%) examined the
incidence of mesh-related complications in either group (40).
The authors report erosion of mesh in 2.4% of cases using
Figure 24.5 Ventral mesh rectopexy.
synthetic mesh and 0.7% of those using biologic mesh. The
median time to mesh erosion was 23 months. Recurrences of
anterior rectum and is ultimately fixed to the sacral prom- rectal prolapse were not included in the outcomes and were
ontory (Figure 24.5). Drains are not routinely placed. not reported in either group.
In most mesh procedures, the mesh is extraperitoneal- The use of an Ivalon sponge was later abandoned fol-
ized, typically by closing the peritoneum with a running, lowing a randomized control trial demonstrating signifi-
absorbable suture. This step theoretically minimizes the cant complication rates, including increased constipation
potential for small bowel obstruction secondary to adhe- and pelvic abscess formation (29). The principles of this
sions to the mesh as well as minimizes the potential for mesh technique have persisted, however, with use of various syn-
erosion and fistula formation. Every patient with rectal pro- thetic and absorbable mesh products, used in both open and
lapse has, by definition, an element of pelvic floor dysfunc- laparoscopic approaches (44–47). Due to increased rates of
tion. The various components of pelvic floor dysfunction complications, we believe that mesh should likely only be
(e.g., rectocele, enterocele, cystocele, uterine or vaginal pro- utilized in a laparoscopic ventral rectopexy procedure, or in
lapse) often coexist. Another advantage of a ventral recto- a repair of a recurrent prolapse.
pexy is that if the dissection is carried low enough, it can
concomitantly address a symptomatic rectocele as well with Posterior mesh fixation (Wells operation)
low recurrence rates (26,27).
The technique of posterior fixation of the rectum to the
LAPAROSCOPIC MESH RECTOPEXY sacrum with insertion of an Ivalon sponge was first described
The principles, anatomy, and landmarks for laparoscopic by Wells with excellent results and a low complication rate
mesh rectopexy are identical to those for open mesh rec- (48). The technique of the Wells operation is similar to that
topexy, as previously detailed. One impact the use of mesh of the Ripstein procedure, except the mesh fixation to the
will have on a laparoscopic case may be port selection, as sacral promontory is posterior. Theoretically, this posterior
a 12 mm port is typically required for introduction of the mesh orientation may reduce the problems typically associ-
mesh. While most synthetic mesh products can be lapa- ated with the anterior sling. The procedure was originally
roscopically handled with similar ease as in open proce- described using an Ivalon (polyvinyl alcohol) sponge. In the
dures, many biologic grafts are thick, and the handling and United States, experience with the Ivalon sponge is limited,
suturing of such materials laparoscopically can pose a chal- as it has not been approved for implantation. Instead, many
lenge. For that reason, when selecting a biologic mesh for centers perform the procedure using polypropylene mesh.
laparoscopic rectopexy, a thinner and partially transpar- Full mechanical bowel preparation is performed, and
ent mesh may reduce some of the difficulty and clumsiness the patient is positioned in lithotomy position. The rectum
when handling and securing the mesh. is mobilized down to the pelvic floor. Retracting the rec-
Robotic-assisted surgery in rectal prolapse repair uses tum anteriorly, a 5 × 8 cm piece of mesh is then anchored
the same dissection and points of fixation as a laparoscopic to the sacrum in the midline using nonabsorbable suture.
Operative repairs / Open abdominal repairs 231
The rectum is then retracted cephalad, and the redundancy procedure will be described in more detail under laparos-
is eliminated. With the rectum under traction, the mesh is copy, as it has evolved primarily as a laparoscopic technique.
sutured bilaterally to the lateral rectal mesentery. The mesh
wrap forms a trough around the dorsal half of the rectum Resection rectopexy (Frykman-Goldberg
and does not cover the anterior rectal wall. The peritoneum
procedure)
is then closed over the mesh to exclude it from the abdomi-
nal cavity (Figure 24.6). Constipation clearly worsens after rectopexy alone. Many
With regard to recurrence, the Wells operation has authors advocate sigmoid colectomy with rectopexy to alle-
exceptional results with recurrence rates generally between viate postoperative constipation. This technique, termed the
0% and 5% for most large open series (49–51). While there Frykman-Goldberg procedure, involves full rectal mobili-
are fewer reported mesh complications, these series uni- zation, sigmoid colectomy with colorectal anastomosis, and
formly show a worsening of constipation after the proce- suture fixation of the rectum to the sacrum.
dure (44,49–51). Patients require a complete mechanical bowel preparation
and are positioned in lithotomy. The rectum is completely
Suture rectopexy mobilized to the pelvic floor posteriorly. The lateral stalks are
left intact. The rectum is then retracted into the abdomen,
Prior to the laparoscopic era, suture rectopexy alone was and the posterolateral mesorectum is fixed to the presacral
not a common procedure. This technique involves rectal fascia using nonabsorbable sutures. The sigmoid colon and
mobilization followed by suture fixation to the sacral prom- upper rectum are then resected. Mobilization of the splenic
ontory. Its appeal lies in the fact that no foreign bodies are flexure is usually not required, as the redundant sigmoid
used, thus negating the complications of mesh infection and colon allows for resection and subsequent anastomosis with-
erosion. A prospective randomized trial comparing open out tension. The anastomosis is created with an EEA stapler.
suture rectopexy to the Wells operation found no differ- The original description of this procedure involved fixation
ence in the two procedures in terms of recurrence (49). This of the anterior rectum to the endopelvic fascia to eliminate
(a) (b)
(c)
Figure 24.6 Ivalalon (polyvinyl alcohol) sponge rectopexy (Wells): (a) polyvinyl sponge being fixed to the sacrum,
(b) sponge in place before fixation to the rectum, and (c) incomplete encirclement of the rectum anteriorly with the
sponge sutured in place.
the cul-de-sac. Most modern proponents of this opera- procedure is most easily accomplished with a 30° camera
tion have abandoned these anterior sutures as they have no to allow for visualization deep in the pelvis. This approach
proven benefit and can be difficult to place safely. requires skill in laparoscopic sewing and knot tying. As
The resection rectopexy has superior results with respect with the open Wells procedure, the recurrence rate is excel-
to both recurrence and constipation. Most large series lent, with recurrence rates ranging from 0% to 4% in recent
report recurrence rates in the low single digits (9,52–54). series (46,47,57). Functional outcomes were also analogous
Morbidity rates range from 0% to 35% and mortality from to the open procedure in these series, with improvement in
this procedure is low (9,52). This remains the only com- continence but worsening of constipation. Morbidity and
monly employed abdominal procedure with significant mortality are low.
improvement in postoperative constipation. One relative Laparoscopists, forever testing the premise that less is
contraindication to resection rectopexy is severe inconti- more, have trended toward more suture repairs without
nence with compromise of the anal sphincter, as sigmoid- mesh. The laparoscopic suture rectopexy is more manage-
ectomy can worsen incontinence in this patient population. able, as it does not require challenging manipulations of
The addition of a sigmoid resection confers a significantly mesh and involves less suturing. Again, three to four ports
increased risk of anastomotic complications when compared are required, and a 30° camera is recommended. After the
to rectopexy alone. Careful adherence to the usual tenets of rectum is mobilized, it should be pulled in a cephalad direc-
a safe colorectal anastomosis (a good proximal and distal tion, and the lateral stalks are sutured to the sacral promon-
blood supply, a tension-free anastomosis, and air testing of tory using nonabsorbable sutures. One suture on each side of
the anastomosis) should allow safe practice of this procedure. the rectum is generally sufficient. The patient’s diet may be
advanced rapidly, and the patient should be ambulated early
LAPAROSCOPY after surgery. It has been our practice to discharge patients
after their first bowel movement; however, many centers per-
Over the last decade, the laparoscopic approach to colorec- form this procedure with only a short postoperative stay.
tal diseases has become pervasive. The literature has been The laparoscopic suture rectopexy has been proven effec-
flooded with series reporting the successful treatment of rec- tive in several recently published series, with recurrence
tal prolapse through minimally invasive techniques. Rectal rates from 0% to 6% (58–60). Continence is improved post-
prolapse lends itself extraordinarily well to the laparoscopic operatively, but the benefit of this simple technique may
approach, as the procedure is isolated to one sector of the be found in improvement in postoperative constipation
abdomen, and there is frequently no specimen removal or (59–61). These series provide hope that the suture rectopexy
anastomosis required, avoiding a conventional incision alone, without mesh, may rival the mesh repair in efficacy,
altogether. Recent reports comparing open to laparoscopic without the long-term complication of constipation. This
treatment of rectal prolapse find that there are significant may obviate the need for a concomitant resection and thus
patient benefits to laparoscopy, including decreased pain, decrease the difficulty and morbidity of the repair.
quicker resumption of diet, earlier return of bowel function, Some centers still favor laparoscopic resection recto-
shorter length of stay, reduced hernia rates, and a lower inci- pexy as the primary procedure for rectal prolapse. As with
dence of small bowel obstruction (45,55,56). Mortality rates the open technique, splenic flexure mobilization is usu-
for the laparoscopic approach are low. All of the open proce- ally not required. The addition of sigmoidectomy increases
dures discussed previously can be performed laparoscopi- the operative time relative to suture rectopexy alone by
cally; however, the Ripstein procedure has proven tedious to nearly 100 minutes (61,62). Results, as with the open tech-
complete laparoscopically and is seldom performed. nique, are excellent, with recurrence rates from 0% to 2.5%
In general, these laparoscopic procedures require a steep (4,63). Both constipation and incontinence are improved
Trendelenburg position to keep the small bowel and sigmoid postoperatively.
colon out of the pelvis. The mesorectum is frequently elon- No comparative studies between open and laparoscopic
gated and thin in these patients. The mesorectal peritoneum techniques have proven a significant reduction in morbidity
is scored at the sacral promontory, and the plane behind the or mortality for the laparoscopic approach, but trends seem
superior rectal artery is identified with the aid of pneumo- to favor the laparoscopic approach (64,65). What is clear from
peritoneum. The hypogastric nerves should be spared and the literature is that the minimally invasive approach to rectal
the ureters identified. The initial mesorectal mobilization prolapse is not inferior. The clear benefits of the laparoscopic
should be posterior in the avascular plane. As with the open approach in terms of cost, length of stay, and decreased pain
approach, division of the lateral ligaments is controversial. mandate consideration of this approach when it is feasible.
The author performs a circumferential mobilization to the
pelvic floor, including division of the lateral ligaments. The
editors prefer to leave the lateral ligaments intact.
The Wells repair has proven more amenable to the RECURRENT PROLAPSE
laparoscopic approach than the Ripstein procedure. The
laparoscopic technique is similar to the open technique. Recurrent rectal prolapse occurs with every procedure, and
Three or four laparoscopic ports are required, and the the surgical approach to repair of the recurrence requires
References 233
consideration of the initial procedure. The mean time to The surgeon who treats this disease should possess the
recurrence is between 18 and 24 months. Patients who have flexibility and breadth of skills to tailor the procedure to
recurred require physiologic testing and defecography to the individual patient. Surgeon preference and experience
evaluate for anismus. If anismus is identified, these patients should play a role in the choice of procedure but should not
should be referred for biofeedback prior to any surgical justify a single procedure for a complex disease. An algorithm
therapy. used in our section is to offer laparoscopic suture rectopexy
There is no clear algorithm for management of recurrent as the default technique. If a patient has severe constipation,
prolapse. Some authors advocate for a change in approach, a laparoscopic resection rectopexy is performed. The patient
performing perineal procedures if the initial approach was with a hostile abdomen or the patient who is too infirm
abdominal, and vice versa. Others promote the use of the to undergo an abdominal procedure is offered a perineal
same approach for repair of the recurrence. No definitive proctosigmoidectomy.
published data exist on the proper selection of the second
procedure. The only absolute principle in the treatment of REFERENCES
recurrent prolapse is that if a resection is planned, any prior
anastomoses must be resected in order to avoid an interven- 1. Kairaluoma MV, Kellokumpu IH. Scand J Surg. 2005;
ing ischemic segment. Again, comorbid disease should play 94(3):207–10.
a role in the selection of the procedure. Patients unfit for 2. Mellgren A et al. Dis Colon Rectum. 1997;40(7):
general anesthetic should be offered a perineal approach if 817–20.
at all possible. 3. Dvorkin LS et al. Br J Surg. 2005;92(7):866–72.
The few published series on the treatment of recurrent 4. Ashari LH et al. Dis Colon Rectum. 2005;48(5):982–7.
prolapse offer little to no insight on the best approach. A 5. Kruyt RH et al. Br J Surg. 1990;77(10):1183–4.
series from the University of Minnesota suggests that the 6. Tjandra JJ et al. Dis Colon Rectum. 1993;36(5):
abdominal approach is superior to the perineal approach 501–7.
in terms of re-recurrence (65). The Cleveland Clinic 7. Glasgow SC et al. Dis Colon Rectum. 2006;49(7):
Florida has published one of the larger series on treatment 1052–8.
of recurrent prolapse. Various surgical approaches were 8. Birnbaum EH et al. Dis Colon Rectum. 1996;39(11):
used, and it is not clear how the procedures are selected. 1215–21.
Compared to primary operations for rectal prolapse, there 9. Kim DS et al. Dis Colon Rectum. 1999;42(4):460–6;
was no difference in terms of recurrence, morbidity, and discussion 6–9.
bowel function (66). 10. Bachoo P et al. Cochrane Database Syst Rev. 2000;
A difficult situation arises in the patient who has had a (2):CD001758.
prior abdominal resection rectopexy but is now unfit for 11. Kimmins MH et al. Dis Colon Rectum. 2001;44(4):
general anesthetic. Before undertaking a perineal proctec- 565–70.
tomy in such a patient, the surgeon must be sure the prior 12. Boccasanta P et al. Dis Colon Rectum. 2006;49(5):
anastomosis can be mobilized and resected. If not, the 652–60.
surgeon is left with three less than desirable options. The 13. Whitlow CB et al. J La State Med Soc. 1997;149(1):
patient may be counseled that an operation is not in his or 22–6.
her best interest. A Delorme procedure may be performed, 14. Chun SW et al. Tech Coloproctol. 2004;8(1):3–8;
or the patient may be offered anal encirclement. discussion 9.
15. Marchal F et al. Dis Colon Rectum. 2005;48(9):
1785–90.
16. Watts AM, Thompson MR. Br J Surg. 2000;87(2):
CONCLUSION 218–22.
17. Tsunoda A et al. Dis Colon Rectum. 2003;46(9):
While many procedures exist for rectal prolapse, only a 1260–5.
few offer acceptable results in terms of recurrence, postop- 18. Lechaux JP et al. Am J Surg. 2001;182(5):465–9.
erative bowel function, and morbidity. Of the perineal tech- 19. Lomas MI, Cooperman H. Dis Colon Rectum. 1972;
niques, the Altmeier procedure appears to offer superior 15(6):416–9.
outcomes in terms of these principles. All of the described 20. Corman M. Rectal Prolapse, Solitary Rectal Ulcer,
open abdominal approaches have satisfactory recurrence Syndrome of the Descending Perineum, and
rates, but only the resection rectopexy shows improvement Rectocele, 5th Edition. Philadelphia, PA: Lippincott,
in postoperative bowel function. Laparoscopy, with all of Williams, and Wilkins, 2005.
its inherent advantages, may be the preferred approach. Of 21. Nelson R et al. Tech Coloproctol. 2001;5(1):33–5.
these procedures, the laparoscopic suture rectopexy appears 22. McKee RF et al. Surg Gynecol Obstet. 1992;174(2):
to offer the best hope of achieving favored status, given the 145–8.
relative simplicity of the procedure and its exceptional out- 23. Schultz I et al. Dis Colon Rectum. 2000;43(1):35–43.
comes with minimal morbidity. 24. Winde G et al. Eur J Surg. 1993;159(5):301–5.
25. Snyder JR, Paquette IM. Rectal prolapse and 46. Zittel TT et al. J Gastrointest Surg. 2000;4(6):632–41.
intussusception. In Beck DE, Wexner SD. (eds). 47. Himpens J et al. Surg Endosc. 1999;13(2):139–41.
Fundamentals of Anorectal Surgery, 3rd Edition. 48. Wells C. Proc R Soc Med. 1959;52:602–3.
New York: Springer, In press. 49. Novell JR et al. Br J Surg. 1994;81(6):904–6.
26. Wong M et al. Colorectal Dis. 2011;13:1019–23. 50. Mann CV, Hoffman C. Br J Surg. 1988;75(1):34–7.
27. Maggiori L et al. Tech Coloproctol. 2013;17:431–6. 51. Aitola PT et al. Dis Colon Rectum. 1999;42(5):
28. Sahoo MR et al. J Minim Access Surg. 2014;10:18–22. 655–60.
29. Ripstein CB. Dis Colon Rectum. 1972;15:334–6. 52. Watts JD et al. Dis Colon Rectum. 1985;28(2):96–102.
30. Gordon PH, Hoexter B. Dis Colon Rectum. 1978;21: 53. Huber FT et al. World J Surg. 1995;19(1):138–43;
277–80. discussion 43.
31. Dyrberg DL et al. Scand J Surg. 2015;104:227–32. 54. Husa A et al. Acta Chir Scand. 1988;154(3):221–4.
32. Dulucq JL et al. Surg Endosc. 2007;21:2226–30. 55. Duepree HJ et al. J Am Coll Surg. 2003;197(2):
33. Kupfer CA, Goligher JC. Br J Surg. 1970;57:482–7. 177–81.
34. Roberts PL et al. Arch Surg. 1988;123:554–7. 56. Solomon MJ et al. Br J Surg. 2002;89(1):35–9.
35. McMahan JD, Ripstein CB. Am Surg. 1987;53:37–40. 57. Dulucq JL et al. Surg Endosc. 2007;21(12):2226–30.
36. D’Hoore A et al. Br J Surg. 2004;91:1500–5. 58. Heah SM et al. Dis Colon Rectum. 2000;43(5):
37. Bloemendaal AL et al. Colorectal Dis. 2015;17: 638–43.
O198–201. 59. Kessler H et al. Surg Endosc. 1999;13(9):858–61.
38. Boons P et al. Colorectal Dis. 2010;12:526–32. 60. Bruch HP et al. Dis Colon Rectum. 1999;42(9):1189–94;
39. Consten EC et al. Ann Surg. 2015;262:742–8. discussion 94–5.
40. Evans C et al. Dis Colon Rectum. 2015;58:799–807. 61. Kellokumpu IH et al. Surg Endosc. 2000;14(7):634–40.
41. Randall J et al. Colorectal Dis. 2014;16:914–9. 62. Baker R et al. Dis Colon Rectum. 1995;38(2):199–201.
42. Sileri P et al. J Gastrointest Surg. 2012;16:622–8. 63. Benoist S et al. Am J Surg. 2001;182(2):168–73.
43. Owais AE et al. Colorectal Dis. 2014;16:995–1000. 64. Kairaluoma MV et al. Dis Colon Rectum. 2003;46(3):
44. Allen-Mersh TG et al. Dis Colon Rectum. 1990;33(7): 353–60.
550–3. 65. Steele SR et al. Dis Colon Rectum. 2006;49(4):440–5.
45. Purkayastha S et al. Dis Colon Rectum. 2005;48(10): 66. Pikarsky AJ et al. Dis Colon Rectum. 2000;43(9):
1930–40. 1273–6.
25
Management of diverticulitis
DAVID E. BECK, H. DAVID VARGAS, AND MOLLY M. FORD
the bowel wall, most common colonic diverticulum involve

CHALLENGING CASE a false or “pseudo” diverticulum where the mucosa and sub-
mucosa protrude through the muscular and serosal layers.
A 52-year-old male presents to the emergency This condition was first described by Curveihier in the mid-
department with complaints of left lower quadrant 1800s, while the first significant case series of diverticulitis
abdominal pain for the last 16 hours. The patient did not appear in the literature until 1899 (1,2).
describes the pain as escalating in nature, unrelieved The impact of diverticular disease on health care is sub-
with a bowel movement. History is unremarkable stantial and increasing in both incidence and cost. Peery
except for hypertension, which is treated with a beta- et al. surveyed the Health Care Utilization Project and
blocker. The patient denies any similar symptoms pre- revealed that diverticular disease represented the sixth
viously. Past medical history includes a left inguinal most common gastrointestinal diagnosis among patients
hernia repair 25 years ago. Abdominal exam reveals being seen in outpatient clinics in the United States, exceed-
a mildly distended abdomen, with tenderness to the ing those presenting with inflammatory bowel disease or
left lower quadrant, but no guarding or rigidity. A colorectal cancer (3). A report querying a database of U.S.
basic metabolic profile is normal, and complete blood emergency departments revealed that from 2006 to 2013,
count reveals a leukocytosis at 14,000. Computed emergency department visits for diverticulitis increased
tomography (CT) scan of the abdomen and pelvis by 40% (3). This same survey indicated that admissions to
with oral and rectal contrast demonstrates thickening the hospital remained constant, suggesting a shift toward
of the sigmoid colon with mesenteric thickening but outpatient management of diverticulitis. Health-care cost
no identifiable abscess or perforation. of diverticulitis over this period of time rose accordingly
and can be measured in many ways, including by hospital
admissions, office visits, and total health-care costs.
CASE MANAGEMENT
Treatment should consist of broad-spectrum antibi- EPIDEMIOLOGY

otics, typically, ciprofloxacin and metronidazole; IV
fluids; and bowel rest. Admission to the hospital is This condition afflicts people of Western industrialized
based on physical examination, comorbidities, and countries more frequently than their counterparts in
CT findings. Treatment should be continued until the underdeveloped, more agrarian-based economies, and
patient’s pain has resolved or symptomatic improve- points to the long-standing belief that diets higher in fiber
ment is noted, and then oral intake may resume. content serve as a protective effect (4). In addition to fiber,
Antibiotics are typically continued for 7–10 days fol- newer theories for pathogenesis include increased elastin
lowing resolution of pain. deposition and variations in collagen cross-linking serving
to reduce wall compliance (5). Last, there appears to be a
genetic component that predisposes individuals to diver-
ticulitis (6).
INTRODUCTION A genetic influence in diverticular disease is supported by
the anatomic observation that in Western countries, diver-
Diverticulosis coli represents a physical outpouching of the ticulosis is most often localized to the left colon, whereas
colonic wall. While a true diverticulum involves all layers of it is predominantly localized in the right colon in Asian
235
236 Management of diverticulitis
countries (7). In addition, studies of population migration

suggest that despite populations adopting new environmen- CLINICAL PRESENTATION AND
tal factors, there may not be subsequent changes in diver- DIAGNOSIS
ticular disease incidence. A study of Turkish migrants to the
Zaanstreek region of the Netherlands found a much lower Clinically, patients report abdominal pain in the left lower
incidence of diverticulosis than the native Dutch popula- quadrant or lower abdomen that progresses to increasing
tion of 7.5% compared to 50% (8). In the Japanese popula- pain with movement or palpation. There is often an associ-
tion living in Hawaii and eating a more Westernized diet, ated change in bowel habits preceding the pain, and symp-
diverticula remain predominantly right sided (9). Most toms of fever and chills. Rectal bleeding can coincide with
recently, two large twin studies found that genetic factors pain but is rare and would be the exception rather than
are a strong contributor to developing diverticular disease. the rule.
A study of the Swedish Twin Registry found that the odds During recent decades, a significant change in both
ratio (OR) of developing diverticular disease if one’s co- diagnostics and treatment of acute colonic diverticulitis has
twin was affected was 7.15 in monozygotic twins as com- evolved. The initial evaluation should include a problem-
pared to 3.2 for dizygotic twins (10). The Danish twin study specific history and physical examination, a complete blood
found a relative risk (RR) for diverticulosis in twin siblings count, urinalysis, and abdominal radiographs in selected
was 2.92 as compared to the general population (11). Both clinical scenarios (16). As a supplement to clinical evalu-
studies estimate the contribution of heredity to be roughly ation (history and physical examination), CT scans verify
40%–50%. and determine the stage of acute colonic diverticulitis and
The development of colonic diverticulum increases improve decision-making in nonoperative and operative
with age; at 50 years old approximately 40% of the popu- management (17). CT scan is a cost-effective way to diag-
lation in the United States will harbor diverticulosis, and nose acute abdominal conditions; however, radiation expo-
this rises to 70% by age 80 years (12). Complications of sure remains a concern (18). CT findings in diverticulitis are
diverticulosis coli include bleeding or inflammation. discussed in Chapter 12.
Lifetime risks of such complications are difficult to accu- The rate of complicated acute colonic diverticulitis is
rately estimate but are probably less than the previously highest during the primary admission. Most patients who
held belief of 15%–20%, and recent reports indicate that experience a free perforation do so during their first presen-
a 5% lifetime risk of both bleeding and inflammation is tation. In contrast, the majority of readmitted patients have
more likely (13). uncomplicated disease (19,20). In these patients with clini-
cal recurrent disease and a C-reactive protein <50 mg/mL,
a repeated CT scan may be omitted (21). A clinical diagnosis
of uncomplicated diverticulitis, which comprises the major-
PATHOPHYSIOLOGY ity of these patients, remains a major component of outpa-
tient management (22–24).
Diverticulitis refers to inflammation and initial local sepsis A prospective study on the capability of clinical findings,
of the peridiverticular colonic tissues. The pathophysiologic temperature, C-reactive protein, and white blood cell count
mechanism underlying diverticulitis is not well understood. to discern patients with acute colonic diverticulitis from all
The most commonly held theory involves increased pressure other patients admitted with acute abdominal pain found
in the lumen of the diverticulum that leads to ischemia, per- that evaluation by nonspecialist doctors had a high degree
foration, and inflammation (14). The loss of integrity with of diagnostic precision (25). The accuracy of the clinical
local contamination with colonic microorganisms leads to diagnosis was slightly better in younger patients. Elderly
local inflammation and infection. Predisposing local factors patients tend to have more vague and nonspecific symp-
may facilitate this development. toms, broader alternatives of differential diagnosis, altered
The actual role of infection and sepsis has been chal- clinical signs that do not correlate with disease severity,
lenged, giving rise to treatment paradigms not involving higher incidence of comorbidity and multipharmacy, and
antimicrobials. In addition, randomized prospective trials communication difficulties because of hearing and cogni-
performed in European centers have shown that CT scan tive impairment (26,27). This makes elderly patients more
proven, uncomplicated diverticulitis can be successfully prone to misdiagnosis than younger patients (28).
treated with supportive measures and bowel rest without High-risk patients include the morbidly obese, elderly,
antibiotics (15). This challenges our conventional and long- and immunocompromised (29–31). A review of the
held beliefs of local sepsis and perhaps would indicate that American College of Surgeons National Surgical Quality
there exists a spectrum of the pathologic condition. Symptom Improvement Program data set from 2005 to 2010 found
development and the timing of clinical presentation may that morbidly obese patients undergoing surgery for
identify the process at a preinfectious phase. Such think- diverticulitis were nearly 10 years younger than normal
ing has led to the alternative treatment paradigm involving weight patients and more likely to require emergency sur-
anti-inflammatory agents such as aminosalicylates. gery, ostomy creation, and open surgery, and to undergo
Treatment / Acute uncomplicated disease 237
procedures without an anastomosis (29). Morbidly obese Table 25.3 Definitions of diverticular disease
patients undergoing emergency surgery were also more
Diverticulosis
likely to have preoperative systemic inflammatory response
• Asymptomatic
syndrome, sepsis, and septic shock.
Diverticulitis
Immunosuppression is commonly related to transplant
• Noninflammatory
status and chemotherapy. The surgery rate at the first epi-
• Symptoms without inflammation
sode of acute colonic diverticulitis appears high in this
• Acute
population and results in a high morbidity (30). However,
• Complicated
if medical treatment is successful, this group of patients
Perforation, abscess, phlegmon, fistula bleeding
appears to have a low recurrence rate, and a follow-up elec-
• Uncomplicated (simple)
tive resection is not recommended (31).
Localized, thickening, fat stranding
• Chronic
• Recurring or persistent disease
CLASSIFICATION Symptoms with systemic signs (may be intermittent)
• Atypical
Symptoms without systemic signs
Classification of the disease severity helps determine how to
• Complex
treat patients with diverticular disease. Older classifications
• Fistula, stricture, obstruction, fibrosing
systems were based on barium enema, physical examina-
tion, and pathology reports (32). While all three methods Source: Boulos PB. Best Pract Res Clin Gastroenterol. 2002;16:
are sufficient to make a diagnose, the improved sensitivity 649–62.
and specificity of newer technology have changed the way
we diagnosis, classify, and treat this disease. CT scanning determine if those are predictors of failure of nonoperative
currently provides practical and predictive information that therapy (34).
assists in the classification of severity of the disease process. In 1978, Hinchey classified patients using the findings at
A number of useful classification systems have been devel- surgery and recommended surgical intervention based on
oped to assist the physician in deciding on a course of treat- his classification system (39). His formula divided the intra-
ment (33,34). These classification systems can be based on operative findings into four categories based on the amount
CT scan findings (Ambrosetti, Table 25.1), intraoperative and type of peritonitis (Table 25.2).
findings (Hinchey, Table 25.2), or a more global view of the However, not all diverticular disease can be classified
disease (Thorson and Goldberg, Table 25.3). by CT scan or operative findings. Thorson and Goldberg
Ambrosetti did extensive work on CT findings of diver- described diverticular disease based on the presenta-
ticular disease and developed a classification system based tion, timing and duration of disease, and complexity (40)
on the appearance of the inflamed colon (38). His work is (Table 25.3).
simple and divides patients into two groups; uncompli-
cated or complicated (Table 25.1). Other studies have looked
at the size of the abscess and amount of mesenteric air to
TREATMENT
Table 25.1 Ambrosetti classification of diverticulitis based
on CT findings Diverticulitis patients present with a spectrum of disease
from uncomplicated acute disease, to complicated disease,
Uncomplicated—Colonic wall thickening, pericolic fat to complex or recurrent disease. Treatment should be indi-
stranding, inflammatory changes vidualized to the specific patient and his or her disease.
Complicated—Extracolonic air, abscess, perforation
Source: Aydin HN, Remzi FH. Dig Liver Dis. 2004;36:435–45. ACUTE UNCOMPLICATED DISEASE
Table 25.2 Intraoperative classification: Hinchey Clinically stable, reliable patients with uncomplicated
classifications colonic diverticulitis are usually treated conservatively with
antibiotics. Patients who can tolerate oral antibiotics can
Type I—Diverticulitis with no or local peritonitis be treated initially as outpatients (41,42). A systemic review
Type II—Diverticulitis with a small pericolic abscess of one randomized controlled trial, six clinical controlled
Type III—Diverticulitis with local purulent or fecal trials, and three case series found no difference in failure
peritonitis rates of medical treatment (6.5% versus 4.6%, p = 0.32) or in
Type IV—Diverticulitis with diffuse purulent or fecal recurrence rates (13% versus 12.1%, p = 0.81) between those
peritonitis receiving ambulatory care and inpatient care for uncompli-
Source: Golfieri R, Cappelli A. Tech Coloproctol. 2007;11: cated diverticulitis (43). Ambulatory treatment was associ-
197–208. ated with an estimated daily cost savings of between 600 and
1,900 euros per patient treated. Treatment in an ambulatory (51). Percutaneous abscess drainage is a safe and effective
setting was related to a higher failure rate of medical treat- alternative to surgery for draining infected fluid collec-
ment at immediate follow-up in comparison to an inpatient tions, with a higher success rate (70%–90%), lower compli-
setting (6.5% versus 4.6%). However, all patients who failed cation rate, and shorter hospital stay compared to surgical
after the initial medical treatment were managed success- drainage (35,36). However, 20%–25% of patients are either
fully without surgical interventions. Similar recurrence not suitable for radiological drainage (multiloculated,
rates at longer periods of follow-up were noted between both anatomically inaccessible) or do not respond to drainage
those who were ambulatory and those in an inpatient set- and will require surgical intervention (37). Deep pelvic
ting (13% versus 12.1%). Jackson and Hammond conducted abscesses not accessible for transabdominal percutane-
a systemic review that showed similar favorable outcomes ous drainage are managed by transvaginal or transrectal
with IV compared to oral antibiotics in an inpatient setting drainage under radiological (ultrasound/CT/fluoroscopy)
and concluded that these results justify ambulatory treat- or endoscopic guidance.
ment of uncomplicated diverticulitis (44). Gastrointestinal hemorrhage associated with diverticu-
However, the role of antibiotics in diverticulitis has lar disease is covered in Chapter 37.
been disputed (45) The Swedish AVOD study prospectively
compared 623 inpatients with CT-confirmed uncompli- EVALUATION AFTER RECOVERY FROM
cated left-sided diverticulitis with IV fluids to IV fluids ACUTE DIVERTICULITIS
with antibiotics and found that antibiotics did not prevent
complications, accelerate recovery, or prevent recurrences After resolution of an episode of acute diverticulitis, the
(46). A subsequent Cochrane Review of three randomized colon should typically be endoscopically evaluated to con-
trials also found no significant difference between antibiot- firm the diagnosis, if this is a first episode or recent colo-
ics and no antibiotics for the treatment of uncomplicated noscopy has not been performed (16). The purpose of the
diverticulitis (15). investigation is to exclude other diagnoses, because patients
After resolution of an episode of diverticulitis, a vari- with simple thickening on imaging may be found to have
ety of agents may be effective in preventing future attacks. ischemia, inflammatory bowel disease, or neoplasia (53).
Supplemental fiber, rifaximin, antispasmodics, mesala- Although the discovery of a mass lesion associated with
mine, and probiotics have been studied in randomized, colon wall thickening is highly suggestive of an underlying
controlled trials as well as in less rigorous studies. Many of neoplasm, the absence of a mass on CT does not preclude
these studies included heterogeneous patients and poorly neoplasia (54,55). When fat stranding is more severe than
characterized the history of diverticulitis in the study sub- expected for the degree of bowel wall thickening, an inflam-
jects. Although some of the literature suggests a protective matory condition such as diverticulitis is most likely (56).
benefit for these agents, their role in prevention of diver- Patients with presumed diverticulitis who have not had a
ticulitis remains to be defined (47). Another study suggests recent colon evaluation should undergo colonoscopy, typi-
that a family history of diverticulitis may predict recurrence cally within 6–8 weeks following resolution of the acute
(48). episode (although data supporting this time interval are
lacking). The absence of neoplasia on colonoscopy may con-
ACUTE COMPLICATED DISEASE firm the diagnosis of diverticulitis suspected on CT (57).
Long-term perforations or abscesses may develop into
Patients with complicated disease (i.e., free perforation, fistulas. The common areas affected by fistula from diver-
larger abscesses, fistula, or stricture) who cannot toler- ticular disease are the bladder (colovesical), the vagina
ate oral hydration, who have relevant comorbidities, or (colovaginal), and the skin (colocutaneous). Symptoms
who do not have adequate support at home require hospi- will be determined by the organs involved: stool or gas
tal admission and, typically, IV antibiotics and bowel rest. per vagina for colovaginal and skin for colocutaneous,
Antibiotics should cover gram-negatives and anaerobes. pneumoturia or fecalurrea, or urinary tract infections
Multidisciplinary, nonoperative management of inpatients for colovesical fistula. Many of these can be confirmed by
with acute diverticulitis is successful in as many as 91% of contrast-enhanced CT scans or contrast enemas. Bullous
patients (49). edema on cystoscopy can also confirm colovesical fistula.
Patients with free perforations and diffuse peritonitis are The primary treatment is resection of the involved seg-
resuscitated and taken for urgent/emergent surgery. A select ment of bowel.
group of patients with free and localized peritonitis may be
managed with close observation and antibiotics (50). ACUTE/URGENT SURGERY
CT-guided percutaneous transabdominal drainage of
intraabdominal abscesses was introduced in the 1980s and Peritonitis, free intraabdominal air with diffuse peritoni-
has been widely adopted (51,52). Large localized abscesses tis, or obstruction unrelieved by other methods is an indi-
(>4–5 cm) are primarily drained by a percutaneous cation for operation. Patients with signs of peritonitis or
approach in order to resolve the sepsis. Those with smaller hemodynamic instability are not candidates for medical
abscesses are often treated with IV antibiotic therapy alone management and should be resuscitated and taken to the
Treatment / Risks associated with Hartmann reversal 239
operating room. Three different operations have been pro- including higher restoration of continuity rate, less hospi-
posed for the treatment of complicated diverticulitis with talization, and fewer infectious complications.
peritonitis. The first operative approach described was the Multiple studies have evaluated the morbidity and
three-stage procedure encompassing drainage with stoma, mortality of the Hartmann procedure as well as the risks
followed by resection and anastomosis with continued incumbent with takedown. Most seasoned surgeons real-
diversion, and finally by restoration of continuity. The sec- ize that at times restoration of continuity can be more of
ond approach involved resection and diversion or the tra- a challenge to both patient and surgeon than the original
ditional Hartmann procedure. However, this approach is operation. This was demonstrated in a multicenter prospec-
being challenged by the third approach of resection with tive trial involving 415 patients with complicated divertic-
primary anastomosis. Primary resection with anastomosis ulitis (61). In this trial, 248 patients underwent resection
(PRA) can be performed with or without a covering stoma, with primary anastomosis. The other 167 had a Hartmann
and/or on table lavage. The three-stage procedure will not procedure. The mortality rate for those undergoing pri-
be discussed here as it is not considered standard of care and mary anastomosis was 4%, while those with resection and
should be used only in infrequent situations. diverting colostomy was 23.4%. After case adjustment, the
In 1921, Hartmann advocated his two-stage resection that data suggested that the Hartmann procedure was associ-
was superior and quickly became the standard of care (58). ated with a 1.8-fold increase in likelihood of death. This
However, early in the 1960s, there were eight reports with a was not statistically significant. However a 2.1-fold increase
total of 50 patients who underwent resection and primary in morbidity was found between the two groups, and this
anastomosis for generalized peritonitis with a low mortal- was significant. In part this is due to the fact that surgeons
ity of 10%. Not much debate is raised now with respect to typically reserved a Hartmann procedure for those older
patients presenting with recurrent or chronic diverticulitis. patients with more comorbidities and thus predisposed to
They are typically managed in an elective fashion with pri- a poorer outcome.
mary anastomosis. Patients are given a bowel preparation
prior to surgery, although the role of mechanical prepara- RISKS ASSOCIATED WITH HARTMANN
tion continues to be discussed. REVERSAL
Patients who present with acute symptoms, typically
Hinchey stages III or IV, are taken to the operating room Reversal of a Hartmann colostomy also carries with it a
urgently. These patients constitute approximately 3.2 per significant risk that must be entertained when considering
100,000 patients (59). These patients present a dilemma, this operation for patients who will desire continuity in the
because typically they are older, have a high number of future. Failure to reverse the colostomy has been reported
comorbidities, and suffer a greater number of complica- in 20%–50% (62) of patients, and leak rates on reversal are
tions. In a review by Salem and Flum of 98 articles on the around 2%–30% (60,62). Mortality has been reported any-
outcome of complicated diverticulitis based on the type of where from 0% to 10%, and wound infection rates range
operation performed, the authors identified 1,051 patients from 12% to 50%.
who underwent a Hartmann procedure from 54 studies, A strong interest in primary anastomosis has been
and 569 patients having undergone a primary anastomosis revived in the literature with papers describing the success-
from 50 studies (60). Of the patients undergoing a primary ful outcomes of patients undergoing this type of operation.
anastomosis, 16% had covering stomas and 10% had on- However, few papers are prospective and less are random-
table lavage. The mortality rates of those in the Hartmann ized, and such a trial is still needed today to definitively
group (19.6%) were much higher than those undergoing a answer the questions of safety and efficacy. Multiple trials
primary anastomosis (9.9%). The anastomotic leak rate in have shown that the outcomes of primary anastomosis are
patients with a primary anastomosis ranged from 6.3% indeed as safe as a Hartmann and in many cases better. In a
to 19.3%. If a diverting proximal stoma was performed at recent review, Constantinides et al. reviewed the outcomes
the time of a primary anastomosis, the anastomotic dehis- of patients undergoing Hartmann (63), PRA (135 patients),
cence rate fell to 6.3%. Wound infections were also more and primary resection with anastomosis and diversion (126
frequently seen in the Hartmann group (24.2%) versus the patients) (59). Patients undergoing a Hartmann procedure
primary anastomosis group (9.6%). Again, patients with had a morbidity and mortality of 35% and 20%, respec-
covering stomas had the lowest wound infection rate at 4%. tively. Primary anastomosis showed a slightly higher mor-
Patients undergoing a Hartmann procedure also required bidity and mortality at 55% and 30%, while those with a
a larger second operation than those who had PRA with primary anastomosis with diverting stoma demonstrated
or without a covering stoma. Complications from a morbidity and mortality rates of 40% and 25%, respectively.
Hartmann reversal were associated with a mortality of Stomas were permanent in 27% of patients undergoing a
0.8%, a wound infection rate of 4.9%, and an anastomotic Hartmann procedure and 8% of those having a primary
leak rate of 4.3%. These patients also experienced stoma anastomosis with diversion. They concluded that primary
complications (10.3%) that required medical attention. anastomosis with defunctioning stoma may be an optimal
The conclusion was that primary anastomosis is no worse strategy for selected patients. The Hartmann procedure
than a Hartmann procedure and has several advantages, should be reserved for patients with an extremely high risk
of perioperative complications and only after consideration elective surgery was most pronounced for younger patients
of long-term implications. and those with complicated disease.
Patients undergoing on-table lavage have been analyzed Young age has traditionally been considered a risk fac-
as well, which showed similar outcomes to those who did not tor for more virulent disease, but increasing evidence has
undergo on-table lavage. Regenet described 60 patients, all accumulated to contradict this theory. Two meta-anal-
Hinchey III or greater: 27 underwent primary anastomosis yses suggest that younger patients are at increased risk of
with intraoperative lavage and 33 had a Hartmann procedure recurrence, but the relationship between age and risk of
(63). In this prospective observational study, they found that emergency operation remains unclear. Katz et al. found
the Hartmann procedure took much less time to perform, no association between young age and risk of emergency
but that the mortality and morbidity for both groups were operation following initial nonoperative management, and
equal. Three patients in the intraoperative lavage group had two studies by Li and colleagues did not find any associa-
an anastomotic leak (11%). A Hartmann reversal occurred tion between young age and risk of emergency operation at
in 69% of the patients. The reversal had its own associated index presentation or during follow-up after nonoperative
morbidity of 24%, an anastomotic leak rate of 7%, and no management (74–76). These studies suggest that young age
deaths. Postoperative stay after primary anastomosis and does not appear to confer a more virulent disease trajectory,
intraoperative lavage was 18.4 days and Hartmann proce- and the practice of elective surgical resection based on age
dure was 38 days. They concluded that primary anastomo- criteria alone should be discouraged (16).
sis with intraoperative lavage and a Hartmann procedure Complicated diverticulitis accounts for up to 35% of
are both adequate approaches for generalized peritonitis admissions for diverticulitis, yet little is known about the
complicating diverticulitis. Covering stomas have been clinical course of such patients managed nonoperatively
recommended by most studies when primary anastomo- (75). Older practice guidelines recommend elective colec-
sis is performed because of the variable anastomotic leak tomy following nonoperative management of complicated
rate. Both diverting colostomies and ileostomies have been diverticulitis, supported by evidence suggesting that such
described with equal success. Most of the poor outcomes patients are at increased risk of recurrent disease and sepsis
noted are not necessarily due to the operation performed (16,34,77). However, the risk of subsequent disease-related
but to the comorbidities and peritonitis associated with the events in this subset of patients has not been well studied,
patient and disease. These risks play more into the outcome with data limited to a few single or multi-institutional cohort
of patients than the type of operation performed. studies with small sample sizes and high loss to follow-up,
with reported recurrence rates ranging from 24% to 53%
ELECTIVE SURGERY (34,78,79). More recently, the success of managing compli-
cated diverticulitis without elective colectomy has led some
In the past, there was a fear that subsequent attacks of diver- to question the necessity of routine elective surgery, par-
ticulitis carried increased risk of perforation and the poten- ticularly in patients at high operative risk (68,69,80). In a
tial need for stoma creation. This widely held belief motivated study by Li and colleagues, the 5-year cumulative incidence
surgeons and patients to proceed with prophylactic elective of readmission and emergency surgery was only 12% and
resection to reduce the need for an urgent operation and the 4.3% among patients with complicated diverticulitis (76).
limited operative options therein. Subsequent studies of the This suggests that elective colectomy may not be necessary
natural history of recurrent diverticulitis and the likelihood for many patients with complicated disease. Surgical rec-
of perforation found the risk to be low even in elderly popula- ommendations may remain for high-risk patient subgroups,
tions (64). Traditionally, elective colectomy has been recom- such as patients with multiple recurrent episodes, persistent
mended for young patients (<50 years at initial presentation), abdominal symptoms, high-risk medical comorbidities, and
patients who have had complicated disease (abscess, fistula, large pelvic abscesses. Certainly, the rationale for resection
or perforation), and patients who have experienced two or must be balanced by the risk of complication of surgery and
more episodes of uncomplicated diverticulitis (65,66). the best estimate of actual reduction of risk of recurrent epi-
However, there is increasing evidence to suggest that sodes. With regard to operative risk, Salem et al. examined
the natural history of the disease may be more commonly a decision analysis tool that revealed a minimum of three to
benign, and the risk of experiencing complicated recurrence four attacks could justify the operative risk (81).
requiring emergency operation may be low, leading many to
challenge the traditional indications for elective colectomy Technical considerations
(67–72). These data are reflected in more contemporary prac-
tice guidelines that favor individualized treatment decisions The extent of elective resection is determined intraopera-
and more selective use of elective colectomy (12,16). tively based on the anatomy and the quality of the tissues.
A population-based analysis by Li and colleagues found The distal margin is an important determinant in mini-
a significant decline in the use of elective colectomy follow- mizing the recurrence of diverticulitis and must extend to
ing diverticulitis; the proportion of patients undergoing the proximal rectum to enable a colorectal anastomosis,
elective colectomy within 1 year of discharge declined from because a colo-colonic anastomosis significantly increases
9.6% in 2002 to 3.9% by 2011 (73). The observed decrease in the risk of recurrence (82,83). Patients in whom the
Complications / Failure to reverse 241
proximal rectum is secondarily inflamed may require more result from cardiovascular problems. However, patients
extensive rectal resection with a lower rectal anastomosis. involved with complicated diverticulitis face greater risks
The proximal extent of resection in the descending colon that can be evaluated with numerous scoring systems.
is chosen by the absence of thickened, hypertrophic tissue Mortality rates range from 0% to 36% in patients presenting
and inflammation. Although it is not necessary to remove with peritonitis and depend greatly on their comorbidities
all diverticula-bearing colon, care should be taken to avoid and time to operation (96).
incorporating any false diverticula in the proximal side of
the anastomosis, because this will increase the risk of leak. FAILURE TO REVERSE
Randomized controlled trials demonstrate that laparo-
scopic colectomy by experienced surgeons is safe and results Maggard looked at colostomy reversal at the population
in better short-term outcomes compared with open surgery. level for the state of California (97). Of the 1,176 patients
Specifically, laparoscopy is associated with decreased oper- who had a Hartmann procedure for diverticular disease,
ative blood loss, less pain, shorter hospitalization, reduced only 65% had a reversal at a mean of 143 days. Younger men
duration of ileus, reduced complication rates, and improved were more likely to have their ostomy reversed, as opposed
quality of life (84,85). A meta-analysis of 25 randomized to older patients, and women. Patients with more comor-
controlled trials comparing open and laparoscopic colorec- bid risk factors also had fewer reversals. When evaluating
tal resection for any indication also documents superior all patients, 35% never had their ostomy reversed during
short-term outcomes associated with the laparoscopic the 4-year study. Complication rates following Hartmann
approach (86). National inpatient sample data also strongly reversal were quite high and included an overall rate of
support laparoscopy over open elective colectomy for diver- 57.4%. Infection (9.1%), aspiration pneumonia (8.7%), pul-
ticulitis (87). Although the majority of published reports monary edema (6%), and acute renal failure (4.9%) were all
included patients with uncomplicated disease, the surgical problematic.
literature supports the laparoscopic approach to compli- Most of the literature quotes a 20%–50% failure of rever-
cated diverticulitis as well (88–90). Hand-assisted laparo- sal rate on patients for a number of factors, including comor-
scopic colectomy may be particularly useful in this setting bidities, age, and failed attempts at reversal (97–100). Boland
(91). Long-term follow-up data from a previously published et al. found that 38% of patients suffered a major compli-
open versus laparoscopic randomized controlled trial with cation after their reversal (98). Failure to restore continuity
a median follow-up of 30 months reported comparable gas- in their population was 10.3%. Due to the morbidity of the
trointestinal quality of life index scores and comparable Hartmann reversal as well as the number of patients who
diverticulitis recurrence rates after surgery (92). In addition, either are not reversed or fail an operative attempt at rever-
the hernia rate in patients who had laparoscopic resection sal, they recommended always trying a primary anastomo-
was one-third of the hernia rate in patients who had open sis first with diversion if possible. In another similar study,
or converted operations. Laparoscopic sigmoid resection Aydin et al. found that Hartmann reversal was associated
for diverticulitis is technically challenging and requires with a higher prevalence of surgical or medical complica-
training and adequate experience. The open approach to tions when compared with primary resection and anasto-
diverticulitis should be performed at the discretion of the mosis (99). The overall postoperative morbidity and 30-day
surgeon as determined by unique patient factors and the mortality rates for Hartmann reversal were 48.5% and 1.7%,
individual surgeon’s judgment and experience. respectively. Patients undergoing a primary resection with
anastomosis suffered a morbidity rate of 26% and mortality
rate of 0.7%. Having controlled for the number of comorbid
conditions, extent of diverticular disease, severity of peri-
COMPLICATIONS toneal contamination, and operative urgency, patients who
underwent Hartmann reversal were 2.1 times more likely
ANASTOMOTIC LEAK to have an adverse surgical event during their postoperative
period.
Elective colectomy has been well documented to carry a very The difficulty with these comparative studies is that
low anastomotic leak rate, of about 1%–3% (60). However, despite attempting to find similar cohorts, patients who
in the face of active inflammation or peritonitis, attempts undergo a Hartmann procedure are usually older, frailer,
at performing a primary anastomosis carry a higher risk of and sicker than those who undergo a primary anastomosis.
anastomotic dehiscence. Primary anastomosis in the set- Surgeons generally wish to correct the problem as fast as pos-
ting of Hinchey stage III or IV carries a leak rate from 8% to sible and get the patient off of the operating room table. This
22% (60,61,93–95). creates the possibility of bias in evaluating the literature, as
patients undergoing Hartmann’s versus primary anastomo-
MORTALITY sis typically have a worse outcome. However, when added
with the risks of a second complex and morbid operation
Elective colectomy also carries with it a low mortality rate, of future stoma takedown, primary anastomosis and diver-
typically less than 1%. A majority of postoperative deaths sion with a loop ileostomy appears much friendlier. If the
patient is able to tolerate the extra 30 minutes required to the downside of a second major operation. Certainly, the
perform a primary anastomosis, one should be performed trends today for diverticular disease are to be less aggressive
with diversion. with operative management, and treat each individual case
based on its own merits as opposed to the more stringent
RECURRENCE OF DIVERTICULITIS AFTER guidelines of the past.
PREVIOUS SURGICAL RESECTION
REFERENCES
Recurrence of diverticulitis or its symptoms following
resection has been reported in 3%–13% of elective cases 1. Cruveillhier J. Paris. 1849;1:590.
(83,101,102). Factors that have been found to contribute 2. Graser E. München Med Wchnschr. 1899;46:721–3.
to the recurrence of diverticulitis after a resection include 3. Peery AF et al. Gastroenterology. 2012;143(5):1179–
shorter resection length and the leaving behind of a cuff 87.e3.
of distal sigmoid (101,102). Most recently, Thaler demon- 4. Hobson KG, Roberts PL. Clin Colon Rectum. 2004;
strated that the level of the anastomosis is the only signifi- 17:147–53.
cant determinant of recurrence after laparoscopic resection 5. Wess, L et al. Gut. 1995;37:91–4.
(83). The practice parameters of the American Society of 6. Sheth A et al. Am J Gastroenterol. 2008;103:1550–6.
Colon and Rectal Surgeons set out several general recom- 7. Rajendra S, Ho JJ. Eur J Gastroenterol Hepatol. 2005;
mendations regarding resection of diverticular disease. For 17:871–5.
elective resection, all thickened, diseased colon, but not 8. Loffeld RJ. Colorectal Dis. 2005;7:559–62.
necessarily the entire proximal diverticula bearing colon, 9. Stemmermann GN. Arch Environ Health. 1970;20:
should be removed. It may be acceptable to retain proxi- 266–73.
mal diverticular colon as long as the remaining bowel is not 10. Granlund J et al. A P & T May. 2012;35(9):1103–7.
hypertrophied. Distally, all of the sigmoid colon should be 11. Strate LS et al. Gastroenterology. 2013;144(4):e14.
removed to the level of the rectum (12). 12. Rafferty J et al. DCR. 2006;49:939–44.
13. Somasekar K et al. J R Coll Sur Edinb. 2002;47:481–4.
TIMING OF CLOSURE 14. Gordon PH. Diverticular disease of the colon. In
Gordon PH, Nivatvongs S (eds). Principles and
Timing of closure continues to be a contentious issue and Practice of Surgery for the Colon, Rectum, and Anus.
has not been fully settled. Traditional teaching is to wait New York, NY: Informa Health Care, 2007, pp. 909–70.
3–4 months to allow the inflammatory process to subside 15. Shabanzadeh DM, Wille-Jorgensen P. Cochrane
and the patient to heal prior to performing another major Database Syst Rev. 2012;11:CD009092.
operation. Mean time intervals in the literature range from 16. Feingold D et al. Dis Colon Rectum. 2014;57:284–94.
120 to 210 days. One study did compare closure at 4 and 17. Andeweg CS et al. Dig Surg. 2013;30(4–6):278–92.
8 months. Complication rates associated with timing of 18. Brenner DJ. Rev Environ Health. 2010;25(1):63–8.
reoperation were 2.5 and 5 times higher at 4 and 8 months, 19. Talabani AJ et al. Int J Colorectal Dis. 2014;29:
respectively (99). Complications from the reversal included 937–45.
anastomotic leak, and rectovaginal fistulas in women. These 20. Ritz JP et al. Surgery. 2011;149(5):606–13.
fistulas are attributed to improper dissection of the vagina 21. Nizri E et al. Tech Coloproctol. 2014;18(2):145–9.
and failure to carefully mobilize the rectum. 22. Laméris W et al. Dis Colon Rectum. 2010;53(6):
896–904.
23. Isacson D et al. Scand J Gastroenterol. 2014;49(12):
1441–6.
CONCLUSION 24. Isacson D et al. Int J Color Dis. 2015;30(9):1229–34.
25. Talabani AJ et al. Int J Colorectal Dis. 2017;32:41–7.
Diverticular disease appears to be increasing in incidence in 26. Lyon C, Clark DC. Am Fam Physician. 2006;74(9):
an ever-widening spectrum of ages throughout the United 1537–44.
States and other developed countries. However, with more 27. Chang C-C, Wang S-S. Int J Gerontolog. 2007;1(2):
experience with the disease process, coupled with better 77–82.
medical therapies and diagnostic measures, more patients 28. Laurell H et al. Gerontology. 2006;52(6):339–44.
are able to be managed conservatively than ever before. 29. Bailey MB et al. J Am Coll Surg. 2013;217(5):874–80.
Uncomplicated diverticular disease may be treated medi- 30. Biondo S et al. Am J Surg. 2016;212(3):384–90.
cally without fear that recurrent episodes will lead to more 31. Biondo S et al. Am J Surg. 2012;204:172–9.
complicated findings. Complicated disease is being man- 32. Nelson RS, Thorson AG. Operative and nonopera-
aged medically more aggressively than ever before in an tive therapy for diverticuar disease. In Whitlow CB,
effort to prevent emergent operations. Primary anastomo- Beck DE, Margolin DA, Hicks TC, Timmcke AE. (eds).
sis with diversion as opposed to the traditional two-staged Improved Outcomes in Colon and Rectal Surgery.
Hartmann procedure appears to be equally effective without London: Informa Healthcare, 2010, pp. 249–62.
References 243
33. Rafferty J et al. DCR. 2006;49:939–44. 65. Wong WD et al. Dis Colon Rectum. 2000;43:290–7.
34. Kaiser AM et al. Am J Gast. 2005;100:910–7. 66. Köhler L et al. Surg Endosc. 1999;13:430–6.
35. Aydin HN, Remzi FH. Dig Liver Dis. 2004;36:435–45. 67. Biondo S et al. Br J Surg. 2002;89:1137–41.
36. Golfieri R, Cappelli A. Tech Coloproctol. 2007;11: 68. Chapman J et al. Ann Surg. 2005;242:576–81.
197–208. 69. Gaertner WB et al. Dis Colon Rectum. 2013;56:
37. Boulos PB. Best Pract Res Clin Gastroenterol. 2002; 622–6.
16:649–62. 70. Janes S et al. Br J Surg. 2005;92:133–42.
38. Ambrosetti P et al. Dis Colon Rectum. 2000;43: 71. Chapman JR et al. Ann Surg. 2006;243:876–80.
1363–7. 72. Pittet O et al. World J Surg. 2009;33:547–52.
39. Hinchey EJ et al. Adv Surg. 1978;12:85–109. 73. Li D et al. Dis Colon Rectum. 2016;59:332–39.
40. Thorson, AG, Goldberg SM. Benign colon: 74. Katz LH et al. J Gastroenterol Hepatol. 2013;28:
Diverticular disease. In Wolff BG, Fleshman JW, 1274–81.
Beck DE, Pemberton JH, Wexner SD (eds). The 75. Li D et al. Dis Colon Rectum. 2014;57:1397–405.
ASCRS Textbook of Colon and Rectal Surgery. New 76. Li D et al. Ann Surg. 2014;260:423–30.
York: Springer-Verlag, 2007, p. 271. 77. Mueller MH et al. Eur J Gastroenterol Hepatol. 2005;
41. Alonso S et al. Colorectal Dis. 2010;12:278–82. 17:649–54.
42. Etzioni DA et al. Dis Colon Rectum. 2010;53:861–5. 78. Elagili F et al. Dis Colon Rectum. 2014;57:331–6.
43. Balasubramanian I et al. Dig Surg. 2017;34:151–60. 79. Eglinton T et al. Br J Surg. 2010;97:952–7.
44. Jackson JD, Hammond T. Int J Colorectal Dis. 2014; 80. Nelson RS et al. Am J Surg. 2008;196:969–73.
29:775–81. 81. Salem L et al. JACS. 2004;199:904–12.
45. Isacson D et al. Scand J Gastroenterol. 2014;49(12): 82. Dozois EJ. J Gastrointest Surg. 2008;12:1321–3.
1441–6. 83. Thaler K et al. Dis Colon Rectum. 2003;46:385–8.
46. Chabok A et al. Br J Surg. 2012;99:532–9. 84. Klarenbeek BR et al. Ann Surg. 2009;249:39–44.
47. Maconi G et al. Dis Colon Rectum. 2011;54:1326–38. 85. Gervaz P et al. Ann Surg. 2010;252:3–8.
48. Hall JF et al. Dis Colon Rectum. 2011;54:283–8. 86. Schwenk W et al. Cochrane Database Syst Rev. 2005:
49. Dharmarajan S et al. Dis Colon Rectum. 2011;54: CD003145.
663–71. 87. Masoomi H et al. World J Surg. 2011;35:2143–8.
50. Sallinen VJ et al. Dis Colon Rectum. 2014;5(7):875–81. 88. Scheidbach H et al. Dis Colon Rectum. 2004;47:
51. Siewert B et al. Am J Roentol. 2006;186:680–6. 1883–8.
52. Singh B et al. Ann R Coll Surg Engl. 2008;90(4): 89. Bartus CM et al. Dis Colon Rectum. 2005;48:233–6.
297–301. 90. Jones OM et al. Ann Surg. 2008;248:1092–7.
53. Wolff JH et al. J Clin Gastroenterol. 2008;42:472–5. 91. Lee SW et al. Dis Colon Rectum. 2006;49:464–9.
54. Eskaros S et al. Emerg Radiol. 2009;16:473–6. 92. Gervaz P et al. Surg Endosc. 2011;25:3373–8.
55. Moraitis D et al. Am Surg. 2006;72:269–71. 93. Stumpf MJ et al. Am Surg. 2007;73:787–91.
56. Pereira JM et al. Radiographics. 2004;24:703–15. 94. Gooszen AW et al. Eur J Surg. 2001;167:35–9.
57. Lau KC et al. Dis Colon Rectum. 2011;54:1265–70. 95. Richter S et al. World J Surg. 2006;30:1027–32.
58. Wong WD et al. Dis Colon Rectum. 2000;43:290–7. 96. Landen S, Nafteux P. Acta Chir Belg. 2002;102:
59. Constantinides VA et al. Ann Surg. 2007;245:94–103. 24–9.
60. Salem L, Flum DR. DCR. 2004;47:1953–64. 97. Maggard MA et al. Am Surg. 2004;70:928–31.
61. Constantinides VA et al. BJS. 2006;93:1503–13. 98. Boland E et al. Am Surg July. 2007;73:664–8.
62. Wigmore SJ et al. Br J Surg. 1995;82:27–30. 99. Aydin HN et al. DCR. 2005;48:2117–26.
63. Regenet N et al. Int J Colorectal Dis. 2003;18:503–7. 100. Oomen JT et al. Dig Surg. 2005;22:419–25.
64. Buchs NC et al. World J Gastrointest Surg. 2015;7(11): 101. Benn PL et al. Am J Surg. 1986;151:269–71.
313–8. 102. Leigh JE et al. Am J Surg. 1962;103:51–4.
26
Abdominal surgery for colorectal cancer
JASON F. HALL AND ROCCO RICCIARDI
Performance Measures, Payment, and Performance

CHALLENGING CASE Improvement Project. With an aim toward equitable and
reliable high-quality care, our nation has started down the
A 65-year-old woman with no significant past medical road of more measurement in order to gain better outcomes.
history underwent a laparoscopic sigmoid colectomy Quality has always been a major focus of attention for
for colon cancer. The patient had delayed return of surgeons, but with the advent of pay for reporting, attempts
bowel function and on postoperative day 7, devel- to measure quality of care have accelerated. Presently, the
oped a fever as well as increasing abdominal pain. A focus of quality assessment is on process measures such
contrast enema revealed an anastomotic leak. as antibiotic prophylaxis and venous thrombosis prophy-
laxis. In time, however, it is likely that morbidity, mortality,
quality of life, and patient satisfaction will become increas-
ingly important components of our assessment. Many of
CASE MANAGEMENT the aforementioned metrics are currently in use in the field
of colon and rectal surgery. Oncologic measures are under
The patient was returned to the operating room for active development and will likely standardize the way we
a laparotomy, abdominal washout, and end colos- manage colon and rectal cancer. It is for this reason that
tomy. The anastomosis had a pinpoint defect poste- our chapter on improving outcomes for abdominal surgery
riorly without evidence of tension or ischemia. She in colorectal cancer is particularly important to readers of
responded well to antibiotics and was discharged this book and, in particular, surgeons who perform these
16 days following her initial procedure. procedures. Procedures of the colon and rectum are charac-
terized by high rates of morbidity and mortality; thus, tech-
niques to reduce the burden of disease from colon and rectal
surgery are important.
INTRODUCTION Surgical resection remains the standard of care for cura-
tive treatment of colorectal cancer. In order to improve out-
The need for improved quality in health care has reached comes, numerous surgical techniques have been proposed
the consciousness of policymakers, providers, payers, and to reduce complications. However, as with all surgical
patients. Following the publication of two Institute of procedures, complications related to preoperative, intra-
Medicine reports: To Err Is Human: Building a Safer Health operative, and postoperative factors occur more frequently
System and Crossing the Quality Chasm: A New Health than most practitioners recognize. As the collective inter-
System for the Twenty-First Century, the quality of our est in surgical outcomes has increased, surgeons have been
nation’s health care has been critically examined while out- charged not only with improving their outcomes but also
comes have been questioned. Increased attention to health with documenting that improvements in surgical care are
care has led to demands for better access and distribution real and generalizable to the population served. This chap-
of care. Rigorous monitoring and quality assessment, as ter provides a systematic overview of the major recent devel-
well as pay for reporting and potentially pay for perfor- opments in outcomes and quality in colorectal cancer. It
mance are likely inevitable. In order to accelerate the diffu- also provides an evidence-based platform for the minimiza-
sion and pace of quality improvement efforts, the Institute tion and management of common surgical complications in
of Medicine launched the Redesigning Health Insurance order to improve outcomes.
244
Surgical outcomes in colorectal cancer 245
distal bowel margin is generally adhered to for colon can-

ONCOLOGIC OUTCOMES IN cers, despite data suggesting that mural tumor migration
COLORECTAL CANCER rarely occurs beyond a 2 cm margin in either the proximal
or distal direction to cancer (10). In the setting of rectal can-
A number of factors affect prognosis after colorectal resection cer, a distal margin of as little as 1–2 cm has become the rule
for cancer. Generally, oncologic outcomes for colorectal can- in lower tumors, while it is important to note that survival is
cer resection are inversely proportional to the patients’ stage adversely affected by a distal margin of less than 0.8 cm (11).
of disease. Following attempted curative resection, survival Controversy regarding the point of ligation of the vascu-
parallels the TNM stage (I, well above 90%; II, 65%–90%; lar pedicle has been a point of contention for years. Many
and III, 45%–75%) (1,2). More specifically, local extent of dis- surgeons argue that a high ligation of the vascular pedicle
ease, the presence of metastatic disease, nodal involvement, is critical to ensuring good oncologic outcomes. However, a
adequacy of regional node harvest, incomplete resection, comparison of high versus more distal ligation for left-sided
preoperative carcinoembryonic antigen level, tumor grade, cancers by the French Association for Surgical Research
and tumor biology have all been correlated with oncologic demonstrated no difference in survival (12). Despite the
outcomes (3–7). Conversely, tumor size and gross tumor con- fact that advocates of proximal vascular ligation have little
figuration have not been correlated with prognosis following evidence to support improved outcomes, the net effect of
surgery. Despite these patient and tumor factors, the surgeon high ligation may be to improve lymph node sampling. An
can greatly influence oncologic outcomes by performing a adequate mesenteric resection to include an appropriate
proper preoperative oncologic evaluation, adequate tumor nodal sample importantly predicts survival (5). Given the
resection, and satisfactory nodal harvest. overwhelming data in this area, the College of American
Pathologists has recommended additional techniques to
PREOPERATIVE EVALUATION enhance nodal recovery if less than 12 nodes are identified
on initial examination (6). Similarly, several national orga-
Following the diagnosis of colorectal cancer, the surgeon nizations have proposed setting benchmarks of 12 lymph
should assess the patients’ surgical risk while determin- nodes as a proxy of an adequate oncologic resection for
ing local and distant extent of disease. Despite the lack of colorectal cancer (13–17). This proposed benchmark may
consensus for preoperative testing of colorectal cancer, we someday serve as an important quality measure to compare
adhere to a thorough evaluation of the abdomen and chest surgeons and providers treating colorectal cancer. At this
to rule out distant disease while determining the local time, however, advocates of lymph node benchmarks have
extent of disease, especially when planning a laparoscopic no evidence that such thresholds will result in real measur-
resection. Most importantly, however, prior to completing able improvement in patient outcome.
a colorectal resection, colorectal cancer patients should Surgical advances in laparoscopic technique have resulted
have a complete assessment of the colon. Approximately 5% in rising enthusiasm for applications of this technique to
of patients will have a synchronous colorectal cancer, and cancer resections. These advances have resulted in some
25%–76% patients will have a synchronous adenomatous improvement in short-term outcomes, such as postopera-
polyp (8). Colonoscopy is advantageous for identifying the tive pain, length of ileus, and hospital stay. Most importantly,
position of the lesion while permitting tattooing for intra- multicenter trials of laparoscopic resection versus open resec-
operative localization during laparoscopic resections. Also, tions for colon cancer reveal no compromise in oncologic
some lesions are prohibitively large, and thus the planned outcomes (18). Despite early reports of port site recurrences
operation may require alteration to ensure a complete from laparoscopic oncologic resections, well-designed trials
resection. Virtual computed tomography (CT) and barium support the oncologic equivalency of laparoscopic colectomy
enema are other methods that may be applied if a colonos- in the hands of experienced surgeons (18,19). Thus, oncologic
copy cannot be performed for technical reasons. In addi- outcomes are comparable for the experienced laparoscopic
tion, we have had excellent results with intraoperative CO2 surgeon as compared to open surgery for colon cancer, with
colonoscopy. Thus, an adequate assessment of the patients’ the proviso that population-based data are unavailable at this
colonic disease reduces the likelihood that the patient may time. At this time, oncologic results for laparoscopic proc-
require another procedure for missed lesion. tectomy for rectal cancer have been concerning (20). Further
data are needed to demonstrate oncologic equivalency for
SURGICAL TECHNIQUE laparoscopic approaches to proctectomy.
In addition to an adequate preoperative assessment, ade-

quate tumor resection is critical to ensuring a good prog-
SURGICAL OUTCOMES IN
nosis. Appropriate surgical margins critically influence
outcome after colorectal cancer resection. In addition to the
COLORECTAL CANCER
proximal and distal bowel margins, the radial or circum-
ferential margin is particularly important in the treatment There are a number of complications that occur following
of both colon and rectal cancers (9). A 5 cm proximal and colorectal resections for cancer. Although some are difficult
246 Abdominal surgery for colorectal cancer
to avoid, we stress methods to reduce the frequency of these can affect outcome. Attention should be focused on the
complications and improve outcomes. patient’s preoperative nutritional status. Golub et al. have
demonstrated that an albumin concentration less than 3 g/L
ANASTOMOTIC COMPLICATIONS is associated with anastomotic leakage (22). In a multivari-
ate analysis, these authors also demonstrated a relationship
Anastomotic complications are some of the most feared between preoperative corticosteroid use, peritonitis, bowel
complications in colorectal surgery patients. Although rare, obstruction, and chronic obstructive pulmonary disease, as
the development of anastomotic complications results in a well as perioperative transfusion and the incidence of anas-
prolonged postoperative stay with a high cost to the patient, tomotic leakage (22). Other authors have identified peri-
the health care system, and society. Perioperative anasto- operative conditions that increase the risk of anastomotic
motic complications can also lead to long-term consequences leakage, and they include obesity, malnutrition, weight loss
including stricture, abdominal wall hernia, permanent greater than 5 kg, and use of alcohol (27). If the patient’s
diversion, poor functional outcome, and need for reopera- nutritional status is in question or the local conditions are
tive therapy. A number of variables have been linked to the not favorable, it advisable not to construct an anastomosis.
development of anastomotic complications, particularly the Once the unfavorable circumstances have been corrected,
technique employed, the conditions under which the anasto- the patient can undergo restoration of intestinal continuity
mosis is constructed, and other patient characteristics. under more favorable circumstances.
Bowel preparations have traditionally been employed to
Anastomotic leak clear the bowel of feces before colorectal operations. This
practice was thought to decrease the likelihood of anas-
One of the most devastating outcomes of a new anastomo- tomotic leak by limiting the passage of stool through the
sis for colorectal cancer is anastomotic leak, occurring in newly constructed anastomosis. More recent data reveal
3%–6% of all colorectal cases. These leaks occur more com- an increase in anastomotic complications with the routine
monly with more distal resections and are reported to be as use of bowel preparations (28). Others have similarly dem-
high as 15.3% for low rectal reconstructions (21). Although onstrated more wound infections in addition to increased
the development of leak is frequently attributed to surgeon rates of anastomotic leak in patients receiving mechanical
error, patient characteristics also importantly influence the bowel preparations (29). Because of the temporary starva-
development of leaks. For example, renal failure, chronic tion and electrolyte imbalance sometimes associated with
obstructive pulmonary disease, steroid use, elevated white mechanical bowel preparations, it is not clear that this prac-
blood count, and malnutrition have all been attributed to tice represents anything more than surgical dogma. Despite
anastomotic leak (22,23). In addition, operative factors such the growing body of evidence against the routine use of
as low rectal anastomoses, intraoperative septic conditions, bowel preparation, surgeons in North America seem slow
difficulties encountered during the anastomosis, and use to move away from this long-held practice, with many sug-
of blood transfusion have been implicated (22). Although gesting that colonic manipulation of an unprepped bowel
surgical construction of the anastomosis is an important during laparoscopic surgery is difficult.
variable, there is no difference in the development of anas- Once an anastomotic leak develops, it usually becomes
tomotic leak whether the reconstruction is stapled, hand- evident within 5–8 days following the procedure. However,
sewn in one layer, or even two layers (24,25). in a recent series of 1,223 patients with intestinal anastomo-
With respect to anastomotic technique, emphasis should ses, 36% were identified more than 30 days postoperatively
be placed on providing an adequate blood supply and ensur- (30). The diagnosis is usually suspected by clinical factors
ing a tension-free anastomosis. Adequate blood supply can and often confirmed by radiologic examination. Patients
be confirmed with multiple methods: by dividing the mar- with early clinical evidence of anastomotic leak can pres-
ginal artery of Drummond or other arcades and encounter- ent with fever, tachycardia, abdominal distention and
ing pulsatile bleeding or by confirming bleeding at the cut tenderness, ileus, early diarrhea, or possibly septic shock.
edge of the colon. A tension-free anastomosis is also critical Depending on the patient’s clinical condition, the pres-
and can be facilitated by high ligation of the feeding vessel, ence of any one of these factors is indication for examina-
although this maneuver is not always critical. Other tech- tion with a radiologic study. If there are obvious signs of
niques to reduce tension and increase mobility include sep- peritonitis or hemodynamic collapse, then urgent explor-
aration of the greater omentum from the transverse colon atory laparotomy is often preferable. Radiological investi-
and adequate mobilization of the approximating ends. gation can be performed with either abdominopelvic CT
In the setting of low pelvic anastomoses, especially those or with a soluble contrast enema (Figure 26.1). There have
anastomoses constructed following the use of neoadjuvant been conflicting reports as to the superiority of each tech-
therapy, a protective proximal intestinal stoma should be nique (31,32); however, CT scans have the additional benefit
considered in order to reduce the life-threatening conse- of demonstrating other intraabdominal pathology such as
quences of anastomotic leak (26). hematoma or abscess.
In addition to lack of tension and adequate blood flow, If an anastomotic leak is demonstrated by clinical or
the local conditions under which an anastomosis is created radiologic means, antibiotics should be administered and
Surgical outcomes in colorectal cancer / Anastomotic complications 247
asymptomatic but may also present with partial or com-

plete large bowel obstruction. The mechanism of stricture
formation is not completely understood but may be related
to anastomotic leakage, pelvic sepsis, radiation injury, or
local ischemia (36,37). Numerous nonrandomized trials
and meta-analyses have demonstrated a higher rate of ste-
nosis with end-to-end stapled anastomosis. The presence of
a proximal diverting colostomy also seems to increase the
risk of stricture, possibly due to the lack of dilation by the
fecal stream (38–40). However, patients who undergo proxi-
mal diverting ileostomy have generally more difficult anas-
tomoses, and often the ileostomy is formed because of the
potentially higher risk for leakage. Thus, the higher risk of
leakage in these patients may be confounded by local fac-
tors or patient variables that have been difficult to describe
to date.
Strictures typically present within the first postoperative
year (41) and can be managed with a number of different
modalities depending on the anastomotic site. Many are
asymptomatic and will resolve on their own as the bulk of
the fecal stream slowly dilates the stenosis over the course
of a few months. Before any therapeutic procedure is under-
taken, the operator should be confident that a new stricture
does not represent a cancer recurrence. If there is uncer-
tainty by visual inspection, the diagnosis can easily be ascer-
Figure 26.1 Contrast radiograph demonstrating anasto- tained through a formal biopsy of the anastomosis. Most
motic leak. colonic strictures following surgery can be managed with
sequential digital manipulation, bougeinage, or hydrostatic
the patient resuscitated. Surgical exploration is then indi- balloon dilation. Patients often require several treatments
cated to wash out the abdominal cavity and examine the before complete resolution as repeat dilations are more com-
anastomosis. If conditions are favorable, an intraperitoneal mon following resections for cancer than for benign causes
anastomosis can be reconstructed, although most anasto- (42). Newer techniques involve the use of endoscopic stents
motic leaks generally require the construction of a stoma. as well as endoscopic transanal resection of strictures. Small
Alternatively, the management of left-sided and low pelvic published series using these techniques report safe and sat-
anastomoses is more complex. If the leak is secondary to isfactory long-term outcomes (43). Surgical revision of the
bowel necrosis or ischemia and there is not sufficient bowel anastomosis is occasionally necessary when the stricture is
for reanastomosis, a colostomy should be created with a not accessible by endoscopic means or recurs after numer-
Hartmann pouch. This procedure should be performed ous less invasive procedures.
with consideration of the fact that the majority of these
patients will be left with permanent stoma (33). If there is Anastomotic bleeding
sufficient bowel for reanastomosis and local conditions are
favorable, revision of the anastomosis can be considered Postoperative bleeding related to intestinal anastomosis is
with the protection of a proximal loop ileostomy. If there a relatively rare but potentially serious event. The incidence
is dense inflammation surrounding the anastomosis, a loop in stapled colorectal anastomoses is 1.8%–5% (44,45). The
ileostomy should be constructed and a large drain placed in method of construction, whether stapled or hand-sewn,
the perianastomotic area. A number of anastomotic leaks appears to be unrelated to the development of this complica-
associated with localized peritonitis and abscess formation tion (46). The diagnosis of anastomotic bleeding is typically
can often be managed nonoperatively (32). If there is a con- inductive as patients will pass variable amounts of maroon-
tained collection, the patient should be treated with IV anti- colored blood with their first bowel movement. More active
biotics, and drainage of the abscess should be considered by bleeding commonly presents with large amounts of blood
radiologic means (34). per rectum. To prevent this complication, we perform a sim-
ple examination of the staple line through the enterotomy.
Anastomotic stricture All visible bleeding is controlled with sutures rather than
electrocautery as applying thermal energy to the staple line
Anastomotic stricture is a common occurrence follow- may increase the likelihood of a full-thickness burn injury.
ing colorectal anastomosis and has been reported in up The initial management of postoperative hemorrhage
to 30% of cases in some series (35). Strictures are often is typically nonoperative. Greater than 80% of patients
will stop bleeding without intervention, but nearly 50% (21,57,58), although these figures likely underestimate the
of patients will require a transfusion (47). Often, simple incidence of the problem as conservatively managed splenic
techniques such as correcting the coagulopathy and halt- injuries are rarely reported. Splenic injury is associated with
ing unfractionated or low molecular weight heparin are the proximity of the lesion to the splenic flexure as trac-
sufficient. Alternatively, treatment for hemodynamically tion on the peritoneal band attaching the greater omentum
stable patients includes endoscopic electrocoagulation of and spleen appears to be the most common mechanism
the anastomotic line (48) or injection of the staple line with by which the spleen is injured (59–61). Other mechanisms
epinephrine or clips (49). Although others have proposed of injury include retractor and direct instrumental dam-
that proximal colonic anastomoses should not be treated age (62). Langevin reported no injuries to the spleen in 733
endoscopically (48), the data against this belief are minimal. procedures in which the splenic flexure was not mobilized,
If endoscopic methods fail, some patients are candidates but 3.1% of patients requiring splenic flexure takedown sus-
for angiographic embolization or vasopressin treatment. tained splenic injuries (21). Mortality rates are higher in
Obviously, angiographic options should be exercised with patients who sustain splenic injury after gastrointestinal sur-
care as embolization may interrupt the blood supply to the gery, particularly colorectal surgery (63,64). Splenic injury is
anastomosis and thus result in bowel infarction as well as associated with a higher incidence of early infections, poten-
anastomotic leak (50). In addition, the use of vasopressin is tially from hematoma formation and subsequent superin-
also associated with myocardial and intestinal ischemia and fection (58) or loss of splenic function. In a recent review of
should be employed with caution in patients with risk fac- California Cancer Registry and California Patient Discharge
tors for heart disease (51,52). Failure of the aforementioned Data, patients undergoing colorectal cancer resection with
hemostasis methods will often require exploratory laparot- inadvertent splenectomy had an increased length of stay and
omy and revision of the anastomosis. a 40% increase in the probability of death (65).
There are few evidence-based recommendations for
PELVIC HEMORRHAGE avoiding intraoperative splenic injury, but basic surgical
principles are obviously essential. To maximize exposure,
Massive pelvic bleeding is a difficult complication that can the surgical incision should be appropriately elongated
occur rarely during proctectomy or retroperitoneal dissec- in order to obtain adequate, tension-free visualization of
tion. This bleeding usually results from inadvertent viola- the appropriate structures in open surgery. All hand-held
tion of the avascular presacral plane and resultant damage and self-retaining retractors should be placed with care
to the presacral veins. Presacral venous hemorrhage is dif- and under direct visualization. Some authors have recom-
ficult to control and can be a significant source of post- mended a modified lithotomy position with the surgeon
operative morbidity and mortality (53). Conventional standing between the patient’s legs during flexure mobi-
methods of hemostasis rarely are effective and usually result lization. This positioning permits clearer visualization of
in increased bleeding. If encountered, bleeding should be the structures in the left upper quadrant (21). Unnecessary
controlled with direct pressure while the anesthesiology traction on the transverse and left colon should be avoided.
team appropriately resuscitates the patient. Laparotomy Consideration should be given to dividing the lienocolic lig-
sponges are used to tamponade bleeding, while microfibril- aments before commencing any left colonic resection (62).
lar collagen and absorbable gelatin can be used. If simple If there is suspicion of tumor invasion into the spleen, an
tamponade does not control the bleeding, then sterile tita- en bloc resection should be performed. There are some data
nium thumbtacks can be inserted into the bleeding point on to suggest the benefits of laparoscopy in mobilization of the
the sacrum (54). In addition, endoscopic multifeed staplers spleen. Malek reported a review of iatrogenic splenecto-
used in laparoscopic mesh hernia repairs can be used (55). mies following open and laparoscopic colon resections. The
Others have described fixing a 4 cm2 piece of rectus mus- authors found 13 iatrogenic splenectomies in 5,477 open
cle to the bleeding vessel while applying a high-frequency resections, but none following 1,911 laparoscopic resections
electrical current to the muscle until it adheres to the pre- (66). Although the authors were unable to adjust for opera-
sacral fascia (56). Alternatively, bonewax can be used with tive difficulty, it would seem that laparoscopic mobilization
some efficacy on the sacrum. In the most difficult circum- may clarify splenic flexure visualization (Figure 26.2).
stances, the pelvis is packed with sponges, and the patient Prompt recognition at the time of surgery is the first step
returned to the operating room in 1–2 days for laparotomy to the successful management of iatrogenic splenic inju-
pad removal (57). ries. Once an injury to the spleen is recognized, there are
two options, either splenectomy or splenic preservation.
SPLENIC INJURY Timely management allows the surgeon to manage bleed-
ing at the first operation, while delayed recognition results
Iatrogenic injury to the spleen is a potentially serious com- in reduced chances of splenic salvage (67). Optimally, the
plication of colectomy with significant long-term adverse surgeon should attempt to salvage the spleen unless blood
consequences. It is defined as any injury to the spleen caused loss prohibits the more time-intensive salvage methods.
by the operating team during a surgical procedure. Splenic Techniques for splenic salvage are generally extrapolated
injury occurs during 1.2%–8% of colorectal resections from the trauma literature. Minor bleeding from capsular
Surgical outcomes in colorectal cancer / Autonomic nerve injury 249
Figure 26.2 Laparoscopic mobilization of splenic flexure.
tears generally improves with gentle tamponade, whereas surgical procedures often result in a very satisfactory out-
more active bleeders may require more intervention to come in the about 94% of cases (71).
achieve hemostasis. There is a long experience with various In order to prevent ureteral injuries, patients with dif-
hemostatic agents such as thrombin, absorbable regener- ficult anatomy, such as extensive pelvic adhesions after
ated cellulose, and microfibrillar collagen. These are often proctectomy, a large pelvic mass, or a phlegmon that makes
placed on top of the bleeder and underneath a surgical pack identification of normal anatomy difficult, should be con-
for tamponade (68). If these simple measures fail, bleeding sidered for preoperative stent placement. Ureteral stents
can be controlled by segmental ligation of the feeding hilar permit quicker intraoperative ureter identification but do
vessels or splenorrhaphy. With severe splenic injury, com- not completely eliminate the risk of injury; rather, ure-
plicated by continued hemorrhage and hemodynamic insta- teral stents permit quicker recognition of ureteral injuries
bility, the surgeon should obviously consider splenectomy. and immediate repair. Since these repairs can be techni-
If splenectomy is performed, the patient should be admin- cally challenging, they should be performed by a surgeon
istered pneumococcal, meningococcal, and Haemophilus who is well versed with ureteral repair techniques. General
influenzae vaccine in order to prevent overwhelming post- guidelines include debridement of necrotic tissues, ensur-
splenectomy sepsis (69). ing excellent blood supply, and performing a tension-free
anastomosis. More distal injuries of the pelvic portion of the
URETERAL INJURIES ureter may be handled by reimplantation (72). Additional
discussion on urologic injuries is presented in Chapter 41.
Due to the proximity of the ureters to the colon, injury is
a common concern during colorectal operations. Although AUTONOMIC NERVE INJURY
injuries to the ureters are uncommon during simple resec-
tions, when they occur, they can lead to significant diffi- Genitourinary function can be greatly altered by injury to
culty. The ureters are most commonly injured in colorectal the pelvic parasympathetic and sympathetic nerves dur-
procedures during one of several maneuvers: while ligating ing colorectal resections. Proper oncologic resection for
the inferior mesenteric artery or dissecting at the sacral rectal cancer has been associated with a significant inci-
promontory or laterally in the pelvis during division of the dence (10%–69%) of urinary and sexual dysfunction (73).
lateral stalks of the rectum. Unfortunately only 20%–30% Although urinary dysfunction is often limited to the first
of intraoperative ureteral injuries are recognized at the time few postoperative days, sexual dysfunction may persist for
of the transgression (70). Despite the fact that few injuries months or indefinitely. Both forms of postoperative dysfunc-
are recognized intraoperatively, ureteral injuries are best tion are related to the patients’ preoperative function. Total
treated during the initial operation as the local conditions mesorectal excision with autonomic nerve preservation has
are likely to be the most favorable for a successful repair. been advocated as an effective approach to the minimiza-
Prompt diagnosis and institution of appropriate corrective tion of pelvic nerve injury. This technique mobilizes the
Mesorectum
Lahey clinic
Autonomic nerves
Figure 26.3 Rectal mobilization demonstrating autonomic nerves.
mesorectum circumferentially with sharp dissection along status, use of chemoradiation, anastomotic technique and
the correct pelvic parietal planes while avoiding the pelvic local factors, and the development of anastomotic compli-
nerves (Figure 26.3) (73). In addition, damage to the sym- cations. Today, sphincter preservation procedures are being
pathetic plexus is often encountered during high ligation performed with increasing frequency for the management
of the inferior mesenteric artery. The hypogastric nerves of midrectal and low rectal cancers. However, preservation
should be identified as they course over the sacral promon- of intestinal continuity frequently leads to continence dis-
tory and preserved. Anterior dissection should be avoided turbances, which range from inadvertent passage of flatus to
when unnecessary as dissecting in Denonvilliers fascia frank leakage of stool necessitating pad use (76–82). Patients
places the nervi erigentes at risk. We preserve Denonvilliers with a straight low anastomosis may also suffer from urgency,
fascia unless the tumor is anterior or circumferential. When frequency, and clustering of bowel movements. Poor func-
total mesorectal excision and autonomic nerve preservation tion after sphincter salvage largely results from a combina-
are combined, several authors have demonstrated a low tion of four factors: damage to the sphincter complex, loss
frequency of bladder and sexual dysfunction (74,75). Both of normal anorectal sensation, a reduced rectal capacity
of these techniques should be considered standard when and compliance, and a reduction in large intestine length
undertaking resection of the rectum. resulting in more liquid effluent reaching the anal canal (83).
Increased effort has thus been exerted to minimize dysfunc-
tion following proctectomy with a focus toward good tech-
nique and reconstruction using a neorectum.
FUNCTIONAL OUTCOMES In recent years, improved functional outcomes have
been reported following anastomotic reconstruction with
Gastrointestinal function following rectal cancer resection a colonic J-pouch or coloplasty. The most studied and
is quite variable depending on the patients’ preoperative accepted reconstruction option at this time is the colonic
References 251
J-pouch, which is associated with improved physiological complications, methods to improve outcomes, surgical
and functional outcomes as compared to the straight anas- outcome metrics, and the future of patient-centered out-
tomosis (83). Until recently, the advantages of a J-pouch comes in colorectal cancer surgery. We have particularly
were thought to be short-lived, but a recent multicentered emphasized the impact of the quality movement and the
study revealed sustained functional advantages after 2 role of outcomes on quality measurement and assurance.
years postoperatively compared to both the straight colo- The information presented in this chapter is critical as
anal anastomosis and the coloplasty (84). Another common quality metrics and measurement are likely to become
technique is the use of a side-to-end Baker anastomosis, more and more important to the individual practitioner.
which in the short term has equivalent functional results to Given emerging refinements of open as well as laparo-
the colonic J-pouch or coloplasty (85). Unfortunately, recent scopic techniques, as well as postoperative care, outcomes
data suggest difficulty with complete evacuation of the measurement will become more and more important
Baker anastomosis as compared to the colonic J-pouch (85). as we prove to our patients, payers, Congress, and our-
It is for these reasons that we recommend colonic J-pouch selves that our outcomes are sufficient. Although payers
reconstruction for low anastomoses at 6 cm or closer to the and other government groups have become the drivers of
anal verge. quality improvement, it is our duty to measure our own
outcomes, assess the quality of care that we provide, and
compare our own results with those of our colleagues.
Internal efforts to improve quality are the most likely
PATIENT-CENTERED OUTCOMES to bring about real meaningful changes in outcomes for
colorectal cancer.
There has been a growing interest in medical and surgi-
cal outcomes that are most important to patients rather
than traditional measures of morbidity and mortality. REFERENCES
This interest has developed from the growing concern that
medical care fails to properly assess all of the needs of the 1. Jagoditsch M et al. World J Surg 2000;24:1264–70.
patient. Patient-centered outcomes, such as patient satis- 2. Mcdermott FT et al. Br J Surg 1981;68:850.
faction or quality of life, are particularly meaningful for 3. Chapuis PH et al. Br J Surg 1985;72:698–702.
colorectal cancer patients. Despite the embryonic status 4. Tominaga T et al. Five-year follow-up report. Cancer
of patient-centered outcomes in surgical fields, the Agency 1996;78:403–8.
for Healthcare Research and Quality has developed the 5. Swanson RS et al. Ann Surg Oncol 2003;10:65–71.
Consumer Assessment of Healthcare Providers and Systems 6. Compton CC et al. College of American Pathologists
(CAHPS) program, a joint public and private initiative to Consensus Statement 1999. Arch Pathol Lab Med
develop standardized surveys of patients’ experiences with 2000;124:979–94.
ambulatory and facility-level care (86). CAHPS surveys 7. Wolmark N et al. Ann Surg 1984;199:375–82.
provide information about patients’ care experiences rather 8. Langevin JM, Nivatvongs S. Am J Surg 1984;147:
than traditional clinical performance indicators, such as 330–3.
cured of disease or morbidity and mortality. 9. Nagtegaal ID, Quirke P. J Clin Oncol 2008;26:303–12.
Surgeons’ ability to measure and understand quality of 10. Quirke P et al. Lancet 1986;2:996–8.
life and other patient-centered outcome would be of great 11. Vernava AM et al. Surg Gynecol Obstet 1992;175:
value to the colorectal cancer patient undergoing surgery. 333–6.
In practical terms, patient expectations would be clearer. 12. Rouffet F et al. Dis Colon Rectum 1994;37:651–9.
Unfortunately, few validated and useful patient-centered 13. Nelson H et al. J Natl Cancer Inst 2001;93:583–96.
metrics have been evaluated and even fewer are in use today 14. Otchy D et al. Dis Colon Rectum 2004;47:1269–84.
(87). Despite the lack of real progress in this area, insurers, 15. Hermanek P. Langenbecks Arch Chir Suppl
patients, and others are very interested in determining what Kongressbd 1991;277–81.
patients think of the treatments we provide them. At this 16. ASCO/NCCN Quality Measures: http://ascopubs.
time, surgeons need to work closely with others to provide org/doi/abs/10.1200/jco.2008.16.5068
more comprehensive and nontraditional outcomes following 17. Compton CC. Arch Pathol Lab Med 2000;124:
surgical care. 1016–25.
18. Bonjer HJ et al. Arch Surg 2007;142:298–303.
19. The Clinical Outcomes of Surgical Therapy Study
Group. NEJM 2004;350:2050–59.
CONCLUSION 20. Guillou PJ et al. Lancet 2005;365:1718–26.
21. Langevin JM et al. Surg Gynecol Obstet 1984;159:
In summary, this chapter on improving outcomes for 139–44.
abdominal surgery in colorectal cancer provides an 22. Golub R et al. J Am Coll Surg 1997;184:364–72.
overview of potential complications, methods to reduce 23. Alves A et al. World J Surg 2002;26:499–502.
24. Suturing or stapling in gastrointestinal surgery: A 54. Khan FA et al. Surg Obst Gyn 1987;165:275–6.
prospective randomized study. Br J Surg 1991;78: 55. Hill AD et al. JACS 1994;178:183–4.
337–41. 56. Xu J, Lin J. JACS 1994;179:351–2.
25. Docherty JG et al. West of Scotland and Highland 57. Civelek A et al. Surg Today 2002;32:944–5.
Anastomosis Study Group. Ann Surg 1995;221: 58. Konstadoulakis MM et al. Eur J Surg 1999;165:583–7.
176–84. 59. Lord MD, Gourevitch A. Br J Surg 1965;52:202–4.
26. Walstad PM. Am Surg 1974;40:586–90. 60. Cioffiro W et al. Arch Surg 1976;111:167–71.
27. Makela JT et al. Dis Colon Rectum 2003;46:653–60. 61. Olsen W, Beaudoin D. Surg, Gynecol, Obstet 1970;
28. Bucher P et al. Arch Surg 2004;139:1359–64. 131:57–62.
29. Bucher P et al. Br J Surg 2005;92:409–14. 62. Cassar K, Munro A. J R Coll Surg Edin 2002;6:
30. Hyman N et al. Ann Surg 2007;245:254–8. 731–41.
31. Nicksa GA et al. Dis Col Rectum 2007;50:197–203. 63. Fabri PJ et al. Arch Surg 1974;108:569–75.
32. Hyman N et al. Ann Surg 2007;245:254–8. 64. Rodkey GV, Welch CE. Ann Surg 1984;200:466–78.
33. Khan AA et al. Colorectal Dis. 2008;10(6):587–92. 65. McGory ML et al. Arch Surg 2007;142:668–74.
34. Schecter S et al. Dis Col Rectum 1994;37:984–8. 66. Malek MM et al. Sur Laparosc Endosc Percutan Tech
35. Lutchfeld MA et al. Dis Colon Rectum 1989; 32: 2007;17:385–7.
733–6. 67. Falsetto A et al. Ann Ital Chir 2005;76:175–81.
36. Tuson JR, Everett WG. Int J Colorectal Dis 1990;5: 68. Scheele J et al. Surgery 1984;95:6–13.
44–8. 69. Working Party of the British Committee for
37. Chung RS et al. Surgery 1988;104:824–9. Standards Clinical Hematology Task Force. Br Med J
38. Graffner H et al. Dis Colon Rectum 1983;26:87–90. 1996;312:430–3.
39. Virgilio C et al. Endoscopy 1995;27:219–22. 70. Higgins CC. JAMA 1967;199:82–8.
40. Waxman BP, Ramsay AH. Aust N Z Surg 1986;56: 71. Al-Awadi K et al. Int Urol Nephrol 2005;37:235–41.
797–801. 72. Hamawy K et al. Seminar Colon Rectal Surg 2000;11:
41. Matos DDM et al. Cochrane Database Syst Rev 163–79.
2001;(3):CD003144. 73. Pocard M et al. Surgery 2002;131:368–72.
42. Suchan KL et al. Surg Endosc 2003;17:1110–3. 74. Heald RJ, Ryall RD. Lancet 1986;1479–82.
43. Forshaw MJ et al. Tech Coloproctol 2006;10:21–7. 75. Nesbakken A et al. Br J Surg 2000;87:206–10.
44. Cirocco WC, Golub RW. American Surgeon 1995;61: 76. McDonald PJ, Heald RJ. Br J Surg 1983;70:727–9.
460–3. 77. McAnena OJ et al. Surg Gynecol Obstet 1990;170:
45. Malik AH et al. Colorectal Dis 2008;10(6):616–8. 517–21.
46. Choy PYG et al. Cochrane Database Syst Rev 2007; 78. Batignani G et al. Dis Colon Rectum 1991;34:329–35.
3:CD004320. 79. Lewis WG et al. Br J Surg 1992;79:1082–6.
47. Murray JJ, Schoetz DJJr. Stapling techniques in 80. Karanjia ND et al. Br J Surg 1992;79:114–6.
rectal surgery. In: Fazio VW (ed). Current Therapy 81. Lewis WG et al. Dis Colon Rectum 1995;38:259–63.
in Colon and Rectal Surgery. Philadelphia, PA: BC 82. Miller AS et al. Br J Surg 1995;82:1327–30.
Decker, 1990, pp. 384–90. 83. Brown SR, Seow-Choen F. Semin Surg Oncol 2000;
48. Chardavoyne R et al. Am Surgeon 1991;57:734–6. 19:376–85.
49. Chassin JL et al. Surg Clin N Am 1984;64:441–59. 84. Fazio VW et al. Ann Surg 2007;246:481–90.
50. Charlmers AG et al. Clin Radiol 1986;37:379–81. 85. Machado M et al. Ann Surg 2003;238:214–20.
51. Dubois JJ et al. Military Med 1989;154:505–7. 86. Consumer Assessment of Healthcare Providers and
52. Atabek U et al. Report of two cases. Dis Colon Systems. https://www.cahps.ahrq.gov/default.asp.
Rectum 1992;35:1180–2. 87. Morris AM. Surg Oncol Clin N Am 2006;15:
53. Wang QY et al. Arch Surg 1985;120:1013–20. 195–211, viii.
27
Abdominal restorative surgery for rectal cancer
CHARLES C. VINING AND NIJJIA N. MAHMOUD
sphincter preservation. While there is still a role for abdom-

CHALLENGING CASE inoperineal resection, the use of this technique is now
reserved overwhelmingly, especially among experienced
A 54-year-old male presents with an adenocarci- rectal surgeons in specialty centers, for those who have anal
noma of the middle rectum on colonoscopy for rectal sphincter involvement, preoperative incontinence, or severe
bleeding. Computed tomography (CT) of the chest, comorbidities precluding sphincter conservation. The abil-
abdomen, and pelvis documented the tumor without ity to reestablish gastrointestinal continuity has been rising
evidence of metastatic disease. Magnetic resonance over the last 80 years. Efforts today focus on reduction of
imaging (MRI) documented a T3N1 tumor. both short- and long-term morbidity associated with rectal
resection and anastomosis.
CASE MANAGEMENT
The patient was presented to a multidisciplinary

A CENTURY OF RECTAL CANCER
tumor board who recommended a course of preop- TECHNICAL IMPROVEMENTS
erative chemoradiotherapy followed by a low anterior
resection and diverting loop ileostomy. There has been continuous evolution and advancement over
the course of the twentieth century in the treatment of rec-
tal cancer, especially with regard to improvement in qual-
ity of life, sphincter conservation, and oncologic outcomes.
INTRODUCTION Advances in rectal cancer are based on improved compre-
hension of the disease biology and technical advancements.
The morbidity associated with proctectomy, including These improvements can be directly attributed to pioneers
incontinence, sexual dysfunction, functional disruption, in the field of colorectal surgery.
and the possibility of permanent ostomy has been the driver Until the late 1800s, rectal cancer surgery was performed
of development of new techniques in rectal cancer surgery solely for symptom palliation (Figure 27.1). In 1826, Lisfrank
and improvement of old ones. While the description of total described a perineal approach to remove the distal 3–4 cm
mesorectal excision (TME) in the 1980s in conjunction with of rectum along with the sphincter complex, essentially
improved neoadjuvant therapy contemporized the field and converting the anus to a stoma (2). Rectal resection with
is the cornerstone for modern-day rectal cancer surgery, the a combined rectal and abdominal approach was described
avoidance of permanent ostomy has motivated several gen- in 1884 by Czerny, and then again by Mayo in 1904. Both
erations of surgeons to explore better ways of reestablishing recurrence rates and morbidity were dismally high. From
gastrointestinal continuity without compromising onco- 1899 to 1906, Miles performed 57 perineal resections for
logic results, while minimizing impact on function. rectal cancer. Of those, 54 of 57 (95%) had early recurrences.
This chapter discusses the advances in abdominal restor- Miles did postmortem examinations of these patients and
ative surgery for rectal cancer with a focus on the contri- saw that the recurrences were composed of pelvic implants,
butions of the use of neoadjuvant chemoradiotherapy, mesocolon nodules, and enlarged, malignant lymph nodes
technical considerations, oncologic outcomes, and func- along the inferior mesenteric artery (IMA). In 1908, he
tional results. Careful preoperative planning with a mul- revised his approach based on these observations, detailing
tidisciplinary team can improve outcomes and chances of an abdominal approach to rectal surgery with low ligation
253
254 Abdominal restorative surgery for rectal cancer
100
80
5-year survival
60
%
40
20 APR
Recurrence rate
0
1888: 1908: 1913: 1926: 1982:

1879: 1975: 1991:
Hochenegg Miles Strauss Lockhart-Mummery 1950: Heald
Gussenbauer circular Jacobs
pull-through APR with en bloc electro- APR in 2 stages Deddish total
Hartmann- stapler laparoscopic
procedure lymphadenectomy coagulation APR with mesorectal
procedure (Russia) total mesorectal
lateral lymph excision
node dissection excision
2008
1739: 1874: 1884: 1896: 1910: 1923: 1948: 1972: 1980: 1985: 1995:
Faget Kocher Czerny Quénu Balfour Hartmann Dixon Parks Moriya Buess Enker
perineal posterior rectal APR APR anterior abdominal confirmation low nerve sparing transanale nerve sparing
resection resection with resection resection safety of spincter anterior rectal resection endoscopic total
sacrectomy with with anus preserving resection with lateral microsurgery; mesorectal
anastomosis in situ resection lymph node (neo-)adjuvant excision
dissection (chemo-)radiation
Figure 27.1 Milestones in the development of restorative proctocolectomy for rectal cancer. (From Lange M. et al. Eur J
Surg Oncol 2009;35[5]:456–63.)
of the IMA, sigmoid diversion, and rectal dissection done where access and exposure are difficult at best (1). Claude
transabdominally prior to completing the anal resection Dixon, at the Mayo Clinic, describes eloquently his pelvic
transperineally. This technique eventually became the gold anastomotic technique—he put curved clamps side by side
standard operation for rectal cancer, because local recur- on both the stump of the rectum as well as the colon. Since
rence rates were dramatically reduced (29.5%) as the lymph the posterior aspect of the rectum is thin, he placed a single
node basin was excised along with the entire rectum and row of posterior interrupted sutures there. Anteriorly, an
sphincter complex. Despite this major breakthrough, early inverting row of sutures was placed as well. He commented,
morbidity and mortality were high with an estimated 1-year “Any so-called aseptic anastomosis in this region is a gym-
survival of 58% (1). nastic feat attended by difficult clamp maneuvers deep in the
It would take another decade of what amounted to direct pelvis, danger of tearing the rectum, the hazard of catching
observation and trial and error for the treatment of rectal the opposite mucosa to create a persistent diaphragm or of
cancer to include attempts at sphincter conservation and perforating the rectum in forcing the anastomosis open, and
consideration of functional results. In the 1920s, Dr. Henri finally the possibility of a poor anastomosis” (3).
Hartmann introduced “anterior resection” of the rectum, pre- Until the late 1930s, rectal resection was performed with
serving the distal third of the rectum and sphincter complex an intraoperative change in the patients’ position from
with the creation of a sigmoid colostomy for proximal rectal supine to the prone or semiprone position. The introduction
tumors with the goal of decreasing the complications and of leg rests allowed for simultaneous performance of the
mortality associated with perineal wounds while improving abdominal and perineal dissections. Introduction of sim-
lymphadenectomy by including a high IMA ligation. This ple stirrups also facilitated the advent of restorative tech-
technique mandated ostomy creation, because pull-through niques that circumvented the need for a permanent ostomy.
methods for sphincter preservation, still in their infancy, Dixon presented results for restorative anterior resection for
were plagued with high leak rates and perioperative mortal- tumors of the proximal rectum and distal sigmoid in 1948.
ity rates as high as 20%. Reliability of the anastomosis was He demonstrated a 64% 5-year survival, prompting wide-
still an issue, as were the technical aspects of establishing spread adoption of this technique for treating cancer of the
gastrointestinal continuity in the confines of the deep pelvis proximal rectum while preserving continuity by creating a
Margins / Distal margin 255
hand-sewn distal anastomosis. Of 426 cases in his series of and pelvis, and either MRI of the pelvis or endorectal ultra-
anterior resection of lesions between 6 and 20 cm from the sound (ERUS). Regional imaging with MRI (most com-
dentate line, there were 25 deaths (5.9%). Eighteen occurred monly performed) or ERUS assesses depth of invasion and
after the first stage of the operation and seven following nodal involvement and is essential in determining need for
closure of the stoma. Seven of the 18 deaths followed peri- preoperative treatment either with or without radiation
tonitis, two after pelvic abscesses, and then the rest from depending on protocol used and size, symptomatology, and
perioperative issues unrelated to the technical aspects of the characteristics of the tumor as well as clinical stage.
case. Of the seven deaths after closure of the stoma, four The ultimate goal of preoperative assessment is to most
resulted from peritonitis. This represents a 3% operative accurately determine stage, characterize the lesion, assess
mortality rate—remarkable for its time. For lesions proxi- resectability, and determine the need for neoadjuvant
mal to 10 cm from the dentate line, the resectability rate treatment. Large administrative cancer databases such as
was 10% by anterior resection. For lesions 10 cm from the Surveillance, Epidemiology, and End Results and National
dentate, it dropped to 80%, and for those at 8 cm it dropped Cancer Database analysis demonstrate that sphincter pres-
to 44%. Dixon felt that 8 cm was the lowest level of general ervation varies based on surgical volume, education, region,
applicability of the operation or his ability to accurately pre- and sociodemographic factors. It has been demonstrated
dict his ability to reestablish continuity. Below 8 cm, there that age, insurance status, National Comprehensive Cancer
must be favorable anatomy—“a thin person with a broad, Network institution, tumor fixation, level of tumor, and
flat pelvis.” Above 8–10 cm, involvement of adjacent organs history of preoperative radiotherapy are all predictors of
did not impact resectability. Almost all of these cases were sphincter preservation. Standardization and dissemination
diverted—a transverse loop colostomy was the most com- of knowledge regarding approach to evaluation, staging,
mon strategy. Analysis of other series published at the time and surgery may help (4,5).
demonstrated that with anterior resection without colos-
tomy, 10%–20% of cases were complicated by the need for
an emergent colostomy in the early postoperative period
because of anastomotic leak. Stricture of the anastomosis MARGINS
occurred in five cases, and long-term outcomes as well as
the nuances of postoperative function were not reported (3). The discussion of margin adequacy has been an area of evo-
lution in the past 20 years. A satisfactory oncologic resec-
tion is predicated on an R0 resection with clear distal and
circumferential margins. That said, the minimum accept-
PREOPERATIVE EVALUATION—IS able distal margin has steadily shrunk with better data
SPHINCTER CONSERVATION POSSIBLE? documenting acceptable clearances less than 1 cm and the
importance of the circumferential radial margin (CRM) as
In Dixons’ 1948 analysis of rectal cancer surgery, he empha- a prognostic indicator has become manifest. Surgical tech-
sized the need for surgeons to preoperatively physically niques allowing lower connections and the advent of neoad-
evaluate the distance of the tumor from the anal verge and juvant chemotherapy facilitating shrinkage of bulky tumors
dentate line by digital rectal exam and proctoscopy. Those made ultralow anastomoses possible and drove research to
guidelines have not changed 70 years later. Physical exami- determine the absolute distal and radial spread of tumor.
nation should include digital rectal examination and direct
visualization via proctoscopy or flexible sigmoidoscopy DISTAL MARGIN
to assess the size, degree of luminal obstruction, location
with regard to the anal margin or anorectal ring/sphincter, In the past, a minimum of 2–5 cm distal margin on the rec-
distal-most aspect of the tumor, and orientation and fixa- tum was recommended for nonirradiated patients because
tion to surrounding structures such as the vagina, prostate, of the belief that intramural tumor extension was present.
and sphincter complex. Digital rectal exam can also be In 1951, Golligher, Dukes, and Bussey demonstrated local
beneficial in assessing sphincter tone and function. These tumor spread in the vast majority of rectal cancer did not
factors can help determine the need for neoadjuvant treat- exceed 2 cm from the tumor margin by showing that tumor
ment and address the possibility of maintaining continu- cells >2 cm from margin were seen in only 2% of specimens.
ity. Examination is an important aspect of clinical staging. It was therefore hypothesized that a 5 cm margin was more
Approximately 6% of patients with colorectal malignancies than sufficient from an oncologic perspective (6). Studies by
have a synchronous colorectal neoplasm. Complete colo- Williams and colleagues in the early 1980s, however, revealed
noscopy to assess the colon preoperatively is necessary. If that in a cohort of more than 50 nonirradiated abdominal
colonoscopy cannot be completed, a virtual colonoscopy perineal resection (APR) specimens, 76% of patients had no
or barium enema can substitute. A complete blood count, intramural spread of tumor beyond the cancer itself, and
basic metabolic panel, liver function tests, and baseline car- in 14%, the spread was less than 1 cm distal to the tumor
cinoembryonic antigen should be obtained. Clinical stag- and most commonly in the submucosa compared with the
ing should be completed with a CT of the chest, abdomen, muscularis propria. Only five patients had tumor extension
beyond 1 cm, and those patients had large poorly differenti- (where TME was carefully controlled for) reported a recur-
ated neoplasms (7,8). Other studies and reviews done con- rence rate of 16% versus 6% if the margins were less than or
temporaneously by Pollett and Nichols as well as the Large greater than 2 mm, respectively. Additionally, the widespread
Bowel Cancer Group Project demonstrated similar results in adoption of TME correlated with a dramatic drop in the rate
patients who had not been pretreated with chemoradiother- of APR (15). Recognition that the proper TME plane extends
apy. Still other data demonstrate that there is no difference to the intersphincteric groove coupled with the rise of spe-
in either 5-year survival rates or local recurrence (LR) in cific technology to allow gastrointestinal continuity set the
patients who had <2, 2–5, or >5 cm distal margins. Others stage for transanal-transabdominal as well as transanal-TME
found that while there is a difference in survival and LR in (Ta-TME) techniques for resection of low rectal cancers.
those with positive versus negative distal margins, there is
no difference in LR between those with negative margins
of <1 cm or >1 cm. Most recently, margins of <1 cm have
TECHNICAL ADVANCES ALLOW LOWER
demonstrated no difference in oncologic outcomes (9–11).
Data published within the past 5 years demonstrate that
ANASTOMOSES
margins of <5 mm are acceptable in those with good risk
pretreated tumors. Whole-mount pathologic examination The development of stapling devices in the late 1960s
of specimens in patients who underwent chemoradiother- changed nearly every surgical field, including colon and
apy (CRT) followed by resection found intramural exten- rectal surgery. Specifically, this technology improved tech-
sion beyond gross tumor in only 2 of 109 patients (1.8%) niques for tissue division, construction of anastomosis,
with a maximum distance of 0.95 mm (12). Moreover, the and improved time efficiency. Ravitch, in 1972, published
rate of local recurrence and 3-year recurrence-free sur- Technics of Staple Suturing in the Gastrointestinal Tract and
vival are similar in patients with distal margins greater or triggered the near universal adoption of stapling devices.
less than 1 cm following CRT. In contrast, patients with Although the development of stapling devices had been
advanced disease (T4 or N1+ tumors) are more likely to underway for several years by this time, it was this seminal
have intramural extension beyond 1 cm (9–11). All of these publication that helped disseminate the technology (16).
factors have facilitated creative and innovative approaches In 1975, Fain demonstrated that use of the circular sta-
to sphincter conservation in the past decade by alleviating pler for low colorectal anastomoses had a similar leak rate
oncologic concerns and highlighting the necessity of good compared with hand-sewn anastomoses. These staplers
patient selection. lay down two staggered rows of staples to create a circular
anastomosis and cut simultaneously (17). Colorectal sur-
CIRCUMFERENTIAL RADIAL MARGIN gery typically requires anything from a 25 mm to a 33 mm
end-to-end anastomosis stapler. Circular staplers come
CRM and the fascia propria emerged as very important in a variety of sizes and must be fit to both the size of the
anatomic landmarks when the importance of TME was descending colon and according to what the anal canal/rec-
recognized. Negative CRM is a major prognostic factor for tal stump can accommodate. The double stapling technique
rectal cancer as local recurrence and long-term survival are for low colorectal anastomosis was introduced in 1980,
highly correlated with the presence or absence of tumor at which accelerated the procedure, minimized intraoperative
the mesorectal margin. The CRM is defined as the shortest contamination, and avoided the disadvantage of joining
distance between the outer edge of the tumor and the meso- bowel segments with significant size disparities. The fol-
rectal fascia (fascia propria). Quirke reviewed 52 specimens lowing decade saw the introduction of techniques aimed at
and found that 25% had positive margins and almost 80% of creating ever lower connections with the advent of the colo-
those recurred locally. A CRM of <1 mm is considered a pos- anal anastomosis, the intersphincteric transanal/transab-
itive margin. Studies have shown that margins <1 mm have dominal rectal dissection with hand-sewn anastomosis, and
a local recurrence rate of 25%–75%. By contrast, those with introduction of the colonic pouch in an attempt to improve
CRM >1 mm have a LR rate of less than 10%. Additionally, functional results that were (and are) frequently compro-
CRMs 1 mm or less were associated with a higher rate of dis- mised. These techniques and instruments allow modern
tant metastasis and shorter survival. Finally, radial margin surgeons to perform anastomoses far below the 8 cm cut-
invasion risk is increased with lack of response to neoad- off described by Dixon with a high degree of reliability, low
juvant chemoradiotherapy, palpable fixation on DRE post- complication rate, and maximized oncologic outcomes.
treatment, and proximity to the dentate line (11,13,14).
Recognition of the role of total mesorectal excision was
facilitated and complemented by pathologic data document-
ing the role of CRM in the rate of local recurrence. Careful
NEOADJUVANT CHEMORADIOTHERAPY
dissection in the plane between the fascia propria of the rec- AND SPHINCTER CONSERVATION
tum and the presacral fascia investing the sacrum dramati-
cally decreased the rate of positive CRM, and subsequently While oncologic outcomes remain the main goal of rec-
improved local recurrence rates. The Dutch TME study tal cancer treatment, avoiding permanent stoma and
Neoadjuvant chemoradiotherapy and sphincter conservation 257
maximizing functional outcomes is a recent additional as well. A comparison of patients receiving combined pre-
quality of life–related outcome most patients demand. There operative radiotherapy (50.4 Gy) with or without con-
is no question that advancements in technology and data- current chemotherapy (5-FU) demonstrated that 90% of
driven reassessment of distal margin requirements have patients who were initially unresectable converted to resect-
made sphincter conservation possible. Neoadjuvant therapy ability versus only 64% of those in the radiation-only group.
in rectal cancer treatment has improved clinical outcomes, Additionally, combined chemoradiotherapy demonstrated
facilitated resection to help with safe sphincter conserva- complete pathological response in 20% of patients versus
tion, and become the standard of care in advanced disease. 6% of those receiving radiation alone. This in addition to
It has the potential to downstage tumors, shrink locally other trials confirm the enhancement of downstaging when
advanced neoplasms, and improve R0 resection rates. Its chemotherapy is administered concurrently with radiation
role in sphincter conservation for low-lying tumors is con- for locally advanced rectal cancer (20–22).
troversial and complex, dependent on the position and loca- The German Rectal Cancer Group compared preop-
tion of the tumor, the body habitus of the patient, the skill erative and postoperative long-course CRT combined with
and training of the surgeon, and the patient’s preoperative TME for patients with stages II and III disease who were
anorectal function. candidates for low anterior resection and found that pre-
Over the last several decades, randomized trials have operative CRT demonstrated improved 5-year local recur-
investigated the role of neoadjuvant radiotherapy (RT) in rence rate (6% versus 13%) and remained significant after 10
the treatment of rectal cancer. Neoadjuvant RT has the years without an effect on overall survival. They also noted
potential for less acute and late toxicity with more patients less acute and chronic toxicities associated with preopera-
receiving full-dose radiation therapy compared with the tive versus postoperative CRT. Therefore, neoadjuvant CRT
adjuvant (postoperative) setting. Preoperative radiation can followed by TME is the treatment of choice for patients with
be administered via two techniques: the intensive short- locally advanced rectal cancer. Local recurrences with pre-
course radiation with large fractions (5 × 5 Gy) for 1 week operative CRT and TME are now <10%, and in patients
followed by resection within 1 week—or 5–6 weeks of con- with complete pathological response, 10-year disease-free
ventional fractions (1.8–2 Gy), with concurrent chemo- survival is >80%. Moreover, patients with complete or
therapy followed by resection in approximately 6–10 weeks. nearly complete tumor regression with neoadjuvant CRT
The Swedish Rectal Cancer Trial conducted between 1987 demonstrate better long-term survival (23).
and 1990 randomized 1,168 patients to preoperative radio- Gerard and colleagues recently did a meta-analysis of 17
therapy (25 Gy in five fractions, short course) followed by randomized clinical trials of rectal cancer treatment involv-
surgery in 1 week versus surgery alone. The group that was ing chemoradiotherapy in an attempt to analyze sphincter
randomized to RT demonstrated decreased local recurrence conservation (24). Only three of the studies included sphinc-
at 5 years (11% versus 27%; p < 0.001), and increased overall ter conservation as an endpoint. Although there was sig-
survival (58% versus 48%; p = 0.004). This benefit was seen nificant heterogeneity among the studies, some generalized
in all ages and supported the idea that oncological survival conclusions could be drawn regarding the role of chemora-
is predicated on improved local control. Additionally, the diotherapy. Rates of sphincter conservation in general have
Uppsala trial between 1980 and 1985 randomized patients gone up considerably from the late 1970s when it was less
to preoperative short-course (5 × 5.1 Gy) versus postopera- than 30% to the present where it exceeds 75% (Figure 27.2).
tive conventional (total 60 Gy) radiotherapy. Preoperative Predictably, short-course radiation and chemotherapy has
radiotherapy was associated with significantly decreased no impact on sphincter conservation, since no time inter-
local failure (13% versus 22%; p = 0.02) without a signifi- val is allowed for tumor downsizing prior to surgery. In
cant improvement in 5-year survival. Duplicating the find- the five large trials of long-course chemoradiotherapy with
ing that radiotherapy confers a survival benefit in other more than 5 weeks between the end of therapy and surgery,
trials has been elusive (18). no increase in sphincter conservation was seen despite an
Criticisms of short-course radiation include an insig- increased rate of pathologic complete response up to 16%–
nificant time for downstaging, particularly in bulky tumors 19%. Alternatively, it is evident from these trials that the
with threatened margins, increased acute and late toxicity, 5-year local recurrence rate has decreased from about 25%
and inability to add concurrent chemotherapy. A French to close to 5%. While the role of preoperative chemoradia-
trial randomized patients with low-lying rectal tumors to tion in local control is well supported by all trials thus far,
either short (within 2 weeks) or long (6 weeks) intervals even controlling for good surgery, the role of downstaging
between radiation and surgery. The long interval group in the facilitation of restorative surgery is unproven (25,26)
demonstrated a significantly better clinical tumor response (see Chapter 34).
(71.7% versus 51.3%; p = 0.007), improved pathological The increase in sphincter preservation appears to be most
downstaging (26% versus 10.3%; p = 0.005) and a trend dependent on advances in surgical techniques. The expertise
toward increased sphincter preservation (76% versus 68%; of the operating surgeon is important—advanced training
p = 0.27) (19). in rectal cancer surgery is directly correlated with improve-
Neoadjuvant chemoradiotherapy with the goal of down- ment in sphincter conservation and the ability to perform
sizing clinically unresectable rectal cancers was investigated restorative surgery. Evolving emphasis on multidisciplinary
Differences in sphinter saving surgery rate (odds ratio) and 95% confidence intervals
Study name Statistics for each study Odds ratio and 95% Cl
Odds Lower Upper
ratio limit limit
EORTC 22921 [8] 0.60 0.34 1.07

Uppsala [23] 0.91 0.60 1.38
Manchester [17] 0.25 0.13 0.48
Swedish [24-25] 0.13 0.89 1.44
Lyon R90.01 [18] 1.47 0.78 2.76
Dutch [6] 0.92 0.75 1.12
CAO/ARO/AIO [7] 0.90 0.66 1.23
Polish [19-20] 0.88 0.55 1.40
LyonR96.02 [22] 3.90 1.55 9.86
EORTC 22921 [8] 1.11 0.87 1.43
FFCD 9203 [9] 0.99 0.74 1.33
Scandinavian [23] 2.18 1.22 3.91
CR07 [13] 0.92 0.73 1.16
NSABP R03 [27] 1.45 0.86 2.44
STAR [14] 1.12 0.79 1.59
ACCORD 12 [15] 1.12 0.77 1.64
Australian NZ [16] 1.41 0.81 2.45
0.15 2.00 4.00 6.00 10.00

Favours ctrl group Favours exp group
Test of heterogeneity p < 0.001
Figure 27.2 Sphincter preservation and neoadjuvant treatment. (From Gerard JP et al. Crit Rev Oncol Hematol
2012[1]:21–8.)
assessment of patients preoperatively by all practitioners cancers (stage 1) can be definitively treated by surgery alone.
involved in the care of rectal cancer patients impacts patient More advanced neoplasms should be managed by a multi-
selection for restorative surgery and local control. disciplinary team for best outcomes. When deciding on
There remains controversy regarding early low-risk stage treatment modality, staging and patient factors as well as
II patients (small tumors with good histopathologic charac- baseline anorectal function must be considered.
teristics) regarding whether neoadjuvant CRT is beneficial.
Unfortunately, limitations in clinical staging make it diffi- GENERAL CONCEPTS AND PREPARATION
cult to determine those with low-risk T3 disease. It should
be noted that approximately 22% of patients who undergo Rectal neoplasms exist in the narrow confines of the deep
preoperative imaging with MRI or ERUS diagnosed with pelvis. Obese body habitus, abundant visceral fat, and nar-
T3N0 disease are found to have positive lymph nodes on row android pelvis all challenge sphincter conservation.
final pathology. Moreover, approximately 18% of patients Besides the preoperative treatment and evaluation already
diagnosed with T3N0 disease on preoperative imaging are discussed, a patient undergoing rectal cancer surgery should
overstaged and found to be T2N0 on final review. At some be preoperatively marked and counseled by a stoma thera-
institutions (e.g., Europe), patients with tumors with clear pist, and the concept of postoperative functional changes
circumferential margins on MRI will be offered surgery with- should be thoroughly discussed. A bowel continence his-
out neoadjuvant CRT. However in the United States, neoadju- tory and assessment of mobility, ability to manage and
vant CRT is still favored for T3N0 disease due to concern for handle a temporary ileostomy, renal function, and general
understaging, inferior local control, increased toxicity, and performance status is essential.
worse functional outcomes with postoperative CRT (27). In the operating room, lithotomy positioning for most
approaches is necessary. Special attention to padding the
extremities to prevent nerve injury is crucial. If approach-
ing the dissection in an open or laparoscopically assisted
SURGICAL TECHNIQUES FOR fashion, lighted retractors such as the St. Mark’s retractor,
SPHINCTER CONSERVATION or a headlight, long instruments, and access to help with
exposure via good retraction are essential. If approaching
The surgical management of rectal cancer should be indi- the tumor purely laparoscopically or robotically, many of
vidually tailored to the patient and disease. Early rectal these same concepts apply. A specially trained team that
Surgical techniques for sphincter conservation / Total mesorectal excision 259
understands how to set up and troubleshoot the robot is

a necessity.
Radical resection involves removal of the tumor and
rectum en bloc with the lymphatics and vascular supply
located in the mesorectum and is known as a TME. In radi-
cal rectal surgery with intent to cure, the goals include pres-
ervation of the autonomic nerves, negative circumferential
and distal margins, preservation of sphincter function, and
restoration of continuity when possible. Radical resections
include sphincter-preserving surgery such as low anterior
resection (LAR) and transanal transabdominal ultra-low
hand-sewn approaches also described as intersphincteric
proctectomy (ISP) and the sphincter-sacrificing surgery
APR. As long as technically and oncologically appropriate,
LAR or one of its modifications is the preferred approach Figure 27.3 Total mesorectal excision specimen
to radical resection. Technical limitations include visceral with a smooth, intact fascia propria. (Courtesy of the
obesity, a narrow and long pelvis, and low tumor location author.)
<1 cm from the sphincter complex. Anterior location can
also challenge sphincter conservation. If the tumor invades
either the sphincter complex or levators, sphincter preser- improvement in sexual and urinary function. In the past,
vation is contraindicated. Moreover, patients with baseline blunt dissection was associated with sexual dysfunction
poor or marginal anorectal control are inappropriate for in up to 75% of men and nearly half of women. Urinary
LAR due to worsened postoperative anorectal function. retention or incontinence was seen in 80% of patients (29).
Currently, rates of sexual dysfunction are less than 10%–
TOTAL MESORECTAL EXCISION 20%, and long-term bladder dysfunction is 5% or less (30).
Heald deserves credit for popularizing the components of
The pillar of modern rectal cancer surgery is based on the TME; however, several contemporary surgeons at institu-
work of Heald who popularized TME as a key concept in tions such as the Cleveland Clinic were performing rectal
1982 (28). Previous to TME, many surgeons performed resections using the techniques of TME.
low anterior resections with a blunt dissection technique TME begins with mobilization of the sigmoid colon,
of the mid- and distal rectum as described by Dixon. This careful identification of ureters and hypogastric nerves,
technique was burdened with poor 5-year disease-free sur- and high ligation of the inferior mesenteric pedicle. Medial
vival for all stages treated with curative intent (<50%) and mobilization starting with ligation of the pedicle is per-
local recurrence up to 20%. The poor outcomes were due fectly acceptable. For the vast majority of patients requiring
to breaches in the mesorectal fascia (fascia propria) during restorative surgery, mobilization of the splenic flexure is a
the blunt dissection with positive circumferential margins must—while this can be done after the TME and vascular
seen in up to 85% of local recurrences. Heald described “one ligation, it typically precedes the pelvic operation when
distinct lymphovascular entity” in which the rectum and approached in a minimally invasive fashion. After vascular
mesorectum, derived from the same embryologic origin, ligation and division of the bowel at the distal descending
are removed en bloc from the surrounding parietal fascia colon/sigmoid border, the TME proceeds by sharp dissec-
covering the sacrum and pelvic sidewall via the avascular tion under direct visualization, either open, with the use of
loose areolar holy plane via sharp dissection under direct retractors, or in a minimally invasive fashion. Either way,
vision (Figure 27.3). Heald realized that by staying away staying in the avascular plane defined by the fascia propria
from the lateral sidewalls, bleeding, nerve injury, and posi- of the rectum and the presacral fascia is essential. As the
tive margins were minimized, thereby dropping rates of operation extends downward, the fascia propria investing
local recurrence, improving bladder and sexual functional the mesorectum becomes somewhat thinner, particularly
outcomes, and dramatically increasing rates of sphincter anteriorly and laterally. At the levators, the mesorectum
preservation. Although there has never been a prospec- ends and the rectum itself extends through the levator hia-
tive randomized trial investigating the TME technique, it tus as a bare muscular tube. The muscularis propria of the
is consistently associated with lower local recurrence and rectum is a direct extension of the internal sphincter. It is at
improved 5-year survival. Heald reported outcomes for the point just above the levator plate and directly adjacent to
519 patients in 1998 demonstrating an astounding 3% local and distal to the mesorectum that the bowel is divided for
recurrence rate in patients treated with curative intent. a complete TME (Figure 27.4). Typically, this division takes
In this study, only 49 patients had undergone chemora- place with a 30 mm TA or PI stapling device and leaves a
diotherapy. Also, only 37 patients needed an APR. Strict small amount of distal rectum right at the levators about
attention to nerve preservation and failure to disrupt the 1.5–3 cm above the dentate line depending on the body hab-
periprostatic plexus and nervi erigentes results in vast itus of the patient and the length of the anal canal.
Mesorectum
Lymph
nodes
Fascia
propria Tumor
Tumor
(b)
Levator
ani muscle
(a) (c)
Figure 27.4 Appropriate dissection planes for a low anterior resection. (a) Coronal view with distal margin delineated.
(b) Sagittal view of dissection plane in male. (c) Sagittal view of dissection plane in female. (From Bordeianou L. et al. J
Gastrointest Surg 2014;18[7]:1358–72.)
Anterior resection in upper rectal cancers (Figure 27.5). Surgery commences in the prone position,
although it can also be done in lithotomy—transanal TME
In low and middle rectal cancers, a complete TME is per- (TaTME) cases start in lithotomy as a rule. The advantage
formed, whereas in tumors of the upper rectum, a more of this approach is that a negative margin can be established
localized TME can be performed with a distal division immediately under direct visualization. An incision is made
approximately 5 cm from the tumor margin. Upper rectal in the intersphincteric plane or just proximal to it with elec-
cancers are generally those defined as the upper one-third trocautery. It is relatively straightforward to get into the meso-
of the rectum, above the peritoneal reflection or above a cer- rectal plane from this approach, because the intersphincteric
tain distance such as 12–13 cm, which is somewhat arbitrary. plane is contiguous with the plane between the mesorectum
Generally, those tumors in the upper rectum that are smaller and the presacral fascia, and the internal sphincter is in con-
(<3 cm) can undergo partial TME, dividing both the rectum tinuity with the rectal wall. The surgeon can dissect all the
and the mesorectum at minimum 5 cm distal to the bottom way up into the pelvis, beyond the levator plate with this
edge of the tumor. This leaves about 5–7 cm of rectum to per- dissection. A Lone Star self-retaining retractor can be quite
form the colorectal anastomosis. It is critical that the meso- helpful as can anorectal retractors such as Hill-Ferguson or
rectum is completely and cleanly resected at the same level as Pratt anal bivalve instruments (Figure 27.6). After dissecting
the rectum to ensure a good resection of the mesorectal lymph beyond the levators circumferentially, the rectum is closed
node basin and reduce the incidence of local nodal recurrence. with a running suture to prevent contamination, and a moist
gauze is inserted into the anal canal. The patient is placed into
Transanal transabdominal or the lithotomy position, and a standard laparoscopic or open
intersphincteric proctectomy LAR with TME is performed. The entire specimen is brought
out the abdominal incision after the abdominal TME plane
Transanal transabdominal (TATA) or ISP is simply a way of meets the perineal TME plane. The colon is then mobilized
reestablishing gastrointestinal continuity in patients whose such that the end of the colon is able to be pulled all the way
tumors are so low that standard staplers cannot be placed out of the anus from the perineal side without tension. The
below them with any confidence that negative margins retractor is placed back in, and orientation and pull-through
would be obtained. In general, these tumors exist at the top of the colon take place. After a centimeter or two of colon is
of the anal canal at the very edge of the sphincter complex gently pulled out of the anus, the staple line is removed, and
Surgical techniques for sphincter conservation / Total mesorectal excision 261
(a) (b)
1 cm
T
1 cm
Figure 27.5 Appropriate planes for intersphincteric resection—TME plane created from above intersects intersphincteric
plane. (a) Partial ISP. (b) Complete ISP. (From Bordeianou L. et al. J Gastrointest Surg 2014;18[7]:1358–72.)
Local recurrence after ISP is not statistically different

from that of LAR or APR. Additionally, ISP and LAR have
shown equivalent 5-year local recurrence-free survival sug-
gesting that ISP is a suitable technique from an oncologic
perspective. Those patients with bulky tumors and those
with microscopic positive resection margins are more likely
to have local recurrence after ISP (31–33).
Postoperative defecation functional outcomes such as
frequency, urgency, ability to distinguish between flatus and
stool, and perianal skin irritation have been demonstrated
to be similar between ISP and LAR. A study by Yamada
showed that patient age was the only factor associated with
risk of fecal incontinence following ISP. Ito reported neo-
adjuvant CRT as a risk factor in ISP for poor postoperative
anal function. Koyama reported no difference in average
defecation frequency between LAR and ISP, no difference
in urgency or differentiation between gas and stool, and
no difference in the use of antidiarrheal agents or perianal
erosion. Overall, bowel functional results between ISP and
LAR were not statistically different (31–33).
Figure 27.6 Intraoperative positioning and retraction
for establishing a hand-sewn anastomosis follow- Transanal total mesorectal excision (TaTME)
ing an intersphincteric proctectomy. (Courtesy of the
author.) Challenging circumstances associated with restorative rec-
tal cancer surgery were the main drivers of development of
the TaTME approach, which is quite similar to the TATA
the colon is gently pushed back into the anal canal until the (ISP) procedure conceptually. Male gender, high body mass
cut edge of the well-vascularized, viable distal rectum is at the index, visceral obesity, bulky distal tumors, and narrow
level of the distal margin. An interrupted hand-sewn anasto- pelvic anatomy all contribute to the difficulty of minimally
mosis is then performed with 3 or 2 Vicryl suture. This tech- invasive approaches to rectal cancer when approached
nique is challenging in individuals with narrow pelvises and transabdominally. The TaTME technique was identified as a
a large amount of visceral fat that make it difficult to mobilize way to overcome some of these anatomic and tumor-related
the colon and pull it through the anal canal. challenges by starting transanally and completing the pelvic
aspect of the dissection from this approach, thereby elimi- margin and high ligation of the IMA in both laparoscopic and
nating the need for deep pelvic retraction for visualization open cases has been debated, with some advocating for a selec-
necessary for laparoscopic, robotic, and open approaches. tive approach (36,37). Because the sigmoid colon is frequently
In 2014 at a consensus conference of experts in the tech- partially in the field of neoadjuvant radiation and can be thick-
nique, it was agreed that the TaTME approach works well ened, affected by diverticulitis or diverticulosis, and margin-
for male gender, narrow or deep pelvic anatomy, low and ally perfused after high IMA ligation, most surgeons favor
mid-rectal cancers 12 cm or less from the anal verge that routine resection of the sigmoid colon for proctectomy involv-
are bulky, those with prostatic hypertrophy, high BME, and ing an anastomosis. Resection of the sigmoid requires SFM
visceral obesity, and those whose tumors are 1.5 cm from in order to create a tension-free, well-vascularized anastomo-
the dentate line or top of the anorectal ring who may require sis. SFM requires additional time (15–45 minutes) and may
ISP/TATA. Contraindications to TaTME are obstructing require an additional laparoscopic port in minimally invasive
rectal cancers, emergency surgery, and most T4 tumors (34). cases or a longer incision in open cases. Those who argue for
TaTME is dependent on a combination of laparoscopic selective mobilization point to data showing no increased
and transanal techniques to get into the appropriate mesorec- anastomotic leak rate and comparable oncologic outcomes.
tal plane from a transanal approach. The patient is typically Data on selective use of SFM are mostly case series (36,37). The
put into lithotomy position and a purse-string suture using a technique does not lend itself to randomized testing. Although
1.0 or 0.0 monofilament is placed >1 cm from the gross distal debate remains, a selective approach to SFM requires experi-
margin of the tumor. Carbon dioxide (CO2) is used to insuf- enced intraoperative judgement, adequate assessment of the
flate via a platform similar to transanal endoscopic microsur- vascularity, and a tension-free anastomosis. Over 75% of low
gery instrumentation. Dissection with standard laparoscopic anterior resections require SFM to achieve these goals.
instruments takes place first posteriorly, where dissection Mobilizing the colon to reach the pelvis requires both
starts below the level of the mesorectum and follows the SFM in most cases as well as medial mobilization with liga-
mesorectal plane posteriorly until visualization of the fibers tion of the left colic artery as it comes off the IMA and liga-
of Waldeyer fascia is observed. Anterior mobilization follows tion of the IMV at the lower border of the pancreas. The
taking great care to stay in the plane between the rectal wall blood supply to the left colon is therefore based off the mid-
and the prostate or vagina. Injury to anterior structures is a dle colic vessels, and medial mobilization that extends to the
hazard, and careful attention to the plane of dissection, espe- base of these vessels is often necessary to achieve this. This
cially with anterior tumors, is paramount. Most advocate for extensive mobilization results in “straightening” of the left
lateral dissection last as it can be most difficult to identify colon while preserving the marginal artery and blood sup-
the lateral planes in very distal dissections. Dissection too ply. Occasionally, in cases of reoperative surgery, or when the
laterally, a common problem with perineal approaches, can left colon is exceptionally short, the middle colic vessels can
result in injury to the nervi erigentes and result in urinary be ligated so that the entire blood supply to the colon is based
and sexual dysfunction. Hand-sewn or stapled anastomoses off the ileocolic pedicle. Placing a clamp on the middle colic
are possible depending on the level of initial dissection. The pedicle prior to ligation with monitoring of the marginal
insufflation of CO2 in the developing TME plane makes visu- arterial blood flow for a few minutes is helpful to establish
alization clear and actually aids in dissection. The entire pel- that flow is sufficient. Creating a window in the mesentery
vic portion of the TME takes place in this manner, obviating of the terminal ileum (ileal window) can help with extend-
the need for a transabdominal pelvic approach. ing the reach of the colon through rather than over the small
The main disadvantage of TaTME currently is that there is bowel mesentery, thereby saving several centimeters of valu-
little data on the oncologic safety of the approach, although in able length. As stated before, ensuring good blood supply
many ways, it mimics conceptually the ISP/TATA technique and pulsatile flow in the marginal artery is essential. Direct
that has been in use since 1994. There have been many small observation and palpation are key, and Doppler confirma-
institutional reviews of data (more than 36 to date) which, tion of flow is necessary selectively in bulky mesenteries
cumulatively have resulted in over 500 evaluable patients. where visualization and palpation may be difficult.
These data have demonstrated rates of morbidity and mor-
tality comparable to laparoscopic TME. The positive CRM
rate was 5%, and the distal margin rate was 0.3% with intact
MINIMALLY INVASIVE TECHNIQUES
mesorectal fascia seen on 94% of specimens. The technique
shows promise and has good utility in selected patients (35).
FOR RESTORATIVE RECTAL CANCER
SURGERY
SPLENIC FLEXURE AND LEFT COLON
MOBILIZATION Laparoscopic and robotic radical rectal resections have had a
large impact on patient quality of life outcomes in the last 15
Splenic flexure mobilization (SFM) in rectal cancer is used to years. Decreased pain scores and length of ileus have translated
obtain additional length, ensure a tension-free anastomosis, into decreased length of hospital stay, improved mobility, and
and preserve blood supply. The routine use of SFM for anterior faster return to normal activities of daily living. Challenges
or low anterior resection after obtaining an adequate surgical to transabdominal minimally invasive techniques revolve
Temporary diversion 263
around the difficulty presented by operating in the deep pel- proctocolectomy after LAR. Reconstruction techniques fol-
vis. Many studies have reported on differences between tra- lowing LAR are most commonly done as a straight colo-
ditional open LAR versus laparoscopic or robotic techniques. anal anastomosis (SCA) either end to end or end to side.
In spite of their obvious advantages, minimally invasive Frequently, however, if anatomy is favorable, creation of
techniques are still on trial with regard to their oncologic a colonic reservoir with a colonic J-pouch (CJP) or trans-
outcomes. The recent randomized clinical trials ACOSOG verse coloplasty pouch (TCP) is possible. A concern with
Z6041 and ALaCarT failed to demonstrate noninferiority of colonic reservoir techniques is they can be technically chal-
laparoscopic TME over the standard open approach (38,39). lenging and may be associated with a higher leak rate. A
The rate of CRM positivity was 7 compared to 12.1% (lap) meta-analysis of reconstruction techniques following LAR
and 3 compared to 7.7% (lap) for each trial, respectively. The reported a decreased frequency in antidiarrheal medications
results were quite consistent. Predecessor trials have also and improved function in the first postoperative year with
found that CRM positivity can be challenged by minimally CJP, whereas TCP and SCA had similar functional outcomes
invasive approaches with rates in some studies as high as (44). All the techniques had similar leak rates. Additionally,
16%, but this has not been consistently worse than open prospective randomized trials have demonstrated that CJP
approaches and translation into longer-term cancer-specific reconstruction is associated with better short-term func-
outcomes such as disease-specific survival and local recur- tional outcomes (urgency and frequency), but long-term
rence is either not available or not significantly different. continence and overall quality of life are similar to a straight
The prospective randomized COLOR II trial reported on anastomosis. A CJP is typically about 5 cm in length and cre-
the differences between open versus laparoscopic LAR in ates a small, bulbous reservoir just above the anastomosis.
patients with rectal cancer within 15 cm of the anal verge This reconstruction may not be possible in those patients
from 30 international centers (40). Local recurrence, dis- with narrow pelvic anatomy or bulky, foreshortened mes-
ease-free survival, and overall survival were nearly identical enteries. Although the TCP technique helps overcome these
between groups at 3 years. Additionally, the COREAN trial limitations, functional outcomes have not been as promising
compared open versus laparoscopic LAR following neoad- or consistent.
juvant CRT and found that 3-year disease-free survival and
overall survival were similar (41). The CLASICC trial from the
UK Medical Research Council compared open versus lapa-
roscopic resection for rectal cancer. There was no difference TEMPORARY DIVERSION
between groups with regard to local control, 3-year disease-
free survival, or overall survival, although a subgroup analysis Temporary diversion with a proximal diverting loop
of the laparoscopic LAR group noted a trend toward wors- ileostomy (DLI) is employed in rectal cancer surgery
ened sexual function in men following laparoscopic LAR. to decrease the rate of symptomatic anastomotic leak.
Additionally, there was a trend toward higher positive CRMs Anastomotic leak following restorative LAR in a large
in the minimally invasive compared to the open group (42). population-based study suggested male gender, low anas-
Rates of conversion from a minimally invasive to open tomosis (<6 cm from the anal verge), preoperative radio-
technique by experienced surgeons are in the range of 15% therapy, and adverse intraoperative events as factors that
and seem to be lower in those employing robotic instrumen- increase leak rate.
tation. Better retraction and resulting improved visualiza- Several randomized studies have suggested that a proxi-
tion seem to favor a robotic approach for minimally invasive mal DLI may prevent anastomotic leakage and rate of
pelvic surgery, and ongoing laparoscopic versus robotic rec- reoperation. In a randomized multicenter Swedish trial, a
tal cancer surgery trials will help define the respective roles defunctioning stoma decreased the leak rate from 28% to
of these technologies (43). The development of TaTME was 10.3% in mid- and low rectal cancer patients. Moreover, this
a direct response to the challenges of transabdominal mini- trial reported a low likelihood of stoma reversal in patients
mally invasive approaches and an attempt to improve visu- who underwent subsequent proximal diversion following
alization and decrease conversion rates. It is quite clear that anastomotic leak (45). Forty percent of these patients were
the considerable advantages of minimally invasive surgery reconstructed with a colonic J-pouch. This was reinforced
make it compelling, useful, and necessary. Directions for the in another study where 256 patients were randomized to
future clearly point to development of technical advances to diversion or none with a straight anastomotic technique
facilitate easier, more reliable deep pelvic dissection. and reported a significantly lower symptomatic anasto-
motic leak rate among patients who were defunctionalized
(46). In general, diversion with a low pelvic anastomosis is
recommended.
RESERVOIR AND POUCH Prior to reversal of the ileostomy, a water-soluble contrast
RECONSTRUCTION enema and visual examination of the anastomosis should be
performed to rule out subclinical anastomotic leak or stric-
The reservoir techniques were developed in an attempt ture. Assuming no leak and normal anatomy, ileostomy
to improve functional outcomes associated with reversal can be scheduled for 6–12 weeks following LAR.
result in increased bladder tone, decreased capacity, and

FUNCTIONAL OUTCOMES retrograde ejaculation. Parasympathetic nerve fibers from
S2 to S4 ventral roots join the hypogastric nerves on the pel-
Technical and medical advances in rectal cancer treatment vic sidewall and form the inferior hypogastric plexus. Injury
and surgery have improved oncological outcomes to the to the parasympathetic plexus can result from traction in
extent that relatively small gains in outcomes from techni- the deep pelvis or encroachment on the deep lateral pelvic
cal advancements are expected. Therefore, posttreatment sidewalls. Parasympathetic fiber injury can result in urinary
quality of life has become an increasingly important focus retention from increased bladder neck tone. Additionally,
of emerging research in this field. Investigation of bowel parasympathetic fiber injury can result in erectile dysfunc-
and sexual functional outcomes is specific to restorative tion in men and impaired vaginal lubrication in women.
rectal cancer surgery.
SEXUAL FUNCTION
BOWEL FUNCTION
Following rectal cancer surgery and chemoradiation for
While sphincter preservation for low rectal cancers has advanced tumors, sexual function can be significantly
increased dramatically, and approaches to resection and altered. Current literature suggests that APR, low rectal
sphincter conservation are multiplying, the same basic cancers, and the use of radiation are risk factors associated
problems plague all techniques. All result in removal of with sexual dysfunction (48,49). Typically, men and elderly
the rectal reservoir; take place in proximity to sympathetic patients are more likely to report postoperative sexual prob-
and parasympathetic nerves affecting sexual, bowel, and lems. There are several possible causes of sexual impairment
urinary function; and result in a low connection that chal- following rectal cancer surgery, but intraoperative injury to
lenges continence and sensation. While patients experience the autonomic plexus is the most important factor, as previ-
a wide range of symptoms and are differentially affected, ously described.
nearly all experience some disruption in normal func- Laparoscopy may be associated with increased sexual
tion. Low anterior resection syndrome (LARS) describes dysfunction in men. Jayne reported similar bladder func-
incomplete evacuation, clustering, frequency, and stool tion following laparoscopic versus open rectal cancer oper-
and gas incontinence after sphincter preservation. Shorter ations, with overall sexual function and erectile function
colonic length, small or nonexistent reservoir with dimin- worse in men following laparoscopic surgery compared to
ished capacity, and nerve impairment affecting sensation an open approach (50). This is contrary to findings by oth-
account for these changes. Most studies report rates of ers who reported a lower incidence of erectile and bladder
LARS in excess of 50%, but rates widely range from 10% dysfunction in patients who undergo laparoscopic TME.
to 74% (47). Moreover, nearly half (43%) of patients are dis- Theoretically, increased exposure, illumination, and mag-
satisfied with their functional outcome in the short term nification can help facilitate the identification and preser-
(48). Improvement in function does take place over time vation of the pelvic autonomic nerves (51). Better data are
with better function evolving over the course of 24 months. needed specifically examining functional outcomes and
Creation of a colonic J-pouch is associated with early reme- reported rates of intraoperative autonomic nerve identifi-
diation of LARS, but long-term results are similar to a cation. Intraoperative nerve identification and preserva-
straight or end-to-end anastomosis (49). tion have been proposed as an intraoperative quality metric
for rectal cancer surgery by the American Society of Colon
AUTONOMIC NERVE PRESERVATION and Rectal Surgeons Quality Committee (52,53). The rec-
ognition of nerve and sphincter preservation as markers of
Preservation of autonomic nerve function is important to quality surgery is a step forward for patients suffering from
patient overall quality of life and satisfaction. In the early rectal cancer and its effects.
1990s, Warren Enker, a colorectal surgeon at Memorial
Sloan Kettering Cancer Center, was a vocal proponent of REFERENCES
the preservation of the autonomic nerve sympathetic fibers,
which arise from T12 to L3 ventral nerve roots. He recog- 1. Lange M et al. Eur J Surg Oncol 2009;35(5):456–63.
nized the functional consequences of nerve injury and pro- 2. Corman M. Dis Colon Rectum 2000;43(Supp 6):
moted active intraoperative recognition and preservation S1–29.
of these structures. He showed that nerve preservation and 3. Dixon CF. Ann Surg 1948;128(3):425–42.
oncologic efficacy were not mutually exclusive. These sym- 4. Abdelsattar ZM et al. Ann Surg Oncol 2014;21(13):
pathetic fibers join to form the superior hypogastric plexus, 4075–80.
and distal to the aortic bifurcation give off the hypogastric 5. Temple LK et al. Ann Surg 2009;250(2):260–7.
nerves. Injury to these nerves can occur during ligation of 6. Goligher JC et al. Br J Surg. 1951;39(155):199–211.
the IMA or during the TME if the plane is dissected too 7. Ludwig KA. Sphincter sparing resection for rectal
close to the sacrum. Injury to the sympathetic fibers can cancer. Clin Colon Rectal Surg 2007;20(3):203–212.
References 265
8. Bordeianou L et al. J Gastrointest Surg 2014;18: 33. Schiessel R et al., Feil W, Urban M. Dis Colon Rectum
1358–72. 2005;48(10):1856–65.
9. Pollett WG, Nicholls RJ. Ann Surg 1983;198(2): 34. Motson RW et al. Colorectal Dis 2016;18:13–8.
159–63. 35. Penna M et al. Clin Colon Rectal Surg 2017;30(5):
10. Bujko K et al. Ann Surg Oncol 2012;19(3):801–8. 339–45.
11. Quirke P et al. Lancet 1986;2(8514):996–9. 36. Park J et al. Surg Laparosc Endosc Percutan Tech
12. Guillem J. et al. Ann Surg 2007;245(1):88–93. 2009;19(1):62–8.
13. Nagtegaal I et al. Am J Surg Pathol 2002;26(3):350–7. 37. Marsden M et al. Colorect Dis 2012;14(10):1255–61.
14. Wibe A et al. British J Surg 2002;89(3):327–33. 38. Fleshman J et al. The ACOSOG Z6051 Randomized
15. Kapiteijn E. et al. N Engl J Med 2001;345(9):638–46. Clinical Trial 2015;314(13):1346–55.
16. Ravitch M, Steichen F. Ann Surg 1972;175(6):815–37. 39. Stevenson AR et al. ALaCaRT investigators. JAMA
17. Fain S. Arch Surg 1975;110(9):1079. 2015;314(13):1356–63.
18. Cedermark B et al. N Engl J Med 1997;336(14):980–7. 40. Van der Pas MH et al. Colorectal cancer Laparoscopic
19. Francois Y et al. J Clin. Oncol 1999;7(8):2396. or Open Resection II (COLOR II) Study Group. Lancet
20. Minsky BD et al. Int J Radiat Oncol Biol Phys 1997; Oncol 2013;14(03):210–8.
37(2):289–95. 41. Jeong SY et al. Lancet Oncol 2014;15(7):767–74.
21. Frykholm G et al. Dis Colon Rectum 1993;36(6): 42. Jayne DG et al. J Clin Oncol 2007;25(21):3061–8.
564–72. 43. Jayne D et al. JAMA 2017;318(16):1569–80.
22. Trakarnsanga A et al. J Natl Cancer Inst 2014;106(10): 44. Brown CJ et al. Cochrane Database Syst Rev 2008;(2).
[Abstract]. 45. Matthiessen P et al. Ann Surg 2007;246(2):207–14.
23. Sauer R et al. N Engl J Med 2004;351(17):1731–40. 46. Chude GG et al. Hepatogastroenterology 2008;
24. Gerard JP et al. Crit Rev Oncol Hematol 2012;81(1): 55(8687):1562–7.
21–8. 47. Bryant CL et al. Lancet Oncol 2012;13(9):e403–8.
25. Bujko K et al. Radiother Oncol 2006;80(1):4–12. 48. Lundby L et al. Dis Colon Rectum 2005;48(7):1343–9.
26. Baker B et al. Surg Oncol 2012;21(3):103–9. 49. Fish D, Temple LK. Surg Oncol Clin N Am 2014;23(1):
27. Smallwood N, Fleshman J. Clin Colon Rectal Surg 127–49.
2015;28:5–11. 50. Jayne DG et al. Br J Surg 2005;92(9):1124–32.
28. Heald R et al. Br J Surg 1982;69(10):613–16. 51. Wallner C et al. J Clin Oncol. 2008;26(27):4466–72.
29. Heald RJ, Ryall RD. Lancet 1986;1(8496):1479–82. 52. Manwaring ML et al. Dis Colon Rectum. 2012;55(3):
30. Ho VP et al. Dis Colon Rectum 2011;54(1):113–25. 294–301.
31. Schiessel R et al. Br J Surg 1994;81(9):1376–8. 53. Glasgow SC et al. Dis Colon Rectum 2016;59(7):
32. Kuo LJ et al. J Surg Res 2013;183(2):524–30. 601–6.
28
Transanal approaches to rectal cancer
CHARLES B. WHITLOW AND LARA MCKEAN BASTÉ
or dehiscence, genitourinary dysfunction, low anterior

CHALLENGING CASE resection syndrome, and need for temporary or permanent
ostomy.
A 64 year-old moderately obese man is found to have The significant morbidity resulting from proctectomy
a 1 cm moderately differentiated adenocarcinoma of makes less invasive and less morbid treatment attractive.
the rectum approximately 9 cm above the anal verge. In order to be an acceptable option as a treatment for cure,
Intra-anal ultrasound and MRI stages the lesion as the oncologic outcomes of any treatment and salvage after
T1N0. failure should be similar to the outcomes of initial proctec-
tomy. In addition to treatment for cure, these less morbid
procedures have found a place in the treatment of patients
who are poor candidates for proctectomy because of comor-
CASE MANAGEMENT bidities or metastatic disease. Transanal fulguration, endo-
cavitary radiation, transsacral and transsphincteric surgical
After presentation at a multidisciplinary tumor confer- approaches have been described, but local excision (LE) by
ence, surgical options are discussed with the patient, one of several techniques has been the most widely accepted.
eg. Low anterior resection or Transanal minimally Standard transanal excision (TAE), transanal endoscopic
invasive surgery (TAMIS). The patient selected a microsurgery (TEM), transanal minimally invasive surgery
TAMIS. (TAMIS), and more recently endoscopic submucosal dis-
section (ESD) all have been used to treat rectal cancer with
acceptable morbidity and mortality.
Several organizations have published statements on the
INTRODUCTION appropriateness of LE. National Comprehensive Cancer
Network (NCCN) guidelines for rectal cancer recommend
Rectal cancer is diagnosed in roughly 39,000 patients every observation for patients who, after appropriate pretreatment
year in the United States. The management of rectal cancer staging, undergo LE and on pathology are found to have
is multidisciplinary, but surgical resection is still the main- T1NX adenocarcinoma without high-risk features (posi-
stay of treatment. The incidence of early colorectal cancers tive margins, lymphovascular invasion, poor differentia-
has increased in the past years as a result of the expansion tion, and sm3 invasion) (2). The Practice Parameters for the
of advanced endoscopic techniques and the widespread American Society of Colon and Rectal Surgeons (ASCRS)
use of colorectal screening. Standard of treatment of rectal state that “local excision is an appropriate treatment modal-
cancer involves major radical abdominal surgery including ity for carefully selected T1 rectal cancers without high-risk
total mesorectal excision (TME) to treat locally advance features” (well to moderately differentiated, no lympho-
or node-positive rectal cancer. This technique achieves vascular invasion, no perineural invasions, less than 3 cm
removal of the primary tumor and the mesorectum with diameter, and less than one-third rectal luminal circumfer-
the associated regional lymph nodes. It has demonstrated ence) (3). The European Association for Endoscopic Surgery
excellent long-term results that approach 4.5% local recur- (EAES) published a consensus statement on this topic. This
rence rates at 5 years and 90% 5-year disease-free survival statement describes LE as a “valid treatment option for early
rates (1). However, proctectomy is associated with signifi- rectal cancer” which should be offered for T1N0 cancers
cant morbidity, mortality, and impact on quality of life from with favorable features. Depth of invasion beyond sm1 was
complications such as anastomotic leak, wound infection considered unfavorable. LE alone for any lesions outside of
266
Local recurrence and survival 267
the described criteria “should be considered only as a com- unacceptably high risk for morbidity and mortality from
promise procedure” (4). proctectomy may elect local excision for palliation or pre-
Though the guidelines discussed previously describe nar- vention of symptoms with the understanding that local
row selection criteria and the literature as a whole advises recurrence may require additional treatment. Patients who
caution in local excision except for superficial lesions with- are being considered for local excision with curative intent
out risk factors, the use of local excision in the United States should be counseled thoroughly on the risk, benefits, and
has continued to increase. Data from the National Cancer alternatives to this treatment. Because the choice between
Data Base (NCDB) from 1998 to 2010 reported 46.5% of T1 local excision and radial resection frequently represents a
and 16.8% of T2 rectal cancers were treated by LE (5). The choice between quality of oncologic treatment versus qual-
increase in the use of LE from 1998 to 2010 was dramatic, ity of life, the value the patient places on these factors is an
with an increase of 40%–62% for T1 tumors and 12%–24% important part of the decision. Additionally, adjuvant treat-
for T2 tumors. ment, surveillance strategies, and possible salvage treatment
require a commitment from the patient. This is a lengthy
and complicated discussion that challenges the surgeon
to present the salient data in an understandable form. The
INITIAL EVALUATION ensuing paragraphs deal with many of these issues.
Patients with rectal cancer come to surgical evaluation via

several avenues, the most common being a mass biopsied or
excised as part of a complete colonoscopy or a mass detected LOCAL RECURRENCE AND SURVIVAL
on rectal exam or imaging study. A histologic diagnosis of a
cancer is mandatory prior to initiation of neoadjuvant treat- Recurrence after local excision has been attributed to several
ment or radical excision. A complete colonoscopy should be factors: imprecise preoperative staging, incomplete exci-
performed if not previously done. The surgeon who will be sion, tumor factors, and failure to treat regional/systemic
making the recommendations for surgical treatment should disease (especially mesosrectal lymph node metastases) at
perform a physical exam to include proctosigmoidoscopy time of presentation. The presence of lymph node metas-
and digital rectal exam to assess tumor location, size, tases (LNM) increases with tumor depth of invasion. T1
mobility, relation to levators, and anal sphincter function. rectal cancers as a whole have a LMN incidence of approxi-
Absence of induration and ulceration on exam suggest there mately 10%–13%, and for T2 rectal cancers the incidence is
is no invasive component, and lack of fixation indicates a 17%–23% (6). The depth of invasion in the submucosa can
superficial tumor. be further used to predict the risk of LNM. While subjective
Local staging of the tumor can be accomplished with terms such as “scant” or “slight” invasion have been used,
endorectal ultrasound (EUS) or magnetic resonance imag- more objective descriptions of the amount of submucosal
ing (MRI). While both of these modalities assess T stage, invasion are preferred. However, there has not been uni-
EUS appears to be better for superficial tumors, and MRI is form reporting of this data that has been published by the
superior for deeper ones. A recognized limitation of EUS is sm1, sm2, sm3 categorization or by absolute depth of inva-
for very distal tumors where the ability to get an acceptable sion. The absolute depth of invasion is typically reported in
image is compromised by anatomic features. Both MRI and microns, but the depth that correlates to lower risk of LNM
EUS have a lack of specificity and sensitivity for detecting has been variously reported as <500, <1,000, <1,500, or
metastases to lymph nodes. Local availability and expertise <2,000 (7). When depth alone is considered, superficial T1
dictate which imaging technique is preferred or they may tumors are associated with a risk of LNM of 3% or lower.
be used as complementary exams. Additional or alternative Other histopathologic tumor features associated with
assessment of tumor depth can be acquired by examining LNM include poor differentiation, lymphovascular or peri-
tumor morphology and using enhanced endoscopic tech- neural invasion, and tumor budding. Location in the distal
niques (such as magnifying chromoendoscopy) to examine third of the rectum is an independent risk factor for LNM.
mucosal pit patterns. Because tumors may be heterogenous, biopsy alone may not
Computed tomography (CT) is less accurate than EUS or demonstrate the presence of adverse features that are found
MRI for local staging. However, it is the most common imag- on examination of a complete excised tumor. Glasgow et al.
ing modality for staging of distant metastases. CT of the chest, performed a meta-analysis of histopathological features of
abdomen, and pelvis is routinely performed as part of com- primary colorectal cancers that predict lymph node metas-
plete staging. The exams of the chest and abdomen give infor- tases. In looking at rectal cancers specifically, they reported
mation not included in the pelvic MRI. The pelvic CT exam that tumor stage, tumor differentiation, vascular invasion,
is complementary to other pelvic imaging. Positron emission lymphatic invasion, and differentiation at the invasive
tomography is not indicated in routine initial staging. front were statistically significant predictors of lymph node
In addition to tumor staging, individual patient fac- metastases (8).
tors, such as comorbidities and sphincter function, impact Most of the data on LE alone for treatment of rectal can-
the decision-making for local excision. Patients with an cer are from standard transanal excision, and no prospective
268 Transanal approaches to rectal cancer
randomized data exist on this topic. The data for LE of T1 consistently in the same position regardless of the positional
cancers from retrospective single institution studies show orientation of the lesion, while others prefer to position the
5-year local recurrence (LR) rates ranging from 13% to 18% patient so that the lesion is in the dependent position. For
and overall survival (OS) rates at 5 years range from 72% to extremely distal tumors, standard lighted half-cylinder
89%. Both ranges are lower than comparisons to standard anoscopes (e.g., Hill-Ferguson) provide adequate exposure.
resection—0%–3% and 80%–96%, respectively. Similarly, Removal of larger and more proximal tumors is facilitated
several large registries have reported 5-year LR rates of by placement of a Lone Star retractor with the hooks placed
7%–12% and 5-year OS of 70%–87%. LE alone for T2 can- in the anal canal at the level of the dentate line. Additional
cers has a LR rate of 20%–30% with OS of 63%–75% (9–11). exposure is obtained with plastic cylindrical Ferguson ano-
Radical surgery performed immediately after finding scopes, which are available in a variety of lengths and diam-
adverse pathologic features by LE results in outcomes simi- eters. The mucosa is scored with electrocautery to outline
lar to initial radical resection. However, the results of radi- the extent of the excision with a 1 cm margin of grossly nor-
cal surgery as a salvage procedure have not been as good as mal mucosa. In general, cancers should be removed with a
radical surgery as initial treatment. Median time to recur- full-thickness excision of the rectal wall using electrocau-
rence varies from 11 to 35 months. While most recurrences tery, ultrasonic shears, or bipolar vessel sealers. Hemostasis
are identified within the first 36 months, recurrence after is obtained and then full-thickness wounds are typically
that is reported and highlights the need for prolonged sur- closed in a transverse manner using an absorbable suture.
veillance. Microscopically clear margins (R0) are obtained The authors’ preference is to use a running barbed suture
in 82%–94% of salvage resections, although Madbouly with a loop at the tail end and a clip at the other, obviating
reported that only 40% of patients underwent surgery for the need for knot tying. While closure can be performed via
curative intent (12). Weiser et al. reported 55% of their radi- a plastic Ferguson anoscopy, switching to a Hill-Ferguson
cal salvage procedures required extended dissections com- anoscope is sometimes helpful in the distal rectum. Stenosis
pared to standard resections, but pelvic exenteration has of the rectal lumen should be assessed by anoscopy, proc-
only been necessary in 7%–8% of cases (13). 5-year OS rates toscopy, or digital exam. If the distal edge of the tumor is
of 53%–62% are reported. Stipa et al. reported 144 patients more than 1 cm from the dentate line, no local anesthetic is
with rectal cancer initially treated by TEM of which 86 needed. For lower lesions, bupivacaine (liposomal or plain)
were T1, 38 were T2, and 20 were T3 (14). Twenty-four of 26 is infiltrated to perform a complete anal block. No pack-
recurrences were endoluminal. ing is needed. Patients are observed overnight for evidence
of bleeding or fever. Antibiotics are not given beyond the
initial perioperative dose except in the case of fever on the
night of surgery. Patients are allowed liquids immediately
TECHNICAL after surgery and advanced to a regular diet as tolerated.
Complications after TAE are uncommon but include
PREOPERATIVE PREPARATION immediate and delayed bleeding, pain, abscess, distur-
bances of continence, and on rare occasion perineal sepsis.
The preoperative evaluation of patients undergoing LE for Abscesses are treated by opening the rectal wall suture line,
rectal cancer involves the standard assessment of comorbid- and perineal sepsis is treated with antibiotics or incision,
ities as any other major surgical procedure. Additionally, a drainage, and debridement if indicated.
precise description of the location of the tumor should be
written in the preoperative note, as well as an assessment TRANSANAL ENDOSCOPIC
of anal sphincter function, invasion into adjacent struc- MICROSURGERY AND TRANSANAL
tures, and relationship to the anterior peritoneal reflection.
MINIMALLY INVASIVE SURGERY
It is the authors’ preference to have patients complete a full
mechanical bowel prep and oral antibiotic prep on the day Both TEM and TAMIS use endorectal carbon dioxide (CO2)
prior to surgery. Standard preoperative IV antibiotics for insufflation and an occluding transanally placed platform
bowel surgery and appropriate venous thromboembolism to allow access to the rectum. TEM was first described by
prophylaxis are given immediately prior to beginning the Buess in 1984 (15). A 40 mm metal operating rectoscope is
procedure. secured to the table via a mechanical adjustable mount. A
detachable faceplate acts as the working attachment and an
TRANSANAL EXCISION insufflation port, a channel for a standard laparoscope (or
stereoscopic telescope), and two 5 mm channels for instru-
While this procedure is commonly performed under gen- ments. Proprietary instrumentation for grasping, cautery,
eral endotracheal anesthesia, TAE can be accomplished and suturing is available for the commercially available
under local anesthetic with monitored anesthesia care. TEM devices.
Positioning is based on individual surgeon preference. TAMIS differs in that a single access port is used, which
Most lesions can be addressed from prone jackknife, lat- does not require attachment to the operative table. Some
eral, or lithotomy. Some surgeons prefer to have all patients authors recommend that all such procedures be performed
Adjuvant treatment 269
in the lithotomy position and in slight Trendelenburg posi- cannot be accomplished transanally, laparoscopy or lapa-
tion (16,17). The commonly available ports for TAMIS rotomy should be performed and adequate bowel closure
are the GelPOINT Path (Applied Medical, Rancho Santa completed.
Margarita, California) and the SILS Port (Covidien,
Mansfield, Massachusetts). Standard laparoscopes and
laparoscopic instrumentation including alternative energy
sources are used through the working channels. COMPLICATIONS
Two technical notes are of use in TEM and TAMIS
with regard to optimal visualization. First, neuromuscu- Postoperative bleeding (up to 9%), urinary retention, or
lar blockade of the patient is critical to maintaining ade- urinary tract infection (up to 11%) are the most common
quate pneumorectum. Second, an insufflation system that postoperative complications (18). Fecal incontinence is
produces constant intrarectal pressure avoids the bellows infrequent and typically temporary. Abscess may occur
effect encountered with standard insufflators used in lapa- with defects that are closed. If abscess develops in this situ-
roscopy, for example, SurgiQuest AirSeal (ConMed, Utica, ation, the suture line should be opened to obtain adequate
New York). drainage. Perineal sepsis has been reported in up to 3% of
Regardless of the platform used, the technique of exci- patients and is associated with lesions within 2 cm of the
sion is the same. Typically the mucosa is scored with cautery dentate line. The senior author has seen two such cases. Both
to outline the extent of the excision. An epinephrine-con- of these occurred in elderly infirm patients with tumors that
taining solution can be injected into the submucosal plane if extended distally to the proximal anal canal. Because of this
a mucosal resection is being performed but is not necessary location, the distal dissection was performed by TAE and
for full-thickness excision. Traction is placed in a manner the proximal resection with TAMIS. In one case the wound
that facilitates perpendicular dissection for the initial por- was left open, and in the other it was closed. Because the
tion of dissection as a full-thickness incision is made using dissection in the distal portion of the rectum and anal canal
electrocautery. This incision can be extended partially or is below the mesorectal fat, full-thickness closure is recom-
completely around the circumference of the lesion, but care mended in this location.
should be taken to avoid “coning” and thus decreasing the
margin on the deeper portion of the lesion. Mesorectal fat is
encountered in full-thickness excisions and is the visual cue
to begin excision in a plane parallel to the rectal wall. This COMPARISON OF TECHNIQUES
can be done with monopolar electrocautery A bipolar seal-
ing device or ultrasonic shears are sometimes used, as this A meta-analysis demonstrated similar complication rates
part of the dissection can produce substantial hemorrhage. between TAE and TEM (16). Additionally, tumors removed
After complete excision, the lesion should be pinned to avoid by TEM were more likely to have negative microscopic
contraction during fixation. Hemostasis should be obtained. margins and less likely to undergo fragmentation during
As with TAE, it is not necessary to close extraperitoneal removal. These factors may result in lower local recurrence
defects; however, many surgeons choose to do so. A variety with TEM. The largest TAMIS experience to date reported
of laparoscopic closure devices have been used including on 200 patients for benign and malignant lesions. The posi-
Endostitch (Covidien, Norwalk, Connecticut) and RD-180/ tive margin rate was 7%, and fragmentation occurred in 5%.
TK Knot Device (LSI Solutions, Victor, New York). Suturing Overall morbidity was 11% with urinary retention, hemor-
with laparoscopic needle drivers is technically more chal- rhage, and subcutaneous/scrotal emphysema being the most
lenging with TEM as opposed to TAMIS, especially with common complications (19).
knot tying. The use of barbed suture with a looped tail is
particularly useful in this setting. Alternatively, a suture
fixation device such as silver beads or Lapra-Ty (Ethicon,
Cincinnati, Ohio) can be used. If standard laparoscopic
ADJUVANT TREATMENT
suturing is done, a knot pusher may be useful. More distal
defects can frequently be approached using the transanal Data on adjuvant radiation therapy +/− chemotherapy
“open” approach. Again, the lumen should be inspected (typically radiosensitizing 5-FU) is retrospective commonly
after closure to check for stenosis. The senior author has from small series. A meta-analysis that included 14 studies
experienced clinically significant stenosis with large defects showed weighted local recurrence rates of 10% for T1 tumors
that were closed or left open. The postoperative care after and 15% of T2 tumors treated with chemoradiation after LE
TEM or TAMIS is similar to TAE. (20). The recurrence rates for tumors treated by TEM after
While these techniques allow for removal of proximal LE were 6% for T1 cancers and 10% for T2 lesions. Pooled
rectal or rectosigmoid lesions, there is at least a theoretical survival data were not reported, but the 5-year disease-free
concern about tumor spread with a full-thickness intra- survival was 75%–100% for patients receiving adjuvant
peritoneal injury and ongoing exposure to CO2 insufflation. radiation (six studies). For patients treated with TEM, two
Peritoneal entry mandates complete defect closure. If this studies reported 5-year and 10-year disease-free survival
270 Transanal approaches to rectal cancer
of 94% and 86%, respectively. Rackley et al. reported on 93 95% for patients with ypT0 tumors. ypT1 and higher tumors
patients who received adjuvant radiation following LE. They had a local recurrence rate of 21.9% and a 68% median dis-
found T1 cancers had a 5-year overall survival of 84% and ease-free survival. The pooled complication incidence was
5-year local control of 92.5% (21). 23.2%. Smith et al. reported data from 25 studies with 1,001
patients (24). Their main conclusions were the significant
variability they identified with regard to many important
technical issues including pathologic assessment, appropri-
ENDOSCOPIC TREATMENT ate staging, and pretreatment marking. They noted substan-
tial morbidity from TAE in the radiated rectum including
Endoscopic treatment of rectal adenocarcinomas is being abscess, fistulas, and severe pain. Additionally, subsequent
reported with greater frequency. Endoscopic mucosal resec- radical salvage surgery for local recurrence after local exci-
tion (saline lift polypectomy) results in a higher proportion sion is technically challenging because the planes of dissec-
of piecemeal resections than any of the surgical excision tion have been violated. Additionally, the practice of watch
techniques or endoscopic submucosal dissection. The tech- and wait for complete clinical responders would seem to
nique of submucosal dissection involves using a needle question the value of the practice of LE after radiation. The
knife to incise circumferentially around the lesion. Several conclusions from this report were that if this practice is
techniques are used to achieve mucosal lift, including the going to continue, additional study is warranted along with
use of a hybrid needle knife with cautering and cutting standardization.
capability combined with an injection port (Erbejet 2, Erbe
Elekromedizin GmbH, Tübingen, Germany). An increasing REFERENCES
number of tools are available to accomplish the actual dis-
section in the submucosal plain. Typically, the procedure is 1. Mellgren A et al. Dis Colon Rectum 2000;43:
performed using a plastic cap on the end of the scope that 1064–71.
facilitates visualization and dissection. One meta-analysis 2. National Comprehensive Cancer Network. Rectal
showed a lower recurrence rate for benign rectal lesions Cancer (Version 3.2017). https://www.nccn.org/
removed by TEM versus those removed by ESD (22). This professional_gls/pdf/rectal.pdf. Accessed March 25,
comparison showed no differences in en bloc resections, 2017.
R0 resections, and complications. Several authors have 3. Monson JRT et al. Dis Colon Rectum 2013;56:535–50.
commented on the technical challenges of ESD in the dis- 4. Morino M et al. Surg Endosc 2015;29:755–73.
tal rectum and proximal anal canal. The ESD rectal can- 5. Stitzenberg KB et al. J Clin Onc 2013;31:4276–82.
cer literature is immature, and there are many case reports 6. Ricciardi R et al. Clin Gastroenterol Hepatol 2006;4:
describing excision, recurrences, and complications that are 1522–7.
already well described in the surgical literature. It is inevita- 7. Maeda K et al. Surg Today 2014;44:2000–14.
ble that ESD for rectal cancer will be used more frequently. 8. Glasgow SC et al. J Gastrointest Surg 2012;16:
The advent of new instrumentation may make it technically 1019–28.
easier than the current version. There is no reason to expect 9. You NY. Semin Radiat Oncol 2011;21:178–84.
the complications related to this approach should be any 10. Althumairi AA, Gearhart SL. J Gastrointest Oncol
more common or severe than surgical techniques. Those 2015;6:296–306.
who use this technique should have a thorough understand- 11. Heafner TA, Glasgow SC. J Gastrointest Oncol 2014;
ing of the nuances of the management of superficial rectal 5:345–52.
cancers as has been outlined above. 12. Madbouly KM et al. Dis Colon Rectum 2005;48:
711–21.
13. Weiser MR et al. Dis Colon Rectum 2005;48:1169–75.
14. Stipa F, Giaccaglia V. Dis Colon Rectum 2012;55:
LOCAL EXCISION AFTER 262–9.
CHEMORADIATION 15. Buess G et al. Chirurg 1984;55:677–80.
16. deBech-Adams T, Nassif G. Clin Colon Rectal Surg
A complete discussion of this topic is beyond the scope 2015;28:176–80.
of this chapter. However, there has been interest in full- 17. Gill S et al. J Gastrointest Surg 2015;19:1528–36.
thickness rectal wall excision following a complete clini- 18. Lee L et al. Ann Surg 2017;267(5):1.
cal response to neoadjuvant chemoradiation. The rationale 19. Clancy C et al. Dis Colon Rectum 2015;58:254–61.
for this is that residual tumor in the rectum is a source of 20. Borstlap WAA et al. Br J Surg. 2016;103(9):1105–16.
potential local recurrence. Two systematic reviews have 21. Rackley TP et al. Dis Colon Rectum 2016;59:173–8.
been published. Hallam et al. identified 20 studies, which 22. Arezzo A et al. Surg Endosc 2014;28:427–38.
encompassed 1,068 patients (23). They reported a local 23. Hallam S et al. Dis Colon Rectum 2016;59:984–97.
recurrence rate of 4% and a median disease-free survival of 24. Smith FM et al. Dis Colon Rectum 2017;60:228–39.
29
Abdominoperineal resection
W. BRIAN PERRY AND HUISAR DAO CAMPI
CHALLENGING CASE HISTORY
A 64-year-old woman is 7 days after an abdomino- Early in the twentieth century, most patients with rec-
perineal resection for a T2N1 rectal adenocarcinoma. tal cancer underwent palliative perineal procedures to
She had received preoperative care. Her perineal address advanced disease. These included the transcoccy-
wound has developed increased tenderness, is swol- geal Kraske approach and the transsphincteric approach
len, and is draining pus. developed by Bevan in America, later attributed to A. York
Mason. Patients were often left with profound sphincter
dysfunction or fistulae following a protracted recovery. A
two-staged operation, consisting of an initial laparotomy
CASE MANAGEMENT and colostomy followed by perineal excision, was used until
the 1930s with reasonable results.
The patient’s wound is opened and the patient is In 1908 Miles first described the operation we now
started on three times a day dressing changes. After 2 know as APR, but initial reports showed a high operative
days the wound is clean and a VAC dressing is placed. mortality, up to 42%. Refinements in technique were made
through the first half of the twentieth century. Gabriel
described a one-stage operation with the abdominal por-
tion done supine and the perineal portion done in the left
INTRODUCTION lateral position. Lloyd-Davies’ synchronous approach to the
abdomen and perineum with the patient in the lithotomy
Abdominoperineal resection (APR) completely removes the position eliminated the cumbersome and sometimes dan-
distal colon, rectum, and anal sphincter complex using both gerous need to reposition the patient while under anesthesia
anterior abdominal and perineal incisions, resulting in a (4). Recent advances include total mesorectal excision and
permanent colostomy. Developed more than 100 years ago, methods to enhance perineal wound healing, especially in
it remains an important tool in the treatment of rectal cancer patients who have received neoadjuvant chemoradiation.
despite advances in sphincter-sparing procedures. Recent Minimally invasive techniques are also being applied to
reports have noted an increase in the use of sphincter-sparing APR, with good results.
options for patients diagnosed with rectal cancer. Abraham
and colleagues found a 10% decrease (60.1%–49.9%) in the
rate of APR from 1989 to 2001 as compared with low ante-
rior resection (LAR) using national administrative data (1).
PATIENT PREPARATION AND
When controlled for several variables, including patient POSITIONING
demographics and hospital volume, patients were 28% more
likely to have an LAR later in the study period. Schoetz notes Preparation for abdominoperineal resection starts with
that LAR outnumbers APR three to one in the submitted marking the ideal placement of the colostomy by the primary
case logs of recent colorectal fellows (2). This ratio is simi- surgeon or enterostomal nurse (5). Patients take a mechani-
lar to that found in the Swedish rectal cancer registry, where cal bowel preparation the day before surgery, typically poly-
approximately 25% of over 12,000 patients with rectal cancer ethylene glycol. Placement of an epidural catheter may be
underwent APR from 1995 to 2002 (3). In no study or regis- considered to improve postoperative analgesia and to reduce
try, however, has APR been eliminated. postoperative ileus (6). Prior to induction of general anesthesia,
271
272 Abdominoperineal resection
No pressure over fibular head

No hyperextension of the hip patient in lithotomy position rather than lateral semi-
prone position. A nonabsorbable purse-string suture is
placed around the anus. The abdomen and perineum are
prepared with antiseptic solution and draped with open-
ings to allow synchronous dissection. The abdomen and
pelvis are accessed through a midline hypogastric inci-
sion that extends to the right of or through the umbilicus.
The abdomen is explored for metastatic disease and syn-
chronous colon lesions. After confirmation of resectabil-
ity, a self-retaining retractor is placed.
Angle should not exceed The small bowel is packed into the upper abdomen with
60 degrees from the horizontal a moist towel. The sigmoid and descending colon is mobi-
lized from the retroperitoneum in the left lateral paracolic
gutter. The left ureter is identified and preserved, and the
descending colon is adequately mobilized for creation of
No forced abduction or adduction No stretching of
of the foot adductor muscles an end colostomy. The peritoneal incision is carried ante-
riorly followed by incision of the right lateral peritoneum.
The right ureter is identified and preserved, and the peri-
toneal incisions are connected anteriorly at the base of the
bladder. For convenience, the proximal sigmoid can be
divided with a linear stapling device and the cut end used
as a handle to aid with the dissection. A finger is passed
below the inferior mesenteric vessels preserving the sig-
moid branches, minimizing vascular compromise of the
stoma. It is unnecessary to ligate the inferior mesenteric
artery at its origin as this has not been shown to increase
survival (11).
The superior hemorrhoidal vessels are transected. The
presacral space is entered with preservation of the invest-
Figure 29.1 Leg positioning for abdominoperineal
ing mesorectal fascia consistent with Heald’s description
resection.
of total mesorectal excision (12). A lighted St. Mark’s
retractor holding the mesorectum anteriorly helps iden-
intermittent pneumatic compression devices are placed on tify the avascular plane of dissection. As the dissection
the lower extremities to reduce the risk of venous thrombo- continues distally, Waldeyer fascia is divided sharply to
embolism (7). IV antibiotics with efficacy against enteric flora avoid injuring the presacral venous plexus minimizing
administered 60 minutes before incision decrease the rate of bleeding.
surgical site infection (8). The abdomen and perineum are The lateral ligaments are divided close to the pelvic side
prepped, and appropriate monitoring is placed. wall to maximize the radial margins. Denonvilliers fascia
After induction of anesthesia, a urinary catheter is in males is dissected down to the pelvic floor anteriorly.
inserted; ureteral stents should be considered if the patient Unless the tumor is anterior, it is not necessary to expose the
has had prior pelvic surgery, tumor extension into the uri- seminal vesicles thus avoiding injury to the nervi erigentes.
nary tract, or prior pelvic radiation. The patient is placed in In females, the presence of an anteriorly based tumor may
the lithotomy position with symmetric hip extension, knee require a posterior vaginectomy. When the pelvic floor is
flexion, and thigh abduction using Allen stirrups with pad- reached circumferentially around the rectum, the abdominal
ding to prevent lower limb acute compartment syndrome portion of the dissection is completed, a colostomy is cre-
(9) (Figure 29.1). A rectal exam is performed under anesthe- ated, and the abdomen is closed.
sia followed by irrigation with dilute betadine solution to
remove any residual stool.
LAPAROSCOPIC ABDOMINAL
OPERATIVE TECHNIQUE
OPEN ABDOMINAL OPERATIVE
TECHNIQUE The indications for laparoscopic APR are the same as for
the open technique; it is employed in distal rectal or anal
The operative technique used today varies little from malignancies with sphincter invasion or in those patients in
Ernest Miles’ description in 1908 (10). Unlike Miles’ whom an acceptable functional result cannot be achieved by
method, we prefer the two-team approach with the the use of sphincter-sparing surgical techniques.
Preservation of sexual and urinary function 273
The use of laparoscopy to treat patients with colon cancer disadvantages tied to the use of robotic platforms include
has increased significantly since the publication of the COST loss of haptic feedback, increased costs, and increased oper-
trial (13,14). Data analyzing the outcomes of those patients ative time.
undergoing laparoscopic surgery for rectal cancer are less Robotic-assisted APR follows the same principles as
clear with regard to long-term oncologic outcomes, and its laparoscopic APR except for port placement, as this will
adoption has been slower among surgeons. A recent publica- vary depending on the robotic system employed. Once the
tion by Fleshman et al. states that laparoscopic resection of abdomen has been entered and ports placed, the patient is
patients with stages II–III rectal cancer failed to meet their set placed on steep Trendelenburg position with the right side
criteria for noninferiority pathologic outcomes when com- down to aid in exposure of the pelvis. The abdominal cav-
pared to the open technique (15). Other trials, with longer ity is inspected, and the pelvis is emptied laparoscopically,
patient follow-up, state that oncologic outcomes are similar proceeding afterward to docking the robotic unit between
between laparoscopic and open surgical techniques (16,17). the patient’s legs. It is important to know that once the robot
The operation starts by entering the abdomen under is docked, the table cannot be adjusted without undocking
direct vision in the supraumbilical area; three additional the robotic system. A standard medial-to-lateral dissection
ports are placed, a 12 mm port in the right lower quadrant, is started by incising the peritoneum at the level of the sacral
a 5 mm port in the right upper quadrant, and a 5 mm port promontory. Once this plane is entered, the superior rec-
in the left lower quadrant. The dissection follows a standard tal artery is dissected and the left ureter and gonadal ves-
medial to lateral technique. The peritoneum at the level of sel identified. TME is completed in the same fashion as in
the sacral promontory is incised, and the dissection contin- the open or laparoscopic technique. The initial dissection is
ues medially, identifying the ureter, gonadal, and iliac ves- posterior, moving toward the lateral stalks, and is completed
sels. The IMA is identified, and ligation is performed distal anteriorly. Once at the level of the levators, the dissection
to the takeoff of the left colic artery, this step can be achieved plane proceeds in an extralevator fashion, to avoiding con-
by the use of either an endoscopic stapler or a vessel-sealing ing at this level to avoid a “waist” in the specimen. Once the
device. High ligation of the IMV and full mobilization of abdominal portion of the surgery is concluded, the robot is
the splenic flexure is seldom needed in these patients. After undocked and the perineal dissection performed until the
the medial dissection has been completed, the lateral peri- two planes of dissection meet.
toneal attachments of the colon are incised completing full
colonic mobilization. The sigmoid colon is then transected
with the aid of an endoscopic stapler.
Pelvic dissection is started by entering the presacral space PRESERVATION OF SEXUAL AND
with great care not to violate the endopelvic fascia and pre- URINARY FUNCTION
serving an intact mesorectum. Dissection is carried posteri-
orly to the level of the levator ani complex. Once this has been Sympathetic nerve fibers travel through the lumbar
achieved, the lateral ligaments are mobilized as close to the splanchnic nerves to the superior hypogastric plexus and
pelvic side wall as possible to maximize the radial margin. then divide into two hypogastric nerves. Parasympathetic
Last, anterior mobilization is performed by entering the ante- fibers emerge from the second, third, and fourth sacral
rior areolar plane and carrying the dissection down to the spinal nerves as the pelvic splanchnic nerves and join the
pelvic floor. Once this step is achieved, the abdominal por- hypogastric nerves to form the inferior hypogastric (pelvic)
tion of the operation is concluded, a colostomy is created, the plexus. The pelvic plexus is rectangular, and its midpoint
abdomen is closed, and the perineal dissection commences. is located at the tips of the seminal vesicles on either side
of the rectum (Figure 29.2). The most caudal portion of the
pelvic plexus travels at the posterolateral border of the pros-
tate, lateral to the prostatic capsular arteries and veins, and
ROBOTIC ABDOMINAL OPERATIVE reaches the hilum of the penis (20).
TECHNIQUE The rate of urinary dysfunction and impotence after
rectal surgery ranges from 33% to 70% and 20% to 46%,
Since the introduction of robotic-assisted surgery, this respectively, while 20%–60% of potent patients are unable
technology has been increasingly applied to the specialty to ejaculate (21). A surprisingly large proportion of patients
of colon and rectal surgery. The advantages of using the suffer various urinary tract problems and sexual problems
robotic platform for rectal operations are improved ergo- due to extended lymphadenectomy involving the hypo-
nomics for the surgeon, excellent three-dimensional deep gastric nerve plexus. Therefore, preservation of the pelvic
pelvic visualization, and articulated instruments that allow autonomic nerves lowers the incidence of sexual and uri-
for an improved freedom of movement in a small space, as nary morbidity. With preservation of the superior hypogas-
it is the deep pelvis (18). From an oncological standpoint, tric nerve plexus, ejaculation is maintained in 90% of the
robotic-assisted operations appear to have similar oncologic patients (22).
outcomes, in terms of circumferential margins and lymph Utilizing precise dissection with preservation of auto-
node harvest, as open or laparoscopic operations (19). The nomic nerves, Kim et al. noted an erection rate of 80%,
Sympathetic
trunk
Hypogastric
plexus
Lumbosacral
plexus
Vesicle
plexus
Pelvic nerves
(Nervi erigentes)
Pudendal N.
Dorsal N.
of penis
Inferior
Perineal N. rectal N.
Figure 29.3 Thumbtack occlusion of bleeding basiverte-

bral vein.
Figure 29.2 Nerve supply to the rectum.
Ideally, by taking sharp dissection down the presacral plane,

penetration ability rate of 75% with only 5.5% of patients there should be little to no bleeding (24,25). Unfortunately,
in their study reporting complete inability for erection and this space may be nonexistent in certain patients or obliter-
intercourse (22). Study by Shirouzu et al. showed oncologic ated in an irradiated field. Bleeding from the sacrum can be
equivalence between previously described extensive resec- controlled by packing, suture ligation, electrocautery, finger
tion pre-1984 and plexus preserving low rectal surgery compression, or thumbtack compression.
post-1985 with local recurrence rates of 9.1% and 3.9%, Thumbtack compression is a quick, safe, and effective
respectively, and 10-year, disease-free survival rates of 77% method of controlling sacral bleeding. There are several
and 81.5%, respectively. No significant difference was noted commercial application devices available; however, using
among the groups (23). a clamp or forceps with finger applications works equally
as well (Figure 29.3). Thumbtacks also prevent damage to
the surround venous plexus that may occur when using
the other methods of attempting hemostasis, such as direct
INTRAOPERATIVE HEMORRHAGE suture ligation or excessive cauterization (25,26).
Hemorrhage during surgery can usually be attributed to an

error in technique, but when faced with a pelvis that had
previously received radiation therapy, hemorrhage may be
STANDARD PERINEAL OPERATIVE
unavoidable. TECHNIQUE
Bleeding may occur when dissection begins at the sig-
moid. This is usually easily identified and controlled. In the When the abdominal operator has determined that the lesion
previously irradiated pelvis, planes become distorted, mak- is resectable, the perineal dissection may begin simultane-
ing it difficult to identify vital structures. It is easy to stray ously with the abdominal portion of the case or sequentially
laterally, which may result in iliac vessel injury. These must if necessary. The perineal dissection begins with an ellipti-
be repaired immediately to avoid prolonged hemorrhage. In cal incision from the perineal body in males or the posterior
a pelvis that has not received radiation or if there is minimal vaginal introitus in females to a point midway between the
fibrosis, meticulous dissection in the proper plane down to anus and coccyx. The incision should include the entirety of
the lateral stalks usually yields minimal bleeding. the external sphincter muscle but does not need to extend
The most troublesome bleeding in the pelvis comes from laterally to the ischial tuberosities. Dissection is carried
the posterior dissection along the sacrum. Very rarely there down to the levator ani muscles with cautery to minimize
will be a prominent medial sacral artery that may be injured. bleeding. The inferior hemorrhoidal arteries located poste-
More commonly, the bleeding on the sacrum will come from rior laterally are ligated. Using a finger on the tip of the coc-
the venous plexus. If present, the basivertebral vein, which cyx as a guide, the posterior dissection is directed anterior
connects the internal vertebral venous system to the presa- to the coccyx, and the anococcygeal raphe is divided. When
cral system, can bleed profusely and be difficult to control. all that remains is the anterior attachments, the specimen
Methods of closure 275
is drawn through the opening and used to provide traction without continuous irrigation, and omental plugging may
to continue the remaining dissection. The specimen is then also be considered.
removed and the pelvis is irrigated. If sufficient levator mus- Rates of primary healing after perineal wounds are closed
cle remains, the pelvic floor is reapproximated to reduce the range from 4% to 92% (10,31,32). Open packing relegates all
risk for perineal herniation. Drains are placed and secured wounds to secondary healing, is inconvenient, and is often
followed by closure of the skin with interrupted permanent painful but may result in a lower rate of chronic perineal
or absorbable monofilament suture in a vertical mattress sinus formation (33). Closure of the pelvic peritoneum has
fashion. been advocated to prevent perineal evisceration and post-
operative small bowel obstruction. However, it may prevent
obliteration of the pelvic cavity, leading to formation of a
persistent perineal sinus (34). Loops of small bowel may also
EXTRALEVATOR OPERATIVE become incarcerated in small defects in the peritoneal clo-
TECHNIQUE sure, resulting in postoperative bowel obstruction.
Two studies compared various methods of peritoneal
Since the description of total mesorectal excision by Heald and perineal closure. Irvin and Goligher (33) prospectively
in 1982, the local recurrence and survival rates from rectal randomized 106 patients undergoing proctectomy to one of
cancer have improved greatly (12). However, those patients three methods of perineal closure: open packing of the peri-
undergoing APR had worse oncologic outcomes than those neal wound; primary closure of the perineal wound with-
patients undergoing sphincter sparing procedures, mainly out closure of the pelvic peritoneum with suction drainage
due to higher rates of inadvertent tumor perforation and of the pelvis; and primary closure of the peritoneal and
positive circumferential margin (27). perineal wounds. The overall complication rate was high;
Extralevator APR has been credited with oncologic repeated surgery was necessary in 21% of patients in the
superiority with reduction of local recurrence and improv- open packing group, most often because of hemorrhage, and
ing survival. This has been attributed to the wider resection in 25% and 19% of the two closed groups, most commonly
margin that avoids “wasting” at the level of the levators and for drainage of abscesses. Primary healing occurred in 45%
increases the circumferential margin of the specimen (28). of the patients with primary closure of both the perineum
Patients undergoing extralevator APR can be placed in and peritoneum and in 43% of patients with open peritoneal
lithotomy position, or they can be flipped into prone jack- and closed perineal wounds.
knife position once the abdominal phase of the operation is In a prospective study part of a multicenter trial in
concluded. The anus is closed with a purse-string suture and Germany, Meyer et al. published a standardized technique
the perianal skin incised. Once the skin is incised, expo- of perineal closure that reduced wound complication rates
sure can be aided by the use of a commercial self-retraction from 17% to 5.4%. The perineal wound was closed tightly in
device. multiple layers utilizing the levator muscles and subcuta-
The anterior margin of the dissection will be the trans- neous fat, which helped to avoid the accumulation of fluid
verse perineal muscle, and the posterior margin the tip of within the wound cavity. Any residual fluid was removed
the coccyx. Lateral dissection is extended into the ischiorec- by closed suction drainage. Additionally, it is thought that
tal fossa on both sides. Dissection is continued laterally to the addition of antibiotic carriers provides local infectious
the pelvic side wall and includes the soft tissues around the prophylaxis leading to lower rates of perineal wound infec-
anorectal ring. Dissection can be performed with a combi- tion (35). This has also been demonstrated in two other
nation of monopolar energy and a vessel sealing device. prospective randomized studies and can be considered an
Extralevator abdominoperineal resection will leave a adjunct in decreasing the overall morbidity of the perineal
large perineal defect in patients with prior history of radia- wound (36).
tion therapy. The perineal wound can be closed primarily in Myocutaneous flaps have been increasingly utilized
multiple layers, but this is often not possible. For this reason, in the initial closure of the perineal defect, especially in
closure of the defect can be aided by the use of biologic mesh patients who have had preoperative radiation therapy.
or by the use of muscle-cutaneous pedicle flaps, minimizing Chessin et al. at Memorial Sloan Kettering reviewed their
tension at the closure site, with decreased healing time and experience with rectus abdominis myocutaneous flap clo-
wound failure rates (29,30). sures of the perineal defect. They found the incidence of
perineal wound complications was 15.8% in the rectus
abdominis myocutaneous flap group compared to the 44.1%
in historical controls (37). Butler et al. also looked at verti-
METHODS OF CLOSURE cal rectus abdominis myocutaneous flaps in previously irra-
diated patients undergoing APR. There was a significantly
The perineal wound can be packed open, partially closed, lower incidence of perineal abscess (9% versus 37%), major
or completely closed. The peritoneal defect above the pel- perineal wound dehiscence (9% versus 30%), and drainage
vic space can also be sutured closed or left open. Adjunctive procedures required for perineal or pelvic fluid collections
procedures such as drainage of the pelvic space, with or (3% versus 25%) (38).
In an effort to fill the pelvic space after rectal resection, the most mobile. They contend that the pelvic space after
Page advocates an omental plug. The omentum is mobi- APR is filled not with granulation tissue but with a combi-
lized on the left gastroepiploic arterial pedicle and placed nation of upward migration of the perineal soft tissues and
in the pelvis, which increases local blood flow, increases descent of the peritoneal contents and argue that any forces
lymphatic drainage, and helps obliterate the pelvic space. that produce a fixed fibrotic cavity are likely to result in a
The omental plug also keeps the small bowel out of the pel- nonhealing perineal wound (34). Artioukh et al. reviewed
vis, decreasing the chance of radiation enteritis in patients their series of APR nonhealing wounds and found several
who require postoperative radiotherapy. Primary healing possible contributing factors, including distant metastases,
was noted in 26 of 34 patients (77%) (39). Nilsson reviewed excessive alcohol consumption, cigarette smoking, transfu-
all available English-language publications on the use of sion requirement, and chemoradiation (45).
omentoplasty in APR wound closure with primary wound Other studies have also observed the increased risk in
healing as the primary outcome measure. Most authors perineal wound infection and nonhealing in those who
reported positive results after omentoplasty; one study had radiotherapy. The Swedish Rectal Cancer trial showed
showed significant improvement in perineal healing rate an increase in wound infection from 10% to 20%, and the
at 6 months. Significant reduction in sinus formation and Dutch Colorectal Cancer Group had a 31% perineal com-
wound dehiscence also was reported (40). Despite these plication rate even in those exposed to short-course radia-
promising results, there need to be randomized trials with tion (46,47).
well-described patient categories, endpoints, and follow-up Silen and Glotzer recommended that the peritoneal
to firmly assess whether omentoplasty should be a standard contents be allowed to descend into the pelvis, the space
part of the wound closure. be kept irrigated and well drained to prevent fluid accu-
mulation, and any packing used in the perineal wound be
removed early to prevent development of fibrotic wound
edges. Despite the excellent description of perineal healing
ABSCESS by Silen and Glotzer and the development of multiple tech-
niques for perineal closure, nonhealing perineal wounds
Abscess formation, either intraperitoneal or perineal, is the remain a common problem. Bacon and Nuguid noted a 40%
most common major complication after APR (31). Incidence incidence of persistent perineal sinus in 1,042 patients after
of abscess formation ranges from 11% to 16% (31,32,41). In rectal resection (48). In almost 500 patients who underwent
some small series, the incidence of perineal wound infec- APR at the Lahey and Mayo Clinics, 14%–24% had unhealed
tion is 100% (33). This can be attributed to the large dead perineal wounds at 6 months.
space remaining after resection of the rectum and from Many techniques have been developed to ensure com-
fecal contamination. plete healing. Early efforts included operative debridement
Incision and drainage with local wound care are the treat- with wide drainage, including coccygectomy and even par-
ment of choice for local perineal wound abscesses. There is tial sacral resection (34). These measures were designed to
a small increased risk of developing a perineal sinus after eliminate the rigid fibrotic space that always accompanies a
opening the skin of a subcutaneous abscess (42). Thus, if the nonhealing perineal wound. Often these measures resulted
incision is healing well, the abscess may be amenable to per- in eventual healing but required extensive wound care for
cutaneous drainage. In addition, percutaneous drainage is many months. Despite this treatment, some wounds failed
the preferred treatment of presacral and pelvic abscesses (43). to heal.
Oomen et al. published a set of guidelines in treating
persistent perineal sinuses or complex perineal wounds
with an overall 80% success rate in healing. Their algorithm
NONHEALING WOUND AND PERINEAL consisted of VAC therapy for large defects prior to plac-
SINUS ing muscle flaps in order to decrease the size of the defect.
Depending on sinus length, they either placed a transposi-
Perineal sinus is defined as a perineal wound that remains tion of rectus abdominal muscle (for sinuses >10 cm) or a
unhealed for a minimum of 6 months. Characteristics include gracilis muscle/gluteal thigh flap (sinus <8 cm). Initially the
a fixed fibrotic pelvic cavity, a long, narrow track lined with a success rate was 57%, but after secondary surgery in some of
thick unyielding peel, and a small external opening (44). the patients, their success rate increased to 80%. Ultimately,
Silen and Glotzer compared the pelvic space after APR the best outcomes were in patients who received the gracilis
with the fixed pleural space after pneumonectomy. The or gluteal thigh flap (49).
pelvic space is bound posteriorly and laterally by the rigid The VAC closure system has been used to deal with com-
bony pelvis, anteriorly by the relatively unyielding genito- plex perineal wounds following extensive operative debride-
urinary structures, inferiorly by the slightly mobile peri- ment for persistent perineal sinuses. Pemberton at the Mayo
neal floor, and superiorly by the peritoneal contents. Of Clinic (50) showed that in difficult perineal sinuses requir-
all these borders, certainly the peritoneal structures are ing debridement and removal of the coccyx and caudal part
References 277
of the sacrum, the VAC system had complete resolution of imaging (MRI). A comparative study of dynamic MRI and
the sinus in nearly all of their patients. While their evidence dynamic cystocolpoproctography showed that MRI was the
is anecdotal, there are documented reports with healing only modality that identified levator ani hernias (54).
rates up to 95% (51,52). There are few large published series to describe which
Omentoplasty has been evaluated in both the primary technique of perineal defect closure is superior. Various case
repair of the perineal wound as well as in complex perineal reports and retrospective reviews provide much of the lit-
sinus disease. Yamamoto et al. reported six patients with per- erature in this respect. In a review of the literature, closure
sistent perineal sinuses who underwent omentoplasty. The techniques have ranged from the use of simple suture closure,
perineal sinus tract was completely excised and communica- prosthetic mesh, human dura mater allograft (62), gracilis
tion with the pelvis attained. The left or right gastroepiploic myocutaneous flap (63), gluteus flap, and retroflexion of the
vessels were then ligated and the omentum brought down to uterus or bladder (64). So et al. described their experience
the perineum where it was lightly sutured to the skin. After with closures and ultimately found that recurrence rates were
a 28-month follow-up period, 83% of the patients had com- equal (20%) between simple and mesh closures (60). Their
pletely healed wounds without any c omplications (53). repair consisted of simple closure of the levator defect with
nonabsorbable sutures. The approach to the repair was also
felt to be a point of consideration in planning the operation.
For the most part, a perineal approach was adequate with
PERINEAL HERNIA AND EVISCERATION the abdominal approach reserved for recurrent hernias, or
those in whom laparotomy is necessary for other reasons.
Perineal hernias are fortunately very rare and often trou- The abdominal approach also provides good visualization
blesome to diagnose. Perineal hernia after abdominoperi- when suturing the mesh to the bony pelvis. A combined AP
neal resection is defined as bulging of peritoneal contents approach is rarely necessary except under unusual circum-
through an intact perineal wound, and perineal eviscera- stances. Skipworth et al. published their experience and
tion describes extrusion of small or large bowel through technique of perineal hernia repair using Permacol mesh.
an open perineal wound. However, other unusual contents Using a perineal approach, they isolated and ligated the sac
have been described, including a leiomyoma, an aggres- in the standard fashion before proceeding to close the peri-
sive angiomyoma, and a large bladder diverticulum (54). neal defect. The mesh was then fashioned to the contours of
Evisceration typically occurs immediately after surgery the defect and sutured in place, tension free, with interrupted
and necessitates repeat surgery with reduction of intestines sutures. A small suction drain was then left superficial to
and repeat packing. Perineal hernias are a rare complica- the mesh and the thin, residual perineal fascia closed. They
tion and occur in about 1% of patients after APR. This figure reported no recurrence in the 18 months following the repair.
increases to 3% after pelvic exenteration. Initial symptoms There are also a growing number of case reports and pro-
include perineal bulging, often associated with fullness or spective studies in the use of laparoscopy for perineal hernia
pain on sitting (55,56). Occasionally, patients complain of repairs. Dulucq et al. describe their experience in a prospec-
voiding problems if herniated bowel compresses the bladder tive study done over the course of a year with three patients
(57). Rarely, skin breakdown occurs, resulting in exposed who had received laparoscopic mesh repairs of their perineal
bowel in the perineum. Perineal hernias, like parasto- hernia defects. A composite mesh was fixed laterally to the
mal and incisional hernias, do not always require repair. border of the levator muscle, anteriorly to the posterior face of
Indications for surgery are similar for all three postopera- the vagina with nonabsorbable sutures and posteriorly with
tive hernias: patient discomfort refractory to conservative tacks to the sacral periosteum. One suction drain was placed.
therapy, bowel obstruction, incarceration, and impending The reported benefits include adequate visualization of pelvic
skin loss. Cosmesis alone should rarely merit surgical repair. anatomy, the ability to look for recurrence, and fast recovery.
Risk factors that predispose patients to developing peri- Long-term results have yet to be published for laparoscopic
neal hernias are not entirely clear. Coccygectomy, previous perineal hernia repairs, but this may be an attractive option
hysterectomy, pelvic irradiation, excessive length of the for patients and surgeons as it often avoids making large inci-
small-bowel mesentery, the larger size of the female pel- sions in areas that have already been irradiated and can there-
vis, and possibly the failure to close the peritoneal defect fore be difficult to heal (65). Prior to embarking on a repair of
have been implicated as possible causes (58,60). So et al. any postoperative perineal hernia, it is imperative to exclude
described 80% of their patients having perineal wounds the possibility of cancer recurrence.
that were laid open or had multiple large drains inserted
through the wound, which they postulate may weaken the
wound and allow hernia formation (61). REFERENCES
Diagnosis of perineal hernias can be difficult as traditional
fluoroscopic imaging techniques often do not identify them. 1. Abraham NS et al. Aliment Pharmacol Ther. 2001;
Other modalities have been used to include herniography, 21(1):35–41.
computed tomography, and dynamic magnetic resonance 2. Shoetz Jr. DJ. J Am Coll Surg. 2006;203(3):322–27.
3. Swedish Colorectal Cancer Registry. Available at: 30. Devulapalli C et al. Plast Reconstr Surg. 2016;137(5):
http://kvalitetsregister.se/englishpages/findaregistry/ 1602–13.
registerarkivenglish/swedishcolorectalcancerregistry 31. Murrell ZA et al. Am Surg. 2005;71(10):837–40.
scrcr.2156.html. Accessed June 1, 2018. 32. Rosen L et al. Dis Colon Rectum. 1982;25:202–8.
4. Ruo L, Guillem JG. Dis Colon Rectum. 1999;42(5): 33. Irvin IT, Goligher JC. Br J Surg. 1975;62:287–91.
563–78. 34. Silen W, Glotzer DJ. Surgery. 1974;75:535–42.
5. American Society of Colon and Rectal Surgeons 35. Meyer L et al. Tech Coloproctol. 2004;8(Suppl. 1):
Committee Members; Wound Ostomy Continence s230–4.
Nurses Society Committee Members. J Wound 36. Gruessner U et al. Am J Surg. 2001;182:502–9.
Ostomy Continence Nurs. 2007;34(6):627–8. 37. Chessin DB et al. Ann of Surg Onc. 2005;12(2):104–10.
6. Marret E et al. Postoperative Pain Forum Group. 38. Butler CE et al. J Am Coll Surg. 2008;206(4):
Br J Surg. 2007;94(6):665–73. 694–702.
7. Geerts WH et al. Chest. 2008;133(Suppl. 6): 39. Page CP et al. Dis Colon Rectum. 1980;23:2–9.
381S–453S. 40. Nilsson PJ. Dis Colon Rectum. 2006;49:1354–61.
8. Bratzler DW, Houck PM. Clin Infect Dis. 2004; 41. Pollard CW et al. Dis Colon Rectum. 1994;37:866–74.
38:1706–15. 42. Baudot PE et al. Br J Surg. 1980;67:275–6.
9. Beraldo S, Dodds SR. Dis Colon Rectum. 2006; 43. Michalson AE et al. Radiology. 1994;190:574–5.
49(11):1772–80. 44. Anthony JP, Mathes SJ. Arch Surg. 1990;125:1371–7.
10. Miles WE. Lancet. 1908;2:1812. 45. Artioukh DY et al. Colorectal Dis. 2007;9(4):362–7.
11. Corman ML. Ed. Carcinoma of the rectum. In Colon 46. Fasth S et al. Ann Chir Gynaecol. 1977;66:181–3.
and Rectal Surgery, 5th Edition. Philadelphia, PA: 47. Scott H, Brown AC. Am Surg. 1996;62:452–7.
Lippincott, Williams, and Wilkins, 2005, pp. 905–1061. 48. Bacon HE, Nuguid TP. Dis Colon Rectum. 1962;5:
12. Heald RJ et al. Br J Surg. 1982;69:613–6. 370–2.
13. Clinical Outcomes of Surgical Therapy Study Group. 49. Oomen JW et al. Int J Colorectal Dis. 2007;22(2):
N Eng J Med. 2004;350(20):2050–9. 225–30.
14. Moghadamyeqhanez Z et al. Dis Colon Rectum. 50. Permberton JH. Colorectal Dis. 2003;5(5):486–9.
2015;58(10):950–6. 51. Argenta LC, Morykwas MJ. Ann Plastic Surg. 1997;
15. Fleshman J et al. JAMA. 2015;314(13):1346–55. 38:563–77.
16. Ng SS et al. Surg Endosc. 2014;28(1):297–306. 52. Deva AK et al. Med J Aust. 2000;173:128–31.
17. Kang SB et al. Lancet Oncol. 2010;11(7):637–45. 53. Yamamoto T et al. Am J Surg. 2001;181(3):265–7.
18. Yamaguchi T et al. Surg Today. 2016;46(8):957–62. 54. Skipworth RJ et al. Hernia. 2007;11:541–5.
19. Barnajian M et al. Colorectal Dis. 2014;16(8):603–9. 55. McMullin ND et al. Aust N Z J Surg. 1985;55:69.
20. Kyo K et al. World J Surg. 2006;30(6):1014–9. 56. Rutledge RN et al. Am J Obstet Gynecol. 1977;
21. Moriya Y. Int J Clin Oncol. 2006;11(5):339–43. 129:881.
22. Kim NK et al. Dis Colon Rectum. 2002;45: 57. Brotschi E et al. Am J Surg. 1985;149:301–5.
1178–85. 58. Cattell RB, Cunningham RM. Surg Clin North Am.
23. Shirouzu K et al. Dis Colon Rectum. 2004;47(9): 1944;24:679–83.
1442–7. 59. Kelly AR. Aust N Z J Surg. 1960;29:243–45.
24. Wang O et al. Arch Surg. 1985;120:1013–20. 60. Frydman GM, Polglase AL. Aust N Z Surg. 1989;59:
25. Arnaud JP et al. Dig Sur. 2000;17(6):651–2. 895–7.
26. Nivatvongs S, Fang DT. Dis Colon Rectum. 1986;29: 61. So JB et al. Dis Colon Rectum. 1997;40:954–7.
589–90. 62. Delmore JE et al. Obstet Gynecol. 1987;70:507–8.
27. Wibe A et al. Dis Colon Rectum. 2004;47(1):48–58. 63. Bell JG et al. Obstet Gynecol. 1980;56:377–80.
28. Marr R et al. Ann Surg. 2005;242(1):74–82. 64. Remzi FH et al. Tech Colo-proctol. 2005;9:142–4.
29. Howell AM et al. Int J Surg. 2013;11(7):514–7. 65. Dulucq JL et al. Surg Endosc. 2006;20(3):414–8.
30
Management of rectal cancer after
complete clinical response to neoadjuvant
chemoradiotherapy
RODRIGO O. PEREZ AND LAURA MELINA FERNANDEZ
of tumor response following nCRT and prior to radical sur-

CHALLENGING CASE gery may identify patients with complete clinical response
who could be managed nonoperatively with strict follow-up,
A 69 yr women with a clinical T3N1 rectal cancer the “watch and wait” (WW) strategy, which in select cases
undergoes neoadjuvant chemoradiation therapy after may help to avoid unnecessary postoperative morbidity and
review at a multi-disciplinary tumor conference (MDT). maintain good long-term oncological outcomes and excel-
Eight weeks after completing the therapy, a digital and lent functional results (3). In addition, close surveillance
endoscopic evaluation reveals no residual tumor. An may allow for early detection of local recurrences and sal-
MRI and PET scan fails to demonstrate any tumor. vage alternatives without oncological compromise (4). The
present chapter discusses the management of rectal cancer
following complete clinical response after nCRT.
CASE MANAGEMENT
The patient’s case is again discussed at a MDT. She

RATIONALE FOR ORGAN
appears to have a complete clinical response. She is
PRESERVATION IN RECTAL CANCER
offered an oncologic resection or enrollment into a
strict surveillance program. Different organ-preserving strategies in the treatment of rec-
tal cancer have gained popularity in recent years. The main
reasons for avoiding a proctectomy include the significant
postoperative morbidity, including long-term urinary, sexual,
INTRODUCTION and fecal continence dysfunction in addition to the require-
ment for temporary or definitive stomas associated with the
Incorporation of new treatment modalities has significantly procedure. Also, depending on associated comorbidities and
increased complexity in the management of rectal cancer patient’s age, postoperative mortality may also be significant
(1). Surgical treatment is still the main pillar in the manage- (5). Therefore, in selected patients with evidence of complete
ment of rectal cancer. We continue to have a better under- primary tumor regression, less invasive surgical and even
standing of the different approaches for total mesorectal nonsurgical approaches have been suggested (6).
excision (TME), including standard, laparoscopic, robotic, The observation that rectal cancers could develop sig-
and, most recently, transanal TME. In addition to evolving nificant tumor regression with reduction in primary tumor
surgical approaches, neoadjuvant therapy and the manage- size (downsizing), depth of tumor penetration, and even
ment of patients after assessment of tumor response to neo- potential nodal sterilization (downstaging) set the ideal
adjuvant therapy are also undergoing widespread changes. stage for organ-preserving alternatives, including local exci-
Neoadjuvant chemoradiation therapy (nCRT) may lead to sion of small and superficial residual tumors (7). In some
significant tumor regression, and ultimately complete path- cases, regression of the primary tumor can result in com-
ological response in up to 42% of patients (2). Assessment plete disappearance of the tumor in the resected specimen
279
280 Management of rectal cancer after complete clinical response to neoadjuvant chemoradiotherapy
(complete pathological response [pCR]). In a subset of these (cEMVI+), and ≥3 positive lymph nodes (cN2). In addition,
patients, complete disappearance of the primary tumor may radical surgery after nCRT was shown to result in worse
be clinically detected prior to surgical resection, referred to functional outcomes and increased surgical morbidity
as complete clinical response (cCR) (8). These patients with when compared to surgery alone (20,21). Altogether, these
complete tumor regression of their primary rectal cancers to data suggested that nCRT was to be restricted for high-risk
nCRT may be ideal patients to consider nonoperative organ- patients (also referred to as the “ugly” tumors) for the devel-
preserving strategies (9). In order to consider this approach, opment of local recurrence only. Considering that baseline
colorectal surgeons have to take into consideration several staging features may influence development of complete
aspects of the disease, the patient, and treatment modalities response to nCRT, one could expect that very few patients
that may be quite relevant during their clinical decision- with such advanced disease would do so.
making process. However, the possibility of avoiding radical surgery and
its related comorbidities after a cCR raised the issue of offer-
ing nCRT to more early stage rectal cancers, particularly for
PREDICTION OF RESPONSE TO nCRT the most distal tumors. Ultimately, patients with cT2N0
or early cT3N0 are more likely to develop a cCR following
AND INTRATUMORAL HETEROGENEITY
nCRT and could benefit the most from nCRT if organ pres-
ervation is considered (22–24).
Several studies have attempted to provide a clinically useful The use of nCRT should be considered only for high-risk
tool based on molecular biology features of rectal cancers patients (cCRM+, cEMVI, and cN2) if radical surgery is to
undergoing nCRT to predict response to treatment up front be performed regardless of response to treatment. However,
(10). This would allow more precise selection of patients if organ-preserving strategies are an option according to
who would benefit the most from CRT, spare patients from tumor response, nCRT may be offered to most rectal can-
potentially unnecessary treatment, and identify ideal cancers (except for cT1N0) (25). Here tumor location or height
didates for nonoperative management. Unfortunately, may be of significant importance. As discussed in the fol-
however, these studies have failed to provide any clinically lowing sections of the chapter, clinical assessment including
relevant information to be implemented into clinical prac- digital rectal examination (DRE) is crucial for the identifi-
tice so far. First, published gene signatures rarely present cation and surveillance of cCR, and only baseline cancers
specific genes overlapping between them. Second, valida- accessible to DRE (usually up to 7–8 cm from the anal verge)
tion of findings between these signatures in independent would be appropriate candidates for organ-preserving strat-
cohorts often results in inaccurate identification of com- egies without immediate surgery (1).
plete responders to nCRT (10–13).
The presence of significant intratumoral heterogeneity NEOADJUVANT TREATMENT OPTIONS
may have accounted at least in part for these disappointing
results (14,15). The coexistence of subpopulations of cancer Specific features of a neoadjuvant therapy regimen may
cells within a single rectal cancer with distinct morphologi- ultimately affect the odds of developing a cCR and should
cal features and genetic mutations may render single biopsy be considered in the setting of organ-preservation strate-
samples simply not representative of the entirety of the pri- gies. Initially, it was thought that long-course CRT was the
mary tumor. Therefore, a single biopsy sample from one only strategy that could result in significant rates of com-
area of the primary tumor may contain cancer cells that are plete response, whereas short-course CRT would only rarely
resistant to nCRT, while biopsy taken from other areas may have such clinical outcome. However, with the understand-
contain cancer cells that are sensitive to nCRT (16). ing of the influence of time in the development of complete
response to therapy, it has been suggested that short-course
BASELINE STAGING AND INDICATIONS RT followed by delayed assessment of response may result
FOR NEOADJUVANT CHEMORADIATION in similar rates of complete response to the observed after
long-course CRT (2,26,27).
Following the results of the German trial, chemoradiation The dose of radiation therapy (RT) may also influence the
was considered the preferred initial approach for cT3–4 or odds of patients with rectal cancer in developing complete
cN+ rectal cancers due to the potential benefits in local dis- response to treatment. Dose-escalation studies have demon-
ease control after radical surgery (17,18). However, data from strated progressive increase in complete response rates with
the Mercury study suggested that after proper or optimal higher doses of RT delivered to the primary tumor (28,29).
total mesorectal excision, local recurrence was unlikely to In addition to the actual dose delivered, the method of deliv-
develop for most cT3 cancers, even in the presence of nodal ery may also affect the development of a complete response.
disease (cN1) (19). Instead, patients at higher risk for local Therefore, the combination of external beam RT or intensity
recurrence, and therefore, those who would most benefit modulated RT with endorectal brachytherapy or even with
from nCRT in order to improve local disease control would contact RT may further increase the total dose of radiation
include radiological evidence of a positive circumferential delivered, maximizing the chances of developing cCR and
margin (cCRM+), presence of extramural venous invasion still avoiding major treatment-related toxicity (30–32).
Prediction of response to nCRT and intratumoral heterogeneity / Timing for the assessment of tumor response 281
More recently a strategy has been suggested to provide with 6-week intervals presented significantly more tumor
neoadjuvant therapy with chemotherapy alone prior to regression after nCRT (37). Due to this study, a 6-week time
RT in an attempt to avoid the toxic and potential morbid- interval between nCRT completion and performance of
ity resulted from RT in these patients (33). The delivery radical surgery has been considered the standard of care
of chemotherapy alone would allow the control of possi- for many years. However, retrospective studies consistently
ble micrometastatic foci of the disease while still provid- reported that patients undergoing radical surgery after lon-
ing significant response to the primary tumor in a good ger than 6–8 weeks from nCRT were more likely to develop
proportion of patients. Standard CRT could be restricted pCR (38–41). One of these studies suggested that the rates
to patients showing minimal response to chemotherapy of pCR after nCRT may keep rising after nCRT for as long
alone, therefore minimizing the amount of patients receiv- as 12 weeks from treatment completion (39). However,
ing RT (34). there was a question whether these prolonged intervals
Finally, combinations of standard CRT and more aggres- from nCRT would result in excessive tissue fibrosis in the
sive chemotherapy regimens have been suggested that area included in the RT field that could lead to increased
include additional cycles of chemotherapy being delivered technical difficulty and postoperative morbidity after radi-
during the resting period after RT completion in stan- cal surgery. One study included patients in nCRT regimens
dard CRT regimens (consolidation CRT regimens). One with progressively longer interval periods prior to surgery.
study adding additional cycles of 5FU-based chemotherapy Even though this was not a randomized study, patients in
during the resting period after 54 Gy of RT suggested an different groups were comparable (74). Curiously, patients
increase of CR rates to >50% in patients with T2/T3 rectal undergoing surgery after 12 weeks developed similar post-
cancer (25,35). Data from a prospective study using stan- operative complication rates when compared to the stan-
dard CRT followed by progressively higher numbers of dard 6-week interval. The study then kept on recruiting
FOLFOX cycles during the resting periods after RT comple- patients for progressively longer intervals: 6, 12, 18, and 24
tion have demonstrated a significant increase in pCR rates weeks between nCRT and surgery. Even though additional
after radical surgery (42). systemic chemotherapy has been offered to patients under-
Altogether, these data may suggest that if organ preser- going surgery after longer interval periods, delaying surgi-
vation is an option, optimization of RT and chemotherapy cal resection to ≥20 weeks resulted in significantly higher
should be considered up front rather than after standard CRT. pCR rates with no negative impact on postoperative mor-
bidity (42). Altogether, these data seem to suggest that the
ASSESSMENT OF TUMOR RESPONSE longer you wait, the more tumor regression is observed, and
that longer intervals than 6–8 weeks would clearly benefit
Considering that patients may develop significant tumor patients after nCRT. However, another recently published
regression after nCRT, which may provide an appropriate randomized study failed to demonstrate the benefits of lon-
setting for an organ-preserving strategy, one issue becomes ger intervals after nCRT. In this study, patients undergoing
crucial in this process: assessment of tumor response. 7-week intervals developed similar pCR rates to patients
However, assessment of tumor response may be quite chal- undergoing 11-week intervals. Moreover, patients undergo-
lenging due to numerous uncertainties including optimal ing 11-week intervals developed increased rates of postop-
timing and clinical/radiological tools for such purpose. erative complications and ended up with worse quality of
Assessment of tumor response is also recommended the resected specimen (quality of the mesorectum), suggest-
even if an organ-preserving strategy is not being considered. ing the detrimental effects of prolonged time after nCRT on
Even if the plan after nCRT is a radical resection, one needs fibrotic changes in the surgical and previously irradiated
to consider that after nCRT the surgeon may be facing a fields (43).
considerably different tumor. Knowing this potentially new The optimal interval after nCRT remains undetermined,
“anatomy” ahead of time may allow the surgeon to optimize and additional ongoing trials will definitely provide more
intraoperative surgical strategy and to know in advance data to allow us to understand the benefits and risks of using
what challenges could be anticipated during the procedure prolonged intervals after treatment. In fact, it may be the
(36). Therefore, the reassessment of tumor response should case that a single and fixed interval may not be appropriate
be performed in all patients. for all patients. Instead, patients/tumors may respond differ-
ently as a function of time to nCRT. Ultimately, responsive
TIMING FOR THE ASSESSMENT tumors may require and actually benefit from prolonged
OF TUMOR RESPONSE intervals from nCRT, whereas unresponsive tumors may
not. It is likely that responsive tumors that are being con-
The grade of tumor regression after nCRT appears to be a sidered for organ-preserving strategies should have their
time-dependent phenomenon. The first randomized trial assessment of response and ultimately surgical strategy
to consider the effect of different time intervals in the decision deferred to longer than 12 weeks. But tumors with
response to CRT was a French study comparing 2 versus little response that still require radical TME may benefit
6 weeks from nCRT. In this study, all patients underwent from 6- to 8-week intervals between nCRT completion and
radical surgery after these two time intervals, and patients radical surgery (44).
TOOLS IN ASSESSMENT OF TUMOR the finger or the proctoscope, particularly regarding nodal
RESPONSE or mesorectal status of the disease. Currently, however, sig-
nificant developments in imaging definition and interpreta-
Clinical and endoscopic assessment tion have resulted in significant increases in accuracy for
the assessment of response not only within the mesorectum
Clinical assessment is one of the most important tools to compartment, but also within the rectal wall.
evaluate tumor response. Commonly, patients with tumor High-resolution magnetic resonance (MR) is now rou-
regression would have relief of their symptoms. DRE is an tinely used for the assessment of response. The ability to
irreplaceable tool for the evaluation of response. The strin- discriminate between fibrosis and residual disease has
gent criteria to consider a cCR include the absence of any improved with advances in technology, placing the reso-
irregularity, mass, ulceration, or stenosis during the DRE. nance as an essential tool to confirm clinical and endoscopic
The surface has to be regular and smooth (8). findings of a cCR (46). MR may provide an accurate radio-
Endoscopic evaluation of the area harboring the origi- logical (mrTRG) estimate of the pathological tumor regres-
nal tumor is the remaining key component of clinical sion grade (TRG). The utilization of this mrTRG score may
assessment. It is important to look for any irregular- identify good and poor responders with significant impact
ity or superficial ulcers missed during DRE. A flat white in disease-free and overall survival (47,48) (Figure 30.2).
scar and telangiectasia are common endoscopic findings Even though clinical and endoscopic assessment using
among patients with a cCR (Figure 30.1). Even though flex- stringent criteria will result in high specificity rates for the
ible scopes may provide photographic documentation of detection of a pCR, a significant amount of patients with
endoscopic response, rigid proctoscopy may suffice for the incomplete clinical response will still harbor complete path-
majority of patients (8). ological response (49,50). In fact, it seems that the majority
In the presence of a cCR by DRE and proctoscopy, endo- of patients with pCR after nCRT have incomplete clinical
scopic biopsies are not recommended. Even in the setting response after 8–12 weeks from nCRT (50). Therefore, there
of incomplete clinical response, endoscopic biopsy results is a potential role for MR studies to identify patients with
should be interpreted with caution. Among patients with incomplete clinical response who may ultimately harbor
significant response, negative predictive values of these pCR. Currently, these patients would be referred to imme-
endoscopic biopsies have been reported to be consistently diate radical surgery. However, radiological tools may be
low (45). Therefore, a negative biopsy in the setting of able to accurately identify these patients and avoid poten-
incomplete clinical response does not rule out microscopic tially unnecessary surgery (51).
residual cancer. Recently, a study that compared mrTRG and residual
mucosal abnormalities following nCRT suggested that
Radiological assessment mrTRG system may identify nearly 10× more complete path-
ological responses compared to clinical endoscopic findings.
Even though historically the definition of a cCR has been
based on clinical and endoscopic findings by direct assess-
ment of rectal wall, radiological studies have always
attempted to provide additional information unavailable to
Figure 30.1 Endoscopic view of a complete clinical Figure 30.2 Magnetic resonance showing low-signal
response with whitening of the mucosa within the area of intensity area in the posterior rectal wall (arrow) consistent
the scar (original area of the primary tumor). with mgTRG1 (suggestive of a complete response).
Prediction of response to nCRT and intratumoral heterogeneity / Complete response: Watch and wait strategy 283
Figure 30.3 Diffusion-weighted magnetic resonance with high-signal intensity area (left-arrow) and its corresponding low-
signal intensity area in the ADC map (right-arrow).
These findings may improve the selection of patients with within areas of interest, and fused images of PET and CT
pCR despite initial incomplete clinical response and patients may indicate precise anatomical areas of residual cancer
who may be appropriate candidates for deferral of surgery (51). cells, even among mucinous histological subtypes (22,54).
Diffusion-weighted MR imaging may add significant Most of the available studies have focused on SUV varia-
functional information to standard MR imaging. The fact tion for the identification of complete responders to nCRT
that diffusion properties of water molecules may vary in using variable interval periods and sequential PET/CT
areas of tissue necrosis, high cellularity (frequently observed imaging (22,55,56). Accuracies, however, have been insuf-
within tumor tissues), or fibrosis, may be used to help assess ficient for its routine recommendation into clinical practice.
tumor response to nCRT (Figure 30.3). The absence of A recently reported study has suggested the role of combina-
restriction to diffusion of water molecules has been associ- tion of SUV variation and volumetric reduction in tumors to
ated with the absence of residual cancer (complete response). predict complete response to nCRT. Using individual tech-
Restriction to diffusion of water molecules (seen as high sig- nical calibration for determining metabolic tumor volumes
nal intensity in the area of the previous tumor) may indicate estimates, variation in total lesion glycolysis (determined by
the presence of residual cancer cells (incomplete response). metabolic tumor volume and mean SUV values) was found
Initial reports with diffusion-weighted MR imaging for to be the best predictor of response to nCRT using sequen-
the assessment of response to nCRT has shown promising tial PET/CT imaging at baseline and 12 weeks from nCRT
results with high accuracy rates and may constitute a useful completion (57) (Figure 30.4).
tool during assessment of response (52,53).
Positron emission tomography/computed tomography COMPLETE RESPONSE: WATCH AND
(PET/CT) imaging has been studied for the prediction of WAIT STRATEGY
response to CRT. The use of molecular imaging may pro-
vide additional information to standard structural/anatom- Watch and wait strategy: Follow-up
ical features to help distinguish between fibrosis or residual
tumor. The use of fluorodeoxyglucose allows for the estima- When a nonoperative strategy for cCR in rectal cancer is
tion of tissue metabolism (standard uptake values [SUV]) considered, a relatively intensive follow-up is certainly
Baseline 12-weeks
FDG uptake No FDG uptake
Figure 30.4 PET/CT images performed at baseline (showing FDG uptake) and after 12 weeks from nCRT completion
(showing no residual FDG uptake) consistent with a complete response to treatment.
Clinical assessment
Gross Re-staging and

Normal Abnormal tumor radical surgery
Subtle
MRI (mrTRG estimation) irregularity
MRI (mrTRG estimation)

PET/CT
mrTRG1 mrTRG2-4
mrTRG1 mrTRG2-4
PET/CT PET/CT normal abnormal
Watch and wait
Negative
Positive
Consider
Radical
reassessment in
surgery
6–8 weeks
Figure 30.5 Flow chart summarizes the Watch & Wait Strategy and assessment of tumor response following neoadjuvant
chemoradiation.
required. Patients should be encouraged to adhere to this Local recurrences after this treatment strategy are still
strict follow-up program in order to allow early recognition a concern and may develop at any time during follow-up.
of any local or systemic recurrence and, therefore, increas- The majority of local recurrences appear to develop within
ing the chance of a successful salvage treatment (Figure 30.5). the first 12 months of follow-up and may represent limita-
After initial assessment of response confirming a cCR, vis- tions in the precise identification of microscopic residual
its should be performed every 1–2 months during the first disease among “apparent” complete clinical responders. For
year, every 3 months during the second year, and every 6 these reasons, these “early recurrences” developing within
months thereafter. DRE, proctoscopy, and carcinoembry- the initial 12 months of follow-up have been called “early
onic antigen level determination are recommended for all regrowths” instead (4,63,64). Still, close and strict follow-up
visits. Timing for radiological assessment during follow-up may allow early detection of regrowths leading to identical
has not yet been standardized. Routine MR for the assess- oncological outcomes to patients with incomplete clinical
ment of the rectal wall, mesorectum, and pelvic nodes every response immediately after 8–12 weeks from CRT com-
6 months for the first 2 years and yearly thereafter has been pletion (65). In addition, local recurrences (late and early
our practice (6). regrowths) are usually amenable to salvage therapies, often
allowing sphincter preservation and being associated with
Outcomes excellent long-term local disease control (4).
Considering that the rate of complete clinical or patho-
Patients managed nonoperatively under the WW strat- logical response was historically <30% of patients across
egy after a cCR following neoadjuvant chemoradiation most of the studies, one could assume that this treatment
were originally reported to have similar long-term onco- strategy could benefit a rather limited proportion of patients
logical outcomes to patients with complete pathological with rectal cancer. However, the observation of increased
response after radical surgery (9). Additional retrospec- rates of complete response (clinical or pathological) using
tive studies reported by others have consistently shown regimens with consolidation chemotherapy and with the
similar oncological outcomes between these subgroups inclusion of earlier stages of disease (cT2N0 otherwise
of patients (31,58–63). These findings further support candidates for ultra-low resections or abdominoperineal
the idea that patients with a cCR may be spared from the resections) may result in nearly 50% that ultimately avoid
surgical morbidity and mortality of radical surgery with surgical resection (25,42). This has been further confirmed
no oncological compromise (5). In addition, functional in a prospective trial including patients with T2 and T3 rec-
outcomes of patients managed nonoperatively not only tal cancers managed by CRT and an additional endorectal
appear to be better than radical surgery but also better high-dose brachytherapy boost (total 65 Gy) that showed a
than other organ-preserving strategies (transanal local 58% cCR rate at 2 years of follow-up without surgical resec-
excision) (3,60). tion (31).
Local excision / Local excision or watch and wait for cCR? 285
Finally, in the era of evidence-based medicine, a ran- wall (ycT0-2N0). However, we would restrict this to patients
domized prospective trial is still lacking to definitively dem- with low residual lesions that would otherwise require an
onstrate the oncological equivalence of WW and radical abdominal-perineal excision or a coloanal intersphincteric
surgery in the setting of a cCR following nCRT (66). Even resection as a definitive procedure (64). Appropriate path-
though such a trial is not likely to be performed, a recent ological information regarding ypT classification, tumor
study using a propensity-score matched cohort analysis regression grade, lymphovascular/perineural invasion, and
comparing WW and radical surgery has been designed to resection margins may allow for a final decision regarding
demonstrate noninferiority of the WW approach. Curiously, the need for total mesorectal excision and protectomy.
however, the comparison between groups demonstrated a It has been our policy to offer strict follow-up to patients
slight superiority of the nonoperative management of these with a final pathological specimen showing ypT0 after this
patients in terms of survival and a clear benefit in colos- “diagnostic” transanal local excision. This is due to the fact
tomy-free survival even when accounting for the develop- that the risk of lymph node metastases among these patients
ment of local recurrences (63). has been shown to be very low in the setting of neoadju-
vant CRT and long (≥8 weeks) intervals. This is already true
for unselected patients with ypT0, where the risk of nodal
metastases are well under 10%, and in most cases less than
ADJUVANT TREATMENT 5% (73,75,76). However, with significant improvements in
radiological imaging, particularly with high-resolution MR
The use of adjuvant systemic chemotherapy following a with the use of a diffusion-weighted series and other lym-
cCR managed nonoperatively is still a matter of contro- photropic agents, the selection of patients with ycT0N0 is
versy. Most studies have not offered adjuvant chemotherapy expected to further improve (68).
to these patients even though several guidelines may rec- There are at least three main drawbacks to this treatment
ommend the use of adjuvant therapy based on pretreat- strategy. First, healing of the rectal defects determined by
ment staging features rather than on response to nCRT. local excision after neoadjuvant CRT is quite challeng-
This means that a baseline cT3N1 would require adjuvant ing and painful, particularly those closer to the anal verge
therapy, whereas a baseline cT2N0 would not, even though (69,77). Healing problems are much more frequent, and
both patients develop cCR. However logical this may seem, it may take as long as 8 weeks to completely heal. Even
there are insufficient data to support either strategy (based though severe complications are not frequent, pain may
on pretreatment or posttreatment status). be quite significant requiring readmission to the hospital.
In a pooled analysis of patients undergoing nCRT fol- The second drawback is that sphincter preservation may
lowed by radical surgery, patients with pCR showed an 11% be compromised after performance of full-thickness local
distant metastases rate (67). Curiously, nearly 40% of these excision in this setting. A few studies have addressed this
patients had received adjuvant 5FU-based chemotherapy. issue and reported that patients requiring radical resection
This compares to a 14% distant metastases rate among after full-thickness local excision frequently ended up with
patients undergoing standard CRT with cCR managed non- an abdominal perineal resection, even though they origi-
operatively without the use of adjuvant chemotherapy (4). nally were considered candidates for a sphincter-preserving
Finally, with the use of consolidation CRT regimens, the procedure (70,78). Finally, completion of TME in this set-
dose of adjuvant systemic therapy may ultimately have been ting may provide an imperfect mesorectal specimen for the
shifted to the neoadjuvant period, rendering the discussion pathologist. A recent review of patients undergoing comple-
of adjuvant chemotherapy meaningless. However, there are tion TME indicated that previous transanal endoscopic
still insufficient data to fully support this. microsurgery (TEM) was a risk factor for poor quality of
the TME specimen (71). All of these issues should be kept in
mind when offering patients “diagnostic” or “therapeutic”
local excision after partial response.
LOCAL EXCISION
LOCAL EXCISION OR WATCH AND WAIT
LOCAL EXCISION AFTER NCRT AS A FOR cCR?
DIAGNOSTIC APPROACH
As mentioned previously, healing of local excision defects
In the presence of an incomplete clinical response and in following neoadjuvant CRT is not as simple as after local
patients who are resistant to radical resection or who are excision alone. The rates of wound dehiscence may be sig-
medically unfit, the least we would offer is a full-thickness nificant (69,72,73). In this setting, not only is pain an issue,
excisional biopsy, preferably with the use of transanal endo- but also significant scarring following delayed healing may
scopic microsurgery. This “excisional biopsy,” otherwise develop, which will make patient follow-up even more dif-
referred to as a local excision, is primarily considered as a ficult. Even though ypT0 may be associated with a lower
diagnostic procedure and may be appropriate for patients risk of local failures, the risk is not zero, and the patient still
with small (≤3 cm) lesions that are restricted to the bowel requires appropriate follow-up. Distinction between local
recurrence in a rectal wall following wound dehiscence 11. Brettingham-Moore KH et al. Clin Cancer Res. 2011;
after a local excision with or without rectal stenosis may be 17(9):3039–47.
quite challenging. Therefore, we believe that follow-up is 12. Watanabe T et al. Cancer Res. 2006;66(7):3370–4.
considerably facilitated by preservation of rectal wall integ- 13. Rimkus C et al. Clin Gastroenterol Hepatol. 2008;6(1):
rity with the WW approach allowing for earlier detection of 53–61.
possible recurrences. 14. Hardiman KM et al. Lab Invest. 2016;96(1):4 –15.
Finally, considering that the ultimate goal of organ pres- 15. Bettoni F et al. Ann Surg. 2017;265(1):e4–6.
ervation is also function preservation, one has to compare 16. Perez RO et al. Dis Colon Rectum. 2016;59(9):895–7.
functional outcomes between local excision and WW in the 17. Sauer R et al. N Engl J Med. 2004;351:1731–40.
setting of neoadjuvant CRT. When patients with cCR man- 18. Sauer R et al. J Clin Oncol. 2012;30:1926–33.
aged by WW were compared to patients with “near-com- 19. Taylor FG et al. Ann Surg. 2011;253:711–9.
plete” response following nCRT by Wexner incontinence 20. Peeters KCMJ et al. J Clin Oncol. 2005;23(25):
scores, manometric findings, and quality of life question- 6199–206.
naires, functional outcomes were significantly better among 21. Loos M et al. Ann Surg Oncol. 2013;20:1816–28.
patients managed nonoperatively, suggesting that local exci- 22. Perez RO et al. Tech Coloproctol 2014;18:699–708.
sion may also risk function preservation despite providing 23. Garcia-Aguilar J et al. Ann Surg Oncol. 2011;19(2):
organ preservation to these patients (3). 384–91.
24. Habr-Gama A et al. Dis Colon Rectum. 2017;60:
586–94.
25. Habr-Gama A et al. Dis Colon Rectum. 2013;56:
CONCLUSIONS 264–6.
26. Bosset J-F et al. J Clin Oncol. 2005;23(24):5620–7.
Organ preservation in the management of rectal cancer 27. Radu C et al. Radiother Oncol. 2008;87(3):343–9.
has become a valid option for select patients after signifi- 28. Wiltshire KL et al. Radiat Oncol Biol. 2006;64(3):
cant response to neoadjuvant CRT. Patients who develop 709–16.
complete tumor regression with no clinical, endoscopic, or 29. Jakobsen A et al. Int J Radiat Oncol Biol Phys. 2012;
radiological evidence of residual cancer may be offered no 84(4):949–54.
immediate surgery and enrolled in a strict surveillance pro- 30. Gerard J-P et al. Acta Oncol. 2015;54(4):545–51.
gram (WW) with excellent functional and acceptable onco- 31. Appelt AL et al. Lancet Oncol. 2015;16(8):919–27.
logical outcomes. Good responders to nCRT (ypT0 or ypTis) 32. Vuong T et al. Clin Oncol (R Coll Radiol). 2007;19(9):
despite incomplete clinical response may warrant local exci- 701–5.
sion as a diagnostic and therapeutic tool, also with good 33. Schrag D et al. J Clin Oncol. 2014;32(6):513–8.
oncological outcomes but at the cost of slightly worse func- 34. Habr-Gama A et al. Colorectal Cancer. 2015;4(1):1–4.
tional outcomes and significant postoperative morbidity. 35. Habr-Gama A et al. Dis Colon Rectum. 2009;52(12):
1927–34.
REFERENCES 36. Patel UB et al. Ann Surg Oncol. 2012;19:2842–52.
37. Francois Y et al. J Clin Oncol. 1999;17(8):2396.
1. Kosinski L et al. CA Cancer J Clin. 2012;62(3): 38. Tulchinsky H et al. Ann Surg Oncol. 2008;15(10):
173–202. 2661–7.
2. Sanghera P et al. Clin Oncol (R Coll Radiol). 2008; 39. Kalady MF et al. Ann Surg. 2009;250:582–9.
20(2):176–83. 40. Evans J et al. Dis Colon Rectum. 2011;54(10):1251–9.
3. Habr-Gama A et al. Dis Colon Rectum. 2016;59(4): 41. Wolthuis AM et al. Ann Surg Oncol. 2012;19(9):
264–9. 2833–41.
4. Habr-Gama A et al. Int J Radiat Oncol Biol Phys. 42. Garcia-Aguilar J et al. Lancet Oncol. 2015;16(8):
2014;88(4):822–8. 957–66.
5. Smith FM et al. Dis Colon Rectum. 2015;58(2): 43. Lefevre JH et al. J Clin Oncol. 2016;34(31):3773–80.
159–71. 44. Perez RO et al. Int J Radiat Oncol Biol Phys. 2012;
6. Habr-Gama A et al. Hematol Oncol Clin North Am. 84(5):1159–65.
2015;29(1):135–51. 45. Perez RO et al. Colorect Dis. 2012;14(6):714–20.
7. Smith FM et al. Br J Surg. 2010;97(12):1752–64. 46. Lambregts DMJ et al. Dis Colon Rectum. 2011;
8. Habr-Gama A et al. Dis Colon Rectum. 2010;53(12): 54(12):1521–8.
1692–8. 47. Patel UB et al. J Clin Oncol. 2011;29(28):3753–60.
9. Habr-Gama A et al. Ann Surg. 2004;240:711–7; 48. Patel UB et al. Ann Surg Oncol. 2012;19(9):2842–52.
discussion 7–8. 49. Nahas SC et al. Dis Colon Rectum. 2016;59(4):
10. Lopes-Ramos C et al. Cancer Genet. 2015;208(6): 255–63.
319–26. 50. Smith FM et al. Dis Colon Rectum. 2014;57(3):311–5.
References 287
51. Bhoday J et al. Dis Colon Rectum. 2016;59(10): 62. Smith RK et al. Int J Colorectal Dis. 2015;30(6):769–74.
925–33. 63. Renehan AG et al. Lancet Oncol. 2016;17(2):174–83.
52. Lambregts DMJ et al. Ann Surg Oncol. 2011;18(8): 64. Perez RO et al. Dis Colon Rectum. 2013;56:6–13.
2224–31. 65. Habr-Gama A et al. Rad Oncol Biol Phys. 2008;71(4):
53. Curvo-Semedo L et al. Radiology. 2011;260(3):734–43. 1181–8.
54. Anjos Dos DA et al. Ann Nucl Med. 2016;30(8): 66. Perez RO. Lancet Oncol. 2016;17(2):125–6.
513–7. 67. Maas M et al. Lancet Oncol. 2010;11(9):835–44.
55. Cascini GL et al. J Nucl Med. 2006;47(8):1241–8. 68. Bach SP et al. Br J Surg. 2009;96(3):280–90.
56. Kristiansen C et al. Dis Colon Rectum. 2008;51(1): 69. Marks JH et al. Surg Endosc. 2009;23:1081–7.
21–5. 70. Morino M et al. Surg Endosc. 2013;27:3315–21.
57. Anjos Dos DA et al. Dis Colon Rectum. 2016;59(9): 71. Hompes R et al. Colorectal Dis. 2013;15:e576–81.
805–12. 72. Perez RO et al. Dis Colon Rectum. 2013;56(1):6–13.
58. Vaccaro CA et al. Cirugía Española. 2016;94(5): 73. Smith FM et al. Dis Colon Rectum. 2017;60(2):228–39.
274–9. 74. Garcia-Aguilar J et al. Ann Surg. 2011;254:97–102.
59. Araujo ROC et al. Eur J Surg Oncol. 2015;41(11): 75. Park IJ et al. Dis Colon Rectum. 2013;56:135–41.
1456–63. 76. Mignanelli ED et al. Dis Colon Rectum. 2010;53:251–6.
60. Maas M et al. J Clin Oncol. 2011;29(35):4633–40. 77. Perez RO et al. Dis Colon Rectum. 2011;54:545–51.
61. Dalton RSJ et al. Colorect Dis. 2012;14(5):567–71. 78. Bujko K et al. Radiother Oncol. 2013;106:198–205.
31
Indications and outcomes for treatment of
recurrent rectal cancer and colorectal liver/lung
metastases
LUANNE M. FORCE AND DAVID J. MARON
CHALLENGING CASE showed an invasive rectal adenocarcinoma into the

vagina and distal sacrum, with margins completely
An 85-year-old female patient presents a previ- free of tumor and no positive lymphadenopathy. The
ous history of T3N1M0 rectal cancer 1 cm from the postoperative course was uneventful. Subsequent
dentate line invasive into the internal and external surveillance PET scan and CEA levels did not show
anal sphincters. She underwent neoadjuvant chemo- any further recurrence.
radiation therapy followed by an abdominoperineal
resection two years prior to presentation, followed
by adjuvant chemotherapy. On surveillance imag-
RECURRENT RECTAL CANCER
ing she was found to have a pelvic mass with a PET
avid lesion in the presacral area, consistent with local
recurrence. MRI of the pelvis demonstrates a pelvic The management of rectal cancer has changed dramatically
recurrence invasive into the posterior wall of the over the past 40 years, resulting in a significant reduction in
vagina and possibly the uterus, as well as the lower the incidence of locally recurrent rectal cancer. Local recur-
coccygeal segment. Metastatic work-up demon- rence rates historically were 20%–40%, but with the use of
strates no distant metastatic disease. improved adjuvant, neoadjuvant, and surgical strategies,
the local recurrence rate has been reduced to 4%–8% (1).
The majority of patients with recurrent rectal cancer will
present within 3 years; 70% will present within 2 years and
CASE MANAGEMENT 85% within 3 years (1). The management of locally recur-
rent rectal cancer is complex, and management needs to be
The patient’s case was discussed at the multidisci- tailored to the individual patient.
plinary tumor board. After reviewing the radiology The risk of local recurrence increases with margin posi-
images, the margins of resection along the pelvic tivity, including distal and radial margins. A margin of
sidewall were deemed to be free from tumor. As she 5 mm has been correlated with a reduced local recurrence
has had previous radiation therapy, she was not a can- rate; however, no additional benefit for margins greater
didate for additional radiation treatment. The patient than 5 mm has been reported (2). Pathologic character-
was brought to the operating room for a complete istics of the tumor, specifically lymphovascular invasion
pelvic exenteration, with en bloc resection of the pel- and poor differentiation, also increase the risk of local
vic mass, uterus, posterior vagina and distal sacrum recurrence (3). Surgeon experience and caseload have
in combination with gynecology and neurosurgery. also been associated with rectal cancer outcomes; higher-
Once the pelvic mass was removed, the pelvis was volume surgeons (>10–12 cases per year) have better local
reconstructed with a vertical rectus abdominis myo- recurrence rates when compared to surgeons who perform
cutaneous (VRAM) flap by plastic surgery. Pathology these operations less often (4). Abdominoperineal resec-
tion, compared to low anterior resection, has also been
288
Initial evaluation 289
shown to have a higher recurrence rate, likely related to a will also characterize the recurrence (10,11). CT scan and
higher rate of circumferential margin positivity (5). There magnetic resonance imaging (MRI) may provide anatomic
is some evidence that colon perforation during resection information regarding local recurrence; however, these
for rectal cancer also significantly increases the risk for methods are not able to identify scar versus a new recur-
local recurrence (6). rence. PET scan may help differentiate these two processes,
with an accuracy of 87% in previous studies (12). PET scan
when combined with CT of the chest, abdomen, and pel-
vis have a combined sensitivity and specificity of 100% and
INITIAL EVALUATION 96%, respectively (9).
Local staging should also be performed with preopera-
Most patients with local recurrence will present within tive imaging. This step is essential to delineate the patient’s
3 years after treatment rectal cancer. About 25% of these anatomy, involvement of surrounding structures, and
patients will be asymptomatic, with the recurrence detected potential resectability. CT scan of the abdomen and pelvis
by surveillance imaging or carcinoembryonic antigen is inferior for local staging when compared to MRI (12). CT
(CEA) levels (7). Patients may present with bleeding, likely scan does not accurately define invasion into surrounding
related to an intraluminal recurrence at the anastomosis. structures likely due to the inability to differentiate scar
Recurrence may also manifest itself as a change of bowel from tumor (13). MRI is more accurate when determining
habits or complete bowel obstruction. Pain as a presenting local invasion, with a specificity of 70%–100%. Signal inten-
symptom is more worrisome and may indicate invasion or sity on MRI provides a more detailed image, allowing better
compression of surrounding structures such as nerves or differentiation between normal tissue, scar, and tumor on
bone (1). Significant pain on presentation has been corre- T1- and T2-weighted images. Recurrent tumor will typi-
lated with worse survival outcomes when compared to those cally have a higher signal intensity on T2-weighted images
patients who do not have pain as a presenting symptom (8). (14). Invasion into surrounding structures can be reliably
The first step in evaluation of a patient with recurrent detected with reasonable sensitivity and specificity for most
rectal cancer is a thorough review of the patient’s medical surrounding structures. An exception is invasion into the
record and treatment history. Operative reports, treatment pelvic sidewall, which has a higher false-positive rate (14).
history, and comorbidities are all essential to plan further Accuracy has been shown to improve when these images are
intervention. Physical examination may reveal lymphade- enhanced with gadolinium (15).
nopathy or hepatomegaly. Rectal exam, when possible, is Confirmation of the recurrence should be performed
also important to characterize the extent and fixity of the with tissue biopsy whenever possible (1,13,12). Prior to
tumor. It has been reported that tumor fixity in more than resection, the presence of a recurrence should be differenti-
one area is correlated with significantly lower survival rates ated from other processes such as fibrosis. If the recurrence
(8). Vaginal exam should be performed in women to deter- is intraluminal, tissue biopsy may be performed by endos-
mine if the vaginal wall is involved. If the patient is too copy. If there is an extraluminal recurrence, CT-guided
uncomfortable to permit an exam in the office, the patient biopsy of the mass should be performed prior to undergoing
should undergo an examination under anesthesia before resection (1). In some cases, tissue biopsy may be inconclu-
any intervention is planned (1). Laboratory values should be sive or not possible. In these situations, PET/CT in combi-
obtained, including a CEA level. A high CEA level at time nation with CEA levels and MRI can accurately detect local
of diagnosis may indicate distant metastatic disease and a recurrence with a sensitivity of 84%–100% and specificity of
poorer prognosis (9). Other laboratory values, including 80%–100% (11).
hemoglobin, nutritional parameters, and kidney function Evaluation of each patient with recurrent rectal can-
should also be investigated prior to any planned interven- cer should include a multidisciplinary team review (12).
tion (1). An elevated creatinine level may suggest ureteral Evaluation by a team of specialists including colorectal sur-
compression or involvement by the tumor and should be geons, oncologists, radiation oncologists, radiologists, and
evaluated preoperatively. Full endoscopic evaluation of pathologists has been shown to provide superior outcomes
the colon should be performed to detect any intraluminal in the management of these highly complex patients. A
recurrence, obtain tissue biopsy, and assess for other colonic recent study by Kontovousnisios et al. showed that the use
lesions prior to intervention. of a multidisciplinary team can improve the outcomes of
Imaging studies are essential in the evaluation and man- patients treated with recurrent rectal cancer, with the ability
agement of recurrent rectal cancer. It is imperative to per- to achieve an R0 resection in 90% of patients (16).
form a metastatic workup, as up to 50% of these patients will Recurrences should be classified based on their location.
present with systemic metastasis (2). Computed tomogra- Axial recurrences are those centrally located, which do not
phy (CT) scan of the chest, abdomen, and pelvis should be involve the anterior, posterior, or lateral walls of the pelvis;
performed to diagnose systemic metastasis, as the presence this includes anastomotic recurrences. Anterior recurrences
of distant metastasis may change management. A positron invade the bladder, prostate, seminal vesicles in males, and
emission tomography (PET) scan should also be performed, the vagina in female patients. Posterior recurrences invade
which will give information about distant metastasis and the sacrum or coccyx. Lateral recurrences involve the lateral
290 Indications and outcomes for treatment of recurrent rectal cancer and colorectal liver/lung metastases
sidewalls of the pelvis, which may involve structures such as locations, and circumferential involvement are all indica-
the ureters, iliac vessels, and pelvic nerves. Classifying the tions that a complete resection cannot be achieved without
tumor based on its anatomic location will help plan opera- causing significant morbidity to the patient. There are several
tive intervention, as well as determine resectability (17). reports of concurrent vascular resection and reconstruction
for rectal cancer, which have been performed in specialized
centers with reasonable results (22). Unresectable distant
metastasis such as lung and liver metastases are also a rela-
MANAGEMENT tive contraindication to curative resection.
The location of the recurrence will determine the opera-
Management of recurrent rectal cancer is best achieved by a tive approach and strategy. These operations often require
multimodal approach. Combined radiotherapy, chemother- multiple specialties including urology, gynecology, and
apy, and surgery offers the best chance to achieve a cura- plastic surgery. Cystoscopy with bilateral ureteral stents
tive resection. In addition, multiple surgical specialties are is often beneficial in assessing bladder involvement and to
required to manage these complex patients. The ultimate facilitate intraoperative identification of the ureters. All
goal is to achieve an R0 surgical resection, which will pro- operations should begin with exploration of the abdomen
vide the patient the best chance for cure. Many factors must and determination of the extent of disease or peritoneal
be taken into account, including the functional status of the carcinomatosis if present. Laparoscopy in this situation can
patient, the feasibility of resection, as well as expected qual- be helpful in avoiding a laparotomy if resection is aborted
ity of life with or without the resection. due to the presence of carcinomatosis. Identification of
Neoadjuvant treatment should be initiated in suit- major structures including vessels and ureters should also
able patients who are being considered for resection. be carried out. Resection of the bladder may be performed
Chemotherapy and radiation may downstage the tumor, en bloc with reconstruction by an experienced urologist.
increasing the likelihood for an R0 resection. If the patient The survival rate after en bloc resection of recurrent cancer
has not undergone previous radiation treatment, pel- and bladder is 61% with negative margins, compared with
vic radiation for a total of 50 Gy should be administered 17% of patients with positive margins (11). In women, inva-
in conjunction with chemotherapy. Surgery can then be sion of the uterus or vaginal structures must be assessed.
performed 6–8 weeks following completion of the radio- Hysterectomy should be performed if there is invasion into
therapy. If the patient previously received radiation to the the uterus (23). Low recurrences in women may necessitate
pelvis, there is evidence that an additional dose of radiation partial or complete vaginectomy. In cases where <50% of
could be beneficial (18). Acute toxicity has been shown to the vagina is resected, the vagina can be closed primarily
be relatively low, up to 7% in some series. The incidence of (23). If a more extensive resection is performed, flap closure
delayed toxicity could be up to 17% (8,18). Reirradiation is by a plastic surgeon may be required. One report showed
associated with a greater chance at achieving an R0 resec- that anterior exenteration along with en bloc resection of
tion, which will impact local control as well as disease-free the pubic bone may also be feasible if an R0 resection is
survival at 42%–60% at 5 years (18). There is limited data achieved (24).
regarding timing and dosing of additional radiation treat- Posterior resections often involve sacrectomy if the
ments, so treatment should be tailored to each individual lesion is located below S2. Identification and protection of
patient. the ureters and major vessels is important when perform-
Surgical intervention is the cornerstone of treatment for ing a sacrectomy. Depending on the level of resection, the
recurrent rectal cancer. The ability to achieve an R0 resec- approach may be a combined abdominal and perineal
tion will confer the best survival benefit to the patient (19). approach. Bilateral internal iliac arteries may be ligated
Planning is key, and imaging should be reviewed to deter- and resected to decrease blood loss during these proce-
mine resectability and the involvement of adjacent struc- dures. Involvement of the ureter will require resection
tures. Care must be taken to avoid R2 resections, which and reimplantation or reconstruction by an experienced
confer no survival benefit to the patient and have a high urologist. These resections will often leave a large perineal
complication rate (19). Anterior recurrences are more ame- defect requiring muscle flap closure by plastic surgery. Once
nable to curative resection when compared to posterior and involvement of the sacrum is identified in the operating
lateral recurrences due to lack of involved bone and vascu- room, an examination of the nerve roots should be per-
lar structures. The ability to achieve a complete resection formed to determine resectability. The ability to achieve R0
in anterior recurrences ranges from 70% to 90%. This is resection at a level above S2 is highly unlikely (23), but there
compared to 6%–36% when the pelvic sidewall is involved have been several small series at specialized centers that
(20). Resection is contraindicated in patients who are medi- demonstrate that it may be feasible (25). If a high sacrectomy
cally unfit to undergo a major operation. Sacral invasion is performed, sacral bony stabilization may be required
above S2 is also a contraindication due to the involvement (25). Once the tumor is completely mobilized and all adja-
of neurovascular structures (21). Resection at this level is cent surrounding organs resected en bloc, the patient will
often incomplete and has significant morbidity. Similarly, need to be placed in the prone position so sacrectomy can
encasement of the external iliac arteries, fixity in multiple be completed (23), as this allows better exposure and a wider
Outcomes 291
excision than what can be done in lithotomy. The colostomy but a large defect exists in the pelvic floor. The mesh may be
may be matured, as well as harvesting of muscle flaps prior placed circumferentially in the pelvis, with an omental flap
to turning the patient prone. Once prone, the perineal dis- if available, to prevent a perineal hernia (31).
section is begun, and the sacrotuberous and sacrospinous Complication rates for recurrent rectal cancer resection
ligaments are taken down. The level of resection may be are high, up to 80% in some series (26). The 30-day mortal-
confirmed by fluoroscopy. Care must be taken to ensure that ity rate ranges among reported series. A recent systematic
the dural sac is closed, usually done in conjunction with an review showed a wide range of mortality and complica-
orthopedic or neurosurgeon (9,23). As a large tissue defect tion rates, 0%–25% and 37%–100%, respectively (32). The
may be left behind, the space may be filled with a pedicled median 30-day survival rate was 2.2% (32). Perineal wound
omental flap, biologic mesh, or muscle flap performed by a complications are the most frequent postoperative com-
plastic surgeon (23). The complication rate for resections plication among this population. These include perineal
with combined sacrectomy is high, 82% in some series (26). wound breakdown, dehiscence, and flap necrosis or fail-
Lateral recurrences have the lowest chance of achieving ure. Wound infection is relatively common. Pelvic accesses
an R0 resection (23). Tumors that invade the lateral walls and fluid collections may also occur, which can often be
of the pelvis have a high incidence of invading structures treated with CT-guided drainage. Concomitant sacrectomy
such as the ureters and iliac vessels. More recently, en bloc increases the complication rate significantly (26). Urologic
resection of one or more iliac vessels has been shown to have complications are also common, including ureteric injury
an R0 resection rate of 40%–53% in experienced centers or stenosis. This is often best treated with ureteral stents.
(22,27). These studies are based on a relatively small sub- Bladder dysfunction is also common.
set of patients, and care must therefore be taken to choose
appropriate patients prior to undertaking resection.
Intraoperative radiotherapy may be beneficial for the
treatment of recurrent rectal cancer. Multiple series have PALLIATIVE TREATMENT
demonstrated a survival benefit, as well as locoregional con-
trol at specialized centers (28). Directed radiation therapy Palliative operations may be performed for symptomatic
minimizes the radiation effect to the surrounding structures patients who are not candidates for curative resection.
and provides concentrated radiation directly to the area of The goal of palliative treatment is to alleviate some of the
treatment. The largest benefit has been shown in patients patient’s symptoms related to the recurrent cancer. Most
who have close margins, <5 mm (1,29). Intraoperative commonly, this is pain related to invasion of bone or nerves,
frozen section is performed to determine the margin of bowel obstruction, or bleeding. Palliative radiation may help
resection. Intraoperative doses of 1000–2000 cGy can be alleviate pain and bleeding (33,34). Intraluminal metallic
administered, depending on the amount of residual disease. stents can be used to alleviate obstruction (33). These can be
Radiation can be administered via external-beam or high- used when the obstruction is at least 5 cm above the dentate
dose brachytherapy, which is dependent on the location of line (35), as patients may experience significant complica-
the treatment area and positioning of the beam. In some tions with stent placement below that level, including pain,
cases, intraoperative radiotherapy cannot be performed tenesmus, and migration (1,36). In cases where stent place-
due to logistical reasons. The brachytherapy applicator may ment is not possible or has failed, diverting colostomy may
conform better to the patient’s anatomy (1). be performed. Ureteral stents may be placed if there is sig-
Perineal reconstruction is often required after cura- nificant ureteral obstruction. Pelvic exenteration for pallia-
tive resection for recurrent rectal cancer. The incidence of tion should be avoided, as symptoms are rarely improved by
wound complications is high, up to 40% of patients. Reasons this technique (37). There is some evidence that an R1 resec-
for perineal wound failure are multiple, including large tion may provide some benefit to quality of life, but there
defects, irradiated tissue, and patient comorbidities. A large was deleterious effect on quality of life if an R2 resection is
defect that cannot be closed primarily will require a muscle performed (38).
flap closure. The most common flap performed is a vertical
rectus abdominus myocutaneous (VRAM) flap (23). This
flap will provide a bulky flap to fill in large defects, such
as after a vaginectomy. The location of any stomas must be OUTCOMES
considered when using a VRAM flap, especially in patients
who will require both a colostomy and a urostomy (30). Survival rates for treatment of recurrent rectal cancer vary
Other flaps include gracilis and gluteus flaps, which may between 25% and 36% at 5 years for patients undergoing
be used to fill in smaller defects. In patients with bilateral multimodal treatment (9). Ability to achieve negative mar-
stomas and large tissue defects, a free flap may be required; gins will improve survival significantly, with reported sur-
the most common free flap is a latissimus dorsi flap (9). vival rates of 37%–60% at 5 years (5,8,19,39). If macroscopic
Another attractive option for perineal reconstruction is the disease is left behind, survival rates drop to 10%–16% (5,8).
placement of a biologic mesh (1,23). This is most feasible Neoadjuvant chemotherapy and radiation treatment has
when there is sufficient tissue to close the perineal defect also been shown to confer a survival benefit in these patients.
A large series from the Mayo Clinic demonstrated an R0 A French multi-institutional randomized trial, however,
resection was able to be achieved in 45% of patients and failed to show any benefit for intraoperative radiotherapy
that survival was significantly decreased for patients with compared to radical resection (44).
residual macroscopic disease (8). Nielsen et al. reported out- Recurrent rectal cancer is a complex clinical problem.
comes for 213 patients, with a R0 resection rate of 61%. The Extensive workup evaluating for distant metastasis and
authors demonstrated a 5-year survival rate of 40% for R0 local invasion is extremely important when planning surgi-
resections, and survival was significantly lower for patients cal intervention. Multimodal treatment with chemotherapy,
with previous abdominoperineal resections (40). A meta- radiation, and surgery, possibly combined with intraopera-
analysis of 22 studies also confirmed a significant survival tive radiation therapy, is the cornerstone of treatment for
advantage for those patients who were able to undergo R0 recurrent rectal cancer. A multidisciplinary treatment team
resections (41). Overall, R0 resection was obtained in 53% is essential to evaluate and plan the treatment of these com-
of patients, which correlated with a survival benefit of an plex patients. Surgery is the mainstay of curative resection,
additional 37 months compared to patients who underwent with significantly better outcomes if an R0 resection can be
an R2 resection (41). Quality of life for patients undergoing achieved. When possible, all invaded organs and structures
resections for recurrent rectal cancer is an important con- should be resected en bloc, and collaboration with multi-
sideration. One study noted a significant decrease in quality ple surgical subspecialists is often required. If an R0 or R1
of life for patients undergoing R1/2 resections when com- resection cannot be performed, palliative options may be
pared to R0 resections (42). undertaken. Palliative chemotherapy, radiation, and endo-
Extensive resection requiring major vascular resection luminal stents may help alleviate some symptoms.
for recurrent rectal cancer has been reported in some small
series. En bloc iliac resection with reconstruction has been
shown to be feasible and safe in specialized centers. One
study by Brown et al. reported a morbidity rate of 52% with COLORECTAL LIVER METASTASIS
a median survival of 26–24 months (27). Another study by
Abdelsattar et al. also demonstrated that an R0 resection Despite improved screening programs, approximately 20%
can be achieved with en bloc resection of tumors involving of all patients presenting with a new diagnosis of colorec-
the aortoiliac axis (22). The authors demonstrated a survival tal cancer will have synchronous liver metastasis (45). The
rate of 45% at 4 years. These small series show that vascular survival of these patients is dependent on the ability to treat
resection and reconstruction can achieve an R0 resection the primary as well as metastatic lesions. Several treatment
with acceptable results in highly selected patients. options exist to manage the outcomes of these patients. The
Outcomes after sacral resection have also been reported. conventional approach involved resection of the primary
Milne et al. reported outcomes for sacrectomy for recurrent tumor followed by adjuvant chemotherapy and resection
rectal cancer, with a R0 resection rate of 74% (26). Patients or ablation of liver lesions. A second option is simultaneous
with an R0 resection were shown to have a survival rate of 45 resection of both the primary and the metastatic lesions.
months, versus 19 months for R1 and 8 months for R2 resec- The third algorithm involves a liver first approach, where
tions. The authors reported a complication rate of 80% (26). the liver metastases are resected and the primary tumor is
Another series of 30 patients undergoing sacral resections left until after adjuvant chemotherapy is completed. This
for recurrent rectal cancer demonstrated a R0 resection rate option may involve neoadjuvant chemotherapy or resection
of 93%, with a 5-year survival rate of 46% for R0 resections followed by chemotherapy (46). In addition, there are several
(43). Another small series of nine patients assessed survival other techniques employed to ablate liver metastases that
for patients undergoing high sacrectomy. Patients had an are not amenable to surgical resection (47). Unfortunately,
average survival of 31 months, with all deaths occurring very few patients meet the criteria for curative resection,
secondary to metastatic disease (25). about 20% of those presenting with liver metastases (48,49).
Intraoperative radiation treatment (IORT) has also been
shown to be beneficial, with an increase in survival of 15%
in some series (23). The survival rate at 5 years has been
reported as 30%–70% for patients treated with IORT with LIVER RESECTION
an R0 resection. This benefit decreases significantly for R1
or R2 resections, with a survival rate at 5 years of 7%–20%. PRIMARY FIRST APPROACH
However, these survival rates are improved over reported
rates in patients who underwent R2 resections without The traditional, primary first, approach involves resection
IORT. Most authors debate the utility of intraoperative of the primary tumor followed by chemotherapy and man-
radiotherapy for resection margins >5 mm, advocating for agement of the metastatic disease (50). Advocates of the
treatment when microscopic margins are close intraopera- primary first strategy seek to avoid complications related
tively (1,29). A recent meta-analysis performed showed that to the primary tumor during adjuvant chemotherapy.
there is significant heterogeneity of studies for IORT, but Complications such as obstruction or perforation often
that overall there is a survival benefit for these patients (28). cause increased morbidity in patients undergoing active
Liver resection / Liver ablation 293
chemotherapy regimens. Another purported advantage of done safely while the patient undergoes chemotherapy (51).
this approach is to rid the patient of the primary tumor, There is some debate in the literature as to whether neoadju-
likely a source for subsequent metastases (46,45). The prob- vant chemotherapy is required prior to resection of the liver
lem with this strategy is that complications related to the metastasis. If the lesion is resectable at initial presentation,
resection of the primary tumor often delay initiation of che- there are several studies that show neoadjuvant chemother-
motherapy in these patients, decreasing overall survival (51). apy may not have a survival benefit for the patient (64,65).
Furthermore, several studies have shown that primary che- If the tumor is unresectable at initial presentation, neoadju-
motherapy for patients with unresectable metastatic disease vant chemotherapy followed by restaging and liver resection
can have direct benefits on the primary tumor (decreased is appropriate (64). A recent systematic review of published
bleeding or obstruction) without surgical intervention (52). studies showed that the liver first approach was feasible and
safe, and had an overall median survival of 40 months (66).
SIMULTANEOUS APPROACH An international consensus group in 2012 noted that the
liver first strategy is as good as other methods for the treat-
The simultaneous approach aims at resecting the primary ment of synchronous colorectal liver metastases (67). The
tumor as well as the liver metastases at the same opera- group advocated for preoperative chemotherapy for as short
tion. Justification for this strategy is to perform the curative a duration as possible, followed by liver resection in select
resection in one operation, followed by adjuvant chemother- cases. In general, the liver first approach is good for patients
apy (53–55). Outcomes for this approach have been variable with large metastatic disease burden in the liver and a rela-
in the literature. Several studies have shown that there may tively asymptomatic primary tumor. The goal is to initiate
be an increased morbidity rate from performing simultane- systemic treatment as soon as possible and achieve an R0
ous bowel and liver resections. Most notably, increases in resection (46,67).
infectious complications and anastomotic leaks have been Overall, multiple systematic reviews have been per-
reported (50,53,56). A study by Broquet et al. demonstrated formed comparing the traditional, simultaneous, and liver
a similar complication and survival rate for simultaneous, first approaches. These reviews have all failed to demonstrate
combined, and reverse liver strategies. All of the strategies that one method is superior over the other in terms of over-
employed had a mortality rate of 3%–5%, with a morbidity all survival for the patient (46,53,54,57,59), and therefore,
rate of 30%–50%. The authors demonstrated a 5-year sur- treatment of synchronous liver metastasis must be tailored
vival rate of 39%–55% for all three strategies (50). A recent to the individual patient. A multidisciplinary evaluation
systematic review by Lykoudis et al. also demonstrated that is essential in determining the optimal treatment plan for
all three strategies had similar outcomes (57). Several other each patient. A traditional primary first approach may be
large studies, including a large meta-analysis and a multi- the most beneficial for patients with a symptomatic primary
institutional analysis also found similar outcomes compar- tumor—those who are at high risk for obstruction, perfo-
ing simultaneous versus staged resections (54,58,59). Many ration, or those continuing to have gastrointestinal bleed-
of the studies on simultaneous resection have a selection ing. A simultaneous approach seems to work best for those
bias toward smaller lesions. A study by Tanaka et al. dem- patients with small disease burden in the liver that require
onstrated that the volume of liver resected had an impact only minor hepatic resections. A liver first approach may
on overall complication rates (53). The authors found that if be best for those patients with a large burden of metastatic
more than a section of liver was resected or a patient’s age disease in the liver, with a relatively asymptomatic primary
was over 70 years, worse outcomes were seen (53). Another tumor (46,49,50,53,54,59).
study demonstrated a cost benefit in those undergoing
simultaneous resection, with 6 days fewer in the hospital LIVER ABLATION
(60). In general, simultaneous resections should be per-
formed for a highly select group of patients, preferably those There are multiple liver ablation techniques employed to
with small resectable liver metastasis and good functional treat liver metastases, including radiofrequency ablation
status (46,58,61). (RFA), ethanol ablation, as well as chemoembolization with
hepatic intra-arterial embolization (HAI). These strategies
LIVER FIRST APPROACH are all employed to treat lesions that may not be amenable
to surgical resection, such as those that will result insuffi-
The liver first approach is a more recent strategy that has cient liver function, those that have hepatic dysfunction, or
been employed to manage colorectal liver metastasis (62). those not suitable to undergo hepatic resection secondary to
The major determinant of overall survival is the control of comorbidities (68). Ablation may be performed open, lapa-
the metastatic disease (50,63). With this approach, the liver roscopically or percutaneously, with the open technique
is managed first with subsequent intervention for the pri- having the best results (69). However, these results are also
mary tumor later (46); however, neoadjuvant chemotherapy dependent on the experience of the physician performing
may be employed prior to the liver resection, particularly if the ablation (69).
the tumor is large. If the primary tumor is relatively asymp- RFA is the most widely utilized of these techniques.
tomatic, delaying the resection of the primary tumor can be Results for RFA as a primary treatment for liver metastases
have been shown to be inferior to curative resection (70,71) patients with metastatic disease (79). Rectal cancer patients
and survival outcomes for those undergoing RFA when are more likely to have lung metastasis than colon can-
compared to surgical resection are significantly lower. One cer patients, a finding that is likely related to the systemic
study compared outcomes for resection, RFA, or a com- venous drainage of the rectum via the inferior and middle
bined approach. The authors found a survival rate of 65% at rectal veins. The incidence of lung metastasis without liver
4 years for those undergoing resection versus 36% for com- metastasis is relatively low, reported to be around 1.7%–7%
bined therapy and 22% for those undergoing RFA alone (70). (79). In patients undergoing curative resection for rectal
While the survival for RFA alone is low, it still represents a cancer, greater than four lateral pelvic lymph nodes involved
significant improvement over no intervention for the meta- or the presence of lateral pelvic lymph nodes bilaterally were
static disease. Another study by Siperstein et al. showed a shown to be risk factors for pulmonary metastases (80). In
similar outcome. The authors found that the number and cases where there is metastatic disease to the lung, only 10%
size of metastases as well as CEA levels were strong prog- of those patients will be amenable to surgical resection (81).
nostic factors. They demonstrated an overall survival rate Resection of the metastatic disease, when feasible, is the
of 18.4% for patients undergoing RFA, which was improved best chance for curative treatment. In one population-based
over nonsurgical therapies (72). RFA as a first-line treatment study, survival improved for patients undergoing metastec-
compared to surgical resection has also been evaluated. One tomy for pulmonary metastasis from 11% to 53% over 3 years
study found a significantly higher local recurrence rate for for synchronous metastases and from 13% to 59% for meta-
patients undergoing RFA, however, very little difference in chronous metastases (82). However, only 4% of synchronous
terms of overall survival (47). A systematic review done in and 14% of metachronous metastases were resected for cure.
2010 as well as a Cochrane Review in 2012 concluded that The authors of this study also confirmed that rectal cancer
the research done on RFA is limited to small, mostly retro- patients have a higher incidence of lung metastasis compared
spective data. The reviews found that survival rates in the to colon cancer patients (82). Another study by Kim et al.
literature are highly variable, 14%–55%, and most studies demonstrated a 3- and 5-year survival rate of 54% and 30%,
had highly variable patient selection criteria. These studies respectively, for patients undergoing curative metastectomy.
conclude that more research should be done to recommend The authors also determined that the absence of adjuvant
RFA over resection (73,74). Overall, RFA is a good treatment chemotherapy, extrapulmonary metastases, elevated CEA
technique for those not amenable to resection and can offer level, and absence of pulmonary resection were all indica-
a survival benefit compared to no intervention (68). tors of a poor prognosis (83). Suzuki et al. reported a 5-year
Hepatic chemoembolization techniques are employed survival rate of 45% for patients undergoing surgical resec-
for patients who are not amenable to surgical resection or tion of pulmonary metastases. This study also found that an
RFA techniques (75). This method directly delivers chemo- elevated CEA level was a prognostic factor; patients with a
therapy to the metastatic lesion via the hepatic artery, which normal CEA had a 5-year survival of 57%, compared to 30%
has been shown to minimize systemic effects of the chemo- of those with an elevated CEA level (84). Many of these stud-
therapy (76). Overall, the results in the literature have been ies are small and retrospective; however, there is currently a
highly variable. One multi-institutional trial showed HAI randomized control trial underway (85).
had some survival benefit in those failing systemic chemo- Metastectomy for synchronous liver and lung metasta-
therapy, with a median survival rate of 19 months (77). A ses has been evaluated in some small series. These series
Cochrane Review in 2009 showed a modest survival benefit show the greatest benefit for patients undergoing both liver
for this treatment (78). and lung resection, if the lesions are detected sequentially.
Survival of patients with metastatic colorectal disease Patients who underwent pulmonary and liver metastec-
to the liver is directly dependent on the ability to treat the tomy had a survival rate of 44% if the lesions were detected
metastatic disease. Many techniques have been employed sequentially, compared to 0% when they were detected
to achieve the best survival rate for these patients. Overall, simultaneously (86). Another retrospective study also con-
resection provides the best chance for curative treatment. firmed this observation. The authors found that patients
The timing and method of resection is highly variable, and undergoing pulmonary metastectomy had better outcomes
should be determined on a case-by-case basis in conjunc- for metachronous lesions with a survival rate of 60%, com-
tion with a multidisciplinary team. If a curative resection pared to 0% for synchronous lesions (87). The studies on
is not possible, ablative techniques should be employed to concurrent liver and lung metastases are limited to small,
improve the overall survival for the patient. retrospective series.
Pulmonary metastases, although relatively rare when
compared to liver metastases, also benefit from curative
resection if the patient is a surgical candidate. Unfortunately,
PULMONARY METASTASIS very few patients will meet the criteria for surgical inter-
vention. For those who do meet criteria, there is a relatively
While the liver is the most common site of metastatic dis- good survival rate of 40%–50% in the literature for patients
ease, lung metastases are seen in approximately 10%–15% of undergoing pulmonary metastectomy.
References 295
REFERENCES 44. Dubois JB et al. Radiother Oncol. 2011;98(3):

298–303.
1. Bouchard P, Efron J. Ann Surg Oncol. 2010;17: 45. Manfredi S et al. Ann Surg. 2006;244:254–9.
1343–56. 46. Inhat p et al. World J Gastroenterol. 2015;21:
2. Kim YW et al. J Surg Oncol. 2009;99:58–64. 7014–21.
3. Ogiwara H et al. Ann Surg Oncol. 1994;1:99–104. 47. Otto G et al. Ann Surg. 2010;251:796–803.
4. Stocchi L et al. J Clin Oncol. 2001;19:3895–902. 48. O’Connell JB et al. J Natl Cancer Inst. 2004;96:
5. Heriot AG et al. Dis Colon Rectum. 2008;51:284–91. 1420–5.
6. Bulow S et al. Colorectal Dis. 2011;13:1256–64. 49. Simmonds PC et al. Br J Cancer. 2006;94:982–99.
7. Palmer G et al. Ann Surg Oncol. 2007;14:447–54. 50. Brouquet A et al. J Am Coll Surg. 2010;210:934–41.
8. Hanloser D et al. Ann Surg. 2002;237(4):502–8. 51. Tevis SE et al. Dis Colon Rectum. 2013;56(12):
9. Troja A et al. Int J Colorectal Dis. 2015;30:1157–63. 1339–48.
10. Watson AJ et al. Dis Colon Rectum. 2007;50:102–14. 52. Matsumo T et al. Dis Colon Rectum. 2014;57:
11. Chessin DB et al. J Am Coll Surg. 2005;201:948–56. 679–86.
12. Tekkis P. Br J Surg. 2013;100:1009–14. 53. Tanaka K et al. Surgery. 2004;136:950–9.
13. Beets-Tan, RG et al. Abd Imaging. 2000;25:533–41. 54. Mayo SC et al. J Am Coll Surg. 2013;216:707–18.
14. Messiou C et al. Br J Radiol. 2008;81(966):468–73. 55. Thelen A et al. Int J Colorectal Dis. 2007;22:
15. Colosio A et al. J M Res Imag. 2014;40:306–13. 1269–76.
16. Kontovounisios C et al. Colorectal Dis. 2017; 56. McKenzie SP et al. Int J Colorectal Dis. 2014;29:
331–338. 729–35.
17. Courtney D et al. Langenbecks Arch Surg. 2014;399: 57. Lykoudis PM et al. Br J Surg. 2014;101:605–12.
33–40. 58. Yin Z et al. Hepatology. 2013;57:2346–57.
18. VanderMeji W et al. Dis Colon Rectum. 2016;59: 59. Kelly ME et al. J Surg Oncol. 2015;111:341–51.
148–56. 60. Ejaz A et al. HPB. 2014;16:1117–26.
19. Alberda WJ et al. Dis Colon Rectum. 2015;58:677–85. 61. Weber JC et al. Br J Surg. 2003;90:956–62.
20. Moore HG et al. Dis Colon Rectum. 2006;49:1257–75. 62. Mentha G et al. Dig Surg. 2008;25:430–5.
21. Kanemitsu Y et al. Dis Colon Rectum. 2010;53:779–89. 63. Ayez N et al. Dis Colon Rectum. 2013;56:281–7.
22. Abdelsattar ZM et al. Dis Colon Rectum. 2013;56: 64. Lehman K et al. Ann Surg. 2012;255(2):237–47.
711–6. 65. Lam V et al. HPB. 2014;16:101–8.
23. Mirnezami AH et al. Dis Colon Rectum. 2010;53: 66. Verhoef C et al. Dis Colon Rectum. 2009;52:23–30.
1248–57. 67. Adam R et al. Oncologist. 2012;17:1225–39.
24. Austin KKS et al. Dis Colon Rectum. 2016;59(9): 68. Amersi FF et al. Arch Surg. 2006;141:581–8.
831–5. 69. Hildebrand P et al. Eur J Surg Oncol. 2006;32:
25. Dozois EJ et al. J Surg Oncol. 2011;103(2):105–9. 430–4.
26. Milne T et al. Ann Surg. 2013;258(6):1007–13. 70. Abdalla EK et al. Ann Surg. 2004;239:818–27.
27. Brown KGM et al. Dis Colon Rectum. 2015;58:850–6. 71. Gleisner AL et al. Arch Surg. 2008;143(12):1204–12.
28. Mirnezami R et al. Surg Oncol. 2013;22:22–35. 72. Siperstein AE et al. Ann Surg. 2007;246:559–67.
29. Ferenschild FTJ et al. Dis Colon Rectum. 2006;49: 73. Wong SL et al. J Clin Oncol. 2010;28:493–508.
1257–65. 74. Cirocchi R et al. Cochrane Database Syst Rev. 2012;
30. Moriya Y. Jpn J Clin Oncol. 2006;36(3):127–31. 6:1–63.
31. Jensen KK et al. Colorectal Dis. 2014;16(3):192–7. 75. deGroote K, Prenen H. World J Gastroinest Oncol.
32. Yang TX et al. Dis Colon Rectum. 2013;56:519–31. 2015;7:148–52.
33. Willett CG, Gunderson LL. J Gastrointest Surg. 2004; 76. Boige V et al. Ann Surg Oncol. 2008;15:219–26.
8(3):277–9. 77. Martin RC et al. Ann Surg Oncol. 2011;18:192–8.
34. Saltz LB. J Gastroint Surg. 2004;8(3):274–6. 78. Mocellin S et al. Cochrane Database Syst Rev.
35. Song HY et al. Gastrointest Endosc. 2008;68(4):713–20. 2009;8.
36. Kim EJ, Kim YJ. World J Gastroenterol. 2016;22(2): 79. Tan KK et al. J Gastrointest Surg. 2009;13:642–8.
842–52. 80. Watanabe K et al. Dis Colon Rectum. 2011;54:
37. Quyn AJ et al. Dis Colon Rectum. 2016;59:1005–10. 989–98.
38. Pacelli F et al. Ann Surg Oncol. 2010;17:152–62. 81. Dahabre J et al. Anticancer Res. 2007;27:4387–90.
39. Rahbari NN et al. Ann Surg. 2011;253(3):522–33. 82. Mitry E et al. Gut. 2010;59:1383–8.
40. Nielson M et al. Ann Surg Oncol. 2015;22:2677–84. 83. Kim CH et al. Dis Colon Rectum. 2012;55:459–64.
41. Bhangu A et al. Colorectal Dis. 2012;14(12):1457–66. 84. Suzuki H et al. Ann Thorac Surg. 2015;99:435–40.
42. Pellino G et al. Dis Colon Rectum. 2015;58:753–61. 85. Treasure T et al. Thorax. 2012;67:185–7.
43. Colibaseuanu DT et al. Dis Colon Rectum. 2014;57: 86. Nagakura S et al. J Am Coll Surg. 2001;193:153–6.
47–55. 87. Marudanayagam R et al. HPB. 2009;11:671–6.
32
Evaluation and management of peritoneal
metastatic disease
JAMES FLESHMAN AND KATERINA O. WELLS
than a systemic process (6). In a population-based cohort

CHALLENGING CASE study of CRC patients from Stockholm County, of the 8%
of patients identified as having PCCRC, half presented with
A 56-year-old woman presents with a cecal cancer isolated peritoneal disease without other solid organ metas-
and four 1 cm nodules on the right peritoneum on tases, suggesting that a discrete pathogenesis exists for
computed tomography (CT) scan. Additional meta- this pattern of disease (7). As such, surgical therapies have
static workup is negative. proven to offer some survival benefit over standard systemic
chemotherapy regimens.
CASE MANAGEMENT
ETIOLOGY
An R0 resection is possible with a right colectomy
and en bloc resection of the right peritoneum.
The pathogenesis of PCCRC follows a series of steps called
the “peritoneal metastatic cascade” initiated by destabiliza-
tion of tumor cells from the primary lesion. Exfoliation of
surface tumor cells from a T4 lesion can occur via downreg-
INTRODUCTION ulation of the intercellular adhesion molecule, E-cadherin
(8). In addition to spontaneous exfoliation, tumor dissemi-
It is estimated that 10%–13% of patients with colorectal can- nation is thought to occur via interstitial hypertension,
cer (CRC) present with synchronous or metachronous peri- wherein increased intratumoral oncotic pressure results
toneal carcinomatosis and these patients have traditionally in tumor shedding (9). Iatrogenic trauma can also result in
a poor prognosis (1–4). peritoneal seeding (10).
In the case of peritoneal carcinomatosis secondary to As free tumor cells gain access to the peritoneal cavity, the
CRC (PCCRC) secondary to appendiceal carcinoma, peri- pathways of physiologic peritoneal transport distribute cells
toneal spread can occur without lymphatic involvement as throughout the space and across the serosal linings (9). These
even fairly small and early tumors can incur appendiceal epithelial cells undergo epithelial to mesenchyme transition
perforation with local seeding. In retrospective comparison (EMT), where they employ developmental processes to gain
of appendiceal versus colon cancer, appendiceal cancer had migratory and invasive properties, such as reorganization of
higher rates of perforation at 44.7% versus 1.1% in colon the actin cytoskeleton with formation of membrane protru-
cancer with a higher rate of peritoneal seeding, 25.5% versus sions, filopodia, and lamellipodia, that allow for invasion
2.5% (p = 0.001) (5). PCCRC represents a more advanced to the mesothelium (11). Epithelial cells that lose sufficient
stage of disease, often with lymphatic invasion preced- cell-matrix interactions undergo a unique apoptotic path-
ing peritoneal involvement. This pattern of disease por- way termed anoikis (12). Resistance to anoikis is critical for
tends a uniformly poor prognosis with a median survival free tumor cells to maintain metabolism within the blood-
of 5–9 months (1). Though the presence of carcinomatosis less peritoneal environment. These phenotypic changes are
represents stage IV disease, PCCRC can represent a distinct mediated by a number of growth factors and molecular
pathology more as a regionally disseminated disease rather pathways (13). This pattern of integration is described in
296
Diagnosis / Diagnostic laparoscopy 297
leukocyte migration during peritoneal inflammation and is in scoring of lesions measuring less than 5 cm (p = 0.007)
“believed to be exploited by tumor cells” during peritoneal as well as determining the extent of regional involvement
invasion (11). Once tumor cells invade the subperitoneal (14). In general preoperative CT, SPCI scoring consider-
layer, production of growth factors and angiogenic factors ably underidentified regions with peritoneal involvement
stimulates proliferation and neovascularization (13). with median SPCI scores of 1–2 correlating with median
intraoperative peritoneal cancer index (PCI) scores of 6.
High SPCI scores were associated with poor outcomes
with SPCI >6 predicting an 83% likelihood of incomplete
DIAGNOSIS cytoreduction. Involvement of the ileocecal region on CT
was unfavorable, and this correlated with poor prognosis
IMAGING when ileocecal involvement was also noted intraoperatively
(HR 3.4, p = 0.041) (15). Similarly, in a multi-institutional
CT imaging has significant diagnostic import in the staging study, Esquivel et al. (16) observed that CT scan underes-
of metastatic CRC. CT imaging is useful for the detection timated a true PCI score of >20 in 12% of patients, there-
of both extraabdominal and solid organ metastatic disease. fore incorrectly qualifying patients for CRS intraperitoneal
However, its accuracy is limited in quantifying the extent and chemotherapy (IPC) not truly amenable for complete cyto-
nature of carcinomatosis, as adenocarcinoma tends to prog- reduction (16). These findings suggest that the extent of
ress along the peritoneal surface in thin layers that outline the peritoneal involvement is often underappreciated on preop-
normal contours of intraabdominal structures. CT imaging erative CT, and accuracy of findings is operator dependent;
can quantify mucinous adenocarcinoma with the presence of however, extensive disease on CT may be prognostic.
free colloid within the peritoneal cavity. Tumor nodules can Based on the prognostic impact of PCI, a novel TNM stag-
sometimes be visualized as solid tissue mass if they are large ing classification for PCCRC was developed to practically
enough in size or disrupt the usual contours of the abdominal stratify survival. T-staging refers to the value of PCI with
cavity. Studding of surfaces may be seen (10) (Figure 32.1). T1 (PCI 1–10) to T4 (PCI 30–39). The presence of positive
Due to these limitations, the role of CT in providing lymph nodes is classified as N1, and the presence of extra-
accurate preoperative prognostic data for success of cyto- abdominal metastases is classified as M1. The 5-year survival
reductive surgery (CRS) is poor. The sensitivity of detecting rates for stage I (T1N0M), stage II (T2-3N0M0), and stage
individual peritoneal lesions larger than 5 cm is accept- III (T4N0-1M0-1) are 87%, 53%, and 29%, respectively (17).
able at 59%–67%; however, this decreases to 9%–24% for Positron emission tomography (PET)/CT has a reported
nodules smaller than 1 cm (14). In retrospective review of sensitivity of 58%–100% for the detection of peritoneal
25 patients who underwent CRS and heated intraperitoneal metastasis. Most commonly, abnormally intense fluorode-
chemotherapy (HIPEC), preoperative imaging was analyzed oxyglucose (FDG) uptake is seen along the abdominal wall.
using a simplified peritoneal cancer index (SPCI) and cor- An SUV (Standardised Uptake Value = concentration/
related to operative findings and postoperative outcomes. dose/weight) maximum threshold of >5.1 is 78% accurate
There were statistically significant interobserver differences in identifying metastatic foci (18). In a prospective single-
center study, PET/CT was able to stratify the extent of car-
cinomatosis with a sensitivity of 90% and specificity of 77%.
In patients with PCCRC and negative PET/CT, the probabil-
ity of complete cytoreduction was 80% (19). However, detec-
tion of FDG avidity is poor in lesions <1 cm in size and in
mucinous tumors, limiting its accuracy in these cases.
The sensitivity of magnetic resonance imaging in detect-
ing peritoneal involvement is 87% with a negative predictive
value of 73% per segment of the abdominal cavity (20). As a
limitation, magnetic resonance imaging is not readily avail-
able or cost effective for routine imaging. In addition, the
quality of magnetic resonance imaging is highly operator
dependent and thus subject to variation.
DIAGNOSTIC LAPAROSCOPY
Due to the limitations of conventional imaging, direct visu-
Figure 32.1 CT imaging of peritoneal carcinomatosis.
Evidence of carcinomatosis on cross-sectional imaging
alization is the most accurate way of determining resect-
can include peritoneal implants (thin arrow) and caking of ability. Unfortunately, 20%–40% of patients in whom CRS
the omentum in the left mid-abdomen, and the presence IPC is attempted are ultimately deemed not amenable
of ascites in dependent areas of the peritoneal cavity for complete cytoreduction (2,21). Staging laparoscopy
(thick arrowhead). has been considered as a diagnostic measure to assess for
298 Evaluation and management of peritoneal metastatic disease
resectability prior to undertaking laparotomy; however, its In a 2002 retrospective analysis of 3,019 CRC patients,
role as a standard preoperative measure is controversial. no significant improvement in patient outcomes had been
The benefits of preoperative laparoscopy include the abil- achieved, with median survival approximating 7 months in
ity to assess for negative prognostic factors such as small patients with PCCRC (1). In comparative analysis of three
bowel and mesenteric involvement, to determine PCI and trials of the North Central Cancer Treatment Group using
plan for anticipated multivisceral resection. In a review by different chemotherapy regimens for metastatic CRC, 2,095
Seshadri et al. (22), staging laparoscopy excluded 7%–41% patients with PCCRC were considered with 2.1% (44) of
of unnecessary laparotomies not amenable to complete patients having isolated PC. Treatment with modern regi-
cytoreduction. The feasibility of laparoscopy in the setting mens such as FOLFOX afforded a median overall survival
of PCCRC is limited, as most patients have undergone prior of 15.7 months with 5- and 8-year survival 4.1% and 1.1%,
laparotomy. Thickness of the abdominal wall with tumor as respectively, among patients with PCCRC. This is a signifi-
well as significant adhesions are significant contributors to cantly worse prognosis compared to non-PCCRC patients
unsuccessful laparoscopy. A 37% rate of understaging was having other sites of metastatic disease with median over-
reported. This was primarily due to involvement of carcino- all survival 17.6 months and 5- and 8-year survival 6% and
matosis within the lesser sac and pancreas (2) and along the 3.2%, respectively (28). Patients with PCCRC also fared
ureters, and deep diaphragmatic invasion (21) that could worse than non-PCCRC patients with second-line treatment
not be thoroughly assessed by laparoscopy. following 5-FU treatment, after adjusting for treatment regi-
mens (HR 1.37, p = 0.006). PCCRC also had shorter median
progression-free survival at 5.8 months versus 7.2 months,
p = 0.002, and higher rates of death from all causes com-
TREATMENT pared to non-PCCRC (28). Over the last two decades, despite
improvements to chemotherapy agents and other novel ther-
NCCN guidelines support the use of systemic chemotherapy apies, median survival for PCCRC remains dismal. The poor
in the management of PCCRC and consider more aggres- outcomes afforded by systemic chemotherapy and palliative
sive treatment, specifically CRS IPC, as controversial. CRS surgery alone have stimulated interest in surgical options to
IPC can be considered in experienced centers for selected more proactively manage PCCRC and improve survival.
patients with limited peritoneal metastases for whom R0
resection can be achieved (23). A consensus statement on CYTOREDUCTIVE SURGERY AND
the management of such patients within major peritoneal INTRAPERITONEAL CHEMOTHERAPY
surface malignancy centers offers a clinical pathway to
optimize clinical outcomes (24). Patients with PCCRC and Several institutional studies have demonstrated a benefit
distant sites of dissemination should be considered for best of CRS and IPC in selected patients. The reported median
systemic therapy. Those without distant disease should survival ranges from 19.2 to 47 months with 1- and 5-year
undergo a completeness of cytoreduction assessment at an survival of 72%–90% and 19%–51%, respectively (29–34).
established peritoneal surface malignancy center. This may In 2004, a multicenter study of 506 patients from 28 cen-
involve diagnostic laparotomy or laparoscopy. At the time ters treated with CRS HIPEC and/or early postoperative IPC,
of surgery, if complete cytoreduction is possible, CRS IPC the overall median survival was 19.2 months with overall
is performed followed by best systemic therapy. If complete survival at 1 and 5 years of 72% and 19%, respectively (30).
cytoreduction is not possible, palliative surgery can be con- Elias et al. reported an improved median survival rate of
sidered followed by best systemic therapy. Neoadjuvant sys- 62.7 months and 2- and 5-year overall survival rates of 81%
temic chemotherapy is supported in initially unresectable and 51%, respectively, using oxaliplatin perfusate (35).
patients with reassessment for resectability performed at Complete cytoreduction (CC) is the most consistent
2–3 months. If a substantial improvement is seen, CRS IPC positive prognostic indicator across all reported studies
is a reasonable consideration. However, if minimal response with those patients <CC-1 achieving significantly better
or progression is noted, the patient will not benefit from median survivals of 32.4–62 months versus 5–17.4 months
CRS IPC, and this therapy is associated with a higher mor- among patients with residual disease (29,30,35). Other posi-
bidity and mortality in this subset of patients (25,26). tive prognostic features include age younger than 65 years,
and use of adjuvant chemotherapy (30,33). Poor prognostic
CHEMOTHERAPY ALONE features include high PCI and poor histologic features, spe-
cifically signet ring cell features, M1 liver disease, and N+
PCCRC has a uniformly worse prognosis compared to met- disease (30,33). Involvement of small bowel and small bowel
astatic colorectal cancer confined to the liver or lung. In the mesentery are poor prognostic features owing to the inabil-
era of 5-FU/leucovorin monotherapy, median survival with ity to completely clear these surfaces of macroscopic disease
PCCRC ranged from 5.2 to 7 months (1,4,27). A disease-free and ensure adequate and uniform exposure of HIPEC to
interval of less than 1 year, the presence of ascites and lung small bowel surfaces (29).
metastasis, and the stage of carcinomatosis are poor prog- As evidenced by these predictive factors, the success
nostic features (1,4). of HIPEC is predicated on stringent patient selection and
Treatment / Cytoreductive surgery and intraperitoneal chemotherapy 299
largely favorable pathologies. To address this bias, a ran- ureteral obstruction, one or less foci of intestinal obstruc-
domized controlled study by Veerwal et al. (36) reported on tion, absence of gross small bowel or small bowel mesentery
105 patients (54 CRS and HIPEC versus 51 systemic 5-FU/ involvement, and no more than small volume disease in the
leucovorin with or without palliative surgery). Those treated gastrohepatic ligament (24). Lack of involvement of the car-
with CRS and HIPEC achieved an overall survival 22.3 dia of the diaphragm or the pericardial sac is also important
months versus 12.6 months, following systemic therapy as peritoneal stripping of these areas is technically challeng-
(HR 0.55, p = 0.32). In 6-year long-term follow-up, those ing. Approximately 8% of patients with PCCRC have con-
with complete cytoreduction achieved a median survival of comitant liver metastases (LM). Historically, the presence
48 months and 5-year survival of 45% (37). Completeness of more than three LM was a relative contraindication for
of cytoreduction was also associated with improved sur- CRS IPC (40). CRS IPC for PCCRC and LM remains contro-
vival with the majority of patients who had extensive tumor versial. However, a growing body of literature supports the
burden (more than five regions) seeing a median survival of feasibility of cytoreductive surgery and liver metastectomy
only 5.4 months and contributing to the majority of signifi- followed by HIPEC with better survival outcomes com-
cant postoperative complications (36). pared to chemotherapy alone (41–43).
Complications PCI staging and prognostic impact

Early reporting of complications following the investiga- The PCI is a scoring tool used in the intraoperative setting
tional phase of CRS and HIPEC reported significant major to grade the extent of carcinomatosis and guide operative
morbidity of 27%, including fistula 4.7%, leak 3%, bleeding decision-making (Figure 32.2). The PCI is a composite of
4.5%, peripancreatitis 7.1%, and hematological toxicity 4%. both implant size and distribution. The lesions size score
Duration of operation, extent of peritonectomy, and num- (LS) considers the size of visible nodules. In the case that
ber of suture lines created were closely associated with post- nodules of varying sizes are seen, the sizes of the largest
operative morbidity. Treatment-related mortality was 1.8% nodules are scored with LS-0 designating no visible tumor
largely due to neutropenic sepsis (38). In more recent review deposits, LS-1 designating nodules less than 0.5 cm in size,
of CRS HIPEC-related morbidity and mortality, grade III/ LS-2 designating nodules between 0.5 and 5 cm in size,
IV morbidity rates range from 12% to 52% and mortality and LS-3 for any nodule >5 cm or for tumor confluence.
from 0.9% to 5.8% in tertiary high volume centers. Sepsis Distribution is divided into 13 abdomino-pelvic regions
and multiorgan failure secondary to surgical complications with each region designated an LS score. The sum of scores
were the most common causes of mortality. Unique to IPC, for each region is then calculated with a maximum available
hematological toxicity was reported as 0%–28%. Common score of 39 (13 × 3) (44).
postoperative complications include sepsis (0%–14%), fistula A number of studies have supported the PCI as a prog-
(0%–23%), abscess (0%–37%), ileus (0%–86%), perforation nostic indicator of postoperative outcomes. In a retrospec-
(0%–10%), anastomotic leak (0%–9%), venous thromboem- tive review of 168 patients by Huang et al. (45), patients with
bolism (0%–9%), and renal insufficiency (0%–7%). These PCI <15 and treated with CRS IPC, overall major morbid-
rates approach those seen with other major abdominal ity was 33.9%, and these patients had shortened intensive
surgery suggesting that CRS IPC can be performed with care unit lengths of stay. Median overall survival was 42.1
acceptable morbidity in tertiary high-volume centers (39). months with 1-, 3-, and 5-year overall survival 90%, 55.8%,
Metabolic and hemostatic “derailment” is common follow- and 34.5%, respectively. These rates of survival varied con-
ing HIPEC and should only be performed in a setting with siderably by PCI score with 5-year survival for patients
intensive care unit support and physicians experienced with with PCI <5 at 58.1% and median DFS 11 months versus
this higher level of care. 8.6 months for PCI >15 (p = 0.04). On multivariate anal-
ysis, PCI was an independent prognostic factor for poor
Indications survival adjusting for other high-risk features. Sugarbaker
(10) reports a significant decrement in 5-year survival as
Patient selection is the basis for success to CRS IPC. Those PCI increases with PCI <10 associated with a 5-year sur-
who stand to benefit most from CRS IPC must have mini- vival of 50% versus 20% for PCI 11–20 and 0% for PCI >20,
mal residual disease to <CC-1 that is isolated to peritoneal p < 0.0001.
surfaces that can be completely accessed by topical chemo- In review of patients undergoing second-look cytoreduc-
therapy to allow for complete eradication. Tumor biology tive surgery for PCCRC, PCI was also useful in predicting
largely dictates long-term response with noninvasive tumors long-term survival with PCI <12 associated with 2-year sur-
such as pseudomyxoma peritonei secondary to appendiceal vival of 64% versus 14% for PCI >12, p = 0.066 (3). As such,
neoplasm of low malignant potential and low-grade sarco- some centers limit the option of CRS IPC to patients with
mas responding best (10). Additional variables associated PCI <12 (46). The PCI does have prognostic limitations as
with complete cytoreduction include Eastern Cooperative this scoring system does not take into account tumor his-
Oncology Group (ECOG) performance status of two or less, tology with extensive burden of low-grade or noninvasive
absence of extraabdominal disease, absence of biliary or tumors still allowing for complete cytoreduction with good
Regions Lesion size Lesion size score

0 Central LS 0 No tumor seen
1 Right upper LS 1 Tumor up to 0.5 cm
2 Epigastrium LS 2 Tumor up to 5.0 cm
3 Left upper LS 3 Tumor > 5.0 cm
4 Left flank or conflunece
5 Left lower
6 Pelvis
7 Right lower
1 2 3
8 Right flank
8 0 4
9 Upper jejunum
7 6 5
10 Lower jejunum
11 Upper ileum 11 9
12 Lower ileum
PCI
10
12
Figure 32.2 Peritoneal cancer index (PCI). PCI is a composite of both lesion size (LS) and distribution across abdomino-
pelvic regions. The sum of scores for each region is then calculated with a maximum available score of 39 (13 × 3).
(Reprinted with permission from Sugarbaker PH. Langenbecks Arch Surg. 1999;384[6]:576–87.)
long-term outcomes. Conversely, limited carcinomatosis in survival in symptomatic patients with positive second-look
anatomically complex areas such as the porta hepatis may surgery is reported at 15%. Critics of this approach question
incur low PCI scores but high morbidity due to the inability the benefit of negative second look with its attendant mor-
to achieve complete cytoreduction in these areas (10). In a bidity and mortality (50).
recent study by Faron et al. (47), overall survival followed Second-look CRS IPC is also supported for patients
a linear relationship to PCI, suggesting that no discrete with PCCRC following initial CRS IPC. This is a planned
cutoff in PCI score can independently contraindicate CRS reexploration scheduled 6–9 months after the initial CRS,
IPC. Ultimately, the decision to proceed with CRS IPC must wherein an attempt is made to visualize all peritoneal sur-
weigh a number of prognostic factors. faces, perform repeat cytoreduction and/or visceral resec-
tion to achieve a score of CC-1 or less. If this is achieved and
Second-look surgery the patient had a fairly good response to the original cyto-
reductive surgery, HIPEC is repeated. If a poor response is
Enthusiasts of CRS IPC would argue that patients with early, noted, then the chemotherapy agent is changed (10).
regionalized peritoneal involvement benefit most from CRS
IPC. As such, selective second-look protocols with subse- Palliation
quent CRS IPC as part of routine follow-up of patients fol-
lowing primary resections at high risk for metachronous CRS and IPC is also reported as a palliative treatment for
PCCRC have been supported as an early intervention. In malignant ascites as peritoneal fibrosis induced by HIPEC
retrospective review by Leung et al. (48) of patients with obliterates surface tumor and decrease potential space in
metachronous PCCRC, patients at high risk for peritoneal the peritoneum (26,51). However, use of CRS with IPC is
involvement included those with T4 lesions, macroscopic not routinely supported as a palliative measure owing to its
synchronous peritoneal or ovarian metastatic involvement morbidity.
at time of initial operation, perforated primary, and muci- Regardless of indication, candidates for cytoreduction and
nous or signet ring pathology. The presence of any high-risk IPC must also be physiologically fit to undergo major abdom-
factor is 85.6% sensitive in identifying patients who go on to inal surgery. In a retrospective observation cohort study by
develop PCCRC. Such a protocol would include a planned Malfroy et al. (52), the mean number of organs resected CRS
second-look surgery at 12 months postoperatively and CRS was 4.3 (52). Hepatectomy was performed in 19% of patients.
IPC should PCCRC be present. The detection rate of second- Diaphragmatic peritonectomy correlated with higher likeli-
look surgery is reported at 55% for patients deemed high hood of ICU admission, p = 0.013. Heated IPC induces a sec-
risk despite no clinical evidence of disease recurrence at ond physiologic stress to the patient. An acute compartment
12 months (49). The rate of conversion to long-term disease- syndrome is induced with instillation resulting in reduced
free status following positive second-look surgery in asymp- venous return and cardiac index. The subsequent response
tomatic patients is 17%. The rate of conversion to long-term to hyperthermia induces a systemic inflammatory response
Treatment / Heated intraperitoneal chemotherapy 301
syndrome. In the 24-hour postoperative period, substan- by identifying peritoneal metastases not visible by conven-
tial fluid shifts occur with 24-hour postoperative peritoneal tional methods. Nanoprobe technology directed specifically
drain output approaching 2 L (52). Hypovolemia and vaso- toward tumor tissue similarly serves to accurately identify
pressor requirements can contribute to end organ hypoper- residual tumor tissue at the time of cytoreduction and is
fusion. Delayed IPC delivery through the use of abdominal currently in development (57).
Port-a-Cath does not induce a profound systemic inflamma-
tory response syndrome response and is better tolerated by HEATED INTRAPERITONEAL
less physiologically robust patients (53). CHEMOTHERAPY
CYTOREDUCTION Surgical principles
Goal of cytoreduction The argument against cytoreductive surgery alone and the
rationale to explain the rapid progression of peritoneal-sur-
Completeness of cytoreduction to R0 resection is a consis- face recurrence following cytoreductive surgery stems from
tent and significant prognostic factor that is paramount to the “tumor cell entrapment” theory wherein the surgeon acts
successful long-term outcomes (10,31,35,36). The primary as the promoter through the dissemination of malignant cells
goal of CRS is to remove all visible tumor to a residual tumor from surgical trauma (10). As inflammatory cells migrate
of 2.5 mm or less. The CC score is a clinical tool that objec- to peritonectomized surfaces, free tumor emboli become
tively describes the extent of cytoreduction achieved at the entrapped in fibrin and adhere to stripped surfaces. Growth
time of surgery (44). CC-0 indicates no macroscopic residual factors stimulated by the postsurgical healing process pro-
cancer; CC-1 indicates less than 2.5 mm of residual tumor. mote proliferation of entrapped tumor cells. By this theory,
Achieving a CC score of CC-1 or less is desired as this bur- the addition of IPC at the time of cytoreduction is necessary to
den represents the level of residual disease for which com- both extirpate free tumor emboli as well as reduce the burden
plete extirpation with IPC would be feasible. CC-2 is defined of inflammatory cells that would contribute to postoperative
as more than 2.5 mm but less than 2.5 cm of residual tumor, healing. The procedure for application of HIPEC following
and CC-3 is defined as more than 2.5 cm of residual tumor. cytoreduction has evolved over time to optimize tissue deliv-
A CC score of CC-2 or CC-3 is considered incomplete. ery and decrease complications related to this therapy. The
Complete cytoreduction should be done with the simultane- principles of HIPEC are outlined in the following sections.
ous goal of preserving as much viscera as possible (26).
Tissue penetration
Technique of cytoreduction
The major limitation to HIPEC is the very superficial pen-
The Sugarbaker technique was first described in 1995 (54). etration of cytotoxic drugs into tumor tissue. Based on
One or all procedures are employed depending on the experimental animal models of tissue penetration, this is
extent of peritoneal involvement. Use of laser electrosurgery estimated to be a maximal depth of 1–2 mm (58,59).
is recommended to both dissect viscera and ablate tumor Despite this superficial level of action, local tissue
nodules from surfaces of small bowel and mesentery where absorption is rapid based on pharmacokinetics testing
peritoneal resection is not feasible. The heat necrosis cre- (60,61). Thus, a 2.5 mm diameter nodule for residual tumor
ated at the line of dissection creates a tumor-free margin remnants is the cutoff at which adequate tissue penetration
of resection (55). Sharp dissection of tumor nodules is not can occur (59).
recommended as margins of resection can retain residual
tumor, and cut surfaces serve as sites of bleeding during IPC Hyperthermia
instillation (44).
Application of heat at 41°C–43.5°C induces a profound and
Novel methods of cytoreduction selective tumor necrosis (62). Heat also optimizes the inten-
sity of tissue chemotherapy dose delivered by decreasing
The presence of residual carcinoma following cytoreduction the interstitial pressure of tissues and increasing tumor cell
is a major contributor to treatment failure, disease progres- membrane permeability and transport. Heat may also alter
sion, and mortality, as such novel therapies to identity resid- cellular metabolism of specific agents (63). This synergis-
ual tumor and aid in complete cytoreduction are of interest. tic effect of heated chemotherapy varies by chemotherapy
In a pilot study by Liberale et al. (56), fluorescence of tumor agent with the highest “thermal enhancement” seen with
nodules following indocyanine green injection was used selected alkylating agents (64).
to identify and resect peritoneal nodules. Of the nodules
resected as a result of this technique, 84% were malig- Closed versus open technique
nant. Nonmucinous tumors demonstrated a clear hyper-
fluorescence compared to benign nodules. This application Original protocols of HIPEC involved the administration
served to improve the extent of resection in 29% of cases of heated chemotherapy via a closed circuit of catheters
placed into the peritoneal cavity following cytoreduction Following cytoreduction, inflow and outflow catheters are
(Figure 32.3a). placed through the abdominal wall. The abdominal walls
By this technique, inflow and outflow catheters are are sutured to a self-retaining retractor incorporating a plas-
placed, and the laparotomy incision is sutured closed. A tic sheet into the suture to cover the abdominal cavity. A slit
large volume of perfusate and heated chemotherapy is intro- is made over the plastic sheet to allow for the introduction of
duced via inflow catheters at high pressure. The abdomen is a hand. Once the instillation is complete, the chemotherapy
manually agitated externally to distribute the chemother- agent is drained from the abdomen, and reconstruction is
apy. Once the instillation is complete, the laparotomy inci- undertaken. Limitations in drug distribution by the closed
sion is reopened and surgical anastomoses are completed. technique are addressed by this open technique, allowing
This approach keeps chemotherapy within a closed circuit for gentle manipulation of the viscera during the instilla-
without theoretical exposure to health-care personnel while tion to ensure uniform contact of chemotherapy agent to all
maintaining hyperthermia and intraabdominal pressure to surfaces. The theoretical risk of chemotherapy agent expo-
aid with tissue penetration. The major reported disadvan- sure to health-care personnel is negligible with the use of
tage of this approach is the nonuniform distribution of che- appropriate protective equipment (66). Other methods of
motherapy to peritonectomized surfaces. Gravity also leads delivery that are essentially hybrid open-closed techniques
to pooling and greater exposure of chemotherapy agents to have been described to address the shortcomings of each of
dependent areas within the peritoneum (63). Early experi- these approaches (67).
ence also reports small bowel thermal injury and fistula sec-
ondary to focal high temperatures of perfusate entering at Choice of chemotherapy agents
the inflow catheter tip (65).
The open “coliseum” technique (Figure 32.3b) is the The peritoneal-plasma barrier allows for retention of large
currently accepted technique for HIPEC delivery (44). molecular weight chemotherapy agents into the peritoneal
(b)
Chemotherapy
reservoir
Heat
In-flow Out-flow
exchanger
pump pump
Temperature
(a) probe
Heat
exchanger
Chemotherapy Digital
reservoir thermometer
Clamps Pump
43.0
x2
2 3 Chemotherapy
Temperature Sampling bags
display part
42.5 Temperature probes
Thompson
retractor
Smoke
evacuator
Figure 32.3 (a) HIPEC closed technique. In the closed approach, the laparotomy incision is closed and three intraperito-
neal catheters are used to instill and drain heated chemotherapy in a closed circuit. (b) HIPEC open “coliseum” technique.
In the open approach the abdominal wall is sutured to a self-retaining retractor incorporating a plastic sheet as a cover.
Inflow and outflow catheters circulate heated chemotherapy with manual visceral manipulation.
Summary 303
cavity and dose-intensive therapy with significantly more distribution of IPC occurs when delivered in a delayed fash-
potent concentration of agent to surface tumor cells (68). ion due to the presence of intraabdominal adhesions and
Systemic toxicity is reduced as the peritoneal cavity serves dependent distribution of perfusate. In retrospective case
as a barrier against systemic absorption with the majority series of 31 patients treated with CRS followed by delayed
of drug that enters circulation cleared via first pass hepatic IPC via abdominal Port-a-Cath, peritoneal scintigraphy
metabolism (69). The choice of chemotherapeutic agent for performed prior to chemotherapy administration dem-
IPC varies across centers with selection based largely on onstrated free diffusion of tracer as intended in 58% of
institutional preference and experience. Chemotherapy patients. Limited diffusion due to postoperative adhesions
agents most commonly used for PCCRC are mitomycin C was noted in 35.5% of patients (53).
and oxaliplatin. Mitomycin C was the earliest agent used in Regardless, limited experience with delayed IPC suggests
HIPEC for CRC and has been shown to be effective in ran- long-term outcomes comparable to HIPEC can be achieved.
domized controlled trial when used at a rate of 10–12.5 mg/ In a case-controlled study of 18 PCCRC patients under-
m2 heated to 42°C with a dwell time of 90 minutes (37,70). going CRS and delayed IPC, Mahteme et al. (73) demon-
The efficacy of oxaliplatin as a HIPEC agent is also well strated a median survival of 32 months versus 14 months,
supported in phase II trials with favorable rates of median p = 0.01 in matched controls treated with 5-FU/leucovorin.
overall survival and disease-free survival (35). This regimen, Two- and 5-year survival was 60% and 28%, respectively. In
termed the Elias protocol, is most commonly employed and this experience, IPC was initiated via transabdominal cath-
consists of oxaliplatin 460 mg/m2 in 2 L/m2 of dextrose at eters on postoperative day 1 and daily for 6 days; this was
a temperature of 43°C over 30 minutes. In a retrospective repeated at 4–6 week intervals. In 7/18 patients, treatment
analysis by Leung et al. (46) comparing mitomycin C versus was discontinued due to catheter-related problems, with
oxaliplatin efficacy, oxaliplatin is associated with a greater drain leakage being a source of morbidity. Culliford et al.
unadjusted median survival compared to mitomycin C (56 (74) reports similarly favorable outcomes following delayed
versus 29 months) with oxaliplatin HR 0.59, p = 0.0017. IPC with median survival 34 months and 5-year survival
This survival advantage was most pronounced in patients 28%. Morbidity was 26% with few grade III–IV complica-
PCI >10 (HR 0.47, p = 0.014) and in well-differentiated and tions; however, port malfunction and/or infection requiring
non-signet-ring pathologies in subgroup analysis compared revision and/or removal was the most common late com-
to mitomycin C. As a limitation, oxaliplatin is associated plication. In a more recent randomized controlled trial by
with higher rates of surgical complications and higher mor- Cashin et al. (75) comparing CRS HIPEC to oxaliplatin/5-
bidity of grade III and grade IV complications compared to FU systemic chemotherapy, patients treated with delayed
mitomycin C (71). IPC achieved a median overall survival of 25 months versus
Many specialized centers employ the use of bidirec- 18 months (HR 0.51, p = 0.04) suggesting this treatment
tional, or HIPEC-plus, chemotherapy protocols wherein modality remains superior to modern systemic chemother-
simultaneous intraoperative systemic and intraperitoneal apy regimens. However, this study was terminated early due
chemotherapy infusions are performed to obtain a bidirec- to low accrual. These benefits suggest that delayed IPC is an
tional fluid gradient across the surface tumor cell. This was oncologically acceptable and perhaps better-tolerated thera-
first employed by Elias et al. (35) with the use of IV 5-FU peutic option in centers where HIPEC is not offered.
and leucovorin with intraperitoneal oxaliplatin due to pH
incompatibility of delivering both agents intraperitoneally.
In a pharmacologic study by Speeten et al. (72), intraopera-
tive systemic infusion of 5-FU results in rapid circulation SUMMARY
and equilibration throughout body compartments includ-
ing peritoneal fluid. Circulating and intraperitoneal levels PCCRC carries a uniformly poor prognosis with a median
of 5-FU work synergistically with instilled intraperitoneal survival of 5–9 months (1). A discrete pathogenesis exists for
chemotherapy agents to enhance tumor toxicity. this pattern of metastatic disease, wherein local peritoneal
seeding from perforated appendiceal carcinoma can occur
DELAYED INTRAPERITONEAL in the absence of nodal or systemic disease. Defects in the
CHEMOTHERAPY e-cadherin pathway and cell interactions seem to play a role
in the ability of cells to establish in a bloodless environment.
Delayed recurrent IPC via transperitoneal catheters is an Shedding or iatrogenic dissemination of tumor surface
alternative method of IPC delivery. The rationale for this cells from locally advanced colon carcinoma also results
approach is to provide high concentrations of chemother- in peritoneal seeding; however, peritoneal carcinomatosis
apy to peritoneal surfaces repeatedly over time. Delayed IPC from colon and rectal primary lesions represents a more
also offers shorter operating room times (243 minutes ver- advanced disease, often with lymphatic invasion preceding
sus 440 minutes), less estimated blood loss (770 mL versus peritoneal involvement. Though the presence of carcino-
4000 mL), shorter hospital lengths of stay (11 days versus 21 matosis represents stage IV disease, PCCRC can represent
days), and less morbidity with fewer grade III–IV complica- a distinct pathology more as a regionally disseminated pro-
tions compared to HIPEC (53). As a limitation, nonuniform cess rather than a systemic disease. Systemic chemotherapy
is virtually ineffective. As such, local therapies targeted at to long-term success. Preoperative imaging has limited
cytoreduction of the peritoneum have proven to offer some sensitivity for determining the extent of peritoneal dis-
survival benefit over standard systemic chemotherapy regi- ease. Staging laparoscopy is supported in some centers to
mens. The most important positive prognostic indicator is accurately stage the peritoneum and prevent unnecessary
completeness of cytoreduction, aiming for less than 2.5 mm laparotomy; however, this approach is also limited in the
of residual disease. Stringent selection of patients amenable ability to accurately survey the retroperitoneum and may
to complete cytoreduction is paramount to long-term suc- understage peritoneal involvement. Ultimately the defini-
cess; therefore, emphasis is placed on accurate preoperative tive cytoreductive laparotomy is the most accurate method
staging of peritoneal tumor burden. Preoperative imaging of staging. Following systematic exploration and cytore-
has limited sensitivity for determining the extent of perito- duction, IPC in an immediate fashion effective extirpates
neal tumor burden. CT imaging with IV contrast has been remaining surface and free tumor cells. CRS and HIPEC
shown to have the best chance for accurate staging. PET/CT is a complex abdominal procedure with major morbidity
is of no use in mucinous tumors. Staging laparoscopy is sup- and mortality rivaling other major multivisceral resections.
ported in some centers to accurately stage the peritoneum Alternative methods of delayed IPC are described that
and prevent unnecessary laparotomy in patients who would offer similar rates of median survival in small series with
ultimately be determined unresectable. However, laparos- the added benefit of less short-term morbidity compared to
copy has been found to understage peritoneal involvement HIPEC. Ultimately the surgical management of PCCRC is
in 37% of cases due to limitations in accurately exploring effective at prolonging the median survival of patients com-
the retroperitoneum (2). The PCI determined at the time pared to palliative chemotherapy and should be considered
of laparotomy is the most accurate method of staging. This in patients eligible for complete cytoreduction.
intraoperative tool is used to guide decision-making and is
well supported as a prognostic tool, with PCI >12 having REFERENCES
significantly higher major morbidity and shorter overall
survival compared to PCI <12 (3). Several reported expe- 1. Jayne DG et al. Br J Surg. 2002;89(12):1545–50.
riences from established peritoneal malignancy centers 2. Pomel C et al. Eur J Surg Oncol (EJSO). 2005;31(5):
have served to optimize the systematic approach of CRS as 540–3.
well as the method of delivery and choice of agent for IPC. 3. Portilla AG et al. World J Surg. 1998;23(1):23–9.
Immediate IPC and delayed IPC techniques are described 4. Chu DZ et al. Cancer. 1989;63(2):364–7.
with similar rates in overall survival that far exceed sur- 5. Tae Son Il et al. Surg Oncol. 2016;25(1):37–43.
vival rates achieved with systemic chemotherapy alone. 6. Spiliotis J et al. Curr Oncol. 2016;23(3):266.
Ultimately, effective surgical management of PCCRC pro- 7. Segelman J et al. Br J Surg. 2012;99(5):699–705.
longs median survival in appropriately selected patients. 8. Hirohashi S. Am J Pathol. 1998;153(2):333–9.
9. Hayashi K et al. Cancer Res. 2007;67(17):8223–8.
10. Sugarbaker PH. Langenbecks Arch Surg. 1999;384(6):
576–87.
CONCLUSION 11. Yilmaz M, Christofori G. Cancer Metastasis Rev. 2009;
28(1–2):15–33.
PCCRC carries a uniformly poor prognosis with a median 12. Cao Z et al. Crit Rev Oncog. 2016;21(3–4):155–68.
survival of 5–9 months. A discrete pathogenesis exists for 13. Lemoine L et al. WJG. 2016;22(34):7692.
this pattern of metastases, as in the case of appendiceal car- 14. de Bree E et al. J Surg Oncol. 2004;86(2):64–73.
cinoma where high rates of perforation contribute to peri- 15. De Bree E et al. Eur J Surg Oncol. 2006;32(1):65–71.
toneal seeding. In locally advanced colon tumors, shedding 16. Esquivel J et al. J Surg Oncol. 2010;102(6):565–70.
from the tumor surface due to increased oncotic pressure 17. Yan TD et al. Cancer. 2010;117(9):1855–63.
and/or iatrogenic trauma similarly seed the peritoneum. As 18. Suzuki A et al. Eur J Nucl Med Mol Imaging. 2004;
such, local therapies targeted at cytoreduction of the perito- 31(10):1413–1420.
neum have proven to offer some survival benefit over stan- 19. Passot G et al. Eur J Surg Oncol. 2010;36(3):315–23.
dard systemic chemotherapy alone. With the experience 20. Klumpp BD et al. Abdom Imaging. 2012;38(1):64–71.
of several established peritoneal malignancy centers, the 21. Iversen LH et al. Br J Surg. 2012;100(2):285–92.
management of PCCRC has evolved significantly over the 22. Seshadri RA, Hemanth Raj E. Indian J Surg Oncol.
past two decades to include a systematic approach of cytore- 2016;7(2):230–5.
duction and optimized intraperitoneal chemotherapy agent 23. Benson AB et al. J Natl Compr Canc Netw. 2015;
and method of delivery. The most important positive prog- 12:1028–59.
nostic indicator is completeness of cytoreduction, aiming 24. Esquivel J et al. Ann Surg Oncol. 2006;14(1):128–33.
for less than 2.5 mm of residual disease. Completeness of 25. Esquivel J. J Gastrointest Oncol. 2016;7(1):72–8.
cytoreduction is most likely in patients with low burden of 26. Bhagwandin SB et al. Oncology. 2016;30(11):
disease and favorable tumor biology. Stringent patient selec- 1002–1007.
tion to achieve complete cytoreduction is the cornerstone 27. Sadeghi B et al. Cancer. 2000;88(2):358–63.
References 305
28. Franko J et al. J Clin Oncol. 2012;30(3):263–7. 52. Malfroy S et al. Surgical Oncology. 2016;25(1):6–15.
29. Verwaal VJ et al. Ann Surg Oncol. 2005;12(1):65–71. 53. Fajardo AD et al. Dis Colon Rectum. 2012;55(10):
30. Glehen O. J Clin Oncol. 2004;22(16):3284–92. 1044–52.
31. Franko J et al. Cancer. 2010;116(16):3756–62. 54. Sugarbaker PH. Ann Surg. 1995;221(1):29–42.
32. Esquivel J et al. Ann Surg Oncol. 2014;21(13): 55. Sugarbaker PH. Cancer Treat Res. 1996;82:375–85.
4195–201. 56. Liberale G et al. Ann Surg. 2016;264(6):1110–5.
33. Huang Y et al. Anticancer Res. 2016;36(3):1033–40. 57. Xi L et al. Ann Surg Oncol. 2014;21(5):1602–9.
34. Alzahrani N et al. ANZ J Surg. 2016;86(11):937–41. 58. Ozols RF et al. Cancer Res. 1979;39(8):3209–14.
35. Elias D et al. J Clin Oncol. 2009;27(5):681–5. 59. Van der Speeten K et al. Cancer J. 2009;15(3):
36. Verwaal VJ. J Clin Oncol. 2003;21(20):3737–43. 216–24.
37. Verwaal VJ et al. Ann Surg Oncol. 2008;15(9): 60. Panteix G et al. Oncology. 1993;50(5):366–70.
2426–32. 61. Elias D et al. Ann Oncol. 2002;13(2):267–72.
38. Stephens AD et al. Ann Surg Oncol. 1999;6(8): 62. Cavaliere R et al. Cancer. 1967;20(9):1351–81.
790–6. 63. González-Moreno S. WJGO. 2010;2(2):68.
39. Chua TC et al. Ann Surg. 2009;249(6):900–7. 64. Takemoto M et al. Int J Hyperthermia. 2009;19(2):
40. Elias D et al. Eur J Surg Oncol. 2006;32(6):632–6. 193–203.
41. Lorimier G et al. Eur J Surg 2017;43:150–158. 65. Jacquet P et al. Cancer. 1996;77(12):2622–9.
42. Varban O et al. Cancer. 2009;115(15):3427–36. 66. Kyriazanos I et al. Surg Oncol. 2016;25(3):308–14.
43. de Cuba EMV et al. Cancer Treat Rev. 2013;39(4): 67. Lotti M et al. J Minim Access Surg. 2016;12(1):86–9.
321–7. 68. Jacquet P, Sugarbaker PH. Cancer Treat Res. 1996;82:
44. Sugarbaker PH. Cancer Chemother Pharmacol. 1999; 53–63.
43(Suppl):S15–25. 69. de Bree E, Tsiftsis DD. Recent Results Cancer Res.
45. Huang Y et al. Int J Surg. 2016;32(c):65–70. 2007;169:39–51.
46. Leung V et al. Eur J Surg Oncol. 2017;43:141–149. 70. Esquivel J et al. Ann Surg Oncol. 2007;14(1):128–33.
47. Faron M et al. Ann Surg Oncol. 2015;23(1):114–9. 71. Rouers A et al. Acta Chir Belg. 2006;106(3):302–6.
48. Leung V et al. Eur J Surg Oncol. 2016;42(6):836–40. 72. Van der Speeten K et al. J Surg Oncol. 2010;102(7):
49. Elias D et al. Ann Surg. 2008;247(3):445–50. 730–5.
50. Sugarbaker P. Surgl Oncol Clin N Am. 2012;21(4): 73. Mahteme H et al. Br J Cancer. 2004;90(2):403–7.
689–703. 74. Culliford AT et al. Ann Surg Oncol. 2001;8(10):787–95.
51. Graziosi L et al. Ann Ital Chir. 2016;87:312–20. 75. Cashin PH et al. Eur J Cancer. 2016;53:155–62.
33
Chemotherapy for colon and rectal cancer
JONATHAN LU AND MARC R. MATRANA
CHALLENGING CASE ADJUVANT CHEMOTHERAPY IN

RESECTABLE COLON CANCER
A 53-year-old man undergoes a left colectomy for a
T3N1M0 adenocarcinoma. The administration of adjuvant chemotherapy in the setting
of localized colon cancer is aimed at eradicating any micro-
metastases in order to reduce the chance of disease recur-
rence (5). Patients with AJCC (American Joint Committee on
CASE MANAGEMENT Cancer) stage 1 disease who have undergone surgical resec-
tion have an excellent prognosis. The overall 10-year survival
It is recommended by the Multidisciplinary Tumor for stage 1 colorectal cancer patients is 90% compared to 5%
Committee that he receive postoperative chemother- for stage 4 patients (6). Given the lack of data supporting the
apy with FOLFOX. use of chemotherapy in stage 1 disease and overall excellent
prognosis, patients with stage 1 disease do not require any
adjuvant therapy (5). While patients with stage 3 disease have
a clear benefit with adjuvant chemotherapy, the role of adju-
vant chemotherapy for patients with stage 2 disease is more
INTRODUCTION controversial. The QUASAR trial looked at 2,291 patients
with stage 2 colon cancer who were randomized to chemo-
Colorectal cancer is among the most common and deadli- therapy with 5-FU (5-fluorouracil) and leucovorin (folinic
est malignancies in the United States. There is estimated acid) in comparison to observation alone (7). The results of
to be over 130,000 new cases of colorectal cancer in the the QUASAR trial showed that adjuvant chemotherapy can
United States in 2016 with nearly 50,000 expected deaths increase survival in patients with stage 2 colon cancer, albeit
(1). However, the annual incidence of colorectal cancer only a modest 3.6% absolute improvement (7). With regard to
has been declining over the past few decades as a result stage 3 disease, metastatic and/or local relapse can occur up to
of improved screening, risk factor modifications, and 50% of the time (8). Adjuvant chemotherapy should always be
improved treatment modalities (2). Increased screen- considered in stage 3 colon cancer. The MOSAIC trial looked
ing has led to increased removal of precancerous lesions, at 2,246 patients with both stage 2 and stage 3 colon cancer
thus reducing colorectal cancer rates (2–4). Despite the and showed improved overall survival in stage 3 colon cancer
declines in incidence and mortality, colorectal cancer patients given FOLFOX (leucovorin, 5-FU, and oxaliplatin)
remains the third most common cause of cancer death in (9,10). In addition, data suggest a decreased recurrence rate
both men and women (1). Management of colorectal can- in high-risk stage 2 patients, defined as those with T4 tumors,
cer is dependent on staging and differs greatly between less than 12 lymph nodes examined after surgery, poorly dif-
localized disease compared to metastatic disease. Surgical ferentiated histology, lymphatic invasion, vascular invasion,
resection is the mainstay of therapy for localized disease bowel obstruction/perforation, or microsatellite instability,
and is further discussed in other chapters of this text- who were treated with FOLFOX, even though there was no
book. Adjuvant chemotherapy is often given postop- statistically significant increase in overall survival or disease-
eratively with the intent to eliminate any microscopic free survival (9,10). Therefore, patients with high-risk stage 2
residual disease. disease in particular, should undergo serious consideration
306
Chemotherapy in metastatic colorectal cancer 307
for adjuvant chemotherapy (11). Given the debatable nature and radiation has also been investigated. Initial studies
of adjuvant chemotherapy for stage 2 colon cancer, patients looking at adjuvant chemotherapy with 5-FU after neoadju-
should be encouraged to participate in clinical trials (11). vant chemotherapy and radiation followed by surgery were
There have been many studies looking at the optimal not promising and showed no improvement to overall sur-
chemotherapy regimen in the adjuvant setting. Most of the vival and no benefit to the rate of local recurrence (18,24).
evidence is in the setting of patients with stage 3 disease, but However, further investigations revealed an improvement
the data are often extrapolated for the management of stage in disease-free survival in patients who received adjuvant
2 colon cancer patients. Standard adjuvant chemotherapy is FOLFOX in addition to neoadjuvant chemotherapy and
based on 6 months of a 5-FU–based chemotherapy regimen radiation (22). Even though there is no conclusive evidence
(8). Current recommendations for chemotherapy include for adjuvant chemotherapy, a 6-month treatment regimen
FOLFOX, CapeOx (capecitabine and oxaliplatin), or FLOX of FOLFOX or CapeOx remains the standard chemotherapy
(bolus 5-FU, leucovorin, and oxaliplatin) (9,10,12–14). In of choice in the postoperative setting and should be con-
addition, capecitabine alone or 5-FU/LV (5-FU and leu- sidered for stage 2 and 3 rectal cancer patients. In addition,
covorin) are alternative treatments if oxaliplatin is not an adjuvant chemotherapy should be started as soon as the
appropriate option for the patient (15,16). The elderly popu- patient is medically able after surgery, as delays can worsen
lation, for instance, had no benefit in overall survival or dis- survival (25).
ease-free survival from the addition of oxaliplatin to 5-FU/
LV based on subgroup analysis (17). However, FOLFOX and
CapeOx are the preferred, especially in the setting of stage
3 disease. Studies comparing 3 months of FOLFOX versus
CHEMOTHERAPY IN METASTATIC
6 months of FOLFOX in this setting are being conducted
COLORECTAL CANCER
with no data yet reported.
Greater than 50% of all patients diagnosed with colorec-
tal cancer will develop hepatic metastases, many of which
are deemed unresectable (26). However, with the advances
CHEMOTHERAPY IN RESECTABLE made in therapeutic options, the median survival has
RECTAL CANCER improved from 1 year to more than 30 months (27). The
treatments for metastatic colon cancer and metastatic rec-
As with colon cancer, stage 1 rectal cancer is managed with tal cancer are approached in the same manner. Currently,
surgical excision alone given its good overall prognosis (6). initial management of metastatic colorectal cancer involves
However, in contrast to colon cancer, the administration one of five chemotherapy regimens: FOLFOX, FOLFIRI
of chemotherapy in the setting of locally advanced rectal (leucovorin, 5-FU, and irinotecan), CapeOx, infusional
cancer is often paired with radiation therapy. Specifically, 5-FU or capecitabine, or FOLFOXIRI (leucovorin, 5-FU,
chemotherapy is often given in the neoadjuvant setting in oxaliplatin, and irinotecan). The timing and sequencing of
conjunction with radiation therapy to stage 2 and 3 rec- these therapies have been investigated in a limited manner
tal cancer patients. The combination of neoadjuvant che- without a clearly defined preferred order of administration
motherapy with infusional 5-FU and radiation has been (28–30). In general, however, FOLFOX and FOLFIRI are
showed to improve local control of disease when compared comparable in efficacy and often the preferred regimen for
to radiation alone (18). In addition, when looking at com- initial management of metastatic colorectal cancer (31,32).
bined chemotherapy and radiation, there is evidence from In addition to 5-FU–based chemotherapy regimens, the
the German Rectal Cancer Study Group supporting that antivascular endothelial growth factor antibody bevaci-
neoadjuvant chemotherapy and radiation is superior with zumab are often added in the first-line setting to the various
regard to local recurrence when compared to adjuvant 5-FU–based options, including both FOLFOX and FOLFIRI
chemotherapy and radiation (19). Moreover, neoadjuvant (33). When added to FOLFOX, bevacizumab improves pro-
chemotherapy and radiation were tolerated better than gression-free survival in patients with metastatic colorec-
the combination of adjuvant chemotherapy and radiation tal cancer (34). Similarly, the addition of bevacizumab
(19). Given the parallels between colon and rectal cancer, to FOLFIRI significantly increases clinical improvement
investigators have looked at the role of oxaliplatin in order with regard to overall survival, progression-free survival,
to improve outcomes. However, several trials have shown response rate, and duration of response (35). Moreover,
that the addition of oxaliplatin did not improve clinical cetuximab and panitumumab are monoclonal antibod-
outcomes (20–22). Therefore, current guidelines recom- ies against epidermal growth factor receptor (EGFR) that
mend a 5-FU–based chemotherapy regimen to be given are also utilized in the treatment of metastatic colorec-
concurrently with radiation therapy. Infusional 5-FU or tal cancer. EGFR overexpression can be found in 50% of
capecitabine are the chemotherapeutic agents of choice for colorectal tumors (36). Multiple trials have found improved
concurrent chemoradiation (18,19,23). overall survival, progression-free survival, and response
The addition of adjuvant chemotherapy in the postop- rate in metastatic colon cancer with the use of EGFR
erative setting after undergoing neoadjuvant chemotherapy inhibitors in addition to traditional 5-FU chemotherapy
308 Chemotherapy for colon and rectal cancer
regimens, specifically in the RAS wild-type population (37). chemotherapy (HIPEC) for patients with peritoneal car-
Furthermore, cetuximab has been shown to have provided cinomatosis from colorectal cancer with favorable results
clinical benefit in overall survival and progression-free sur- (46,47). When compared to systemic chemotherapy alone,
vival when utilized alone (38). Therefore, cetuximab and cytoreductive surgery followed by HIPEC and adjuvant che-
panitumumab are options for patient unable to undergo motherapy showed increased survival (47,48). However, the
more intensive chemotherapy. Despite their benefits, when study utilized 5-FU and leucovorin as opposed to the stan-
studies evaluated EGFR inhibitors added to combination dard combination therapy with FOLFOX or FOLFIRI. In
5-FU–based chemotherapy and the vascular endothelial addition, disease recurrence after cytoreductive surgery and
growth factor antibody bevacizumab, the results were not HIPEC is common, ranging from 22.5% to 82%, and periop-
as favorable. The results showed a shorter progression-free erative morbidity is high ranging from 19% to 49% (49,50).
survival and worse quality of life (39); therefore, the addi- Therefore, cytoreductive surgery in combination with
tion of EGFR inhibitors to 5-FU–based chemotherapy and HIPEC should be limited to select patients (see Chapter 32).
bevacizumab is not recommended.
After induction therapy, maintenance therapy has now
become a mainstay of therapy. The phase 3 AIO 0207 trial FUTURE DIRECTION IN COLORECTAL
looked at 472 patients who underwent induction with
CANCER
FOLFOX and bevacizumab or CapeOx and bevacizumab
(40). After induction, the study compared standard main-
tenance with 5-FU plus bevacizumab, bevacizumab alone, Immunotherapy has quickly become a widely used ther-
and observation alone. Treatment with bevacizumab alone apy in many malignancies given its success. The therapy
was shown to be noninferior to the group receiving both is based on the theory that many tumor cells upregulate
5-FU and bevacizumab (40). PD-L1 (programmed death ligand 1) in order to escape the
Treatment choices after disease progression are based host’s immune system (51). There have been limited studies
on previous regimens used. In general, patients who have looking at immunotherapy in the setting of colorectal can-
received FOLFOX or CapeOx for initial therapy will transi- cer. A phase 2 study investigated the relationship between
tion to FOLFIRI. Likewise, patients who received FOLFIRI mismatch-repair status and possible clinical benefit of
for initial treatment should undergo FOLFOX or CapeOx immune checkpoint blockade, specifically with pembroli-
therapy, which are comparable in efficacy (41). EGFR inhibi- zumab (52). Although overall survival and progression-free
tors such as cetuximab or panitumumab are viable options survival endpoints were not reached, the study did show
in the non-first-line setting for patients who have wild- that mismatch-repair status can predict benefit from such
type KRAS (42,43) and have progressed while on intensive agents (52). There are several ongoing trials in this space, the
5-FU–based chemotherapy. results of which may help define the role of immunotherapy
Regorafenib and combination trifluridine and tipiracil for colorectal cancers.
are two additional agents that should be considered for use
as subsequent therapy options. The CORRECT trial looked
at 760 patients who were noted to have disease progression CONCLUSION
after standard therapy and found that patients who received
regorafenib had improved overall survival and progression-
free survival when compared to placebo (44). Looking at the Even with the advances made in therapeutic options and
oral combination trifluridine and tipiracil, the RECOURSE improvements in median survival from 1 year to more
trial examined 800 patients with metastatic colorectal can- than 30 months, colorectal cancer remains the third lead-
cer who had progressed on two lines of therapy and found ing cause of cancer-related death (1,27). There are still many
improved overall survival and disease-free survival (45). As areas of research that need to be investigated in order to
a result of these studies, both regorafenib and combination help guide the management of colorectal cancer. Overall,
trifluridine and tipiracil can be utilized in the third- and it will undoubtedly rely on a multidisciplinary approach to
fourth-line setting. However, there are no direct compari- obtain the best outcomes for patients.
sons between the two regimens, so sequencing is deferred to
the clinician and patient. REFERENCES
1. Siegel RL et al. CA Cancer J Clin. 2016;66(1):7–30.
2. Edwards BK et al. Cancer. 2010;116(3):544–73.
HYPERTHERMIC INTRAPERITONEAL
3. Cress RD et al. Cancer. 2006;107(5 Suppl.):1142–52.
CHEMOTHERAPY IN METASTATIC
4. Siegel RL et al. Cancer Epidemiol Biomarkers Prev.
COLORECTAL CANCER 2012;21(3):411–6.
5. Saltz LB. Surg Oncol Clin N Am. 2010;19(4):819–27.
There have been several studies looking at cytoreductive 6. Patrlj L et al. Hepatobiliary Surg Nutr. 2014;3(5):
surgery in combination with hyperthermic intraperitoneal 324–9.
References 309
7. Quasar Collaborative G et al. Lancet. 2007;370(9604): 31. Tournigand C et al. J Clin Oncol. 2004;22(2):229–37.
2020–9. 32. Goldberg RM et al. J Clin Oncol. 2004;22(1):23–30.
8. Des Guetz G et al. Cochrane Database Syst Rev. 2010; 33. Kabbinavar FF et al. J Clin Oncol. 2005;23(16):
(1):CD007046. 3706–12.
9. Andre T et al. N Engl J Med. 2004;350(23):2343–51. 34. Saltz LB et al. J Clin Oncol. 2008;26(12):2013–9.
10. Andre T et al. J Clin Oncol. 2009;27(19):3109–16. 35. Hurwitz H et al. N Engl J Med. 2004;350(23):
11. Benson 3rd AB et al. J Clin Oncol. 2004;22(16): 2335–42.
3408–19. 36. McKay JA et al. Eur J Cancer. 2002;38(17):2258–64.
12. Schmoll HJ et al. J Clin Oncol. 2007;25(1):102–9. 37. Pietrantonio F et al. Crit Rev Oncol Hematol. 2015;
13. Haller DG et al. J Clin Oncol. 2011;29(11):1465–71. 96(1):156–66.
14. Kuebler JP et al. J Clin Oncol. 2007;25(16):2198–204. 38. Jonker DJ et al. N Engl J Med. 2007;357(20):2040–8.
15. Twelves C et al. N Engl J Med. 2005;352(26): 39. Tol J et al. N Engl J Med. 2009;360(6):563–72.
2696–704. 40. Hegewisch-Becker S et al. Lancet Oncol. 2015;16(13):
16. Marsoni S et al. Lancet. 1995;345(8955):939–44. 1355–69.
17. Tournigand C et al. J Clin Oncol. 2012;30(27): 41. Cassidy J et al. Br J Cancer. 2011;105(1):58–64.
3353–60. 42. Karapetis CS et al. N Engl J Med. 2008;359(17):
18. Bosset JF et al. N Engl J Med. 2006;355(11):1114–23. 1757–65.
19. Sauer R et al. N Engl J Med. 2004;351(17):1731–40. 43. Van Cutsem E et al. J Clin Oncol. 2007;25(13):
20. Aschele C et al. J Clin Oncol. 2011;29(20):2773–80. 1658–64.
21. Gerard JP et al. J Clin Oncol. 2012;30(36):4558–65. 44. Grothey A et al. Lancet. 2013;381(9863):303–12.
22. Rodel C et al. Lancet Oncol. 2015;16(8):979–89. 45. Mayer RJ et al. N Engl J Med. 2015;372(20):1909–19.
23. Hofheinz RD et al. Lancet Oncol. 2012;13(6):579–88. 46. Elias D et al. J Clin Oncol. 2010;28(1):63–8.
24. Bosset JF et al. Lancet Oncol. 2014;15(2):184–90. 47. Verwaal VJ et al. J Clin Oncol. 2003;21(20):3737–43.
25. Biagi JJ et al. JAMA. 2011;305(22):2335–42. 48. Hendlisz A et al. J Clin Oncol. 2010;28(23):3687–94.
26. Yoo PS et al. Clin Colorectal Cancer. 2006;6(3):202–7. 49. van Oudheusden TR et al. Eur J Surg Oncol. 2015;
27. Heinemann V et al. Lancet Oncol. 2014;15(10): 41(10):1269–77.
1065–75. 50. McRee AJ et al. Oncology. 2015;29(7):523–4, C3.
28. Koopman M et al. Lancet. 2007;370(9582):135–42. 51. Topalian SL et al. N Engl J Med.
29. Ducreux M et al. Lancet Oncol. 2011;12(11):1032–44. 2012;366(26):2443–54.
30. Seymour MT et al. Lancet. 2007;370(9582):143–52. 52. Le DT et al. N Engl J Med. 2015;372(26):2509–20.
34
Adjunctive treatment of rectal cancer with
radiation and the adverse effects of radiation
exposure of the rectum
ROLAND HAWKINS
Radiation therapy is also used for the nonsurgical, poten-

CHALLENGING CASE tially curative treatment of cancers of the cervix, uterus,
bladder, prostate, and anus. To achieve cure of these pelvic
A 62-year-old man presents with blood per rectum. cancers without surgical resection, a relatively high dose
He has mild rectal discomfort with bowel movements of radiation is necessary, for instance 70–90 Gy compared
and a feeling of incomplete evacuation. Two years to the 40–54 Gy used for adjunctive treatment and for the
previously he received external beam radiotherapy nonsurgical treatment of squamous cell carcinoma of the
for prostate cancer. His rectal examination is nor- anal canal. The rectal injury that sometimes results from
mal except for some blood on the gloved finger. A exposure of the rectum in the course of treating pelvic can-
flexible sigmoidoscopy demonstrates friable mucosa cers is occasionally referred to the rectal surgeon for treat-
with neovascularity of the distal 4 cm of rectum. The ment. Treatment of rectal injury from radiation exposure is
mucosa is friable with telangiectasia. addressed in the section “Rectal Injury and Its Treatment.”
Adjunctive radiation treatment is employed to eradi-
cate deposits of cancer in soft tissue beyond the margins of
CASE MANAGEMENT resection and in pelvic lymph nodes not removed at surgery.
It is administered either before or following en bloc resec-
The history and endoscopic exam are suggestive of tion of the involved length of large bowel by low anterior
radiation proctitis. Management includes fiber and resection (LAR) or abdominal perineal resection (APR).
topical therapy. The friable areas of the rectum can Preoperative treatment is referred to as neoadjuvant and
be treated with topical application of a large swab postoperative treatment as adjuvant. Radiation treatment
soaked with 10% formalin passed through an ano- is usually administered to patients with locally advanced
scope or proctoscope. Argon plasma coagulation is but resectable stage II or III disease (Table 34.1). Radiation
also effective treatment. treatment is also administered as an adjunct to local exci-
sion of less advanced rectal cancer.
Recurrence after apparently curative surgery for rectal
cancer may develop in structures adjacent to the margin of
resection, in regional lymph nodes in the pelvis, or as metas-
INTRODUCTION tasis to the peritoneum or distant organs. Treatment with
radiation and/or chemotherapy is judged as beneficial inso-
Apart from a few exceptional circumstances, radiation far as it increases patient survival or reduces the incidence
treatment is used as an adjunct to surgical resection in the of local, regional, or distant recurrence. Long-term sur-
potentially curative treatment of adenocarcinoma of the vival is the most important outcome in judging benefit. It is
rectum. The benefits and adverse effects of the adjunctive unambiguously evaluable and reflects the balance of benefit
use of radiation in the surgical treatment of rectal cancer are and potentially life-threatening adverse effects of treatment.
the principal topics addressed in this chapter. Local and regional recurrence is often not salvageable. Its
310
Introduction 311
Table 34.1 Staging of rectal carcinoma  d  T ln 2

BEDα /β = nd 1 + − (34.1)
 α/β  ατ 2
TNM
Dukes group TNM Description
In which T is the time between the first and last radiation
A I T1 N0 M0 Tumor limited to treatments. The value of τ2 is the volume doubling time of
submucosa the irradiated population of cells in the absence of radiation
T2 N0 M0 Tumor into, not through, exposure. It includes the effect of the time from mitosis to
muscularis propria mitosis of cycling cells, the presence of a subpopulation of
B II T3 N0 M0 Tumor through muscularis cells that are not cycling, and cell death from causes other
propria than radiation. The range of values of τ2 for clinically evi-
T4 N0 M0 Tumor invades other dent rectal cancer is from about 20 to 100 days. The values
organs or through of α and β are constants of the linear quadratic relation
peritoneal serosa between dose and the fraction of cells that survive irradi-
C III N1 or N2 N1 (1–3 nodes +) ation with the ability to reproduce enough times to form
Any T, M0 N2 (>3 nodes +) a visible colony. As shown in Equation 34.1, the ratio α/β
D IV M1, any T Distant metastasis determines the influence of variation of the fractional dose
or N d on the cell killing effect of the radiation course. The term
that depends on T represents the increase in the fraction
of cells that survive because of cell replication during the
prevention is important, if not a requirement, for achieving course. The fraction of cells that survive a course of radia-
cure of the disease. It may itself be life threatening and may tion is defined as the number of viable cells present immedi-
act as a source of distant metastasis. Furthermore, uncon- ately after the last exposure divided by the number present
trolled recurrence in the pelvis is detrimental to the qual- immediately prior to the first.
ity of life of patients who are not cured. It may cause pain, The value of the α/β ratio has been measured and tabu-
bleeding, infection, obstruction, and incontinence affecting lated for an array of experimental animal tumors, human
bowel and urogenital organs. Distant recurrence is impor- cancers, and the cells upon which the function of various
tant because it is usually an unsalvageable life-threatening normal tissues and organs depends (5,6). For malignant
form of treatment failure. tumors, α/β is usually estimated to be about 10 Gy. The
Evolution of the method of adjunctive radiation treat- value of α is estimated to be about 0.15 Gy−1 for rectal and
ment of rectal cancer over the past 30 years has produced other carcinomas. With these values and assuming τ2 is
what are now two standard treatment regimens, referred about 30 days, the BEDα/β of the short course of radiation
to here as the short and long treatment courses. The short treatment consisting of five treatments of 5 Gy each over
course has been used only for preoperative treatment. It 7 days is 37.5–1.1 = 36.4 Gy. With these same values, the
typically consists of a dose of 25 Gy in fractions of 5 Gy each BEDα/β of the long course consisting of 28 treatments of
over a period of 5–7 days with surgery following within a 1.8 Gy each over 37 days is 59.5−4.3 = 55.2 Gy. This indi-
week. A variant of the short course has been reported to cates the long course provides a biologically effective dose
give improved disease-free survival (1,2). It consists of that is 52% greater than the short course.
short-course radiation followed by chemotherapy and then However, consider that when given as a neoadjuvant pre-
total mesorectal excision. operative treatment, the cancer cells that survive the long
The long course has been used for both pre- and postop- course are expected to continue replicating during the time
erative treatment. It typically consists of 45–54 Gy in frac- between the last radiation exposure and the day of surgery.
tions of 1.8–2 Gy over a period of 5–6 weeks and includes Therefore, to include the effect of cell replication on the
concurrent chemotherapy with 5-fluorouricil. When used fraction of the cells that survive at the time of surgery, the
preoperatively, surgical resection is usually 6–12 weeks after value of T in Equation 34.1 should be the time from the day
completing the long-course radiation treatment. of the first radiation exposure to the day of surgery. For the
The cell killing effectiveness of a course of radiation long course this is about 90 days. With T equal to 90 and
treatment is expressed by a biologically effective dose with the other constants the same as for the short course, BEDα/β
symbol BEDα/β. The value of BEDα/β is defined so that frac- of the long course is 59.5−13.9 = 45.6 Gy. Doing the same
tionated treatment courses that have the same BEDα/β may thing for the short course and making T equal to about
be expected to kill the same fraction of the population of 12 days gives BEDα/β for the short course equal to 35.6 Gy.
cells of a treated cancer or of an exposed normal tissue or That is, the long course is a biologically effective dose that
organ (3,4). As such, it is used to quantitatively compare the is about 28% greater than the short course with respect to
level of effect of different fractionated radiation treatment fraction of the cancer cells present at the start of the radia-
courses with regard to their likelihood of curing a cancer tion course that are still viable at the time of the surgery. The
or causing a specific organ injury. The value of BEDα/β for a addition of concurrent chemotherapy to the long course is
course of radiation consisting of n treatments (fractions) of expected to make the differential between long and short
dose d Gy is calculated as courses even greater.
312 Adjunctive treatment of rectal cancer with radiation and the adverse effects of radiation exposure of the rectum
With the exception of some lymphocyte subsets that acute and long-term side effects except that the VMAT treat-
die within hours of exposure to radiation, the physiologic ment was associated with significantly less high-grade anal
death, disintegration, and disappearance of nearly all cells incontinence, 4% compared to 16% with p-value of 0.032 (12).
lethally injured by ionizing radiation takes place only after
they and/or their descendants go through one or more,
often aberrant cell divisions. As a result, there is a time lag
BENEFIT OF ADJUVANT AND
between exposure of a cancer and the physiological death
and disappearance of the irradiated carcinoma cells that is
NEOADJUVANT RADIATION
variable and dependent on their mitotic activity. This lag TREATMENT
ranges from a few days to a year or more. A typical time to
manifest the maximal response of a carcinoma or sarcoma Tables 34.2, 34.3, and 34.4 summarize several trials in
to radiation is the order of a few weeks to a month or two. which patients were randomly assigned to treatment con-
The same phenomenon is in part responsible for the delay sisting of various combinations of pre- and postoperative
of 6 months to a year or more in the development of some radiation and chemotherapy. The radiation treatment plans
forms of radiation injury to normal organs. of each assignment are similar to either the short or long
With short-course preoperative radiation, there is little course previously described and can be gleaned from the
time for tumor response before surgery. There is evidence table by noting the dose shown. When the dose is about
that at surgery after short-course irradiation, the aver- 25 Gy, it is a short course and when 40–60 Gy it is similar
age tumor size and average number of lymph nodes with to the long course. Benefits and adverse effects of pre- and
metastatic carcinoma have decreased slightly, but this is not post-operative radiation treatment reported in these studies
enough to produce a change in the distribution of tumor will be examined and compared. Adjuvant treatment after
or nodal stage in a study population (7). With long-course local excision is also discussed.
preoperative irradiation, more time is allowed for response Several randomized trials of postoperative adjuvant ther-
of the disease and downstaging to occur. This is evident in apy in the late 1970s and 1980s listed in Table 34.2 indicate
some of the trials listed in Tables 34.3 and 34.4. This was that postoperative radiation and chemotherapy can lead to
demonstrated in a trial in which all patients were treated statistically significant improvement in overall survival and
with 13 daily fractions of 3 Gy each and randomly assigned the incidence of local recurrence when compared to surgery
to surgery within 2 weeks after the end of radiation or sur- alone. Based on gastrointestinal tumor study group (13),
gery 6–8 weeks after radiation (8). North Central Cancer Treatment Group (NCCTG) (14),
With both long and short courses, radiation is directed at and National surgical adjuvant breast and bowel program
the pelvis with the superior border placed at about the L5-S1 (NSABP) (15) studies, a U.S. National Institutes of Health
interspace. The inferior border is placed at least 3–5 cm consensus development conference in 1990 recommended
below the distal-most extent of tumor or at the inferior that postoperative radiation and chemotherapy be standard
margin of the obdurator foramen. For distal tumors, it may treatment for stages II and III rectal cancer (16). An advan-
include all or part of the anal canal. In earlier studies, treat- tage of postoperative treatment is that selection for adjuvant
ment was restricted to anterior-posterior directed beams treatment can be based on pathologic staging, whereas with
(9). More recently, laterally directed beams that exclude preoperative treatment selection is based on necessarily
some bladder and bowel in the anterior part of the pelvis imperfect clinical staging.
are a standard part of the beam arrangement referred to The use of preoperative radiation has been extensively
as a three-dimensional (3D) conformal plan. Only the vol- evaluated in Europe. From inspection of the randomized
ume that is exposed to all the beams gets the full prescribed trials in Table 34.3, it is evident that preoperative radiation
dose. This includes, in addition to the rectum and mesorec- treatment reliably produces a clinically and statistically
tum, small and large bowel in the posterior pelvis, the pos- significant reduction in the incidence of local recurrence
terior part of the bladder and prostate, the presacral area, by about 50%–60%. This remains true even in the Dutch
the sacrum and sacral canal nerves, the lymph nodes of the colorectal cancer trial, which was designed to minimize the
internal iliac, and the most distal part of the common iliac need for pelvic irradiation by mandating surgery to include
chains. If there is extension to urogenital organs, the lymph a total mesorectal excision (TME) (17). Note that the results
nodes of the external iliac chains are sometimes included. for the surgery-only arms shown in Table 34.3 indicate
After APR, the perineal incision, which tends to be a site of that TME is more rigorously extirpated than the surgery of
recurrence, is included in the treatment volume (10,11). historical practice. Its use reduced the total recurrence at
With the availability of intensity modulated radiation ther- 5 years after surgery alone to 10.4% compared to the 25%–
apy, the clinical treatment volume can sometimes be encom- 28% found in the Stockholm I and II and Swedish rectal tri-
passed with less dose to bowel, bladder, and perineal skin als that did not require TME (9,18,19).
than with the 3D conformal plan. When given as a rotational About 35% of the patients in the Dutch study had disease
arc (volumetric-modulated arc therapy [VMAT]) the inten- found in pelvic nodes making them stage III. Among this
sity modulated treatment has been reported to have efficacy subgroup, 20.6% of those who did not have radiation treat-
equivalent to the 3D conformal treatment. There were similar ment and 10.6% of those who did suffered a local recurrence
Benefit of adjuvant and neoadjuvant radiation treatment 313
Table 34.2 Postoperative adjuvant radiation studies
Number of Pelvic recurrence Survival %, 5

Study patients Therapy arms %, 5 years years Comments
GITSG (13) 202 S 24 46 T3, T4, or N+
S-C 27 56 Semustine + 5-FU
S-44 Gy 20 52
S-44 Gy + C 11 59 (p = 0.07)
NCCTG (14) 204 S-50.4 Gy 25 47 Semustine + 5-FU
S-50.4 Gy + C 13 (p = 0.036) 57 (p = 0.02)
NSABP (15) 555 S 25 43 Semustine + 5-FU
R-01 S-46 Gy 16 (p = 0.06) 41 +Vincristine
S-C 21.4 53 (p = 0.01)
Norway (63) 144 S 30 50 Bolus 5-FU on 6 days
S-46 Gy + C 12 (p = 0.01) 64 (p = 0.05) During radiation
NSABP (64) 694 S-C 14 58 Semustine + 5-FU
R-02 S-50.4 Gy + C 8 (p = 0.02) 58 +Vincristine or
5-FU + leukovorin
MGH and Emory (31) 99 LE (T1) 11 Chemotherapy for some
patients
LE-xrt (T1) None
LE (T2) 67
LE-xrt (T2) 15 (p = 0.004)
RTOG (30) 65 LE, T1 (fav) 14.3 86 Fav = favorable
89-02 LE-xrt (T1,T2,T3) 17.6 72 Features, see text
Table 34.3 Preoperative neoadjuvant radiation studies
Number of Pelvic recurrence Survival %,

Study patients Therapy arms %, 5 years 5 years Comments
Stockholm I (9) 849 S 28 36 No lateral beam
25 Gy-S 14 (p < 0.001) 36
Stockholm II (18) 557 S 25 39a Patients ≤80
25 Gy-S 12 (p < 0.001) 46a (p = 0.03)
Swedish (19) 1,168 S 27 48 Patients ≤80
Rectal 25 Gy-S 11 (p < 0.001) 58 (p < 0.001)
Dutch TME (17) 1,861 S 10.4 64 Patients ≤80
25 Gy-S 5.6 (p < 0.001) 64
Manchester (20) 284 S 36 39a
20 Gy-S (5 Gy x 4) 13 (p < 0.001) 46a (p = 0.03)
MRC II (21) 289 S 48 19 S 4 weeks
40 Gy-S (2 Gy x20) 32 (p = 0.04) 26 (p = 0.09) After external radiation
therapy
Polish (22) 312 25 Gy-S 9 67.2 Mostly TME (T3/T4)
50.4 Gy + C-S 14 66.2 Patients ≤75
EORTC (23) 1,011 45 Gy-S 17.1 63.2 two arms Patients ≤80
45 Gy + C-S 9.6 Without post-op C
45 Gy-S-C 8.7 67.2 two arms
45 Gy + C-S-C 7.6 With post-op C
(p = 0.12)
FFCD (24) 762 45 Gy-S-C 16.5 67.2 Patients ≤75
45 Gy + C-S-C 8.1 (p = 0.004) 66.2
a Surviving percent is for patients who underwent a potentially curative resection.
Table 34.4 Pre- versus postoperative radiation studies
Number of Pelvic recurrence %, Survival %,

Study patients Therapy arms 5 years 5 years Comments
Uppsala (25) 471 25.5 Gy-S 12 44 5.1 Gy × 5
S-60 Gy 21 (p = 0.02) 39 (p = 0.49) 2 Gy × 30
German (26) 823 50.4 Gy + C-S-C 6 74 TME
S-50.4 Gy + C-C 13 (p = 0.006) 76 (p = 0.8) Patients <75
(p < 0.001). About 28% had stage II disease. Among these, an extent that a small incidence of local recurrence in the
the local recurrence rate without radiation was 7.2% and surgery-only arm and its reduction by radiation treatment
with radiation 5.3% (p = 0.331). About 28% had stage I dis- has no statistically significant, or even discernible, impact
ease. Among these a local recurrence rate was 1.7% without on survival. This may be the principal explanation in the
radiation and 0.4% with (p = 0.091). Among 7% of patients Dutch TME trial.
with distant metastasis found at surgery (stage IV) there was The other way impact on survival of a local recurrence
local recurrence in 26.9% without radiation and with 15.9 advantage may be reduced, or lost, is if excess nonrectal
with (p = 0.207). Thus, for all four stages there was less local cancer death is produced in the radiation treatment arm.
recurrence in patients who had radiation, but the differen- This is likely the explanation for limitation of statistically
tial only reach statistical significance for the node-positive significant survival benefit to the subgroup that had cura-
(stage III) subgroup and the entire randomized population. tive surgery in the Stockholm II trial (18). At median follow-
Similarly, it was found that the difference reached statistical up of 8.8 years for this trial, 19% of the radiation arm and
significance in the subgroup that had LAR but not in sub- 12% of the surgery-only arm had died of noncancer causes
groups that had APR or Hartmann pouch surgery and in (p = 0.1). There was cardiovascular death in 13% of the
the subgroup for which the distal tumor edge was between 5 radiation arm and 7% in the surgery-only arm (p = 0.07).
and 10 cm from the anal verge, but not those more proximal This differential was established within the first 6 months
or distal. after surgery during which 5% of radiated patients and 1%
The Swedish rectal study differs from the Dutch study in of the surgery-only patients died from cardiovascular causes
that TME was not required (19). The proportion of patients (p = 0.02). The excess cardiovascular death was predomi-
in each stage was similar, but the differential in rate of local nantly in patients older than 68 years. It is suggested this is
recurrence between arms of the trial was greater and statis- due to change in the coagulation properties of blood dur-
tically significant for all stages. In the stage III subgroup of ing the several months of recovery from pelvic surgery that
the Swedish study, local recurrence was 40% without pre- has followed radiation treatment. This leads to increased
operative radiation and 20% with (p < 0.001). For stage II thrombotic events in the irradiated patients.
it was 23% without and 10% with radiation (p = 0.002). For The only randomized study of preoperative radiation
stage I it was 4% without and 2% with radiation (p = 0.02). with a surgery-only control arm that used a radiation treat-
Comparison of these two studies suggests that benefit ment regimen resembling the long course described above is
from preoperative radiation in preventing local recurrence the MRC II trial (21). Patients were eligible if they had a par-
is maximal if given to patients likely to have node-positive tially or totally fixed rectal tumor on physical examination.
(stage III) disease, expected to have LAR as opposed to APR, The population likely consisted mostly of T3 and T4 tumors.
and with lowest tumor extent in the middle to distal rec- There were likely more locally advanced cancers than in the
tum. However, some reduction in risk of local recurrence short-course trials. As shown in Table 34.3, there was a sig-
may be expected for all patients. nificant decrease in local recurrence in the radiation arm
As shown in Table 34.3, overall survival rate was not and a tendency toward increased survival, though not sta-
affected by the short-course preoperative radiation treat- tistically significant, similar to the findings in several short-
ment in the Dutch TME trial and in the earlier Stockholm I course trials.
trial. In the Swedish rectal trial, the short-course preopera- The Polish trial compares short-course preoperative radi-
tive radiation treatment produced a statistically significant ation with long-course preoperative radiation plus concur-
gain in overall survival. Two other short-course preopera- rent chemotherapy (22). Most of the surgery was with TME.
tive radiation trials, Stockholm II and Manchester, showed Patients were clinically staged with physical examination,
statistically significant improvement in overall survival transrectal ultrasound, and/or magnetic resonance imag-
among the subgroup that actually underwent curative ing. Only those with evidence of T3 or T4 tumors that were
resection but not in all randomized patients (18,20). palpable on digital examination and had no anal sphincter
Failure to improve overall survival even though local involvement were included. Patients found to have involved
recurrence rate is significantly reduced can occur in two nodes at surgery usually received postoperative chemother-
important ways. First, the dominant cause of death may apy. More in the short-course arm were node positive, sug-
be from development of distant metastatic disease to such gesting downstaging by the long-course treatment. There
Benefit of adjuvant and neoadjuvant radiation treatment 315
was no difference in survival between the two arms. There is from randomized trials and other comparative studies are
a suggestive difference in local recurrence favoring the short included. Overall survival, disease-free survival, cancer-
course, but it did not reach statistical significance. There specific survival, local recurrence-free survival, and metas-
was no statistically significant difference in the fraction that tasis-free survival are tabulated for the whole population
received a permanent stoma with a tendency to favor the and for subgroups defined by follow-up time, patient age,
long-course arm for sphincter preservation. BEDα/β, study time period, pathologic stage, radiation regi-
The EORTC trial examined the effect of adding chemo- men, anatomic location in rectum, surgical procedure, and
therapy to long-course preoperative radiation with a finding geographic origin of the included population. The hazard
that if chemotherapy is given concurrently with preop- ratio (HR) of death is less than or equal to one favoring
erative radiation, postoperatively, or both, the rate of local better overall survival with neoadjuvant treatment over
recurrence is reduced significantly relative to the preopera- surgery alone for all patients and subgroups. However, it
tive long-course radiation with no chemotherapy (23). This reaches statistical significance or near significance only
suggests concurrent radiochemotherapy does not contribute for treatment with long-course radiation with concurrent
much if postoperative chemotherapy is given. Alternatively, combination chemotherapy (HR = 0.54, p < 0.06) and for
the FFCD trial in which both arms got postoperative che- the American patient population treated with long-course
motherapy reports a significant decrease in local recurrence radiation and concurrent 5-fluorouricil chemotherapy
if concurrent chemotherapy is given with preoperative radi- (HR = 0.39, p < 0.01). Local recurrence-free survival favors
ation (24). There was no survival difference. neoadjuvant treatment and reaches statistical significance
Two randomized trials listed in Table 34.4 have directly for the population as a whole (HR = 0.63, p < 0.04) and for
compared pre- and postoperative radiation treatment arms. all the subgroups.
In the earlier Uppsala trial, the preoperative arm had the In all of the above trials, surgery consisted of LAR or
short course of radiation (25). Those randomized to the APR. For patients with evidence of stage T1 or T2 rectal
postoperative arm and found to have stage II or III disease cancer distal to the peritoneal reflection, usually within
were treated with long course to a higher dose of 60 Gy in 10 cm from the anal verge, smaller than about 4 cm and
2 Gy fractions. In the recent German trial, the surgery was occupying a limited fraction of the circumference of the
mandated to be with TME, and clinical staging was intended rectal wall, local excision via transanal, transsphincteric
to exclude stage I patients from the study (26). Those ran- (York-Mason), or posterior proctotomy (Kraske) procedure
domized to the long-course preoperative arm and the sub- may be able to achieve en bloc full-thickness excision of the
set of those randomized to the postoperative arm who were tumor with negative margins. This limited surgery may be
proved to have stage II or III disease at surgery, received elected in lieu of APR or LAR to preserve sphincter function
similar regimens of chemotherapy. The radiation courses or to avoid major surgery in those not fit or not willing to
were concurrent with 5-FU chemotherapy and consisted undergo it. Comparison of local excision (LE) with APR or
of 50.4 Gy in fractions of 1.8 Gy except that an additional LAR as to the ability to remove all the carcinoma has not
5.4 Gy to a reduced volume was included in the postop- been established by the randomized trial. Nevertheless, it is
erative treatment. Both of these trials showed statistically expected that limited local excision will not as reliably pre-
significant difference in local recurrence rate favoring the vent local recurrence as the more radical surgery, particu-
preoperative arm and no significant difference in survival larly TME. This is confirmed by the local recurrence rates
when grouped by intention to treat at randomization. It is of reported in the retrospective series shown in Table 34.2,
note that 28% of the postoperative arm of the German trial particularly for T2 disease (30,31). The decrease in local
received no radiation treatment. Of these, in 18% the cause recurrence with adjuvant radiation, with or without con-
was finding pathologic stage I disease, and in 10% the cause current chemotherapy, suggests that the local excision with
was postoperative death or complications, or the finding of adjuvant treatment is efficacious enough to be considered as
stage IV disease at surgery. Patient selection and the treat- an option under some circumstances. Bias in the retrospec-
ment regimen of the preoperative arm of the German trial tive studies would be to select for radiation treatment those
are now standard treatment in many institutions. patients with unfavorable features in the pathology, such as
Several reviews and meta-analyses of neoadjuvant and positive or close margins, lymphovascular invasion, or high
adjuvant radiation treatment of resectable, locally advanced histologic grade. Thus, the benefit from adjuvant treatment
rectal cancer have been compiled (27–29). They indicate that may be more than indicated by the results shown.
radiation approximately halves the rate of recurrence in the The RTOG protocol 89-02 study enrolled patients with
pelvis when given either preoperatively or postoperatively tumors judged by the surgeon to be distal enough to not
and gives a small increase in overall survival. allow clearance by LAR and who underwent local excision
Preoperative irradiation has become the favored way via transanal, transsacral, or transcoccygeal approach (30).
to include radiation in rectal cancer treatment. It is infor- To be eligible the tumor had to be mobile, less than 4 cm in
mative to examine the most recent of these meta-analyses, size and occupy less than 40% of the rectal circumference.
which compares preoperative irradiation, with and with- Those patients with cancer found to be pathologic stage
out concurrent chemotherapy, with surgery alone (29). TI, with histologic grade 1 or 2, excised with at least 3 mm
It is compiled from a database of 41,121 patients. Patients margin in all directions, absent any lymphatic or vascular
invasion and with normal CEA received no postoperative rate of local recurrence as the first event was approximately
treatment. Patients lacking any one of these favorable fea- 9% at 5 years among those who received chemotherapy pre-
tures were treated with radiation to the pelvis with boost to operatively, postoperatively, or both, and 17% in those who
the tumor site to a total dose of 50–56 Gy in 1.8–2 Gy frac- had no chemotherapy (p < 0.002). There was no statistically
tions with concurrent 5-FU chemotherapy. If the margin significant difference in overall survival. This suggests the
was microscopically positive or closer than 3 mm, the dose additional acute toxicity of preoperative concurrent radia-
to the tumor bed was increased to give a total dose of 59.4– tion and chemotherapy over that of preoperative radiation
66 Gy. The local recurrence rate for T2 tumors, all of which alone may not be necessary if postoperative chemotherapy
received adjuvant treatment, was 4 of 25 (16%) that for T3 is to be given. This is contradicted by the FFCD trial (24).
tumors was 3 of 13 (23%). It is not clear what the chance of The incidence of severe diarrhea during postoperative
salvage for local failure with APR is, but it may be as much radiation treatment following LAR or APR depends on the
as 50% (32). The results for local excision shown in Table specific concurrent chemotherapy regimen. For 656 patients
34.2 support the view that local excision with postopera- treated on the phase III NCCTG trial, it was found to be 13%
tive adjuvant treatment with radiation and chemotherapy, for bolus infusion of 5-FU at a dose of 500 mg/m2 on each
although not as likely to be curative as radical surgery, is an of 3 days of the first and fifth weeks. It was 23% for infu-
acceptable option for tumors of the size and position that sion of 5-FU at the rate of 225 mg/m2 per day given continu-
permit it, when there is sufficient reason to avoid radical ously for the entire length of the course of radiation (35).
surgery. The treatment of early rectal cancers has recently Improvement in survival at 4 years of 70% with a continu-
been reviewed (33). ous regimen compared to 60% with bolus infusion was felt
to justify the definite, though modest, increase in toxicity.
The type of surgery was also a significant determinant
of the risk of severe diarrhea. In those who had undergone
ACUTE ADVERSE EFFECTS LAR, there was a 31% rate of severe diarrhea compared to
13% in those who had an APR (p < 0.001). This differential
By acute adverse effect is meant one that develops during is not unexpected as there is a significant rate of diarrhea
a radiation course, or in the 1 or 2 weeks following it, and after LAR in the absence of radiation. In this regard, it is
which resolves within a month or two after the completion of note that the frequency of bowel movements at the time
of the course, without treatment, other than that to relieve of discharge after LAR via total mesorectal excision in 81
the temporary symptoms. The acute adverse effects that patients who were not treated with radiation averaged about
most often require a break from treatment are the perianal 8 per day (36).
skin reaction and diarrhea. A scale adopted by the RTOG In the trial that randomized patients to pre- versus post-
and EORTC for reporting acute effects of radiation of the operative long-course chemoradiotherapy conducted by the
lower GI tract is representative and in use in current trials German rectal cancer study group, the incidence of severe
(34). Grade 1 is given for increased frequency or change in diarrhea among 399 patients randomized to preoperative
bowel habits not requiring medication or rectal discomfort treatment was 12%. Among the 237 patients actually treated
not requiring analgesics. A score of grade 2 implies diar- with postoperative radiation, the rate of severe diarrhea was
rhea requiring Imodium or Lomotil medication, or mucus 18% (p = 0.04) (26). The postoperative arm included some
or bloody discharge not requiring sanitary pads, or rec- 23% that had APR. Thus, among those who had a LAR,
tal or abdominal pain requiring analgesic medication. A and are most comparable to patient’s preoperative arm
score of grade 3 is given for diarrhea requiring parenteral with respect to bowel function, the rate of severe diarrhea
support, mucous or bloody discharge requiring sanitary must have been greater than 18% and the differential in
pads, or abdominal distention with distended bowel loops favor of the preoperative treatment even greater. But if the
on radiograph. Grade 4 implies acute or subacute bowel 110 patients in the postoperative arm who, for one reason
obstruction, or fistula or perforation or GI bleeding requir- or another, had no radiation treatment are included in the
ing transfusion, or abdominal pain or tenesmus requiring toxicity score, there was no difference in rate of severe acute
tube decompression or bowel diversion. Grades 3 and 4 are grade 3 or grade 4 toxicity.
often combined and reported as severe adverse effects. Other grade 3 or 4 acute side effects reported in the
In the EORTC trial, 1,011 patients were treated with German study were hematologic and dermatologic. The
preoperative irradiation to a dose of 45 Gy in 25 fractions percent of grade 3 and 4 hematologic toxicity was 6% in the
over 5 weeks (23). Half were randomly assigned to also have preoperative and 8% in the postoperative arms (p = 0.27).
concurrent preoperative chemotherapy, and half had none. Dermatological toxicity refers to radiation dermatitis in the
Acute grade 2 toxicity was reported in 38.4% of those who perianal skin or perianal crease suture line (Figure 34.1).
received the concurrent preoperative chemotherapy and Grade 3 or 4 radiation dermatitis is reported for 11% of pre-
29.7% of those who did not (p < 0.001). Grade 3 or 4 acute operative and 15% of the postoperative patients who had
adverse effects were reported in 13.9% of those whose treat- radiation (p = 0.09). The rate of grade 3 or 4 acute toxic-
ment included preoperative chemotherapy and 7.4% of those ity of any kind was 27% in the preoperative and 40% in the
who had only preoperative for irradiation (p < 0.001). The postoperative patients who had radiation (p = 0.001).
Surgical complications after preoperative irradiation 317
systems were less frequent than those in the gastrointesti-

nal and neurologic systems.
SURGICAL COMPLICATIONS AFTER

PREOPERATIVE IRRADIATION
Patients treated preoperatively with short-course radio-

therapy in the Stockholm I trial had surgical mortality of
8% compared with 2% in the surgery-only arm (p < 0.01)
(9). Among patients over 75 years of age, the mortality in
the preoperative arm was 16% and again only 2% in the
surgery-only arm. The dominant cause of the increase
in postoperative death was cardiovascular. The radiation
treatment in Stockholm I was specified to be anterior-
posterior and posterior-anterior directed beams only, and
encompassed, in addition to the pelvis, the para-aortic
nodes cephalad to the L2 vertebral level. With the inclu-
sion of laterally directed beams and restriction of the radi-
Figure 34.1 Radiation dermatitis. ated volume to the pelvis as well as exclusion of elderly
patients in the subsequent Stockholm II, Swedish rectal,
and Dutch TME trials, the surgical mortality was not sta-
These results from two randomized studies support the tistically different between pre-op radiation and surgery-
conclusion that preoperative standard fractionated 5–6 only arms (17–19). For instance, in the Dutch TME trial
week radiation treatment with chemotherapy produces less the surgical mortality was 3.5% in the preoperative radia-
diarrhea and other acute adverse effects than in comparable tion arm and 2.6% in the surgery-only arm (p = 0.38) (17).
patients who have had the same treatment after surgery. The The in-hospital death rate was 4% in the pre-op radiation
differential is definitely present. However, it is a modest dif- arm and 3.3% in the surgery-only arm (p = 0.49) and very
ference so that, in itself, it does not provide a compelling strongly correlated with age in both arms. There was no
reason for preferring preoperative neoadjuvant treatment exclusion for age in the Dutch TME trial with the oldest
over postoperative treatment. Furthermore, 28% of patients patient age 92 years.
in the postoperative arm of the German study were spared In the Dutch TME trial, there was no significant differ-
radiation treatment because of the finding of stage I disease ence between the two arms in operating time (180 minutes)
(18%) or distant metastasis (10%) at surgery, and thus had or length of hospital stay (15 or 14 days median) (37). Median
zero adverse radiation effects. blood loss in the pre-op radiation arm was 1,100 mL and
The short preoperative radiation treatment course in the surgery-only arm was 1,000 mL. The percent of LAR
of five fractions of 5 Gy each in 1 week does not usually patients with a diverting stoma increased from 60% to 67%
produce significant adverse effects in the 2–3 weeks dur- in the 60 days following surgery. In the surgery-only arm it
ing the radiation treatment and before surgical resection. increased from 54% to 63% (p = 0.17). A statistically signifi-
In the Dutch TME trial, grade 1 acute gastrointestinal side cant difference in postoperative complications between the
effects were reported in 12%, grade 2 was reported in 2.3%, arms was found for cardiac events, 5% with pre-op radiation
and grade 3 in 1 of 605 patients (17). Acute neurologic and 3% surgery only (p < 0.05); psychologic disorders, 4%
effects of radiation were reported as grade 1 (requiring no with pre-op radiation and 1% surgery only (p < 0.01); and
intervention) in 7.5%, as grade 2 (requiring narcotic pain for any complication, 48% in pre-op radiation arm and 41%
medicine or adjustment of treatment) in 1%, and grade 3 surgery only (p < 0.01). Complications in the APR patients
(intractable severe pain or causing treatment interruption) occurred in 29% of the radiated patients and 18% of sur-
in 2.8%. This has been attributed to radiation-induced lum- gery-only patients (p < 0.01). There was no significant dif-
bosacral plexopathy. It was first reported in patients treated ference in complication rate among LAR patients, 11% and
with a short course in the Uppsala and Swedish rectal trial 12%, respectively, in the radiation and surgery-only arms.
(19,25). It consists of pain in the lower extremities and but- These results indicate that there is the potential for short-
tocks and in a minority of the patients was associated with course preoperative radiation to complicate the ensuing
other lower extremity neurologic signs. In a few patients, surgery and recovery particularly manifest in patients over
the effect persisted or recurred for months to years. Acute the age of 70 and even more so in those over the age of 80.
neurologic effects have not been reported with the lower This is minimized but not eliminated by adherence to the
fractional doses of the long-course preoperative radiation now standard radiation treatment planning specifications
treatment. Acute effects on the genitourinary and other noted in the introduction.
The German trial required TME surgery and excluded

patients over the age of 75. The radiation treatment was the CHRONIC LATE ADVERSE EFFECTS
long course (50.4 Gy in 28 fractions of 1.8 Gy each) with con- OF ADJUVANT AND NEOADJUVANT
current chemotherapy and was given either preoperatively or IRRADIATION
postoperatively (26,38). There was 0.8% surgical mortality in
the preoperative arm and 1% in the postoperative arm indi- By late adverse effect is meant one that persists or develops
cating no increase attributable to the preoperative radioche- 4–6 months or more after the last radiation treatment and
motherapy. The incidence of any postoperative complication becomes a chronic injury. Examples are frequent bowel
was 34.5% in the preoperative arm and 34% in the postop- movements, rectal bleeding, incontinence, constipation,
erative arm. Anastomotic leak occurred in 13% and 12%, and bowel obstruction. The total dose, fractional dose, and
delayed wound healing in 5% and 6% of, respectively, pre- time span of the long and short courses of adjuvant and
and postoperative arms. All other complications occurred in neoadjuvant radiation treatment have been established so
less than 3% of each arm with no significant difference. that chronic injury that significantly affects quality of life
The Polish trial randomized patients between preopera- is unusual, and one that is life altering is rare. Chronic
tive short-course radiation and long-course radiation with injury may be more likely in patients who have had severe
chemotherapy (22,39). Surgery was by TME for the more acute reactions to radiation, termed consequential chronic
distal tumors, and patients over age 75 were excluded. The injury, but may also occur in those who experienced little
overall rate of complication events was 31% in the short- or no acute effect. It also may have increased likelihood in
course arm and 22% in the long-course arm (p = 0.06) show- patients with diabetes.
ing a near significant trend. The overall number of patients Chronic injury occurs as a result of depletion of the pop-
suffering a complication was 27% of the short-course and ulation of stem cells responsible for maintenance of tissue
21% in the long-course arm (p = 0.27). Postoperative death integrity and organ function due to radiation-induced cell
occurred in 0.7% of the long-course and 1.3% of the short- reproductive death. For instance, the death of the stem cells
course arms (p = 1.0). Reoperation was needed in 8.2% responsible for maintaining the microvascular circulation
of the short-course and 9.5% of the long-course patients leads to occlusion of arterioles that is the distinguishing
(p = 0.85). No statistically significant difference and no feature of the histology of irradiated tissue examined more
suggestive trend were found to favor one or the other arm than a few months after radiation exposure. Over ensuing
with respect to other less severe complications. years, the impaired microcirculation may cause progressive
Postoperative complications after preoperative irra- ischemic injury to the exposed tissue manifest principally
diation examined in the meta-analysis (29) include wound as telangectasia, fibrosis, and defective healing. These effects
complications, anastomotic leakage, bowel obstruction, and may be associated with bleeding; incontinence; stenosis of
mortality. Notable findings expressed as a hazard ratio are as bowel, ureter, or urethra; and bowel obstruction. They may
follows. Preoperative short-course radiation versus surgery complicate surgery on irradiated tissue several months to
alone had statistically significant unfavorable outcome for many years after the radiation treatment.
wound complications (HR = 1.65, p < 0.01), over all post- Several long-term quality of life outcomes are reported in
operative morbidity (HR = 1.45, p < 0.01) and postopera- the meta-analysis (29). These are expressed as an odds ratio
tive mortality (HR = 2.15, p < 0.01). The mortality increase (OR) of having the unfavorable complication with preop-
was predominantly from thrombotic vascular events such erative radiation treatment compared to surgery alone. The
as pulmonary embolus, myocardial infarction, and stroke. OR reached statistical significance for use of an absorbent
The long-course preoperative treatment with concurrent pad (OR = 2.4, p < 0.01), incontinence of gas (OR = 1.62,
chemotherapy showed statistically significant increase in p < 0.04), incontinence of stool (OR = 1.81, p < 0.01), fecal
wound complications (HR = 1.52, p < 0.02) and anasto- urgency (OR = 2.4, p < 0.03), and male erectile dysfunc-
motic leakage (HR = 1.22, p < 0.05). For patient population (OR = 2.25, p < 0.01). It did not reach statistical signif-
tions with median age less than 65 years, none of the post-op icance for constipation, diarrhea, or urinary incontinence.
complications showed a statistically significant increase for Patients enrolled in the Dutch TME preoperative short-
either long- or short-course radiation treatment. course radiation trial who are alive with no evident disease
In conclusion, the studies shown in Tables 34.2, 34.3, were sent a questionnaire by mail to assess bowel, stoma,
and 34.4, and the meta-analysis indicate that inclusion of and urinary function (40). A response was obtained from
radiation in preoperative neoadjuvant treatment of rectal 597 (84% of those mailed). Among these the median time
cancer is associated with a statistically significant increase since surgery was 5.09 years. The mean number of bowel
in postoperative complications in elderly patients but not movements during the day among the 362 patients who had
in younger patients. The dividing age between the elderly no stoma was 3.69 in the irradiated patients and 3.02 in the
and the young is somewhere in the 70s. The most important surgery-only patients (p = 0.011). The mean number of noc-
increase is in postoperative mortality, which is found fol- turnal movements was 0.48 in the irradiated patients and
lowing the short-course preoperative treatment. It does not 0.35 in the surgery-only patients (p = 0.207). Daytime fecal
reach statistical significance with the long course. incontinence was reported as 62% of those irradiated and
Chronic late adverse effects of adjuvant and neoadjuvant irradiation 319
38% of the surgery-only patients (p < 0.001), and noctur- A comparative study by phone interview of patients 2 or
nal incontinence in, respectively, 32% and 17% (p = 0.001). more years after they had undergone LAR for rectal cancer
The incontinence also occurred more often and was more at Mayo Clinic reports significantly more bowel symptoms
troublesome in the irradiated compared to surgery-only in the 41 who had also had postoperative long-course pel-
patients. Pads were in use for incontinence and anal mucous vic irradiation and chemotherapy than in the 59 who had
and blood loss in 56% irradiated and in 33% of surgery- only surgery (42). The fraction having more than five bowel
only patients (p < 0.001). Among 235 responding patients movements a day was 37% in the irradiated group and
with a stoma, there was no significant difference between 14% in the surgery-only group (p < 0.001). The fraction of
irradiated and surgery-only patients with respect to stoma patients who reported incontinence was 66% in the irradi-
function. ated group and 7% in the surgery-only group (p < 0.001).
A review of the patients treated on the Dutch TME trial In the irradiated group 41% wore a pad, and in the surgery-
was conducted to determine risk factors for development only group, 10% (p < 0.001). Urgency with the inability to
of fecal incontinence (41). Potential risk factors examined defer defecation for 15 minutes was reported in 78% of the
included age, gender, childbirth, body mass index, cancer irradiated and 19% of the surgery-only patients (p < 0.001).
stage, tumor distance from anal verge, anastomosis distance A retrospective study of 192 patients who had LAR
from anal verge, duration of surgery, blood loss at surgery, with coloanal anastomosis at the Mayo Clinic and had
presence of a pouch, temporary stoma, and an anastomotic preoperative long-course radiation, postoperative radia-
leak. No risk factors emerged as statistically significant among tion, or no radiation reports an anastomotic stricture was
the surgery-only patients. Among the preoperative radiation the most common effect requiring surgical intervention.
patients, only blood loss at surgery and distal tumor mar- This occurred with nearly the same frequency in all three
gin distance from the anal verge were statistically significant groups: 16% no radiation, 14% pre-op radiation, and 15%
risk factors. Blood loss of surgery greater than 1,400 mL had post-op radiation. It was usually managed with dilation and
relative risk (RR) of incontinence of 3.24 (p = 0.005) com- was not a significant cause of permanent fecal diversion.
pared to those with less blood loss. Compared to patients Permanent fecal diversion resulted from recurrence, bowel
with distance of distal tumor margin less than 5 cm from obstruction, incontinence, fistula, stricture, abscess/leak,
the anal verge, a distance between 5 and 10 cm had RR of and patient preference. The 5-year survival without colos-
0.21 (p = 0.016), and a distance greater than 10 cm had RR tomy was 92% in patients who had no radiation treatment
of 0.13 (p < 0.003). The location of distal tumor extent deter- and 72% in those who did (p < 0.001). There was no signifi-
mines the inferior extent of the radiation treatment port. cant difference between the rates in pre- and postoperative
Among those few respondents who had the perineum and irradiated patients.
consequently the entire anal sphincter included in the radia- A scale adopted by the RTOG and EROTC for report-
tion field compared to those who did not, the RR for fecal ing late chronic effects of radiation on the bowel is as fol-
incontinence at 2 years after surgery was 2.64 (p = 0.085) lows (34). Grade 1 implies mild diarrhea, mild cramping,
and at 5 years after surgery the RR was 7.45 (p = 0.059). It less than five movements per day, and slight rectal discharge
was also noted that the fraction of patients reporting fecal or bleeding. Grade 2 implies moderate diarrhea and colic,
incontinence increased after reaching a minimum at 2 years more than five bowel movements per day, excessive mucus,
postsurgery, whereas that in surgery-only patients increased or intermittent bleeding. Grade 3 implies obstruction or
only slightly. This time course is consistent with a late effect bleeding requiring surgery. Grade 4 implies necrosis per-
of radiation on pelvic nerves and fibrosis. foration or fistula. Fecal incontinence was not explicitly
Urinary function was not significantly different in irra- included in the grading criteria.
diated and surgery-only patients. About 39% of patients in The German trial reports grades 3 and 4 long-term gas-
each group reported incontinence of urine. Back and but- trointestinal effects, for example, diarrhea and small bowel
tock pain, hip stiffness, and difficulty walking were not sig- obstruction, in 9% of the pre-op group and 15% of the post-
nificantly different in the two groups, suggesting an absence op group (p = 0.07); anastomotic stricture in 4% of the pre-
of chronic radiation-induced lumbosacral plexopathy in op and 12% of the post-op arms (p = 0.003) (26,38). Bladder
this trial. dysfunction of grade 3 or 4 occurred in 2% of the pre-op
The rate of hospital admission was significantly increased and 4% of the post-op arms (p = 0.21). Any grade 3 or 4
in the irradiated patients compared with surgery-only effect occurred in 14% of the pre-op and 24% of the post-op
patients in the first 6 months after surgery. Admissions were patients (p = 0.01). With the long-course fractionation of
for infection, endocrine, cardiovascular, and gastrointesti- pelvic chemoradiotherapy for adjunctive treatment of rectal
nal diagnoses. Of note among gastrointestinal admissions, cancer, the preoperative irradiation appeared significantly
those for constipation and abdominal pain are significantly less likely to produce severe, chronic, long-term sequelae
increased in the radiated patients, but those for bowel then postoperative radiation.
obstruction were not. The rate of hospital admission more The Polish trial comparing short-course preoperative
than 6 months after surgery was not significantly different radiation with long-course preoperative radiochemother-
for patients in the two groups, including for myocardial apy at median follow-up of 48 months reports the over-
infarction or stroke. all incidence of late toxicity is 28.3% in the short-course
and 27% in the long-course arms (p = 0.81)(43). The inci-

dence of severe late toxicity, presumably grade 3 or 4, was
10.1% in the short-course and 7.1% in the long-course
arms (p = 0.36). Severe gastrointestinal toxicity occurred
in 5.1% of the short-course and 1.4% of the long-course
patients, no P value given. Quality of life questionnaire on
anorectal function including questions on bowel function,
incontinence, and urgency reports no significant difference
between short- and long-course arms (43). For instance, 39%
and 41% of, respectively, the short-course and long-course
patients reported use of pads. In answering the question,
“did your health status and/or treatment cause your sexual
life to decline,” there also was no significant difference in
the two arms. This direct comparison with long- and short-
course preoperative treatment shows no statistically signifi-
cant difference in late toxicity.
The evidence from several trials summarized here indi- Figure 34.2 Radiation proctitis.
cates that both preoperative and postoperative radiation
treatment are associated with increased chance of chronic oral sulfasalazine plus rectal steroids to rectal sucralfate
adverse effect on bowel function. The direct comparison of and oral placebo. The sulfasalazine regimen demonstrated
pre- and postoperative long-course radiochemotherapy in a significant improvement in both clinical symptoms and
the German trial indicates there is less likelihood of this endoscopic findings; however, by comparison clinically this
with the preoperative treatment. The Polish trial comparing was less effective than sucralfate (47). One randomized,
long- and short-course preoperative radiation finds no clear controlled trial found that oral sucralfate decreased diar-
difference. rhea symptoms in both the acute and chronic phases (48).
Short-chain fatty acids act as a major fuel source for rectal
mucosa. Two small randomized placebo controlled trials
using short-chain fatty acid enemas noted improvement in
RECTAL INJURY AND ITS TREATMENT symptoms and endoscopic findings (49,50).
Various endoscopic ablation therapies have been applied
A part of the rectum or the nearby sigmoid colon is sometimes to the treatment of colonic proctitis–related bleeding due
exposed to radiation dose as great as 70–80 Gy with BEDα/β to local telangiectasias. The two most commonly utilized
greater than 100 Gy in the nonsurgical radiation treatment approaches are the laser and the argon plasma coagulator.
of cervical, endometrial, or prostate cancer. As many as 20% There are no prospective randomized trials assessing either
of these patients may develop symptoms of chronic proctitis of these approaches, only several retrospective case series.
with rectal bleeding, urgency, pain, and tenesmus. Some may The largest series reporting on the use of Nd:YAG laser
develop fistulas, strictures, and incontinence (44). It should found excellent response rates and a significant decrease
not be assumed that these symptoms and findings represent in rectal bleeding (51). Rare complications include mucus
radiation proctitis without a thorough evaluation, as recur- discharge, ulcers, or stricture. Similar results were obtained
rent cancer may cause the same symptoms. using an argon plasma coagulator in three treatment ses-
Endoscopy reveals friability, pallor, erythema, or promi- sions (52). However, over 70% required maintenance treat-
nent mucosal telangiectasias (Figure 34.2) (45). Histologic ment over the long term (53).
findings in the chronic phase include vascular changes such For treatment of bleeding related to chronic proctitis,
as the telangiectatic capillaries, platelet thrombi formation, 4% and 10% formalin have been utilized. Two approaches
and lumen of arterioles filled in with fibrin. These changes are commonly utilized: rectal formalin irrigation and dab
are accompanied by fibrosis in the lamina propria and crypt technique utilizing topical application of formalin with
distortion (46). swabs or soaked gauze (54,55). De Paradies et al. reported
Numerous therapeutic agents have been evaluated and a prospective case series using the formalin gauze applica-
are currently utilized against radiation-induced proctitis. In tion and noted a beneficial result in 70% (56). However, sig-
many cases, patients presenting initially with symptoms sug- nificant rates of stricturing and incontinence were reported.
gestive of radiation proctitis will first be offered treatment Numerous other retrospective series have reported good
with anti-inflammatory medications. This most commonly success with formalin. Of those using of gauze or pledget-
involves either oral or enema-delivered steroids or various mediated application, at least a 75% success rate for cessa-
5-amino salicylic acid preparations. Though often utilized tion or improvement in bleeding was reported (57). Many
in both acute and chronic settings, evidence is lacking for required multiple treatments, though complications were
the use of steroid preparations in the treatment of radia- minimal. Due to the small volume used, 10% formalin
tion proctitis. A prospective randomized trial compared is often used. Of those reporting use of formalin rectal
References 321
irrigations, 50 cc aliquots of 4% formalin were utilized up to surgery is eliminated or minimized by restricting the irra-
a total volume of 400–500 cc. Again, >75% success rate was diated volume to those parts of the pelvis at risk for harbor-
noted with this approach, with the most common reported ing disease, possibly by using intensity modulated radiation
complication being anal or pelvic pain occurring in 25% of therapy, and by avoiding the short-course treatment for
those treated (58). patients with age greater than about 70.
There is low-level evidence supporting the use of hyper- The Polish trial, which directly compared short-course
baric oxygen treatments for chronic radiation proctitis with long-course preoperative irradiation, was inconclusive
and a single prospective series, which reported significant (22). The German trial has provided evidence that preopera-
improvement of bleeding, diarrhea, and urgency, but no tive long-course chemoradiation is in balance preferable to
change in rectal pain with oral vitamins E and C (59,60). the similar treatment postoperatively (26,38). The selection
Metronidazole along with anti-inflammatory agents (oral criteria and preoperative treatment arm of the German trial
mesalazine and betamethasone enema) produced a signifi- have become the standard in many treatment centers. The
cantly lower incidence of rectal bleeding and diarrhea in long course can be expected to produce some downstaging
chronic radiation proctitis (61). and facilitate R0 resection.
Despite the numerous medical approaches available for For younger patients, the short-course preoperative
the treatment of radiation proctitis, surgical therapy remains radiation treatment may be considered for several reasons.
an option for refractory cases. The indications for surgery It has a theoretical advantage of more timely removal of all
are most commonly rectum or rectosigmoid stenosis and evident disease than the long course. It has better compli-
rectovaginal fistula, while the most common presenting ance and is more economical. The surgical staging is undis-
symptoms are rectal bleeding, diarrhea, or tenesmus (61). torted by downstaging from the long course of treatment,
The majority of patients undergo diversionary procedures and this may affect the recommendation for adjuvant che-
(proctectomy with colostomy, with or without a Hartmann motherapy. For instance, patients found to have no lymph
rectal stump) with resection performed less commonly. node metastases may be spared adjuvant chemotherapy. But
When continuity is restorative, coloanal anastomosis (with chemotherapy with the full adjuvant regimen, as opposed
or without colonic J-pouch) with proximal covering stoma to 5-FU alone, can begin earlier with the short course than
is the procedure of choice in selected cases. Successful out- with the long course. The high fractional dose of the short
comes with diversion alone are reported in the range of course has the theoretical disadvantage of greater risk of
70%–73% (62). In refractory rectal bleeding, this option has chronic injury, particularly that related to pelvic nerves that
less morbidity. Overall, morbidity with surgical intervention is associated with neuropathic pain, bowel and bladder con-
is extremely high, ranging from 30% to 65% with mortality tinence, and sexual function.
rates in the postoperative period reported at 6.7%–25% (62).
REFERENCES
1. Markovina S et al. Int J Radiat Oncol Biol Phys 2017;
CONCLUSION 99:417–26.
2. Bisscchop C et al. Ann Surg Oncol. 2017;24:2632–8.
Chemotherapy and radiation treatment to the pelvis as an 3. Barendsen GW. Int J Radiat Oncol Biol Phys. 1982;
adjuvant to surgical resection, either individually or when 8:1981–97.
both are administered, reduces the chance of pelvic recur- 4. Jones B et al. Clin. Oncol. 2001;13:71–81.
rence and increases by a small increment the chance of the 5. Bentzen SM, Joiner MC. 2009. The linear quadratic
patient surviving the disease. This has been demonstrated approach in clinical practice. In: Joiner M, van der
in several randomized trials for both the pre- and postop- Kogel A (eds) Basic Clinical Radiobiology, 4th edition,
erative treatment sequences, as noted in the tables and in Hodder Arnold. London, 120–134.
meta-analyses (27–29). 6. Kian Ang K et al. Altered fractionation schedules. In:
That preoperative treatment with radiation can compli- Brady LW and Perez CA (eds) Principles and Practice
cate ensuing surgery and postoperative recovery is illus- of Radiation Oncology 2nd edition. Lipincott-Raven,
trated in the occurrence of additional noncancer, mostly Philadelphia New York, 119–142. 1998.
cardiovascular, death among irradiated patients in the 7. Marijnen CAM et al. J Clin Oncol. 2001;19:1976–84.
immediate postoperative period and in the first 6 months 8. Francois Y et al. J Clin Oncol. 1991;17:2396.
postsurgery in the Stockholm trials. A similar adverse effect 9. Cedarmark B et al. Cancer. 1995;75:2269–75.
was not evident in the later Swedish and Dutch TME tri- 10. Walz BJ et al. Int J Radiat Oncol Biol Phys. 1981; 7:
als that also used the short-course radiation regimen, and it 477–84.
was not evident in the studies that used the long-course pre- 11. Rich T et al. Cancer 1983;52:1317–29.
operative treatment regimens. These later trials were with 12. Regnier A et al. Frontiers Oncol 2017;7:225.
better radiation therapy technique, and all but the Dutch 13. Gastrointestinal Tumor Study Group. New Engl J Med.
TME trial excluded the most elderly patients. The potential 1985;312:1465–72.
for serious adverse effect of preoperative irradiation on the 14. Krook JE et al. New Engl J Med. 1991;324:709–15.
15. Fisher B et al. J Natl Cancer Inst. 1988;80:21–9. 39. Bujko K et al. Colorectal Dis. 2005;7:410–6.
16. NIH consensus conference. JAMA. 1990;264: 40. Peeters KCMJ et al. J Clin Oncol. 2005;23:6199–206.
1444–50. 41. Lange MM et al. Br J Surg. 2007;94:1278–84.
17. Koen CMJ et al. Ann Surg. 2007;246:693–701. 42. Kollmorgen CF et al. Ann Surg. 1994;220:676–82.
18. Martling A et al. Cancer 2001;92:896–902. 43. Pietrzak L et al. Radiother Oncol. 2007;84:217–25.
19. New Engl J Med. 1997;336:980–7. 44. Hayne D et al. Br J Surg. 2001;88:1037–48.
20. Marsh PJ et al. Dis Colon Rectum. 1994;37:1205–14. 45. Reichelderfer M, Morrissey JF. Gastrointest Endosc.
21. Lancet 1996;348:1605–10. 1980;26:41–3.
22. Bujko K et al. Brit J Surg. 2006;93:1215–23. 46. Haboubi NY et al. Am J Gastroenterol. 1988;83:
23. Bosset J et al. New Engl J Med. 2006;355:1114–23. 1140–4.
24. Gerard J et al. J Clin Oncol 2006;24:4620–5. 47. Kochlar R et al. Dig Dis Sci. 1991;36:103–7.
25. Fyrkholn GJ et al. Dis Colon Rectum. 1993;36: 48. Henriksson R et al. Scand J Gastroenterol. 1992;191:
564–72. 7–11.
26. Saur R et al. New Engl J Med. 2004;351:1731–40. 49. Pinto A et al. Dis Colon Rectum. 1999;42:788–95.
27. Colorectal Cancer Collaborative Group. Lancet. 2001; 50. Kennedy GD, Heise CP. Clin Colon Rectal Surg. 2007;
358:1291–1304. 20:64–72.
28. Glimeliius B J et al. Acta Oncologica. 2003;42: 51. Vigiano TR et al. Gastrointest Endosc. 1993;39:513–7.
476–92. 52. Tam W et al. Endoscopy. 2000;32:667–72.
29. Bin Ma et al. Int J Cancer. 141:1052–65. 53. Taylor JG et al. Gastrointest Endosc. 1993;39:641–4.
30. Russell AH et al. Int J Radiat Oncol Biol Phys. 2000; 54. de PV et al. Dis Colon Rectum. 2005;48:1535–41.
46:313–22. 55. Parik S et al. Dis Colon Rectum. 2003;46:596–600.
31. Chakravarti A et al. Ann Surg. 1999;230:49–54. 56. Luna-Perez P, Rodriguez-Ramirez SE. J Surg Oncol.
32. Sharma A, Hartley, Monson JR. Surg Oncol. 2003;12: 2002;80:41–4.
51–61. 57. Dall’era MA et al. J Urol. 2006;176:87–90.
33. Tytherleigh MG et al. Brit J Surg. 2008;95:409–23. 58. Kennedy M et al. Am J Gastroenterol. 2001;96:
34. Cox JD et al. Int J Radiat Oncol Biol Phys. 1995;31: 1080–4.
1341–6. 59. Cavcic J et al. Croat Med J. 2000;41:314–8.
35. Miller RC et al. Int J Radiat Oncol Biol Phys. 2002;54: 60. Pricolo VE, Shellito PC. Dis Colon Rectum. 1994;37:
409–13. 675–84.
36. McAnena OJ et al. Surg Gynoecol Obstet. 1990;170: 61. Anseline PF et al. Ann Surg. 1981;194:716–24.
517–21. 62. Camma C et al. JAMA. 2000;284:1008–15.
37. Marijnen CAM et al. J Clin Oncol. 2002;20:817–24. 63. Tveit KM et al. Br J Surg. 1997;84:1130–5.
38. Sauer R et al. Strahlentherapie Oncol. 2001;177: 64. Wolmark N et al. J Natl Cancer Inst. 2000;92:
173–81. 388–96.
35
Surgical management of ulcerative colitis
SHANNON McCHESNEY AND BRIAN R. KANN
Of these patients, up to 40% will require colectomy at

CHALLENGING CASE some point in their lifetime, despite current medical ther-
apies (4). This chapter explores the indications, methods,
A 45-year-old man with a 20-year history of ulcerative and outcomes of surgery in UC.
colitis (UC) is presenting with perineal pain, tenesmus,
bloody stools, and incontinence. He has a history of
restorative proctocolectomy with ileal pouch anal
anastomosis (IPAA) performed 10 years prior. INDICATIONS FOR SURGERY
URGENT/EMERGENT
CASE MANAGEMENT
The timing of surgery is dependent on severity of disease.
Pouchitis is a complication frequently seen in UC Truelove and Witts classification system is commonly used
patients following restorative proctocolectomy with to categorize mild, severe, and fulminant disease, with the
IPAA. Diagnosis of this condition involved clinical latter described as greater than 10 continuously bloody
assessment, as well as endoscopic examination via stools per day, fever, pulse greater than 90, anemia requir-
pouchoscopy, to confirm mucosal inflammation, ing transfusion, erythrocyte sedimentation rate greater than
and biopsy. Treatment is primarily medical, consist- 30, colonic dilatation, and abdominal distension and tender-
ing of antibiotics, probiotics, and steroids; however, ness (5) (Table 35.1). Since the classification was published
in refractory cases surgical intervention may be in 1955, describing the benefit of steroids for acute colitis,
required. Surgical options include pouch excision with corticosteroids have been implemented for controlling active
diverting ileostomy or redo IPAA. disease, which has resulted in reduced mortality (6).
Urgent or emergent surgery is indicated for patients with
acute toxic colitis and associated complications thereof,
including perforation, hemorrhage, and obstruction. It is also
INTRODUCTION indicated in patients with toxic megacolon, sepsis, or fulmi-
nant disease unresponsive to maximum medical therapy.
UC is a chronic inflammatory condition predominantly Urgent or emergent colectomy for acute colitis is associ-
involving the colon and rectum, characterized by recurrent ated with considerable morbidity and mortality. A systemic
inflammatory episodes. This inflammatory bowel disease review of 29 studies comprising 2,714 patients demonstrated
may present with abdominal pain, rectal bleeding, diar- 50.8% morbidity and 8% in-hospital mortality (7).
rhea, and weight loss. Bowel involvement can vary, from Total abdominal colectomy with end ileostomy is most
isolated rectal disease to pancolitis. Extraintestinal mani- commonly performed in emergent surgery, and recommended
festations may present as well. Diagnosis is most common (8). The colon is divided at the level of the sacral promontory,
in late adolescence or early adulthood but occurs at all ages leaving the rectum in place to safely avoid pelvic dissection
(1). An estimated 3 million U.S. adults have ever received during the period of acute inflammation. The rectal stump
a diagnosis of inflammatory bowel disease (2), with up to should be kept long and may be left in place as a Hartmann’s
1.4 million people in the United States currently living with pouch, exteriorized as a mucous fistula, or placed in an extra-
the disease (1). The prevalence of UC is higher in adults fascial position. Completion proctectomy can be planned for a
than children; 238 versus 28 per 100,000, respectively (3). later date following resolution of acute inflammation.
323
324 Surgical management of ulcerative colitis
Table 35.1 Truelove and witts classification system
Mild Severe Fulminant

Bowel movements/day <4 >6 >10
Rectal bleeding Intermittent Frequent Continuous
Hemoglobin Normal <75% of normal Requiring transfusion
Erythrocyte sedimentation rate (mm/h) <30 >30 >30
Temperature (C) Normal >37.5 >37.5
Heart rate Normal >90 >90
Colonic features on x-ray Thumbprinting, edema Dilatation
Clinical signs Abdominal tenderness Abdominal tenderness and distension
ELECTIVE PROCTOCOLECTOMY WITH BROOKE

ILEOSTOMY
There are several indications for elective colectomy in UC,
including failure of medical management, which is most The proctocolectomy option followed by Brooke ileostomy
common, as well as intolerable medication complication or is a one-stage operation that was originated by Brooke in
side effects, stricturing disease, dysplasia, invasive cancer, 1952. This technique was previously considered the “gold
and growth retardation, the latter more frequently seen in standard” and involves removal of colon and rectum and
the pediatric population. Severe extraintestinal manifesta- formation of end ileostomy. The procedure is considered
tions may be an indication for colectomy; uveitis, monoar- safe, effective, and curative of UC. While this option may
ticular arthritis, and iritis are observed to improve following avoid certain complications associated with anastomoses or
colon resection (9). pouch formation, it does entail a permanent ostomy, which
Colonic strictures develop in 5%–10% of UC patients may be undesirable to patients.
and can pose a risk of large bowel obstruction. While the
majority are benign, as many as 25% will be malignant, with Indications
30% of cancers occurring in UC presenting as such (8). For
this reason, all UC patients with strictures should undergo Proctocolectomy with Brooke ileostomy is indicated in patients
oncologic resection (8). Strictures developing in long-stand- with UC requiring colectomy who are poor candidates for a
ing disease, causing obstruction, or located proximal to the more extensive restorative procedure, such as those with pre-
splenic flexure are more likely to be malignant (10). operative incontinence or perianal disease, those with a distal
Patients with UC are at increased risk of developing rectal cancer, or for those desiring a single-stage operation.
colorectal cancer: 2% at 10 years, 8% at 20 years, and 18%
at 30 years in patients with long-standing disease (11), Technique
with risk factors for malignancy including disease dura-
tion and extent, family history of colon cancer, and con- Prior to operation, it is essential that the patient meet pre-
current primary sclerosing cholangitis (12). Surveillance operatively with an enterostomal therapist to select the ideal
colonoscopy and biopsy are recommended. Patients with ostomy placement. The patient should be assessed in different
extensive disease should undergo colonoscopy after 8 positions, and the site selected should be away from any folds,
years of disease, with surveillance colonoscopy every 1–2 scars, or bony prominences. The patient should also be pre-
years thereafter (8). pared for the operation with mechanical bowel preparation.
A finding of high-grade dysplasia, dyplasia-associated Preoperative antibiotics should be administered in the
lesion or mass, or invasive cancer on colonoscopy is an operating room and the patient placed in lithotomy posi-
absolute indication for surgery (8). The indication for sur- tion. The proctocolectomy may be performed via open or
gery in low-grade dysplasia is debated, as studies are con- laparoscopic approach. Full mobilization from terminal
flicting. The current practice parameters recommend either ileum to rectosigmoid junction is performed, followed by
total proctocolectomy or surveillance endoscopy (8). pelvic dissection to elevate the rectum along a total meso-
rectal excision plane, taking care to identify and avoid the
ureters and the hypogastric nerves. The dissection is car-
ried down to the levators, at which time attention may be
SURGICAL OPTIONS turned to the perineum. For the perineal approach, the skin
is incised circumferentially around the anus, and dissection
Surgical options for UC include a proctocolectomy with carried down until the peritoneum is entered.
Brooke ileostomy, proctocolectomy with continent ileos- Following removal of the colon and rectum, the ileos-
tomy, total abdominal colectomy (TAC) with ileorectal anas- tomy is created. The ileostomy should be placed within
tomosis, and restorative proctocolectomy (RP) with IPAA. the rectus muscle over a relatively flat area, and everted to
Surgical options / Proctocolectomy with continent ileostomy 325
minimize effluent contacting the skin and optimizing the fit (a) (c)
of ostomy bag appliance.
Complications/outcomes
While ileostomy may avoid complications related to pouch
creation, proctocolectomy has associated complications to Nipple valve
inside pouch
be anticipated.
Perineal wound complications may arise, which should
alert the surgeon to ascertain the presence of retained
mucosa, foreign body, or Crohn disease. Management of
this issue ranges from simple wound care to debridement
and flap closure. Additional complications include sexual
dysfunction such as impotence, retrograde ejaculation aris-
ing from autonomic nerve injury, and dyspareunia and
infertility due to scar tissue formation (13). Overall, the
most common long-term complications are ostomy related,
such as skin excoriations, parastomal hernia, stomal pro- (b)
Mesentery Staples
lapse, necrosis, retraction, or stenosis, all of which have
potential to require ileostomy revision. Considering that
patients with ileostomy are more prone to dehydration and
related sequelae such as electrolyte derangements and kid- Mucosa
ney stone formation, it is important to educate patients on
Cross-section
the importance of maintaining adequate hydration.
PROCTOCOLECTOMY WITH CONTINENT

ILEOSTOMY
Another single-stage operation for UC is proctocolectomy GIA-90 stapler
with continent ileostomy, first introduced by Nils Kock in without knife
1969 (14). While initially met with enthusiasm, it became
less utilized with the advent of IPAA.
Indications
The continent ileostomy is an elaborate construction involv- Figure 35.1 Continent ileostomy. (a) S-shaped pouch cre-
ated first with inner posterior suture line, (b) Nipple valve is
ing a reservoir with internal nipple valve. It is more commonly
created with three firings of a non-cutting GIA stapler, (c)
reserved for patients who have failed previous Brooke ileos- Anterior layer of pouch is closed. (From Krane MK, Lange
tomy or IPAA, those who are not pouch candidates, or those EO, Fichera A. Ulcerative colitis: Surgical management. In
with poor sphincter tone, who wish to avoid a conventional Steele SR, Hull TL, Read TE, Saclarides TJ, Sebagore AJ,
ileostomy (15,16). Important contraindications include Crohn Whitloe CB. (eds). The ASCRS Textbook of Colon and Rectal
disease, obesity, and physical limitations or psychological dis- Surgery, 3rd Edition, 2016, Figures 50–5 and 6, p.876.)
abilities that would prevent successful self-catheterization.
Technique Complications/outcome
Following the initial proctocolectomy, a continent ileostomy, Due to the complexity of the procedure, continent ileosto-
or Kock pouch, is constructed from the terminal 50 cm of mies have been associated with a considerable rate of compli-
ileum. There are two main components: a reservoir and an cation and reoperation. The continent ileostomy is vulnerable
outlet valve. A nipple valve is fashioned by the intussus- to both early and late complications. Common early com-
cepting efferent loop of ileum. The opposing intussuscepted plications include valve necrosis, hemorrhage, and leakage.
intestinal walls must adhere to each other to prevent des- Late complications range from valve malfunction (including
susception, which would result in incontinence. There are slippage, prolapse, and stenosis) to herniation, perforation,
several variable techniques for construction, including the fistula, and pouchitis. A retrospective review demonstrated
S-pouch (Figure 35.1), the Barnett continent ileal reservoir, valve slippage to be the most common complication, occur-
and the T-pouch, the latter of which lacks an intussuscepting ring in 29.7% (18). The main causes for reoperation are cited
valve, but was shown to have acceptable outcomes (13,17). as slipped nipple valve, fistula, and stenosis (16).
An evaluation of quality of life after continent ileostomy RESTORATIVE PROCTOCOLECTOMY

of 330 patients reported higher scores when compared to (PROCTOCOLECTOMY WITH IPAA)
patients who had conversion of Kock pouch to Brooke ileos-
tomy (18). Likewise, an evaluation following patients who Considered the most appealing option for patients with UC
underwent continent ileostomy creation following failed requiring surgery, restorative proctocolectomy with IPAA,
IPAA reported good quality of life, despite a 60.9% compli- introduced in 1978 (24,25), effectively removes all diseased
cation rate (15). bowel while preserving continence and avoiding a perma-
Overall, while associated with significant complications, nent stoma.
a continent ileostomy may allow patients freedom from tra-
ditional ileostomy and remains a viable option for selected Indications
patients.
Restorative proctocolectomy with IPAA is indicated in UC
TOTAL ABDOMINAL COLECTOMY WITH patients with adequate sphincter function requiring surgery.
ILEORECTAL ANASTOMOSIS Preoperatively, sphincter function should be thoroughly
evaluated using digital rectal exam. Endoanal ultrasound
Indications and anal manometry may also be utilized. The presence
of preoperative incontinence is considered a contraindica-
Considering that the majority of patients with UC will have tion to IPAA. IPAA is typically avoided in low or mid-rectal
rectal involvement, this surgical option is selected for patients cancers due to the need for adjuvant radiotherapy and the
with indeterminate colitis, rectal sparing, and satisfactory rec- potential effects on the pouch. Neoadjuvant therapy has
tal compliance and adequate sphincter function, or in high- also been shown to have adverse effects on pouch function,
risk patients who are poor candidates for IPAA. Advantages of and significantly contributes to pouch failure (26). Morbid
this option include avoidance of both a pelvic dissection and obesity is considered a relative contraindication to IPAA.
permanent stoma, and a single-stage operation. In a retrospective review of patients categorized by body
Pursuing TAC with ileorectal anastomosis obligates con- mass index (BMI), obese patients were found to have an
tinued rectal surveillance. Endoscopy with rectal biopsies increased risk of pouch-related complications after IPAA
every 6–12 months is recommended. (61% versus 26%) (27), and a retrospective review compris-
ing 1,046 patients evaluated the association of BMI and abil-
Technique ity to technically perform IPAA and found that increasing
BMI was associated with not being able to perform IPAA,
The patient is similarly prepared as described for procto- with chance of unsuccessful pouch rising from 2% at BMI
colectomy placed in lithotomy position. The colon is fully 30, to 15% at 40 (28).
mobilized from the terminal ileum to the rectosigmoid
junction and removed via laparoscopic or open approach. Technique
The ileorectal anastomosis may be performed in a side-to-
side or side-to-end via stapled or hand-sewn fashion. Leak The patient is prepared as described above for proctocolec-
test confirms integrity. tomy with mechanical bowel preparation, perioperative
antibiotics, and placed in lithotomy. Full mobilization to
Complications/outcomes free the colon from retroperitoneal attachments is carried
out from the terminal ileum to the rectosigmoid junction.
In comparison to proctocolectomy, TAC is associated with Dissection is carried along a total mesorectal excision plane,
less infertility and fewer functional complications due to elevating and then dividing the rectum, leaving a short rec-
the avoidance of pelvic dissection (19–21). The main con- tal cuff, and taking care to preserve the sphincter muscle
cern for UC patients undergoing TAC is retained rectum complex.
and potential for persistent inflammation and development The pouch may be configured as a J-pouch, S-pouch, or
of dysplasia or cancer, resulting in ileorectal anastomosis W-pouch, by either a double-stapled or hand-sewn anasto-
(IRA) failures, shown to be 27% and 40% at 10 and 20 years, mosis (Figure 35.2). The most commonly used variety is the
respectively (22). A retrospective study out of the Cleveland J-pouch, which is constructed from the terminal 30–40 cm
Clinic demonstrated rectal dysplasia and rectal cancer to of ileum. The apex of the pouch is selected after identify-
occur in 17% and 8%, respectively, over a 9-year period. ing the longest mesentery. The segment of small intestine
The reported cumulative probability of rectal dysplasia at is folded into two 15 to 20 cm segments, and the two limbs
5, 10, 15, and 20 years was reported to be 7%, 9%, 20%, and are approximated using stay sutures. A longitudinal enter-
25%, respectively. For rectal cancer along the same time otomy is made at the apex, and a side-to-side anastomosis
interval, it was reported to be 0%, 2%, 5%, and 14%, respec- of the two segments is done using a linear stapler, taking
tively. Proctectomy was necessary in 53% of patients (23). care not to involve the mesentery, thus creating a reservoir.
The most common indication for proctectomy after TAC is An end-to-end anastomosis stapler is used to create the
refractory proctitis, followed by dysplasia and cancer (23). pouch-anal anastomosis. The anvil is secured in the apex of
Surgical options / Restorative proctocolectomy (proctocolectomy with IPAA) 327
(a) (b) (c)
II
15 cm
II I III
15–20 12 cm
cm IV III II I
2 cm
Figure 35.2 Ileal pouch configurations. (a) J-pouch, (b) S-pouch, (c) W-pouch.
the pouch with purse-string suture, and the stapler is intro- Long-term studies comparing mucosectomy with
duced transanally. The suture line is inspected endoscopi- hand-sewn anastomosis and double-stapled anastomosis
cally, and a leak test is performed to ensure integrity. have found the latter to be safe and with better functional
The S-pouch involves three limbs of ileum, each outcome (33–35). Fichera et al. demonstrated excellent
12–15 cm in length, anastomosed side-to-side, with a 2 cm functional results long term without new-onset dysplasia,
exit conduit. This variation allows additional length and suggesting that preserving the ATZ is oncologically safe
may be used when a J-pouch will not reach without ten- (33). Ziv et al. likewise described significantly more septic
sion. The W-pouch consists of a quadruplicated reservoir complications and subsequent pouch excisions after hand-
with increased capacity and compliance. A meta-analysis sewn anastomosis in their retrospective review (34).
comprising 18 studies compared the three pouches and Prospective randomized trials comparing mucosectomy
demonstrated increased bowel frequency associated with and hand-sewn anastomosis and double-stapled anastomo-
J-pouch, increased need for pouch intubation with S-pouch, sis have found either no difference or improved outcomes
and no difference otherwise in regard to early postopera- with double-stapled anastomosis (36–39).
tive complications (29). Comparison of S- and W-pouch The Cleveland Clinic prospectively evaluated 3,382
demonstrated decreased bowel frequency, as well as greater patients undergoing IPAA and compared hand-sewn and
efficiency of evacuation in W-pouch (30). In a prospective stapled anastomosis and found significantly greater use
comparison of the J-pouch and the W-pouch, however, no of ileostomy, longer length of stay, anastomotic stricture,
significant difference was identified in regard to bowel fre- septic complications, bowel obstruction, and pouch failure
quency (31). with hand-sewn anastomosis. Additionally, a significantly
larger portion of patients in the hand-sewn group described
Mucosectomy and hand-sewn anastomosis incontinence, seepage, and worse quality of life (26).
versus double-stapled anastomosis Overall, a stapled anastomosis is preferred, as it is not
only easier to perform, but is also associated with better
The type of anastomosis used for IPAA, hand-sewn or sta- functional outcomes and quality of life.
pled, is a controversial issue.
Mucosectomy with hand-sewn anastomosis is included One versus two versus three stages
in the original description of IPAA and involves complete
removal of anorectal mucosa up the dentate line, including The RP with IPAA can be achieved in one, two, or three
the anal transition zone (ATZ) (24). Proponents of muco- stages.
sectomy and hand-sewn anastomosis maintain that com- One-stage procedure consists of proctocolectomy fol-
plete removal of mucosa more effectively eliminates the risk lowed by IPAA without ileostomy.
of malignant degeneration. Two-stage procedure consists of proctocolectomy, IPAA,
Alternatively, performing a double-stapled anastomo- and ileostomy during the initial operation, followed by ile-
sis leaves a short rectal cuff and preserves the ATZ, which ostomy closure. Modified two-stage procedure consists of
thus preserves specialized anoderm and results in improved subtotal colectomy with end ileostomy during the initial
sensation and function. Together with the rectoanal inhibi- operation, followed by proctectomy and pouch creation.
tory complex, the ATZ allows sampling of rectal contents, Three-stage procedure consists of subtotal colectomy and
contributing to continence. Moreover, a double-stapled end ileostomy, followed by proctectomy and pouch creation
anastomosis avoids the extensive manipulation and dilation with diverting loop ileostomy, and finally ileostomy closure
required for a complete mucosectomy, which avoids undue (40). Unlike the modified two-stage, the three-stage proce-
trauma to the sphincter complex (13,32). dure has a diverting loop ileostomy with interval closure
after pouch creation. The three-stage procedure is typically to allow complete division on the terminal ileum mesentery
recommended for patients with active colitis, malnutrition, and subsequent additional length (46).
or taking high-dose steroids. If adequate length cannot be obtained despite implemen-
A retrospective review of over 2,000 patients undergoing tation of each maneuver, the patient can be diverted proxi-
RP IPAA observed that while patients who underwent ileos- mally, leaving the pouch in the pelvis to revisit at another
tomy had an older mean age, were taking greater doses of operation. Gravity, along with the weight of the pouch, may
preoperative steroids, and required more blood transfusions, allow anastomosis at a later date.
there were no differences in the ileostomy and nonileostomy
groups in regard to septic complications, quality of life, or Complications/outcomes
functional outcomes. Early postoperative ileus was found to
be more common in the one-stage group. They concluded Pouch-related complications range from pouchitis to pelvic
that in selected patients, such as those with a stapled, tension- sepsis and pouch failure, requiring intervention spanning
free anastomosis, with intact tissue rings, adequate hemosta- from medical management to redo IPAA.
sis, and without air leak, anemia, toxicity, malnutrition, or Aiming to identify risk factors for pouch failure, a
prolonged steroid course, ileostomy can be avoided (41). single institution retrospectively reviewed 3,754 patients
A retrospective database review of 17 studies includ- who underwent ileoanal pouch between 1983 and 2008,
ing 1,486 patients undergoing RP IPAA from 1978 to 2005 and found the strongest predictors of pouch failure to be
demonstrated single-stage IPAA to have similar outcomes completion proctectomy, hand-sewn anastomosis, diabetes,
to those with ileostomy but was associated with a significant Crohn disease, and age at surgery (47).
increased risk of anastomotic leak (42). A review of outcomes, complications, and quality of life
Proponents of the ileostomy maintain that diversion pre- in 3,707 patients after IPAA from 1984 to 2010 reported that
vents morbidity and mortality associated with anastomotic 33.5% of patients had early perioperative complications, with
leakage, and that ileostomy closure itself is associated with a 0.1% mortality rate, 29.1% had late complications (exclud-
minimal morbidity. ing pouchitis) with a 5.3% pouch failure rate overall, and
A retrospective review of 190 patients undergoing IPAA declared IPAA to be a safe procedure with favorable out-
at a single institution from 1995 to 2003 compared modi- comes and high levels of satisfaction. Pelvic sepsis occurred
fied two-stage IPAA to traditional three-stage IPAA, and in 6.5% of patients, early anastomotic leak in 4.8%, late anas-
found no anastomotic complications among the modified tomotic leak in 1.7%, pouch bleeding in 3.4%, early anasto-
two-stage group, and also concluded the modified two- motic stricture in 5.2%, and late stricture in 11.2% (48).
stage IPAA to be more cost effective (43). A retrospective
comparison of the traditional two-stage and the modified Functional results: Bowel, urinary,
two-stage found that the modified group had a significantly gynecologic, sexual
lower rate of anastomotic leak (44).
In a 5-year review of national trends of restorative proc- A meta-analysis of 96 studies comprising 14,966 patients
tocolectomy for UC, it was concluded that use of two-stage investigated outcomes observed in pre-2000 studies, and
and three-stage has remained stable. It was observed that post-2000 studies. Incidence of pouch failure significantly
patients undergoing three-stage were less likely to have pre- decreased from 8.5% to 4.7%. Other functional outcomes
operative steroid therapy, hypoalbuminemia, preoperative were not significantly different from pre-2000 findings.
sepsis, and weight loss at the time of initial pouch creation. Review of studies since 2000 reflected pelvic sepsis occur-
It was also found that superficial surgical site infections are ring in 7.5%, fistula in 4.5%, stricture in 10.7%, pouchitis in
more common following three-stage surgery (45). 26.8%, sexual dysfunction in 3%, small bowel obstruction
in 11.4%, mild daytime incontinence in 14.3%, severe day-
Reach time incontinence in 6.1%, mild nighttime incontinence in
17.3%, and severe nighttime incontinence in 7.6% (49).
A tension-free IPAA is critical to avoid adverse sequelae. In regard to IPAA in the elderly, studies demonstrate
Certain maneuvers may be performed to achieve adequate worse outcomes but maintain that IPAA can be offered to
length, such as complete mobilization of the small bowel select patients (50–53). Retrospective review of 1,454 patients
mesentery, exposure of the inferior portion of the head of who underwent IPAA for UC observed worse functional out-
the pancreas, scoring the peritoneum of the small bowel comes in older patients compared to younger patients, finding
mesentery on anterior and posterior surfaces, and relax- nocturnal stooling, fecal incontinence, pad, and constipating
ing transverse incisions along the SMA mesentery. Specific medication usage to be higher in patients undergoing surgery
arcades under tension can be identified within the mes- after age 45. Moreover, these issues become more common
entery and ligated, as can the ileocolic vessels. Goes et al. as follow-up duration increases. These outcomes were simi-
describes a technical procedure, performed in cadavers, in lar among both men and women (50). Delaney et al. likewise
which the marginal vascular arcade of the right colon is observed improved functional outcomes in patients under
preserved, while the distal third of the superior mesenteric 45, while describing an overall high percentage of patients
artery, the ileocolic artery, and right colic artery are ligated, reporting satisfaction with the surgery (54). Moreover, a
Surgical options / Restorative proctocolectomy (proctocolectomy with IPAA) 329
survey of 154 patients after IPAA echoed that functional out- by vague complaints, and failure to progress along clinical
comes worsened the further out from surgery, finding that course. Beyond clinical findings, cross-sectional imaging
daytime and nighttime incontinence became significantly is of utility in defining pathology, as is pouchogram and
more common 12 years or more after surgery (51). examination under anesthesia.
Women and men are at risk of developing sexual dys- Patient management depends on source and severity of
function after IPAA. Men are at risk of developing erectile septic complications. In additional to broad-spectrum IV
dysfunction and retrograde ejaculation, while women can antibiotics and close observation, computed tomography–
experience dyspareunia and infertility. Considering that guided or open drainage may be required. Any patient dem-
the majority of women undergoing IPAA are at childbear- onstrating hemodynamic instability should be explored, the
ing age, this is an important consideration (55). The cause of anatomy examined, and washout performed, leaving drains
sexual dysfunction is multifactorial; hypogastric or pelvic in place. Intestinal diversion may be required.
nerve injury during surgery can result in vaginal dryness,
contributing to dyspareunia, while adhesions can cause SMALL BOWEL OBSTRUCTION
occlusion of the fallopian tubes (55). Diagnostic laparoscopy Small bowel obstruction is a frequently seen complication
performed at the time of ileostomy closure in IPAA patients after IPAA, occurring in up to 25% of patients (48,59), and
demonstrated fewer adnexal adhesions in the laparoscopic may be the foremost cause for hospitalization. A systematic
group compared to open, using the American Fertility review of 28 studies describes early small bowel obstruc-
Society adhesion score (56). tion occurring in 2%–12% of patients, with late small bowel
Meta-analysis of 22 studies comprising 1,852 women obstruction occurring more frequently at 17% (60). Small
investigated the effect of RP on urinary, sexual, and gyneco- bowel obstruction may occur due to adhesions, stenosis,
logic function, and found incidence of infertility to increase volvulus, or internal hernia. Several studies have found use
from 12% pre-RP to 26% after. Dyspareunia likewise of diverting ileostomy to be associated with an increased
increased following RP, from 8% preoperatively to 25%. risk of bowel obstruction (61–63), and increased risk of
No association was found between sexual dysfunction and requiring surgery for small bowel obstruction (59). A pro-
pouch function. No significant difference in urinary func- spective randomized study determined the use of an adhe-
tion was found after RP, nor was an increase in pregnancy sion barrier, Seprafilm® (Sanofi, Bridgewater, NJ), reduced
complications observed (57). the incidence of small bowel obstructions requiring sur-
In spite of the pouch-related complications, patients report gery; however, it did not reduce the incidence of small bowel
improvement in quality of life and satisfaction with IPAA (58). obstruction overall (64).
A retrospective review of patients with medically treated
Other complications small bowel obstruction after IPAA demonstrated an
increased total number of bowel movements per day in the
SEPTIC COMPLICATIONS: LEAK AND SEQUELAE first postoperative year compared to patients without small
Presentation of pelvic sepsis typically manifests as fever, bowel obstruction, as well as more daytime seepage at 10
leukocytosis, pelvic and perineal pain, ileus, and purulent years postoperative. Quality of life, however, was compa-
drainage, but may be more conspicuous, characterized rable among groups (65).
only by vague complaints and failure to progress along the The majority of small bowel obstructions resolve without
expected clinical course. The incidence of pelvic sepsis after surgery, with the need for surgery ranging from 3% to 19%
IPAA by meta-analysis is reported to be 7.5% (49). Pouch- (59); however, if no progression is observed, operative inter-
related septic complications include anastomotic leak, vention should be pursued.
pouch abscess, and pouch anal fistula.
A single institution study of 3,707 patients identified a HEMORRHAGE
7.5% incidence of anastomotic leak among IPAA patients Bleeding from the pouch is a less common complication of
(48), whereas meta-analysis reported 7.1% (42). A leak may IPAA and may present from ostomy or anus. Incidence of
occur at the pouch anal anastomosis, as well as any suture or bleeding was shown to be 1.5% in a study of 3,194 patients
staple line, including the tip of the J-portion of the pouch. A (66). Intervention, while frequently required, is typically
leak at this location may be difficult to diagnose, and suture nonsurgical. Endoscopy with clot evacuation and cautery
or staple repair may salvage the pouch. or epinephrine enemas has been shown to be effective (66).
Risk factors for anastomotic leak include anastomosis
under tension and compromised blood supply. Additionally, STRICTURE
a review of 3,233 patients found factors significantly associ- Pouch stricture is a common complication of IPAA. Risk
ated with septic complications to include BMI >30, pathol- factors for stricture include use of small stapling gun, use
ogy consistent with inflammatory bowel disease, and of defunctioning ileostomy, anastomotic dehiscence, and
intraoperative and postoperative blood transfusion (51). pelvic sepsis (67). Two locations prone to stricture forma-
Presentation of pelvic sepsis typically manifests in fever, tion include the pouch-anal anastomosis and the junction
leukocytosis, pelvic and perineal pain, ileus, and purulent of the neoterminal ileum and pouch. Endoscopic dilation of
drainage, but may be less conspicuous, characterized only strictures has proven efficacious (66,67).
Fistula Treatment consists primarily of oral or topical medica-

tions: antibiotics, probiotics, and steroids (76). Should sur-
POUCH VAGINAL FISTULA gical treatment be required for symptomatic, refractory
Pouch vaginal fistula (PVF) is a major cause of morbidity pouchitis, diversion, pouch excision, or redo IPAA can be
and pouch failure after IPAA, with an incidence between 4% performed.
and 16% (68). Pouch failure occurs in 21%–30% of patients
CROHN DISEASE
with PVF, with excision required in 26% (69). Major risk
factors for PVF include pelvic sepsis, anastomotic separa- Crohn disease of the pouch is an unfortunate complica-
tion and stricture, small bowel obstruction, hemorrhage, tion that may occur after IPAA and is a leading cause of
and pouchitis (69), as well as a diagnosis of indeterminate pouch failure (78). Presentation is similar to that of pou-
colitis or Crohn disease (70). chitis, characterized by abdominopelvic pain and diarrhea.
Patients with diagnosis of indeterminate colitis or Crohn Perianal disease and fistula are also suggestive of Crohn
disease have a higher incidence of PVF, as well as a higher disease, especially if occurring greater than 12 months after
rate of morbidity with PVF compared to UC (71). A delayed IPAA, especially in the absence of postoperative complica-
diagnosis of Crohn disease after IPAA is associated with tions such as sepsis, abscess, or leak (79).
worse outcomes and higher rates of pouch failure after Patients found to have Crohn disease after IPAA experi-
treatment (72). ence significant complications. A 93% complication rate was
The type of anastomosis employed at the time of sur- observed in a series of 31 patients with subsequent Crohn diag-
gery has been investigated as a possible cause of PVF. Lee nosis after IPAA. The complications included perineal abscess/
et al. described an association between PVF and hand-sewn fistula, pouchitis, and anal stricture. Pouch excision was neces-
anastomosis (73), as did Mallick et al., whereas Groom et al. sary in 29% (80). In another series, patients with a subsequent
observed an insignificant increased incidence among sta- diagnosis of Crohn disease had a 47.8% pouch excision rate,
pled anastomoses (74). compared to a 10.9% rate for those diagnosed with UC (81).
PVF were found located above the anastomosis in Management of Crohn disease of the pouch depends on
24%–26%, at the anastomosis in 26%–28%, and below the phenotype—inflammatory, fibrostenotic, or fistulizing—
anastomosis in 48% of cases in two separate retrospective and consists of combined medical, endoscopic, and surgi-
reviews (68,75). cal treatment. Inflammatory Crohn disease may be difficult
Surgical repair is varied and includes local repair via a to distinguish from pouchitis or cuffitis and is treated with
transanal approach with either advancement flap or direct oral and topical agents. Fibrostenotic frequently presents
repair in layers, transvaginal repair with advancement flap, with ulcerated strictures in the distal small bowel, afferent
fistulectomy, interposition flap, transabdominal approach, limb, pouch inlet or outlet, or mid-pouch (78); manage-
and redo IPAA. The choice of repair is dependent on loca- ment consists of medication, endoscopic balloon dilation
tion and character of the fistula. It is imperative to treat (82), and bowel-preserving strictureplasty. Of the Crohn
pelvic sepsis prior to definitive repair, and to attempt the disease phenotypes, fistulizing disease may be associated
simplest approach first. Diverting ileostomy is frequently with higher risk of pouch failure (79). Fistulas suggestive of
employed in surgical treatment (70). A retrospective review Crohn disease present in areas outside the anastomosis, tip
by Tsujinaka et al. reported a mean number of 2.2 surgical of the “J” suture line, or dentate line (78). Infliximab has
treatments, with Crohn patients requiring a mean of 3 (75). been shown to be effective in short- and long-term treat-
An algorithm suggested by Lolohea et al. outlines man- ment of Crohn disease of the pouch, with 62% of patients
agement of PVF, with surgical intervention ranging from demonstrating a complete short-term response in a case
local procedure and interposition flap to pouch excision, series of 26 patients (83).
depending on location of fistula (70). Due to the associated morbidity, IPAA is not routinely
recommended for patients with known Crohn disease.
POUCHITIS AND CUFFITIS
Pouchitis and cuffitis are two similar complications unique POUCH FAILURE AND SALVAGE
to IPAA. Cuffitis, referring to the remnant rectal cuff, and Pouch failure is defined as a nonfunctioning pouch or pouch
pouchitis, referring to the pouch itself, are characterized excision within 12 months of IPAA, and occurs in 5.3%
by inflammation, bleeding, abdominopelvic pain, anal dis- (48,84). Risk factors contributing to pouch failure include
charge, incontinence, tenesmus, urgency, and increased Crohn disease, and chronic pouch-related issues, with pel-
stool frequency, in the setting of mucosal inflammation. vic sepsis and pouch fistula reported as the most common
Pouchitis is the most common complication in patients with causes (50,84–87).
UC following IPAA, reported to occur in 14%–59% (48,76). Evaluation of pouch dysfunction should include digi-
Diagnosis involves pouchoscopy and biopsy. Frequently tal and endoscopic examination, as well as cross-sectional
used parameters comprising the pouchitis disease activity imaging and anal manometry (88).
index (PDAI) include stool frequency, rectal bleeding, fecal Salvage procedures can be performed locally at the
urgency/cramping, fever, endoscopic inflammation, histo- perineum, with an abdominal approach, or via a combined
logic findings, and degree of ulceration (77). abdominoperineal approach. Pouch revision, augmentation,
References 331
resection, mobilization, repair, or new construction, with or increased risk in regard to morbidity or mortality, including
without ileostomy, are all options for reoperative pouch sur- anastomotic leak and infections (95).
gery (87). When examining the preoperative serum level of
A retrospective series of 51 patients who underwent a nti-TNF-α medication levels in UC patients, no signifi-
reoperative pouch surgery, either via abdominal or perineal cant increase in postoperative complications was observed
approach, reported a 69.5% or 75% success rate, respec- between the patients with detectable and undetectable serum
tively (85). levels (96).
Meta-analysis demonstrated salvage procedures to be In regard to timing of surgery, Waterman et al. compared
performed an average of 24 months after initial RP, with an postoperative complications between patients taking either
overall successful healing rate of 73.5%, 82.2%, 79.6%, and infliximab or adalimumab within different time intervals
68.4% for overall, redo, revisional, and local perineal proce- prior to surgery: 14 days, 15–31 days, and 31–180 days. The
dures, respectively. Overall functional success was reported study declared that biologic agents alone were not associated
in 71.9% (89). with an increased risk of postoperative complications, but
Following salvage surgery, patients are found to have sig- rather that combination therapy of biologic agents and thio-
nificantly higher daytime and nighttime seepage, as well as purines was associated with increased postoperative com-
daytime pad usage. Quality of life, dietary, work, social, sexual plications. Moreover, shorter time interval did not increase
restrictions, incontinence, and total bowel movements were postoperative complications, suggesting that surgery should
not significantly different compared with those undergoing not be delayed based on use of biologic agents alone (97).
primary restorative proctocolectomy and IPAA (87,89).
REFERENCES
1. Loftus EV Jr. Gastroenterology. 2004;126(6):1504–17.
BIOLOGICS—IMPACT ON SURGICAL 2. Dahlhamer JM et al. Morb Mortal Wkly Rep. 2016;
PROCEDURE AND TIMING 65(42):1166–9.
3. Kappelman MD et al. Clin Gastroenterol Hepatol.
Classically treated medically with corticosteroids, 5- 2007;5(12):1424–9.
aminosalicylates, and immunosuppressants, UC entered 4. Hancock L, Mortensen NJ. Inflamm Bowel Dis. 2008;
the era of biologic treatment with advent of tumor necrosis 14 Suppl 2:S68–9.
factor-alpha (TNF-α) inhibitors such as infliximab in 2005. 5. Truelove SC, Witts LJ. Br Med J. 1955;2(4947):1041–8.
Since then, additional biologics, such as adalimumab, goli- 6. Sobrado CW, Sobrado LF. Arq Bras Cir Dig. 2016;
mumab, and certolizumab, have been introduced and have 29(3):201–5.
helped, as monotherapy or in combination with other drugs, 7. Teeuwen PH et al. J Gastrointest Surg. 2009;13(4):
to induce and maintain remission in UC patients (90,91). 676–86.
Despite the medical therapy available, surgery is still 8. Ross H et al. Standards Practice Task Force of the
ultimately required in 30%–40% of patients with UC (4). American Society of Colon and Rectal Surgeons. Dis
The data regarding surgical complications after treatment Colon Rectum. 2014;57(1):5–22.
with anti-TNF-α therapy are mixed. 9. Beck DE et al. The ASCRS Manual of Colon and
When retrospectively reviewing postoperative complica- Rectal Surgery. New York, NY: Springer 2014.
tions after IPAA, the risk of early and late complications, 10. Gumaste V et al. Gut. 1992;33(7):938–41.
as well as need for three-stage operation, is increased in 11. Eaden JA et al. Gut. 2001;48(4):526–35.
patients treated with infliximab (91). Systemic review of 12. Bernstein CN. Curr Gastroenterol Rep. 1999;1(6):
162 patients receiving biologics echoed this finding, dem- 496–504.
onstrating an increased risk of pouch-related complications 13. Steele SR et al. The ASCRS Textbook of Colon and
with preoperative infliximab (92). Rectal Surgery. New York, NY: Springer Science +
Likewise, meta-analysis comprising five studies and 132 Business Media 2016.
patients treated with infliximab demonstrates an increased 14. Kock NG. Dis Colon Rectum. 1994;37(3):278–85;
risk of short-term complications with preoperative inf- discussion 285–7.
liximab including abdominal wound infection, anasto- 15. Lian L et al. Dis Colon Rectum. 2009;52(8):1409–14;
motic leak, pelvic sepsis, and small bowel obstruction. discussion 4414–6.
Considering that patients with UC receiving systemic cor- 16. Wasmuth HH et al. Colorectal Dis. 2007;9(8):713–7.
ticosteroids >40 mg/day have significantly increased risk of 17. Kaiser AM. Dis Colon Rectum. 2012;55(2):155–62.
pouch-related complications after IPAA (93), and that the 18. Nessar G et al. Dis Colon Rectum. 2006;49(3):336–44.
majority of patients with UC taking infliximab are also tak- 19. Mortier PE et al. Gastroenterol Clin Biol. 2006;30(4):
ing additional medications, it is unclear whether infliximab 594–7.
alone is culpable (94). 20. Saito Y et al. J Gastroenterol. 1995;30(Suppl 8):131–4.
Conversely, retrospective analysis of 142 patients treated 21. Khubchandani IT, Kontostolis SB. Arch Surg. 1994;
with infliximab prior to surgery did not demonstrate any 129(8):866–9.
22. Uzzan M et al. Ann Surg. 2017;266(6):1029–34. 61. MacLean AR et al. Ann Surg. 2002;235(2):200–6.
23. da Luz Moreira A et al. Br J Surg. 2010;97(1):65–9. 62. Marcello PW et al. Dis Colon Rectum. 1993;36:
24. Utsunomiya J et al. Dis Colon Rectum. 1980;23(7): 1105–11.
459–66. 63. Francois Y et al. Ann Surg. 1989;209:46–50.
25. Parks AG, Nicholls RJ. Dis Colon Rectum. 1988;31(10): 64. Fazio VW et al. Dis Colon Rectum. 2006;49(1):1–11.
826–30. 65. Erkek AB et al. J Gastroenterol Hepatol. 2008;23(1):
26. Wu XR et al. J Crohns Colitis. 2013;7(10):e419–26. 119–25.
27. Klos CL et al. J Gastrointest Surg. 2014;18(3):573–9. 66. Lian L et al. J Gastrointest Surg. 2008;12(11):1991–4.
28. Khasawneh MA et al. Dis Colon Rectum. 2016;59(11): 67. Lewis WG et al. Dis Colon Rectum. 1994;37(2):120–5.
1034–8. 68. Mallick IH et al. Dis Colon Rectum. 2014;57:490–6.
29. Lovegrove RE et al. Colorectal Dis. 2007;9(4):310–20. 69. Heriot AG et al. Dis Colon Rectum. 2005;48(3):
30. Sagar PM et al. Gastroenterology. 1992;102(2):520–8. 451–8.
31. Johnston D et al. Gut. 1996;39(2):242–7. 70. Lolohea S et al. Dis Colon Rectum. 2005;48(9):
32. Miller R et al. Dis Colon Rectum. 1990;33(5):414–8. 1802–10.
33. Fichera A et al. J Gastrointest Surg. 2007;11(12): 71. Koltun WA et al. Dis Colon Rectum. 1991;34(10):
1647–52; discussion 1652–3. 857–60.
34. Ziv Y et al. Am J Surg. 1996;171(3):320–3. 72. Shah NS et al. Dis Colon Rectum. 2003;46(7):911–7.
35. Gemlo BT et al. Am J Surg. 1995;169(1):137–41; 73. Lee PY et al. Dis Colon Rectum. 1997;40(7):752–9.
discussion 141–2. 74. Groom JS et al. Br J Surg. 1993;80(7):936–40.
36. Choen S et al. Br J Surg. 1991;78(4):430–4. 75. Tsujinaka S et al. J Am Coll Surg. 2006;202(6):912–8.
37. Luukkonen P, Järvinen H. Arch Surg. 1993;128(4): 76. Hata K et al. Dig Endosc. 2017;29(1):26–34.
437–40. 77. Sandborn WJ et al. Mayo Clin Proc. 1994;69(5):409–15.
38. Reilly WT et al. Ann Surg. 1997;225(6):666–76; 78. Shen B. Inflamm Bowel Dis. 2009;15:284–94.
discussion 676–7. 79. Shen B et al. Dis Colon Rectum. 2007;50(9):1450–9.
39. Kirat HT et al. Surgery. 2009;146(4):723–9; 80. Braveman JM et al. Dis Colon Rectum. 2004;47(10):
discussion 729–30. 1613–9.
40. Sofo L et al. World J Gastrointest Surg. 2016;8(8): 81. Keighley MR. Acta Chir Iugol. 2000;47(4 Suppl 1):
556–63. 27–31.
41. Remzi FH et al. Dis Colon Rectum. 2006;49(4):470–7. 82. Shen B et al. Am J Gastroenterol. 2004;99(12):2340–7.
42. Weston-Petrides GK et al. Arch Surg. 2007;143(4): 83. Colombel J-F et al. Am J Gastroenterol. 2003;98:
406–12. 2239–44.
43. Swenson BR et al. Dis Colon Rectum. 2005;48(2): 84. Lepistö A et al. Dis Colon Rectum. 2002;45(10):
256–61. 1289–94.
44. Zittan E et al. J Crohns Colitis. 2016;10(7):766–72. 85. Shawki S et al. Dis Colon Rectum. 2009;52(5):884–90.
45. Bikhchandani J et al. Dis Colon Rectum. 2015;58(2): 86. Körsgen S, Keighley MR. Int J Colorectal Dis. 1997;
199–204. 12(1):4–8.
46. Goes RN et al. Dis Colon Rectum. 1995;38(8):893–5. 87. Remzi FH et al. Dis Colon Rectum. 2009;52(2):198–204.
47. Manilich E et al. Dis Colon Rectum. 2012;55(4):393–9. 88. Sagar PM, Pemberton JH. Br J Surg. 2012;99(4):
48. Fazio VW et al. Ann Surg. 2013;257(4):679–85. 454–68.
49. de Zeeuw S et al. Int J Colorectal Dis. 2012;27(7): 89. Theodoropoulos GE et al. J Am Coll Surg. 2015;220(2):
843–53. 225–42.e1.
50. Farouk R et al. Ann Surg. 2000;231(6):919–26. 90. Biondi A et al. World J Gastroenterol. 2012;18(16):
51. Bullard KM et al. Dis Colon Rectum. 2002;45(3): 1861–70.
299–304. 91. Mor IJ et al. Dis Colon Rectum. 2008;51(8):1202–7;
52. Kiran RP et al. Ann Surg. 2010;251(3):436–40. discussion 1207–10.
53. Pellino G et al. BMC Surg. 2013;13(Suppl 2):S9. 92. Selvaggi F et al. Inflamm Bowel Dis. 2015;21(1):79–92.
54. Delaney CP et al. Ann Surg. 2003;238(2):221–8. 93. Heuschen UA et al. Ann Surg. 2002;235(2):207–16.
55. Bharadwaj S et al. Inflamm Bowel Dis. 2014;20(12): 94. Yang Z et al. Aliment Pharmacol Ther. 2010;31(4):
2470–82. 486–92.
56. Hull TL et al. Br J Surg. 2012;99(2):270–5. 95. Krane MK et al. Dis Colon Rectum. 2013;56(4):449–57.
57. Cornish JA et al. Dis Colon Rectum. 2007;50(8): 96. Lau C et al. Ann Surg. 2015;261(3):487–96.
1128–38. 97. Waterman M et al. Gut. 2013;62(3):387–94.
58. Hueting WE et al. Int J Colorectal Dis. 2004;19(3): 98. Krane MK, Lange EO, Fichera A. Ulcerative colitis:
215–8. Surgical management. In Steele SR, Hull TL, Read TE,
59. Aberg H et al. Int J Colorectal Dis. 2007;22(6):637–42. Saclarides TJ, Sebagore AJ, Whitloe CB. (eds). The
60. Peyrin-Biroulet L et al. Aliment Pharmacol Ther. ASCRS Textbook of Colon and Rectal Surgery, 3rd
2016;44(8):807–16. Edition, 2016. Figures 50–5 and 6, p.876.
36
Surgery for Crohn disease
EMILY STEINHAGEN AND SHARON L. STEIN
and diarrhea are common symptoms, but patients may also

CHALLENGING CASE initially present with perianal fistulization or with extrain-
testinal manifestations such as joint pain, eye problems, or
A 29 year-old man, with a 10 year history of Crohn’s skin lesions. Symptoms of CD are generally relapsing and
disease, has been maintained on biologics. He remitting.
previously had two ileocolic resections. He recently Surgery does not cure CD. Therefore, the goals of surgery
developed postparandial pain (1 hour after eating) in a patient are to treat symptoms and complications of the
with bloating. On MR enterography, a long fibrotic disease that are not adequately managed with medication.
stricture is found in mid jejunum. The most common indications for surgery are obstruction,
neoplasia, fistulization, or failure to respond to medi-
cal management. Approximately 70% of patients with CD
will require surgical intervention during their lifetime. Of
CASE MANAGEMENT patients requiring an initial surgery, 50% will require a sec-
ond surgery.
A fibrotic stricture will rarely respond to medical
management. Therefore, the patient is offered a
stricturoplasty.
NONSURGICAL MANAGEMENT
OF CROHN DISEASE
INTRODUCTION
BIOLOGICS AND SURGERY FOR CROHN
DISEASE
Crohn disease (CD) is a chronic inflammatory disease of
the intestinal tract. It is hypothesized to be an overreaction The majority of patients are initially and chronically treated
of the immune system to the bacterial biome found within with medical management. Medical management is a com
the intestines, although the precise etiology has not been plex process involving a combination of antibiotics, amino-
elucidated. While CD affects the entire intestinal tract, the salicylates, and immunosuppressants, including steroids,
terminal ileum and colon are the most frequent sites of immunomodulators, and biologics, to achieve a clinical
disease. and endoscopic remission. Medical management is gener-
CD most commonly affects adults, with onset of symp- ally considered to be lifelong, with the goal of preventing
toms generally between the second and fifth decades of life. disease progression. Increasingly, biologics are being used
The incidence of CD has been increasing worldwide. While in the hope of maintaining a durable remission for patients
it was historically diagnosed in patients in North America, with CD.
Western Europe, and Australia, cases are increasingly diag- However, whether the increased use of biologics will
nosed in Asia and South America as well (1). Women and decrease the risks of surgical intervention is unclear. Initial
urbanites are more likely to develop CD. In the United data demonstrated lower rates of surgery for patients who
States, prevalence may be as high as in 330 per 100,000 received biologic therapy. The ACCENT I trial demon-
Americans (2). strated a twofold lower rate of surgery at 54 weeks, and the
Symptoms of CD are heterogeneous, based on the sever- CHARM trial demonstrated a nine times lower rate of sur-
ity and location of disease. Abdominal pain, weight loss, gery at 56 weeks (3,4). Surgical rates during both trials were
333
334 Surgery for Crohn disease
lower than historical data, with a reduction from 7.5% to 3% patients, data need to be reevaluated, rather than assuming
in ACCENT I, and 3.8% to 0.6% in the CHARM trial. A lon- all treatments are the same.
gitudinal study of patients with CD from 1991 to 2014 noted
that the use of biologics increased from 3.1% to 41.2% dur- MULTIDISCIPLINARY MANAGEMENT
ing the study period; during the same period, the frequency OF CROHN DISEASE
of surgical intervention decreased from 42.9% to 17.4% (5).
However, subsequent studies have not maintained these Given the concern regarding increasing risks of periop-
conclusions. A review of data from the National Inpatient erative complications as well as the need for periopera-
Sample did not support decreasing surgical rates in the bio- tive optimization of patients, the use of multidisciplinary
logic era; in this study the frequency of surgical interven- decision-making has become increasingly common in the
tions was unchanged, and patients who did undergo surgery treatment of CD. Similar to the care of patients with cancer,
were more likely to be malnourished (6). A study of 195 patients with benign complications of CD are best treated
patients with CD noted that patients on biologics appeared by a multidisciplinary group of physicians. The goals of
to have a longer lead time from initial diagnosis of CD to therapy, preoperative patient optimization, and timing of
operation (29 to 61 months, p = 0.005) but did not demon- surgical intervention can be optimized through a coordi-
strate a decreased amount of bowel resected when surgery nated care process. Gastroenterology, dietary, psychology,
was required (7). radiology, and surgical services should discuss complex
The longer duration of medical management and the cases of CD with respect to whether continued medical
finding of malnourished patients suggests an additional management, endoscopic evaluation and treatment, or
concern in patients on biologic therapies who do require surgical intervention is optimal. Several papers discuss-
surgery: they may be sicker and at increased risks of periop- ing the psychosocial benefits of multidisciplinary care,
erative complications both from the disease and the signifi- health-care cost benefits, and improved medical outcomes
cant immunosuppression that may also impact healing. A secondary to unified care plan have been published (14,15).
single-center study from St. Marks recently demonstrated a Multidisciplinary decision-making can also help surgeons
hazard ratio of 24.6 for septic complications in CD patients optimize patients prior to surgery.
on preoperative biologics (8). A meta-analysis including
over 5,700 patients noted that patients on infliximab had an PREOPERATIVE EVALUATION OF
increased risk of total complications (odds ratio [OR] 1.45, PATIENTS WITH CROHN DISEASE
confidence interval [CI] 1.04–2.02), infectious complica-
tions (1.47, CI 1.08–1.99), and other complications (2.29, CI Prior to surgery for Crohn disease, a full assessment of the
1.14–4.61) (9). Similar results were noted in several addi- gastrointestinal tract is strongly preferred. Endoscopic eval-
tional studies; Kopylov et al. found a trend toward increased uation is generally a first step. This allows for visualization
rate of total complications, infectious complications, and of affected portions of the stomach and duodenum as well
noninfectious complications (10). A second meta-analysis as the colon and terminal ileum. Although a colonoscopy
by El Hussana noted that patients with exposure to biologics report is useful, a surgeon should examine the colon directly
within 3 months of surgery were more likely to have anas- if an anastomosis is planned; visualization of the quality of
tomotic complications, particularly in trials with lower risk the mucosa for the “landing” zone of the anastomosis is
of bias (OR 1.63, CI 1.03–2.60) (11). However, the data are important to ensuring the integrity of the anastomosis and
still not definitive: a recent multinational, multicenter trial preoperative planning. Not surprisingly, inflamed tissue
recently demonstrated no increased rate of complications, increases the risk of anastomotic leak (16).
intraabdominal sepsis, or anastomotic leak in patients on Since upper endoscopic evaluation is limited in its abil-
immunologics or biologics (12). Notably, this trial did note ity to identify disease in the jejunum and proximal ileum,
that subsets of patients including those who required trans- video capsule endoscopy has added to the field by allowing
fusions, those with perforating disease, and patients with imaging of the small bowel mucosa and determination of
prior surgery were at increased risk for complications when existence, location, and extent of small bowel disease. In
they were on biologics at the time of surgery. This supports capsule endoscopy, a criterion of three or more ulcers is used
the concern that at least for some patients, biologics increase as diagnostic for CD, and sensitivity ranges around 77%,
their surgical risk. with a negative predictive value of 96% (17). Compared with
Most biologics function as antagonists to tumor necro- magnetic resonance imaging (MRI), capsule endoscopy has
sis factor-alpha (TNF-α); these include medications such better sensitivity for proximal disease identification (18,19).
as infliximab and adalimumab. Vedolizumab has an alter- However, capsule endoscopy is limited by its inability to
native mechanism of action and is an anti-integrin that pass through severely strictured areas, which occurs with
inhibits leukocyte migration into the gut. A recent study of some frequency. In early studies, retention rates of up to 13%
vedolizumab demonstrated a significant increase in the rate have been reported (20). If the capsule is retained within
of postoperative infections and superficial site infections, the small intestines, it necessitates surgical intervention for
even when compared to other biologics (13). Although this retrieval. A patency capsule, which dissolves over time, can
represents a small cohort, as new biologics are available to exclude significant stenosis prior to giving the true capsule.
Nonsurgical management of Crohn disease / Preoperative evaluation of patients with Crohn disease 335
Cross-sectional imaging is used with increasing frequency surgical risks. One review demonstrated successful tem-
in CD patients. A recent study noted a 1.9-fold increase in porizing treatment in 74% of cases (26). Patients who were
imaging (computed tomography and MRI) for CD over the more likely to require surgery included those with stenotic
last decade (21). These imaging modalities have the advan- disease, enterocutaneous disease, or refractory active dis-
tage of demonstrating both intraluminal and extraluminal ease. A meta-analysis also demonstrated a reduction in
disease and offer information about spatial relationships that the rate of complications via initial percutaneous abscess
are unable to be demonstrated with traditional barium tech- drainage compared with initial surgery (OR 0.44) (27).
niques. However, dynamic imaging studies such as contrast However, a different meta-analysis demonstrated that 70%
enema can be quite helpful in distinguishing intermittent of patients who presented with an abscess still ultimately
spasm or peristaltic contraction from CD. required surgery and did not find a reduction in complica-
For acute evaluation of patients, CT scans are almost tion, stoma rates, or length of stay with delayed surgical
always performed. They provide relatively low cost, highly intervention (28).
reliable data for rapid diagnosis and are widely available
in most medical centers. CT is extremely helpful for iden- Strictures
tifying obstruction, perforation, and abscess. However,
standard imaging techniques fail to provide optimized dis- For some patients with intestinal stricture, balloon dila-
tension of the bowel or dynamic imaging. Therefore, in the tion has been utilized effectively. This is particularly com-
nonemergent setting, enterography is preferred for a more mon for short-segment ileocolic disease. A systematic
detailed examination and the ability to identify mucosal review of endoscopic balloon dilation in CD that included
abnormalities. over 1,000 patients demonstrated that symptomatic
CT enterography (CTE) is generally also widely avail- response was achieved in 70% of patients, and technical
able, cost effective, and has good sensitivity for CD lesions. response (ability to pass the scope following procedure)
However, it does require an experienced radiologist to inter- was achieved in 90% of cases. Complications occurred in
pret the study accurately. Compared to magnetic resonance 6.4% and perforation occurred in 3% of patients. However,
enterography (MRE), CTE generally has superior interob- the cumulative rate of surgery at 5 years was still 75%.
server agreement (22). However, other studies have demon- Outcomes were not different in de novo versus anasto-
strated improved sensitivity of diagnosis of fistula, stenosis, motic strictures (29).
and abscess with MRE compared to CTE (23). In addition, Recent data indicate that salvage after failed endoscopic
MRE may include the ability to predict response to treat- dilation may be associated with increased risk of diver-
ment for CD in the future (24). Both modalities have the sion as well as deep space and superficial infections (30).
ability to discern whether strictures are inflammatory or However, it is unclear if this is directly correlated with the
fibrotic in nature, which can help drive surgical decision- dilation, or if it is because patients are clinically worse at the
making. Because of the chronic nature of CD, and repeat time of surgery.
imaging requirements for patients over their lifetime, there
is concern about the increased exposure to ionizing radia- Thromboembolic disease
tion (25). For this reason, MRE has become a diagnostic test
of choice. Active inflammation is a risk factor for thromboembolic
When surgery is planned for CD, imaging is crucial. An disease. Studies demonstrate that patients with inflamma-
appropriate evaluation will define the extent and location of tory bowel disease (IBD) have an increased rate of thrombo-
disease, determine the existence of fistulizing disease, dem- embolic disease when compared to the general population.
onstrate the presence and extent of stenotic disease, identify Thromboembolic events are 1.5–3.5 times more likely to
abscess or phlegmon, and prepare the surgeon and patient occur in patients with IBD, with a 2.5-fold increase in mor-
with respect to extent of resection and need for stoma. tality following a venous thromboembolism (VTE) event
(31). Within this patient population, patients with malnu-
Intraabdominal abscess trition, reduced preoperative functional status, chronic
steroid use, and anemia are all at even higher risk of VTE.
As opposed to other types of intestinal perforations, Emergency surgery and prolonged anesthesia time also
disease progression in CD is generally slow—therefore, increased this risk. Patients who experience a VTE are more
patients are more likely to present with a contained per- likely to have prolonged hospital stays, experience a greater
foration or abscess than free intraabdominal perforation. risk of additional complications, and have a higher mortal-
When patients present with intraabdominal abscess, the ity rate (32). After discharge, VTE events may be as high as
first-line treatment is generally antibiotics with or with- 3.3% for these patients. Patients with increased risks include
out radiologic-guided placement of drainage tube, rather those who have a new stoma, are on steroids, and have a lon-
than immediate surgical intervention. This enables sur- ger postoperative hospital stay (33). Based on this evidence,
gical planning as well as patient optimization. Even if it surgeons should consider extended postoperative prophy-
is clear that a patient will require surgery to address the lactic anticoagulation for patients with IBD at the time of
problem, controlling active inflammation can decrease discharge.
STRICTUREPLASTY
SURGICAL MANAGEMENT OF SMALL
BOWEL DISEASE The idea of revision of a segment of CD rather than resec-
tion was first popularized by Lee in the 1970s. Success after
Some patients with CD are at higher risk of requiring sur- increasing luminal diameter had previously been reported
gery than others. While inflammatory disease is most likely in patients with tuberculosis in India, and this experience
to respond to medical management, fibrostenotic disease is was extrapolated to CD patients with multifocal disease.
least likely to respond. Additional risk factors for surgical The most commonly used strictureplasty is the Heineke-
intervention include terminal ileal disease, proximal small Mikulicz (Figure 36.1). This is used for short-segment dis-
bowel disease, and prior appendectomy (34). There are sev- ease without signs of fistulization, cancer, or abscess. It
eral specific technical issues that merit detailed discussion allows for revision and luminal restructuring without resec-
when considering surgery for CD. tion of intestinal mucosal absorptive surface. Generally,
(a) (b)
(c)
Figure 36.1 Heineke-Mikulicz strictureplasty. (a) Bowel is opened along stricture, (b) backwall is sutured together,
(c) Anterior edges are approximated. Insert demonstrates correct suture placement. (Reproduced with permission from
the Cleveland Clinic.)
Surgical management of small bowel disease / Strictureplasty 337
(a) (b)
(c)
Figure 36.2 Finney stricturoplasties. (a) Stricture is opened on antimesenteric border, (b) Traction on sutures approxi-
mates edges of opened stricture and enterotomy is closed with interupted sutures. Insert demonstrates correct suture
placement. (c) Completed one layer anastomosis. (Reproduced with permission from the Cleveland Clinic.)
segments less than 10 cm are ideal for Heineke-Mikulicz through diseased segments, while maintaining an absorp-
strictureplasty. Longer segments of disease can be treated tive surface within the bowel.
with Finney or Michelassi stricturoplasty techniques. Recent data of the recurrence rates for long-segment
Finney stricturoplasties (Figure 36.2) are generally pre- side-to-side isoperistaltic strictureplasties has been favor-
ferred for single long segment of strictured bowel; although able: 90% of patients were shown to have resolution of symp-
a low-flow diverticulum is created, which may be an area of toms, though 45% had recurrence at a mean of 55 months
stasis within the small intestines. Michelassi, or side-to-side after surgery, with 15 of 83 requiring surgery for their
isoperistaltic strictureplasty (Figure 36.3), can be used for recurrence (35). A recent examination of stricturoplasty
lengths of up to 150 cm. The strictured segment of bowel is that utilized the National Surgical Quality Improvement
married to a dilated area of bowel, causing widening of the Program (NSQIP) database demonstrated decreasing use of
entire length of bowel. This allows for restoration of flow strictureplasty from 5.1% in 2005 to 1.7% of surgeries for
(a) (b)
(c)
Figure 36.3 Michelassi, or side-to-side isoperistaltic strictureplasty. (a) Stricture is divided, opened on anti-mesenteric
border and overlaped. Insert demonstrates tapering ends of opened bowel (b) Back wall sutured together. (c) Anterior
layer is sutured together. (Reproduced with permission from the Cleveland Clinic.)
CD in 2012. Over 9,000 patients underwent strictureplasty wedge resection can be performed, rather than segmental
with an average albumin of 3.6%. During the study period, resection, of the bystander bowel. If the secondary segment
risk of complications and reoperation rate remained low appears to have primary CD, it should be resected as well.
(36). A second review of the literature demonstrated early
complication rates of 15% in patients undergoing “conven- LAPAROSCOPIC VERSUS OPEN
tional strictureplasty” (Heinicke Mikulicz, or Finney) with INTERVENTION
long-term complication rates of 29%. Patients undergo-
ing “unconventional strictureplastly” such as isoperistaltic Following the overall rise of minimally invasive surgical
strictureplasty had even lower complication rates: 8% short techniques, laparoscopy has been increasingly applied to
term and 17% long term (37). CD cases. A small trial found that endoscopic recurrence
was more common after laparoscopic surgery than open
ILEOCOLIC RESECTION surgery (58.3% versus 22.7% at 1 year, and clinical recur-
rence rates were 28.6% versus 11.8%) (40). Other stud-
The most common location of CD is the terminal ileum. ies demonstrate equivalence between groups (41), and a
Approximately 40% of patients present with isolated ter- meta-analysis of data found that there was no evidence of
minal ileal disease. Multiple options exist for treatment, increased rate of surgical recurrence (relative risk [RR] 0.78,
depending on severity of disease as well as presence of CI 0.54–1.11, p = 0.17) but did show reduced preoperative
structuring disease. When disease is fibrotic and stric- complications (RR 0.71, CI 0.58–0.86) and decreased rate
tured, dilation and medical management are less likely to of incisional hernia (0.24, 0.07–0.82) (42). A 2011 Cochrane
be effective. A recent randomized controlled study dem- Review evaluating laparoscopic resection for small bowel
onstrated that resection is a reasonable alternative to bio- disease demonstrated trends toward decreased wound
logics for patients with disease segments less than 40 cm. infection rates and reoperation rates, but these differences
In this study, 37% of patients undergoing treatment with were not statistically significant (43).
biologics required surgery within 4 years, while only 26% An advanced laparoscopic skill set is needed to approach
of patients undergoing surgery required escalation to bio- CD. It is important to run the bowel to evaluate for dis-
logics in the same time period (38). Surgical resection of tant disease. The surgeon should be comfortable handling
an isolated segment of terminal ileal disease may be the inflamed bowel and mesentery, adhesions, and possibly fis-
best option and may leave the patient with no visible dis- tulizing disease or abscess. Often, the mesentery will be sig-
ease in these cases. nificantly thickened in the area of CD. Energy devices may
The resection generally includes the terminal ileum and be inadequate in the treatment of this mesentery, and the
the cecum. Even if the disease does not abut the colon, plac- surgeon should be prepared to assess this intraoperatively,
ing an anastomosis in close proximity to the ileocecal valve and vary approaches as appropriate.
is generally discouraged. This may increase the risk of dila-
tion of the new anastomosis, just upstream from a physi-
ologic sphincter. The ascending colon should be maintained
as much as possible, as patients will often require reopera- TYPE OF ANASTOMOSIS
tion for recurrence of disease in the same location. Equally
important is resection back to the area of healthy appearing The type of anastomosis used to restore intestinal continu-
bowel, without external signs of CD (corkscrew vessels and ity may be important both in terms of perioperative risks
creeping fat), and without internal signs of CD (aphthous and long-term disease recurrence. Data from Simillis et al.
ulcers, or cobblestoning). The small bowel should be soft demonstrated that end-to-end anastomosis had increased
and pliable at the transection location. risk of anastomotic leaks in CD as well as increased recur-
Resection of repeated segments of the terminal ileum rence rates (44). However, they also noted a decreased rate
may lead to issues with malabsorption such as wasting of of other postoperative complications and a shorter hos-
bile salts, fat-soluble vitamins (A, D, E, K), and B-12 related pital stay. The explanation for this is unclear. However,
anemia (39). Generally, patients with greater than 20 cm of a meta-analysis demonstrated no change in anastomotic
terminal ileal disease are at greater risk of deficiencies. In leak rate but a decreased rate of recurrence with stapled
cases of repeat resection, consideration of strictureplasty side-to-side anastomosis when compared to other anas-
may be reasonable. Generally, this will be an isoperistaltic tomotic configurations (45). Conversely, a second meta-
anastomosis with the longitudinal opening on the diseased analysis found statistically reduced rates of postoperative
small intestine to the normal ascending colon. complications (0.54, 0.32–0.93), anastomotic leak (0.45, CI
Disease segments in the terminal ileum may be noted 0.20–1.00), and recurrence rates (OR 0.20, CI 0.07–0.55).
to have fistulized to healthy appearing adjacent bowel. This Perioperative morbidity and hospital stay were similar
may include proximal segments of small intestine, or fre- (46). Other data have suggested that the length of anas-
quently as segment of sigmoid colon. Prior colonoscopy, or tomosis may be a key to decreasing clinically significant
if needed intraoperative colonoscopy, can assess the seg- recurrence rates, and this can be achieved by utilizing a
ment of colon to rule out primary disease. In this case, a longer stapler or multiple firings. A more recent technique,
Crohn colitis / Colectomy 339
(a) (b)
(c) (d)
Figure 36.4 Kono side-to-side anastomosis. (a) The intestinal segments are divided with the staple lines perpendicular
to the axis of the mesentery. (b) The supporting column is constructed by suturing the two staple lines together and a
longitudinal enterotomy and colotomy are performed. They start no more than 1 cm away from the supporting column,
extending proximally and distally to allow a transverse lumen of 7 cm on the small intestine and close to 8 cm on the large
intestine. (c) The longitudinal enterotomy and colotomy are closed transversely with an outer layer of 4/0 silk Lembert
interrupted sutures, and an inner layer of running 3/0 absorbable suture starting on the posterior wall. (d) The complete
anastomosis is shown.
popularized by Japanese surgeon T. Kono, has impres- strictures occur in approximately 17% of cases of colonic
sive early results regarding recurrence (Figure 36.4). The CD (48). Endoscopic balloon dilation has been used to
current data evaluating 187 patients undergoing Kono-S manage strictures with a 90% success rate; 28% required
antimesenteric anastomosis demonstrated 98.6% freedom surgery for recurrence at approximately 4 years of follow-
from recurrence requiring surgery at 5 and 10 years after up (49). Endoscopic dilation is most suitable for isolated,
surgery (47). short strictures. Risks include bleeding and perforation,
and the major benefit is potentially avoiding a resection.
Colonic stricturoplasty for CD can be used for short stric-
tures and has equivalent outcomes to resection, but there
CROHN COLITIS is about a 7% chance that a stricture harbors a malignancy
(48,50). Shorter strictures and longer duration of disease are
CD affects the colon in approximately 30% of patients. It both associated with a higher risk. Colonic strictures, then,
may affect the entire colon or be segmental, and it may should be carefully surveyed and biopsied based on the can-
occur in conjunction with other disease distributions. cer risk. Resection is the preferred management strategy for
Perianal disease, in particular, suggests a more aggressive strictures in Crohn colitis (51).
phenotype for Crohn colitis. Indications for surgery for
Crohn colitis include strictures, perforation, hemorrhage, COLECTOMY
severe or fulminant colitis, medically refractory disease,
and neoplasia. When CD affects only a segment of the colon and surgery is
required, segmental resection may be performed when indi-
COLONIC STRICTURES cated without any alteration in complications, incidence of
recurrence, or need for a permanent stoma when compared
Colonic strictures may be inflammatory or fibrotic in to subtotal or total colectomy (52). However, many authors
nature, and both mechanisms often coexist. Obstructive believe that because of the high recurrence rate, when
operating on the colon for CD, a subtotal colectomy or total distention. Any worsening of the patient’s condition should
proctocolectomy should be performed regardless. When prompt operative intervention. Severe colitis may progress
two or more segments of colon are affected, it appears that to perforation, toxic megacolon, hemorrhage, or peritonitis.
subtotal or total colectomy is beneficial. Laparoscopic colec- When surgery is undertaken for severe, medically refrac-
tomy for CD can be performed safely with a reported con- tory disease, a staged approach is logical because of the
version rate of 16%–26% and the benefit of quicker return of added operative risk of proctectomy and the possibility of
bowel function and shorter hospital stay (53–55). An urgent a nonhealing perineal wound. The rectal stump may be left
or emergent operation for colonic CD does not change the closed either with staples or sutures in the pelvis, or tacked
beneficial impact of laparoscopy (56). above the anterior abdominal wall fascia to avoid the risk of
There is a subset of patients who have Crohn colitis with stump disruption and pelvic abscess. There are no data on
relative rectal sparing and no perianal disease who are can- the use of rectal tubes in the setting of subtotal colectomy
didates for colectomy with ileorectal anastomosis. These for Crohn colitis, but some routinely use this approach.
patients should first be evaluated to make sure their rectal Occasionally it is necessary to form a mucous fistula when
compliance is normal, either by distending it via proctoscopy the residual rectosigmoid is too inflamed to hold sutures
or with manometry. Those with a tolerated rectal volume of or staples. Completion proctectomy can be performed at
less than 150 cc do not have good function with ileorectal a later date when the patient is medically and nutritionally
anastomosis (57). Some of these patients, or those with iso- optimized. Patients should be counseled about the options
lated disease in the colon only above the distal rectum, may of completion proctectomy or leaving the residual rectum
be candidates for an ileal pouch, which can be attached to the with surveillance endoscopy due to the risk of neoplasia and
mid- to distal rectum (58), and while they have a relatively the possibility of disease-related symptoms (64,65).
high incidence of disease recurrence and pouch loss (10%–
52%), some are able to avoid a permanent stoma (59–62). COLON CANCER
Most patients with Crohn colitis require total procto-
colectomy with end ileostomy. These patients may have The relative risk of colon cancer in CD is 2.4 times that of the
proctitis, sphincter dysfunction, or severe perianal disease. general population (66). Patients with Crohn colitis have up
There is a significant risk of cancer in the residual rectum, to a 25% chance of dysplasia by their 10th surveillance colo-
so even if it is done as a staged procedure, the entirety of the noscopy (67,68). Patients who have CD that does not affect
rectum should ideally be removed (63). It is also feasible in their colon are not at particularly increased risk of colon
some cases to perform an ultra-low Hatrmann procedure cancer (69). When low- or high-grade dysplasia is biop-
leaving only a few centimeters of rectum so there is less sied, it should be confirmed by a pathologist experienced
residual rectum, and it also enables the surgeon to perform in IBD, as interobserver variability is high. Additionally,
the completion proctectomy from the perineal approach. when dysplasia is found incidentally using standard endos-
This is preferable as it avoids another abdominal operation copy techniques, the examination should be repeated with
and its associated morbidity for the patient. high-definition and/or chromoendoscopy techniques (70).
Dysplasia often, but not always, precedes colorectal cancer,
TOXIC COLITIS and dysplasia does not always progress from low to high
before manifesting as invasive carcinoma in CD. Patients
Severe colitis is a serious and potentially fatal manifestation with high-grade dysplasia on random biopsies have a 73%
of Crohn colitis. A flare accompanied by six or more bloody chance of cancer in the final specimen, and with low-grade
stools per day with evidence of systemic toxicity including dysplasia the risk is 36% (71). When dysplasia is present in
anemia, elevated erythrocyte sedimentation rate (ESR), a visible lesion that can be completely resected endoscopi-
fever, or tachycardia can be a strong indication for surgery cally, close surveillance is appropriate. Multifocal high-
if it does not respond to initial medical management. Those grade dysplasia should be a consideration for colectomy, and
with more than 10 bloody bowel movements per day, con- obviously invasive cancer warrants colectomy. In general,
tinuous daily bleeding, transfusion requirements, elevated patients with CD undergoing surgery for neoplasia should
fever, tachycardia, tenderness and distention, and colonic have a complete proctocolectomy because of the high rate
dilation on imaging are labeled as having fulminant colitis of synchronous lesions as well as the high rate of metachro-
(51). When there is dilation greater than 5.5 cm on imag- nous cancer (72).
ing in association with systemic symptoms, toxic mega-
colon should be considered, and the risk of perforation is
heightened.
Patients should be evaluated for concomitant Clostridium
PERIANAL CROHN DISEASE
difficile and cytomegalovirus infection. Endoscopy can be
carefully performed to evaluate the mucosa. Medical ther- The incidence of perianal disease in CD patients ranges from
apy includes high-dose steroids, antibiotics to minimize 25% to 40% (73–75). It may represent the presenting symp-
sepsis from microperforation, and close serial abdominal tom of CD for a subset of patients (76). Those patients with
examinations with x-rays as necessary to evaluate colonic multiple abnormalities, atypical fissures located laterally,
Perianal Crohn disease / Perianal abscess and fistula 341
cavitating ulcers, complex fistulas, large fibrotic skin tags, CD. While some have demonstrated a decreased need for
and other unusual features of typical perianal problems surgical intervention with infliximab, others have shown
should be suspected of CD. When there is isolated perianal no difference in fistula surgery rates since the introduction
disease at the time of diagnosis, most patients manifest of these agents in population-based studies. However, the
intraluminal disease within 5 years. ACCENT II trial demonstrated that infliximab must be
continued as maintenance therapy to increase the change of
PERIANAL ABSCESS AND FISTULA a sustained response (81). In cases refractory to infliximab,
adalimumab can be used with long-term response rates of
The incidence of perianal fistulas appears to correlate with approximately 41% without surgical intervention (82,83).
the location of luminal disease: 15% of patients with iso- Currently, several sets of guidelines recommend anti-TNF
lated ileocolic disease develop fistulas, but more than 90% agents as the standard for complex anorectal CD (84–87).
of those with Crohn colitis and proctitis are affected (73). In rectovaginal fistulas, the response to anti-TNF agents
Abscess presents with pain, fever, erythema, and fluctu- is decreased compared to other types of fistulas; 14%–38%
ance in any of the potential spaces where a sporadic peri- versus 46%–78% (81,88).
anal abscess may develop. However, in CD, complex fistulas
should always be suspected. In cases where it is not imme- Surgical management
diately apparent that there is a drainable collection, imag-
ing may be helpful. Endorectal ultrasound increases the The goal of treatment of anal fistulas in CD patients should
sensitivity of physical examination in this group of patients be the absence of symptoms rather than the absence of the
(77). CT is easily available and has excellent effectiveness in fistula. Many individuals with fistulae are relatively asymp-
diagnosing abscesses. While MRI is usually beneficial for tomatic. A single institution series with long-term follow-up
delineating fistulas, it is not typically necessary to identify determined that the overall cumulative probability of avoid-
an abscess. ing proctectomy was 91.6% at 10 years and 82.5% at 20 years
An abscess should be treated first with simple incision in this patient group. Resection of all proximal CD did not
and drainage. An attempt should be made to find the asso- ameliorate the anorectal condition, except in those with all
ciated fistula, with care to avoid making false tracts. Long- proximal disease removed who did not have a recurrence (89).
term, continuous drainage can be facilitated by insertion It is important to determine if there is concomitant anal
of a mushroom or Malecot catheter or by the application inflammation related to the CD. Those who do not have
of seton drainage. Antibiotics are often prescribed if there active CD in the anal canal are more likely to have success
is associated cellulitis or signs of systemic illness. Most with any intervention. In those with anal canal disease, it
abscesses in CD patients eventually develop into fistulas. In may be more prudent to manage their symptoms with long-
addition, it is possible for a patient to present with a symp- term seton placement until their anal canal is normal on
tomatic fistula even without ever reporting an abscess. examination and there are no signs of inflammation in their
A loose or noncutting seton, often made from silastic proximal bowel (90).
vessel loops and secured in a circular or tear-shaped loop, Definitive management of fistulas is based on location,
maintains the patency fistula tract and enables it to drain complexity, the presence or absence of proctitis, and anal
freely. The primary goal of the seton is to allow local sepsis canal disease. Complexity is defined as the presence of an
and inflammation to resolve. In many cases, it will be a tem- internal opening above the dentate line, traversing signifi-
porary solution until a suitable time for a more definitive cant sphincter muscle, branching, or multiple openings.
procedure, but for some patients with complex, refractory Rectovaginal fistulas are always considered to be complex
disease, it can be a long-term solution. The combination of fistulas (91).
seton and anti-TNF agents has been very effective for the It is often helpful to perform examination under anes-
treatment of fistulae. Patients treated with infliximab and thesia to fully delineate the fistula prior to determining a
noncutting seton placement have better outcomes than plan for surgical intervention. Proctosigmoidoscopy should
patients treated with infliximab alone in terms of response be performed at the same time to evaluate the presence and
rate, recurrence rate, and time to recurrence (78,79). The activity of inflammatory changes in the rectum and distal
optimal timing for seton removal with this approach is not colon. The presence of an anal stricture is a particularly poor
clearly defined, but initial studies reported that setons were prognostic indicator for both the success of any attempt at
removed after the second infusion with infliximab. repairing the fistula and the possibility of preserving the
Medical management plays an important role in peri- rectum in the long term.
anal fistula management. It has been suggested that met- Imaging can increase the accuracy of the examination,
ronidazole may produce symptomatic improvement in either via endorectal ultrasound or pelvic MRI. Either of
some patients with perianal disease, though it nearly always these modalities has been demonstrated to improve diag-
recurs after the antibiotics are stopped. Immunomodulators nostic accuracy (92,93). MRI can show the path of the
increase the rate of healing in approximately half of fistula fistula, the presence of deeper abscesses not apparent on
patients (80). Furthermore, anti-TNF medications such as clinical examination, and relationship to the sphincters,
infliximab have altered the treatment paradigm for perianal and can differentiate between inflammatory changes and
fibrosis (94). Furthermore, MRI has been shown to sub- repairs, healing up to 91% of fistulas (105). LIFT has not
stantially alter the operating plan with additional or more been well studied in the context of CD; one small series of
complex findings than were demonstrated on EUA in up 15 patients reported 53% healing at 12 months, but three
to 40% of cases (95,96). MRI may also be helpful in assess- patients developed a novel fistula (106). However, because
ing the persistence of fistulas after medical management it is performed entirely outside of the lumen and does uti-
that has resolved symptoms. Despite clinically appearing lize anal or rectal mucosa while also avoiding damage to the
healed, inflammatory changes may persist on imaging and sphincter complex, some practitioners prefer it as an initial
demonstrate that the fistula is still present. For patients who approach in CD-related fistulas.
are considering a change in their therapy or who have been
diverted to allow time to heal, this is of particular impor- Rectovaginal fistula
tance (97).
For simple, low-lying fistulas without associated procti- Rectovaginal fistula (RVF) affects 9% of women with CD;
tis, fistulotomy may be appropriate if there is no threat to they more commonly occur in conjunction with colonic dis-
continence. Success rates for this range from 80% to 100% ease than small bowel disease, following the same pattern as
(91). Timing for fistulotomy is important: there is no associ- other manifestations of perianal CD (107). It has a substan-
ated inflammation to increase the chance of healing. When tial impact on quality of life and can be quite disabling for
performed in the setting of biologic therapy, the rate of heal- those who are affected. The presence of a RVF also increases
ing significantly improves as well (98). the risk of requiring proctectomy for disease control (98).
Both fibrin glue and collagen plugs have been described Common presenting symptoms are air or stool from the
for treatment of CD-related fistulas. The advantage of these vagina, but patients may also report purulence from the
techniques is there is no risk to the sphincter complex. Fibrin vagina, dyspareunia, irritation, and urinary tract infection.
glue is a mix of fibrinogen and thrombin with the goal of Though on examination it may be possible to visualize the
plugging the hole and promoting healing. The best reported fistula, the diagnosis can be made with a high clinical sus-
results are from a study of highly selected CD patients who picion based on symptoms even when it cannot be visual-
had a 38% success rate, with 15% of those recurring by 16 ized in an outpatient setting. In these cases, examination
weeks and no longer-term follow-up, and most authors under anesthesia is mandated prior to planning repair. It is
report significantly lower success rates (99,100). The fistula also possible for fistulae to develop from the colon or small
plug is made of porcine submucosa that is inserted into the bowel to the vagina in CD.
tract. It acts as a scaffolding for healing. A meta-analysis of There are numerous options for addressing RVFs in
the small series that included CD patients found an overall women with CD. The success of the repair depends on over-
success rate of 55%, but with varying follow-up and recur- all CD control, patient comorbidities, inflammation of the
rence rates (101). While early results were encouraging, involved tissue, anal stricture, and the size of the fistula. As
long-term results have been discouraging, particularly in with perianal fistulas, prior to undertaking definitive repair,
CD patients, and these methods are generally not employed. medical management should be optimized, all local inflam-
Endorectal advancement flaps can be used when there mation and infection should be completely resolved, and a
is no proctitis or anal stenosis. It is often the procedure of staged approach may be helpful. It is important to manage
choice for rectovaginal fistulas. The technique is identical patient expectations, as there is a high rate of recurrence
for treating non-CD-related fistulas; a broad flap is mobi- and need for multiple procedures, and for patients who are
lized from mucosa and submucosa. The internal opening asymptomatic or minimally symptomatic, there is no man-
is excised. Some surgeons will close the internal opening. date to treat the fistula. The surgical approach may be trans-
The flap is mobilized until it can be brought down below the anal, transvaginal, or transperineal. Tissue transfer may be
opening and secured in a tension-free manner with inter- used to treat RVF. The use of biologic mesh interposition
rupted sutures. Advancement flaps have a success rate of up has been reported with disappointing results; the failure
to 71% in series with CD patients (102). There is a potential rate was 80% (108). The overall healing rate for all tech-
for alterations in continence with this procedure. While the niques is 63%, and multiple procedures are often required
success rates of this technique for sporadic cryptograndular (107). Consideration for diverting stoma should be given,
fistulas are reportedly 81%, it was 64% in CD patients in a either before the repair to aid in diminishing inflammation
meta-analysis that specifically compared these groups (103). or at the time to allow the repair time to heal.
Ligation of the intersphincteric tract (LIFT) is a more When there is no sphincter involvement, very low sim-
recently described procedure that poses minimal to no dan- ple fistulas can be treated with fistulotomy. Transrectal
ger to the sphincter complex (104). A probe is placed into the approach usually involves an endorectal advancement flap.
fistula tract, and an incision is made in the intersphincteric It is an option only when the anorectal mucosa is healthy.
groove. The fistula tract is then dissected out and tied off at Initial healing rates are approximately 54%–71% (109,110).
the internal and external sphincter; it is then divided. The Transperineal repairs are particularly well suited for RVF
remainder of the external tract is curetted, and the external with concomitant sphincter defects. The first method, epi-
opening is widened. The skin incision is closed. This tech- sioproctotomy involves a linear incision over the perineum,
nique is relatively effective in comparison with other fistula in essence creating a fourth-degree tear. The fistula tract is
Perianal Crohn disease / Anal stenosis 343
debrided back to healthy tissue. The edges of the sphincter long-standing fistulas in patients with severe perineal dis-
are mobilized, the rectal mucosa is repaired, and an over- ease (117). The presentation can range from increasing
lapping sphincteroplasty is performed. The vagina is then pain to pelvic sepsis, or with typical fistula, symptoms such
repaired. The second option for transperineal repair is a as chronic drainage unchanged from baseline. When a
transverse approach, which utilizes a transverse incision patient has a long-standing, quiescent fistula that suddenly
through the perineal body. The plane is dissected proxi- becomes symptomatic, carcinoma should be suspected.
mally, mobilizing the posterior vagina and anterior rectum Adenocarcinoma and squamous cell carcinoma are both
until healthy tissue above the fistula is reached. The tract described in this setting, and treatment ranges from chemo-
is debrided, and then all layers are reapproximated. Both radiation to proctectomy with wide margins in the perianal
of these transperineal approaches have success rates of region (118).
approximately 70% (111). The vaginal approach may be ben-
eficial when there is significant scarring or disease of the FISSURES
anal canal limiting the ability to manipulate the tissues. A
vaginal flap is created similarly to an endoanal advancement Fissures in CD may be located in the anterior or posterior
flap. The layers are separated and repaired at the level of the midline like idiopathic fissures, but can occur in any loca-
fistula, and then the flap is tacked down. Levator muscles tion around the anus. Atypical presentation of a fissure
are closed between the layers as reinforcement. should always raise suspicion that it is the first clinical man-
Tissue transfer typically involves using a gracilis mus- ifestation of CD. The etiology may be related to the inflam-
cle flap or healthy tissue from the labia majora, a Martius matory process, trauma from excessive diarrhea, or from
flap. The gracilis flap is usually used as a second- or third- the same etiology as sporadic fissures.
line intervention, after other techniques have failed (112). Many fissures in CD are painless unless they are associ-
A Martius flap can be combined with transperineal repairs ated with an abscess, and many respond to medical therapy.
(113). A longitudinal incision is made in the labia majora, The same symptomatic therapies used for idiopathic fissures,
and the bulbocavernosus and its associated fat pad are dis- i.e., nitroglycerin, calcium channel blockers, warm baths,
sected out; it remains attached inferiorly. They are passed pain medications, and bowel management, may be trialed,
through a subcutaneous tunnel and brought through to the but their effectiveness in CD has not been evaluated. Topical
fistula where they are tacked posterior to the vaginal wall, mesalamine or hydrocortisone suppositories are often help-
which is closed over it. ful to promote healing, and limiting the amount of diarrhea
either with systemic medications or antidiarrheal agents is
Fecal diversion also beneficial. In general, surgical treatment for isolated fis-
sures in patients with active CD should not be undertaken,
Fecal diversion may be employed to allow fistula repairs as there is a real risk of nonhealing and incontinence.
time to heal, or to allow inflammation to subside. Diversion
alone improves perianal disease in approximately 80% of SKIN TAGS
patients, but about half will relapse once they are reversed
(114). Diversion or proctectomy may be required when Many patients with CD will have characteristic, large, atypi-
perianal disease is so severe that it cannot be salvaged cally shaped skin tags. The etiology is thought to be from
through maximal medical and surgical options. Predictors fissures and chronic inflammation leading to lymphedema.
of diversion include colonic disease, anal strictures, and A conservative surgical approach is critically important
multiple complex fistulas (115). Many patients are reluctant in CD patients. Though these skin tags may be large and
to commit to permanent stoma; for this group of patients, cause hygiene difficulties as well as pruritus, outcomes of
temporary diversion has the benefit of showing them the any attempt to excise them may be worse than the initial
improvement in quality of life and makes it easier for them complaint. Chronic, nonhealing anal or perianal ulcers can
to accept permanent stoma. In patients with severe perianal be painful and more difficult to manage.
disease, a multidisciplinary approach working with plastic
and reconstructive surgeons for flaps is recommended. In ANAL STENOSIS
patients with complex or nonhealing fistulas, fecal diversion
may play a role. This generally improves quality of life, but Anal stenosis or strictures may present later than would be
fewer than 25% of these patients have intestinal continuity expected in CD patients due to softer stool consistency. In
restored after a secondary procedure aimed at resolving the large series from major referral centers, the prevalence of
perianal problem (114,116). Proctectomy is required in 5% anal strictures is up to 22% (89). Most often, they occur in
of CD patients for perianal disease alone. the presence of perianal disease or represent consequences of
long-term inflammation. Patients may report overflow diar-
Carcinoma in a fistula rhea, perineal pain, constipation, and fecal incontinence.
Medical management may be helpful for the inflammatory
Cancer in a chronic perianal fistula is extremely rare in component of the stricture, but surgical intervention should
CD and typically occurs after years of active disease and be limited to patients who have difficulty with evacuation.
Dilation, either with digital dilation or Hegar dilators, is of endoscopic and clinical findings, and also less need for
the most common management technique. This may need additional surgery (122,138). Adalimumab appears to be as
to be repeated at regular intervals. Many patients can be effective as infliximab (139), but there is less data regarding
taught to maintain dilation performed in the office or oper- other anti-TNF agents or other newer medications such as
ating room using a set of Hegar dilators at home on a regu- vedolizumab.
lar basis. Balloon dilation has also been popular but is more
expensive and must be done in an office or endoscopy set- REFERENCES
ting (119). Use of a bougie dilator has also been successfully
described (120). Potential risks associated with stricture 1. Kaplan GG. Nat Rev Gastroenterol Hepatol 2015;
dilation include bacteremia, worsening pain, and perfora- 12(12):720–7.
tion. Fecal incontinence from chronic injury to the sphinc- 2. Molodecky NA et al. Gastroenterology 2012;142(1):
ter, both from the stenosis and the dilation procedure, is 46–54.e42.
also a concern. 3. Hanauer SB et al. Lancet 2002;359(9317):1541–9.
Historically, anorectal strictures in CD have been a poor 4. Colombel J et al. Gastroenterology 2007;132(1):
prognostic indicator and often suggested the need for a 52–65.
stoma (115,121). Even with the use of biologic therapy, which 5. Jeuring SFG et al. Am J Gastroenterol 2017;112(2):
is often effective for other perianal manifestations of CD, 325–36.
long-term outcomes in patients with anal strictures were 6. Hatch QM et al. J Gastrointest Surg 2016;20(11):
not significantly improved (122). 1867–73.
7. de Groof EJ et al. Colorectal Dis 2017;19(6):551–8.
8. Morar PS et al. J Crohns Colitis 2015;9(6):483–91.
9. Yang Z-P et al. Int J Surg 2014;12(3):224–30.
POSTOPERATIVE MEDICAL THERAPY 10. Kopylov U et al. Inflamm Bowel Dis 2012;18(12):
2404–13.
Since surgery for CD is not curative, there should be some 11. El-Hussuna A et al. Dis Colon Rectum 2013;56(12):
consideration of empiric therapy to prevent endoscopic and 1423–33.
clinical recurrence after resection. Up to 80% of CD patients 12. Yamamoto T et al. United Eur Gastroenterol J 2016;
will have an endoscopic recurrence at 1 year after surgery 4(6):784–93.
(123). Factors that should be weighed include both disease- 13. Lightner AL et al. J Crohns Colitis 2017;11(2):185–90.
and patient-specific factors to reach an individualized plan. 14. Mikocka-Walus AA et al. Inflamm Bowel Dis 2012;
Smoking (124,125), history of multiple resections (126,127), 18(8):1582–7.
and a perforating phenotype (128,129) have all been identi- 15. Louis E et al. J Crohn’s Colitis 2015;9(8):685–91.
fied as high risk for recurrent disease (130). Length of resec- 16. Shental O et al. Dis Colon Rectum 2012;55(11):
tion, perianal disease, and short duration from onset to 1125–30.
surgery have also been proposed as high-risk features. 17. Tukey M et al. Am J Gastroenterol 2009;104(11):
If the decision is made to treat prophylactically after sur- 2734–9.
gery, the options for medications are the same as those in 18. Kopylov U et al. Dig Liver Dis 2017;49(8):854–63.
the preoperative setting. In several meta-analyses, mesala- 19. Dionisio PM et al. Am J Gastroenterol 2010;105(6):
mine appears to decrease the risk of clinical recurrence but 1240–8; quiz 1249.
is less favorable in terms of endoscopic recurrence (131,132). 20. Cheifetz AS et al. Am J Gastroenterol 2006;101(10):
Thiopurines are more efficacious in reducing endoscopic 2218–22.
and clinical recurrences but have a higher likelihood of 21. Kordbacheh H et al. Inflamm Bowel Dis 2017;23(6):
medication intolerance and side effects, leading to 22% ces- 1025–33.
sation in one clinical trial of patients in the postoperative 22. Jensen MD et al. Inflamm Bowel Dis 2011;17(5):
period (133,134). Metronidazole and other similar antibi- 1081–8.
otics have been explored in the postoperative period and 23. Qiu Y et al. Aliment Pharmacol Ther 2014;40(2):
found to be relatively well tolerated as a 3-month course but 134–46.
probably do not further reduce risk if used with another 24. Ordás I et al. Gastroenterology 2014;146(2):374–82.e1.
agent (135). Using metronidazole postoperatively but prior 25. Kroeker KI et al. J Clin Gastroenterol 2011;45(1):
to introduction of a biologic agent has not been studied, but 34–9.
starting these agents is often delayed by weeks to months 26. de Groof EJ et al. Dig Dis 2014;32(s1):103–9.
after surgery, and they can be used as a bridge in this setting. 27. He X et al. J Clin Gastroenterol 2015;49(9):e82–90.
Notably, a decision analysis model found that antibiotics are 28. Clancy C et al. J Crohns Colitis 2016;10(2):202–8.
the most cost effective for reducing clinical postoperative 29. Morar PS et al. Aliment Pharmacol Ther 2015;42(10):
recurrence (136). Anti-TNF therapy is the most efficacious 1137–48.
for reducing clinical and endoscopic recurrences (131,137). 30. Li Y et al. Br J Surg 2015;102(11):1418–25;
There are improved short- and long-term outcomes in terms discussion 1425.
References 345
31. Murthy SK, Nguyen GC. Am J Gastroenterol 2011; 63. Cirincione E et al. Dis Colon Rectum 2000;43(4):544–7.
106(4):713–8. http://www.ncbi.nlm.nih.gov/pubmed/10789755.
32. Wallaert JB et al. Dis Colon Rectum 2012;55(11): Accessed January 4, 2017.
1138–44. 64. Lavery IC, Jagelman DG. Dis Colon Rectum 1982;
33. Brady MT et al. Dis Colon Rectum 2017;60(1):61–7. 25(6):522–4. http://www.ncbi.nlm.nih.gov/pubmed/
34. Liverani E et al. World J Gastroenterol 2016;22(3): 7117054. Accessed January 4, 2017.
1017. 65. Guillem JG et al. Dis Colon Rectum 1992;35(8):
35. Fazi M et al. JAMA Surg 2016;151(5):452–60. 768–72. http://www.ncbi.nlm.nih.gov/pubmed/
36. Geltzeiler CB et al. J Gastrointest Surg 2015;19(5): 1644001. Accessed January 4, 2017.
905–10. 66. von Roon AC et al. Dis Colon Rectum 2007;50(6):
37. Campbell L et al. Dis Colon Rectum 2012;55(6):714–26. 839–55.
38. Ponsioen CY et al. Lancet Gastroenterol Hepatol 67. Itzkowitz SH, Present DH, Crohn’s and Colitis
2017;2(11):785–92. Foundation of America Colon Cancer in IBD Study
39. Ward MG et al. Inflamm Bowel Dis 2015;21(12): Group. Inflamm Bowel Dis 2005;11(3):314–21. http://
2839–47. www.ncbi.nlm.nih.gov/pubmed/15735438. Accessed
40. Bellinger J et al. J Laparoendosc Adv Surg Tech January 4, 2017.
2014;24(9):617–22. 68. Friedman S et al. Clin Gastroenterol Hepatol 2008;
41. Stocchi L et al. Surgery 2008;144(4):622–8. 6(9):993–8.
42. Patel SV et al. BMC Surg 2013;13(1):14. 69. van den Heuvel TRA et al. Int J Cancer 2016;139(6):
43. Dasari BV et al. Laparoscopic versus open surgery 1270–80.
for small bowel Crohn’s disease. Cochrane Database 70. Laine L et al. Gastrointest Endosc 2015;81(3):489–
Syst Rev 2011;(1):CD006956. 501.e26.
44. Simillis C et al. Dis Colon Rectum 2007;50(10):1674–87. 71. Kiran RP et al. Ann Surg 2012;256(2):221–6.
45. Guo Z et al. World J Surg 2013;37(4):893–901. 72. Maser E a et al. Inflamm Bowel Dis 2013;19(9):
46. He X et al. Dig Dis Sci 2014;59(7):1544–51. 1827–32.
47. Kono T et al. J Gastrointest Surg 2016;20(4):783–90. 73. Schwartz DA et al. Gastroenterology 2002;122(4)
48. Yamazaki Y et al. Am J Gastroenterol 875–80. http://www.ncbi.nlm.nih.gov/pubmed/
1991;86(7):882–5. http://www.ncbi.nlm.nih.gov/ 11910338. Accessed January 4, 2017.
pubmed/2058631. Accessed January 4, 2017. 74. Beaugerie L et al. Gastroenterology 2006;130(3):
49. Wibmer AG et al. Int J Colorectal Dis 2010;25(10): 650–6.
1149–57. 75. Gelbmann CM et al. Am J Gastroenterol 2002;97(6):
50. Broering DC et al. Int J Colorectal Dis 2001;16(2): 1438–45.
81–7. http://www.ncbi.nlm.nih.gov/pubmed/ 76. Williams DR et al. Dis Colon Rectum 24(1):22–4.
11355323. Accessed January 4, 2017. http://www.ncbi.nlm.nih.gov/pubmed/7472097.
51. Strong S et al. Dis Colon Rectum 2015;58(11): Accessed January 4, 2017.
1021–36. 77. el Mouaaouy A et al. Z Gastroenterol 1992;30(7):
52. Tekkis PP et al. Colorectal Dis 2006;8(2):82–90. 486–94. http://www.ncbi.nlm.nih.gov/pubmed/
53. da Luz Moreira A et al. J Gastrointest Surg 2007; 1509788. Accessed January 4, 2017.
11(11):1529–33. 78. Regueiro M, Mardini H. Inflamm Bowel Dis 2003;9(2):
54. Holubar SD et al. Inflamm Bowel Dis 2010;16(11): 98–103. http://www.ncbi.nlm.nih.gov/pubmed/
1940–6. 12769443. Accessed January 4, 2017.
55. Umanskiy K et al. J Gastrointest Surg 2010;14(4): 79. Topstad DR et al. Dis Colon Rectum 2003;46(5):
658–63. 577–83.
56. Marceau C et al. Surgery 2007;141(5):640–4. 80. Pearson DC et al. Ann Intern Med 1995;123(2):
57. Keighley MR et al. Gut 1982;23(2):102–7. http://www. 132–42.
ncbi.nlm.nih.gov/pubmed/7068033. Accessed January 81. Present DH et al. N Engl J Med 1999;340(18):
4, 2017. 1398–405.
58. Kariv Y et al. J Am Coll Surg 2009;208(3):390–9. 82. Fortea-Ormaechea JI et al. Gastroenterol Hepatol
59. Melton GB et al. Ann Surg 2008;248(4):608–16. 2011;34(7):443–8.
60. Regimbeau JM et al. Dis Colon Rectum 2001;44(6): 83. Echarri A et al. J Crohns Colitis 2010;4(6):654–60.
769–78. http://www.ncbi.nlm.nih.gov/pubmed/ 84. Gecse KB et al. Gut 2014;63(9):1381–92.
11391134. Accessed January 4, 2017. 85. Orlando A et al. Dig Liver Dis 2011;43(1):1–20.
61. Sagar PM et al. Dis Colon Rectum 1996;39(8):893–8. 86. Sciaudone G et al. Can J Surg 2010;53(5):299–304.
http://www.ncbi.nlm.nih.gov/pubmed/8756845. http://www.ncbi.nlm.nih.gov/pubmed/20858373.
Accessed January 4, 2017. Accessed January 4, 2017.
62. Braveman JM et al. Dis Colon Rectum 2004;47(10): 87. Hyder SA et al. Dis Colon Rectum 2006;49(12):
1613–9. 1837–41.
88. Parsi MA et al. Am J Gastroenterol 2004;99(3): 115. Galandiuk S et al. Ann Surg 2005;241(5):796–801.
445–9. 116. Kasparek MS et al. Dis Colon Rectum 2007;50(12):
89. Wolff BG et al. Dis Colon Rectum 1985;28(10):709–11. 2067–74.
http://www.ncbi.nlm.nih.gov/pubmed/4053875. 117. Church JM et al. Dis Colon Rectum 1985;28(5):361–6.
Accessed January 4, 2017. http://www.ncbi.nlm.nih.gov/pubmed/3158499.
90. Thornton M, Solomon MJ. Dis Colon Rectum 2005; Accessed January 5, 2017.
48(3):459–63. 118. Shwaartz C et al. Dis Colon Rectum 2016;59(12):
91. Sandborn WJ et al. Gastroenterology 2003;125(5): 1168–73.
1508–30. http://www.ncbi.nlm.nih.gov/pubmed/ 119. Singh VV et al. J Clin Gastroenterol 2005;39(4):
14598268. Accessed January 4, 2017. 284–90. http://www.ncbi.nlm.nih.gov/pubmed/
92. Schwartz DA et al. Gastroenterology 2001;121(5): 15758621. Accessed January 5, 2017.
1064–72. 120. Kashkooli SB et al. Can J Surg 2015;58(5):347–8.
93. Villa C et al. Eur J Radiol 2012;81(4):616–22. 121. Linares L et al. Br J Surg 1988;75(7):653–5. http://
94. Haggett PJ et al. Gut 1995;36(3):407–10. http:// www.ncbi.nlm.nih.gov/pubmed/3416120. Accessed
www.ncbi.nlm.nih.gov/pubmed/7698701. Accessed January 5, 2017.
January 4, 2017. 122. Uchino M et al. World J Gastroenterol 2011;17(9):
95. Buchanan GN et al. Radiology 2004;233(3):674–81. 1174.
96. Beets-Tan RG et al. Radiology 2001;218(1):75–84. 123. Buisson A et al. Dig Liver Dis 2012;44(6):453–60.
97. Van Assche G et al. Am J Gastroenterol 2003;98(2): 124. Reese GE et al. Int J Colorectal Dis 2008;23(12):
332–9. 1213–21.
98. El-Gazzaz G et al. Color Dis 2012;14(10):1217–23. 125. Kane S V et al. J Clin Gastroenterol 2005;39(1):
99. Grimaud J et al. Gastroenterology 2010;138(7): 32–5. http://www.ncbi.nlm.nih.gov/
2275–2281.e1. pubmed/15599207. Accessed January 5, 2017.
100. Zmora O et al. Dis Colon Rectum 2003;46(5):584–9. 126. Fortinsky KJ et al. Dig Dis Sci 2017;62(1):188–96.
101. O’Riordan JM et al. Dis Colon Rectum. 2012;55(3): 127. Riss S et al. Dis Colon Rectum 2013;56(7):881–7.
351–8. 128. Simillis C et al. Am J Gastroenterol 2008;103(1):
102. Joo JS et al. Am Surg 1998;64(2):147–50. http://www. 196–205.
ncbi.nlm.nih.gov/pubmed/9486887. Accessed January 129. Sachar DB et al. Inflamm Bowel Dis 2009;15(7):
5, 2017. 1071–5.
103. Soltani A, Kaiser AM. Dis Colon Rectum 2010;53(4): 130. DeCruz P et al. Lancet 2015;385(9976):1406–17.
486–95. 131. Singh S et al. Gastroenterology 2015;148(1):64–76.
104. Rojanasakul A. Tech Coloproctol 2009;13(3):237–40. e2; quiz e14.
105. Bleier JIS et al. Dis Colon Rectum 2010;53(1):43–6. 132. Doherty G et al. Cochrane Database Syst Rev 2009;
106. Gingold DS et al. Ann Surg 2014;260(6):1057–61. (4):CD006873.
107. Narang R et al. Dis Colon Rectum 2016;59(7):670–6. 133. D’Haens GR et al. Gastroenterology 2008;135(4):
108. Mege D et al. Color Dis 2016;18(2):O61–5. 1123–9.
109. Kodner IJ et al. Surgery 1993;114(4):682–9. 134. Reinisch W et al. Gut 2010;59(6):752–9.
110. Hull TL, Fazio VW. Am J Surg 1997;173(2):95–8. 135. Mañosa M et al. Inflamm Bowel Dis 2013;19(9):
111. Valente MA. World J Gastrointest Pathophysiol 2014; 1889–95.
5(4):487. 136. Ananthakrishnan AN et al. Am J Gastroenterol 2011;
112. Zmora O et al. Dis Colon Rectum 2006;49(9):1316–21. 106(11):2009–17.
113. McNevin MS et al. Am J Surg 2007;193(5 SPEC. ISS.): 137. Zhao Y et al. Clin Res Hepatol Gastroenterol 2015;
597–9. 39(5):637–49.
114. Yamamoto T et al. World J Surg 2000;24(10):1258– 138. Regueiro M et al. Clin Gastroenterol Hepatol 2014;
62. http://www.ncbi.nlm.nih.gov/pubmed/11071472. 12(9):1494–1502.e1.
Accessed January 5, 2017. 139. Kotze PG et al. J Crohns Colitis 2015;9(7):541–7.
37
Evaluation and management of lower
gastrointestinal bleeding
ARJUN N. JEGANATHAN AND SKANDAN SHANMUGAN
hemorrhage. Classically, the aspiration of nonbloody, bil-

CHALLENGING CASE ious fluid on nasogastric lavage was considered to have
effectively excluded an UGIB; however, due to low sensitiv-
A 78-year-old male on Coumadin for atrial fibrillation ity and the invasive nature of the procedure, it is becoming
presents with massive hematochezia and associated less popular in favor of clinical scoring systems (3). For the
hypotension. He has never had a colonoscopy. purposes of this chapter, we assume UGIB has been effec-
tively excluded, most commonly with an upper endoscopy.
The patient with acute lower GI bleeding traditionally
CASE MANAGEMENT presents with hematochezia, defined as passage of bright
red blood or clots per rectum. A patient with an upper GI
Large-bore IV access is obtained, labs, including a bleed traditionally presents with melanotic (i.e., black and
cross-match, are sent, and he is resuscitated with rapid tarry) stools due to enzymatic breakdown of blood dur-
infusion of lactated Ringer solution. He is given fresh ing the period of intestinal transit. Hemorrhage from the
frozen plasma to normalize his international normal- small intestine distal to the ligament of Treitz is a difficult
ized ratio (INR) and transfused 2 units of packed red clinical entity to diagnose, and the investigational workup
blood cells for a hemoglobin of 6.2 g/dL. His vital signs varies quite a bit from colonic etiologies (4). In general, the
normalize, and the bleeding appears to have slowed. differential diagnosis is affected by age, but the more com-
A computed tomography (CT) angiography is done, mon causes include inflammatory bowel disease, neopla-
which shows a blush of extravasation in the ascending sia, Meckel diverticulum, angioectasia or other vascular
colon, where there appears to be diverticular disease. malformations, and nonsteroidal anti-inflammatory drug
He subsequently undergoes angiographic embolization ulcers. For this chapter, we have chosen to focus our discus-
of the bleeding vessel with cessation of hemorrhage. sion on colorectal etiologies of LGIB.
BACKGROUND ETIOLOGY
Lower gastrointestinal bleeding (LGIB) is classically defined Table 37.1 lists the causes of lower GI bleeding in a retrospec-
as luminal hemorrhage arising distal to the ligament of tive series of 1,112 patients from 1998 to 2006 at an urban
Treitz. The incidence of LGIB has slightly declined, but medical center (5). Each disease entity is briefly described
recent estimates show that LGIB still accounts for 35 hospi- in the following sections and is covered in-depth in prior
talizations/year/100,000 adult patients, with a case fatality chapters; herein, we focus on the management strategies of
rate between 1% and 3% (1). An important consideration to lower GI bleeding.
remember when evaluating a patient with gastrointestinal
(GI) bleeding is that upper gastrointestinal bleeding (UGIB) DIVERTICULAR BLEEDING
still remains the more common cause by at least twofold
(2). Thus, paramount to the treatment of LGIB is the elimi- Diverticular bleeding is the most common cause of hemor-
nation of the stomach and/or duodenum as the source of rhage from the lower GI tract in adults (5). Diverticulosis
347
348 Evaluation and management of lower gastrointestinal bleeding
Table 37.1 Etiologies of lower gastrointestinal bleeding be more occult in nature and is caused by ulceration of the
from 1998 to 2006 in an Urban U.S. Medical Center mucosa. Classically, right-sided lesions present with mela-
notic or maroon stools, while left-sided lesions tend to pro-
Disease Percentage (%)
duce bright red blood. While endoscopic therapies are often
Diverticulosis 37.3 limited for bleeding secondary to neoplasia, the diagnostic
Hemorrhoids 21.1 benefit is immense. Surgical treatment for colorectal neo-
Neoplasia 11.8 plasms will be dictated by clinical staging, but recurrent
Colitis (noninflammatory) 10.7 bleeding can often be an indication for palliative resection
Inflammatory bowel disease 5.4 even in locally advanced malignancies that would otherwise
Vascular malformation 2.3 require neoadjuvant therapy.
Other colonic disease 6.6
Small intestine disease 1.3 COLITIS (NONINFLAMMATORY)
Unknown 3.5
Noninflammatory causes of colitis are extremely diverse but
can include infectious, ischemic, and radiation-induced eti-
is widely prevalent, and its presence on screening colonos- ologies. On endoscopy, the mucosal lining appears erythem-
copy increases with age. Diverticular bleeding results from atous, friable, and ulcerated. Since the treatment is dictated
thinning of the colonic wall with eventual erosion of the by the underlying disease, importance is placed on arriving
penetrating vasa recta. While the presence of diverticulosis at a diagnosis. Infectious colitis often presents with diarrhea
is more common in the left side of the colon, diverticular and malaise. In the United States, the most common causes
bleeding occurs more commonly in the right side of the include Salmonella, Shigella, Campylobacter, and Escherichia
colon (6). In a cohort of 1,514 patients with colonoscopy- coli (10). Diagnosis is best established by stool culture and/or
confirmed asymptomatic diverticulosis, the cumulative toxin assay. Ischemic colitis is more common in the elderly
incidence of diverticular bleeding was 0.21% at 1 year, population and often associated with abdominal pain.
increasing to 9.5% at 10 years (6). Diverticular bleeding Fulminant presentation with bowel necrosis is a surgical
classically presents as painless hematochezia. The bleed- emergency and requires resection of the nonviable segments.
ing is generally self-limited, with 75% of cases stopping Alternatively, ischemic colitis from low flow states without
spontaneously without requiring transfusion (7). During infarction can be managed conservatively with fluid resus-
endoscopy, an active bleeding vessel, a nonbleeding but citation and bowel rest. Last, radiation-induced colitis can
visible vessel, or an adherent clot may be identified and occur early (within 6 weeks) from direct mucosal damage
necessitate endoscopic intervention (8). Current American or as a late manifestation associated with obliterative endar-
College of Gastroenterology (ACG) clinical guidelines favor teritis and chronic mucosal ischemia. Bleeding from radia-
endoscopic clips as they are generally safer than contact tion proctitis generally responds well to a trial of sucralfate
thermal therapy and easier to apply than ligation bands (9). enemas, with formalin instillation and/or endoscopic argon
However, after a second episode of diverticular hemorrhage, plasma coagulation reserved for refractory cases (11).
the risk of a future occurrence is close to 50%; thus, elective
segmental resection can be considered after localization (7). INFLAMMATORY BOWEL DISEASE
HEMORRHOIDS Inflammatory causes of LGIB include Crohn disease and

ulcerative colitis, with the latter more commonly causing
Painless rectal bleeding in a patient younger than 40 years hematochezia. Fortunately, less than 6% of patients have
can most commonly be attributed to a hemorrhoidal source, progression to massive LGIB requiring urgent surgical
while alternative diagnoses must be sought if clinical suspi- intervention (12). Differentiating Crohn disease from ulcer-
cion exists. Bright red blood will generally coat the stool at ative colitis is important when considering the extent of
the end of a bowel movement, drip into the toilet bowl, and/ surgical resection. Classically, subtotal colectomy with end
or stain the toilet paper. Anoscopy with visualization of the ileostomy is indicated for refractory bleeding from ulcer-
hemorrhoidal cushions may identify active bleeding or any ative colitis, while limited resection of affected segments is
stigmata of recent bleeding. Treatment for indolent bleed- preferred for Crohn disease (13).
ing from hemorrhoids is often conservative with dietary
and lifestyle changes, but office-based procedures such as VASCULAR MALFORMATION
rubber band ligation or surgical intervention may occasion-
ally be required. Vascular malformations of the GI tract encompass a myr-
iad of clinical entities and are quite poorly described in the
NEOPLASIA literature due to inconsistent nomenclature (14). Colonic
lesions include vascular tumors/angiomas, anomalies asso-
Colorectal neoplasms are a less common cause of LGIB, ciated with congenital/systemic disease (i.e., hereditary
accounting for 11% of all cases (5). The bleeding tends to hemorrhagic telangiectasia), and acquired/sporadic disease,
Management / Radiologic strategies 349
such as angiodysplasia and vascular ectasia. While rare, ENDOSCOPIC STRATEGIES

angiodysplasia remains the most common vascular mal-
formation. Pathologically, angiodysplasia refers to dilated As supported by the 2016 ACG clinical guidelines, colo-
submucosal vessels, likely caused by degenerative changes noscopy should be the initial diagnostic procedure for
of the colonic walls with aging (15). Bleeding most often nearly all patients presenting with acute LGIB (9). Careful
originates from the cecum or ascending colon but is often inspection of the colonic mucosa and intubation of the ter-
episodic and self-limiting (16). Clinical guidelines favor the minal ileum to rule out more proximal bleeding is gener-
use of noncontact thermal therapy, namely, argon plasma ally recommended (19). Ideally, endoscopic examination
coagulation, due to improved outcomes with decreased should be conducted within 24 hours of presentation after
transfusion requirements (9). a mechanical bowel preparation. A randomized trial of
100 consecutive patients presenting with LGIB found that
urgent colonoscopy (<8 hours), when compared to stan-
dard colonoscopy (1–3 days), was more successful in iden-
INITIAL EVALUATION tifying a source of bleeding, though a clear improvement
in outcomes (namely, rebleeding, transfusion requirements,
The chief complaint of lower GI bleeding is a very common length of stay, and mortality) was not supported (20). Of
clinical entity and mandates a systematic, stepwise approach. note, a recent feasibility study of 12 patients demonstrated
In general, LGIB can be categorized into acute (urgent or that urgent colonoscopy without the use of bowel prepara-
emergent) or chronic. Relevant past medical history should tion was found to be adequate for presumptive identifica-
be elucidated so as to identify potential sources of bleeding, as tion of the source of bleeding in all procedures, without the
well as any prior episodes of LGIB. Concurrent comorbidities need to repeat colonoscopy because of inadequate prepa-
could dramatically alter management strategies and must be ration (21). However, the current ACG clinical guidelines
included in relevant past medical history. Additionally, a care- recommend bowel preparation, consisting of a large volume
ful review of medications must be conducted with emphasis (4–6 L) of polyethylene glycol–based solutions administered
on contributory pharmacologic drugs, such as anticoagulant, rapidly (over 3–4 hours) until effluent is clear (9).
antiplatelet, and nonsteroidal anti-inflammatory agents.
General supportive measures must first be undertaken, RADIOLOGIC STRATEGIES
including establishment of adequate IV access, supplemental
oxygen via nasal cannula, and initiation of resuscitation. The One advantage of radiographic testing for acute GI bleed is
first assessment must include a determination of the severity the ability to assess the entirety of the GI tract, including
of bleeding, including an estimation of blood loss. In a com- small bowel sources. The two most commonly performed
parison of a restrictive (Hgb <7 g/dL) versus liberal (Hgb tests for luminal hemorrhage include nuclear scintigraphy
<9 g/dL) transfusion strategy, a restrictive strategy signifi- (radionuclide scanning) and CT angiography. According
cantly improved outcomes in patients with acute UGIB and to our treatment algorithm, radionuclide scanning is best
LGIB (17). The probability of survival at 6 weeks was higher applied for ongoing bleeding despite a negative colonoscopy,
in the restrictive-strategy group than in the liberal-strategy while CT angiography has applicability for the patient with
group (95% versus 91%). As for management of coagulopa- severe bleeding who cannot be stabilized for colonoscopy
thies, the risk of holding or reversing anticoagulation should or who cannot undergo a rapid bowel prep for colonoscopy.
be weighed against the indicated benefit. The decision must The most sensitive radiographic test for GI bleeding is
be individualized, but, in general, the transfusion of platelets a radionuclide scan, capable of detecting bleeding at a rate
and/or plasma is indicated for patients with large transfusion of 0.04–0.1 mL/min (22). Scintigraphy for GI bleeding was
requirements, as well as coagulopathies (INR >1.5 or plate- first described in 1977 using Tc-99m sulfur colloid but
lets <50,000/microL) with ongoing bleeding. When major mostly has been replaced by Tc-99m red blood cell (RBC)
bleeding was studied in the trauma population, early admin- imaging to allow for longer periods of sequential imaging
istration of plasma, platelets, and red blood cells in a 1:1:1 if needed. While very sensitive in its ability to detect active
ratio compared with a 1:1:2 ratio resulted in increased hemo- bleeding, the major limitation of radionuclide scanning is
stasis and fewer deaths due to exsanguination by 24 hours, its ability to only localize bleeding to a general region of the
but did not result in significant differences in mortality at 24 abdomen and not to a specific anatomic structure. In fact, in
hours or at 30 days (18). a study of 203 patients, tagged RBC scan was positive in 26%
of patients, but incorrectly localized the site of bleeding in 1
out of 4, resulting in 8 misguided surgical procedures (23).
CT angiography for the localization of GI bleeding has
MANAGEMENT become increasingly popular due to widespread availabil-
ity. However, when compared to scintigraphy, CT angi-
Upon completion of the initial evaluation, the management ography lacks the ability to sequentially image over time
of acute LGIB can generally be approached in a stepwise without repeated radiation and contrast exposures. In
fashion, as detailed by Figure 37.1. swine models, researchers had success detecting bleeding
350 Evaluation and management of lower gastrointestinal bleeding
Acute GI bleed
Rule out UGIB
Resuscitation with IV fluid and/or blood products

determine urgency/acuity
Emergent/hemodynamic instability (<2 hrs)* Urgent/hemodynamic stability (>6 hrs)*
Consult surgery Consult GI

CT angiogram for localization colonoscopy for localization
Localized Non-localized Non-localized Localized
Embolization Ongoing bleeding Ongoing bleeding Endoscopic

hemostasis
Surgical exploration Consider push enteroscopy, capsule

endoscopy, and/or scintigraphy
Figure 37.1 Diagnosis and treatment algorithm for acute lower GI bleeding. *Determination of urgency/acuity is a difficult
clinical decision but can generally be stratified into emergent cases who need intervention within 2 hours and cases who
will require urgent colonoscopy but may wait at least 6 hours before intervention.
at a rate of 0.3–0.5 mL/min (24). CT angiography has formation, and renal failure, with bowel infarction occur-
excellent results when studied clinically, with a recent ring in up to one-third of cases (29).
meta-analysis arriving at a sensitivity and specificity of
85% and 92%, respectively, for the detection of GI hemor- SURGICAL STRATEGIES
rhage (25). A positive CT angiography is typically followed
by a therapeutic angiographic procedure in patients who Due to the significant associated morbidity and mortality,
are hemodynamically stable. A study in 2015 concluded urgent operative exploration for LGIB is often reserved for
that preceding visceral angiography with a diagnostic massive hemorrhage with hemodynamic instability despite
study (i.e., CT angiography) improved localization of the resuscitation. Additionally, patients who have failed repeated
site of lower GI hemorrhage when compared with visceral endoscopic or interventional approaches or cannot obtain
angiography alone (26). these less invasive modalities are also surgical candidates.
Diagnostic visceral angiography is typically reserved For ongoing hemorrhage, surgical procedures will occasion-
for patients not favorable for endoscopy, namely, individually be indicated as first-line treatment based on the pathol-
als with severe bleeding with hemodynamic instability and ogy alone, including bleeding from early stage colorectal
high operative risk. Detection as a contrast blush or extrav- neoplasms or benign anorectal disease. The decision to oper-
asation generally requires blood loss of 0.5–1 mL/min (27). ate on massive LGIB is a challenging one based on numerous
In the absence of prior localization, the superior mesenteric clinical entities, but the literature would support that only
artery is generally examined first, followed by the inferior a small percentage (4.7%) require urgent operative proce-
mesenteric artery, due to the predilection of bleeding to dures (30). No single predictor can accurately dictate the
occur in the embryologic midgut distribution. The distinct need for surgical intervention, but ongoing bleeding with
advantage of visceral angiography is that, like colonos- transfusion requirements upward of 4 units of RBCs within
copy, therapeutic interventions may occur. In patients with 24 hours has become a general benchmark (31). While our
active bleeding, superselective embolization of distal ves- outcomes have likely improved, early data suggested that the
sels is successful in 80% with no episodes of rebleeding in overall mortality, while not statistically different, was not
97% of cases (28). However, the major complication profile insignificant for urgent total abdominal colectomy versus
of visceral angiography includes arterial injury, thrombus limited colonic resection (6% versus 15%, respectively) (32).
References 351
In a retrospective study of 77 patients, recurrent bleeding 5. Gayer C et al. Surgery, 2009;146:600–7.

was significantly more common in the segmental colectomy 6. Niikura R et al. Aliment Pharmacol Ther. 2015;41(9):
group than the total abdominal colectomy group (18% ver- 888–94.
sus 4%), and morbidity and mortality were not significantly 7. McGuire HH. Ann Surg. 1994;220(5):653.
different (33). Consequently, total abdominal colectomy with 8. Jensen DM et al. N Engl J Med. 2000;342(2):78–82.
end ileostomy should remain the procedure of choice for 9. Strate LL, Gralnek IM. Am J Gastroenterol. 2016;111:
surgical treatment of massive LGIB without a clearly local- 459–74.
ized source. However, when the bleeding site has been local- 10. DuPont HL. Curr Opin Gastroenterol. 2012;28(1):
ized and confirmed without control, a segmental colectomy 39–46.
with primary anastomosis can be undertaken with rebleed- 11. van de Wetering FT et al. Cochrane Database Syst
ing and mortality rates of 4%–10% and 0%–40%, respec- Rev. 2016;4:CD003455.
tively. A “blind” segmental colectomy without localization 12. Berg DF et al. Am J Surg. 2002;184(1):45–51.
should not be performed. Last, in a retrospective review of 13. Shanmugan S, Stein SL. Lower gastrointestinal
subtotal colectomies performed for LGIB, 55 of 58 patients bleeding. In Cameron JL, Cameron AM (eds).
were treated with a primary ileorectal anastomosis with an Current Surgical Therapy, 302–307, 11th Edition.
overall mortality of 17%, mainly resulting from sepsis due to 2014. Elsevier, Philadelphia, PA.
anastomotic leak (34). The current data would strongly argue 14. Regula J et al. Best Pract Res Clin Gastroenterol.
against reestablishment of GI continuity during the index 2008;22(2):313–28.
emergent operation, while an elective ileorectal anastomosis 15. Boley SJ et al. Gastroenterology. 1977;72(4):650.
at a later date has far fewer consequences. 16. Diggs NG et al. Clin Gastroenterol Hepatol. 2011;
9(5):415–20.
17. Villanueva C et al. N Engl J Med. 2013;368(1):11–21.
18. Holcomb JB et al. JAMA. 2015;313(5):471–82.
SUMMARY 19. Jensen DM. Am J Gastroenterol. 2005;100:2403–6.
20. Green BT et al. Am J Gastroenterol. 2005;100(11):
Key points in the diagnosis and management of LGIB
2395–402.
include stabilization with resuscitation, localiza-
21. Repaka A et al. Gastrointest Endosc. 2012;76(2):
tion of the bleeding source, and cessation of ongoing
367–73.
hemorrhage via endoscopy, angiography, or surgery.
22. Currie GM et al. J Clin Gastroenterol. 2011;45(2):
Given the recent advances in nonsurgical strategies,
92–9.
fewer patients are undergoing subtotal colectomies
23. Hunter JM, Pezim ME. Am J Surg. 1990;159(5):504.
for nonlocalized bleeding. Segmental colectomy
24. Kuhle WG, Sheiman RG. Radiology. 2003;228(3):
should be reserved for patients with a localized bleed-
743–52.
ing site; subtotal colectomy with an ileorectal anas-
25. García-Blázquez V et al. Eur Radiol. 2013;23(5):1181.
tomosis should be reserved for elective patients and
26. Jacovides CL et al. JAMA Surg. 2015;150(7):650–6.
avoided in the emergent settings.
27. Walker TG. Tech Vasc Interv Radiol. 2009;12(2):
80–91.
28. Strate LL, Naumann CR. Clin Gastroenterol Hepatol.
REFERENCES 2010;8(4):333–43.
29. Khanna A et al. J Gastrointest Surg. 2005;9(3):
1. Laine L et al. Am J Gastroenterol. 2012;107(8):1190–5. 343–52.
2. Rotando G. Gastroenterol Clin North Am. 2014;43(4): 30. Newman J et al. Colorectal Dis. 2012;14(8):1020–6.
643–63. 31. Bender JS et al. Am Surg. 1991;57:536–40.
3. Srygley FD et al. JAMA. 2012;307(10):1072–9. 32. Baker R, Senagore A. Am Surg. 1994;60(8):578–81.
4. Gerson LB et al. Am J Gastroenterol. 2015;110(9): 33. Farner R et al. Am J Surg. 1999;178(6):587–91.
1265–87. 34. Plummer JM et al. Int J Clin Pract. 2009;63(6):865–8.
38
Ostomies
DANIELLE PICKHAM AND SUPRIYA S. PATEL
A 55-year-old morbidly obese male undergoes a low Stoma formation is a common procedure for colon and
anterior resection with concomitant loop ileostomy rectal surgeons. It is estimated that approximately 450,000
for a T2 rectal cancer. Six weeks postoperatively, he people in the United States are currently living with an
presents to the clinic with an easily reducible parasto- intestinal stoma, and 120,000 new stomas are created each
mal hernia. He complains of increasing pain, difficulty year (1). There are various types of intestinal stomas created
with application of his ostomy appliances, and symp- for a broad spectrum of diseases and clinical situations.
toms of intermittent obstruction. When successful, stomas are designed to save lives or ulti-
mately enhance a patient’s quality of life. However, they can
be fraught with a number of complications, resulting in sig-
CASE MANAGEMENT nificant economic, physiologic, and psychological impact.
The focus of this chapter is to review complications associ-
The optimal management for a symptomatic para- ated with stoma formation and discuss how to prevent or
stomal hernia of temporary stoma includes reversal manage these adverse outcomes.
of the ostomy after ensuring that the distal anasto-
mosis has healed. Anastomotic integrity is confirmed
by a contrast study, often a Gastrografin enema or
computed tomography scan with rectal contrast. An HISTORICAL PERSPECTIVE
ostomy reversal ameliorates and addresses all of the
symptoms including the hernia, obstruction, and pain. To botanists a “stoma” refers to a pore in the epidermis of a
After reversal, the skin of the ostomy can be partially plant that participates in gas exchange. Similarly, in medi-
closed; however, extreme vigilance of the wound is cine, a stoma refers to a surgically created opening of the
necessary secondary to an increased rate of local intestine through a patient’s anterior abdominal wall. The
wound infection. Depending on the size of the fascial first recorded stoma surgeries were in the 1700s (2). Based
defect and corresponding hernia, additional mesh on the observation that patients were able to survive intes-
may be needed for hernia repair. Due to increased tinal rupture due to formation of cutaneous fistulas, surgi-
risk of infection of most prosthetics, biologic materi- cal techniques were devised to intentionally create those
als should be considered as a first option. For patients fistulas in similar, emergent settings (2). In the 1800s, the
who are not candidates for ostomy reversal, various first elective stoma surgery was described by Freer, the loop
options are available and include both open and lapa- colostomy was introduced, and the concept of a temporary
roscopic approaches. These options include primary stoma was devised (2). Since then, surgical techniques have
fascial repair, repair with biologic or prosthetic mesh, evolved to include the Hartmann end colostomy (1923) and
and finally, stoma relocation. The approach and the the Brooke ileostomy (1952) (2). Currently, there are a wide
method of repair are dependent on the surgeon’s variety of stomas, and multiple techniques have been devel-
preference and experience. Certainly, observation oped to assist in the prevention and management of stoma-
for minimally symptomatic parastomal hernias is the related complications.
preferred option until stomal takedown is possible.
352
Considerations / Primary anastomosis with diversion 353
that may exist in patients with cancer or chronic obstruc-

CONSIDERATIONS tions. However, endoscopic stenting, subtotal colectomy, or
on-table colonic lavage are alternative options to achieve pri-
SURGICAL DECISION-MAKING mary anastomosis and have been demonstrated to be safe, but
discussion of them is beyond the scope of this chapter (3–6).
Stomas are utilized over a broad range of clinical sce- Perforated diverticulitis is another common clinical sce-
narios, including lower intestinal cancer, diverticulitis, nario in which a primary anastomosis has historically been
inflammatory bowel disease, bowel perforation or obstruc- avoided. However, in recent years this has been challenged,
tion, trauma, and fistulizing disease. There are two general and primary anastomosis, with or without a temporary
indications for stoma creation: (1) to provide a solution diverting ileostomy, is being utilized with no significant
for elimination when intestinal continuity is not feasible increase in morbidity or mortality, evidence of higher stoma
and (2) to provide fecal diversion in the setting of distal reversal rates (7–10), and decreased cost and hospital stay
obstruction or to allow additional time for healing. Multiple (11). In-depth discussion of management in these emergent
different stoma options exist with respect to type (ileos- settings is beyond the scope of our chapter, but we encour-
tomy/colostomy), duration (temporary/permanent), and age the reader to always consider the option of primary
configuration (end/looped). Individualizing the stoma to anastomosis when reasonable to do so.
each patient and circumstance is critical to preventing or
minimizing future complications. PRIMARY ANASTOMOSIS WITH
The following case study illustrates the multitude of sur- DIVERSION
gical options that may be present in any one clinical scenario.
When an anastomosis is feasible, but the risk of an anasto-
motic leak is increased, a diverting stoma should be con-
CASE STUDY sidered. Loop ileostomies are favored over loop colostomies
due to an overall lower complication rate and ease of closure
A 72-year-old female presents with an obstructing (12–14). Harris et al. demonstrated loop colostomies having
rectosigmoid mass. Colonoscopy is performed and the highest complication rate at 38% and end ileostomies the
confirms the diagnosis of adenocarcinoma. She has lowest (16%); however, skin-related issues (more common in
no prior surgical history but is obese with hyperten- ileostomies) were not taken into account in this study (13).
sion, diabetes, and coronary artery disease. Multiple One of the most common settings for a diverting loop
surgical options can be considered: ileostomy is rectal cancer. Surgical resection results in the
creation of a colorectal anastomosis, which is often diverted
a. Resection with primary anastomosis
with an ileostomy, the benefit of which has recently been
b. Resection with primary anastomosis and diverting
challenged. Multiple studies (15–17), including a recent
stoma
meta-analysis by Gu including 13 publications and over
c. Resection with end colostomy or ileostomy
8,000 patients, demonstrated that diversion significantly
d. Diverting loop colostomy or ileostomy
reduced post-op anastomotic leakage and reoperation after
After taking into account patient-related fac- low anterior resections (17). This was supported by a few
tors such as age, comorbidities, patient’s functional other studies (14,18,19), but others have shown similar anas-
status including mobility and continence, stage of tomotic leak rates and need for reoperations with and with-
the cancer, and overall prognosis, the surgeon should out stomas (20). It is also acknowledged that the addition of
also consider what surgical options, including type of an ileostomy is not without risk of other adverse outcomes.
stoma, will offer the patient the lowest risk of compli- Diverting ileostomies have been shown to be associated
cations and the best quality of life. with increased hospital stays (15), increased risks of acute
renal failure (18), and stoma-related complications (20),
which will be outlined in more detail within this chapter.
PRIMARY ANASTOMOSIS While the benefit of a diverting ileostomy in this set-
ting has been challenged and is clearly controversial, it may
Primary anastomosis is always favored whenever feasible, but help to consider factors that increase the risk of anastomotic
the risks of anastomotic leak and alterations in bowel func- leak, thereby favoring the formation of a diverting stoma.
tion must be weighed against the potential complications and These include technically challenging case, male gender,
lifestyle modifications required with a stoma. In our clini- colorectal anastomosis <10 cm or coloanal anastomosis (14),
cal scenario, primary anastomosis may be avoided due to a musculoskeletal disorders, and high American Society of
variety of factors, such as concern for increased risk of infec- Anesthesiologists score (21). In our practice, we routinely use
tion due to unprepped bowel, concern for increased risk of a diverting stoma in those patients who have received neoad-
anastomotic leak due to size discrepancy and poor viability juvant chemoradiation and those with an anastomosis below
of a chronically dilated proximal segment, and decreased 6 cm. We would encourage surgeons to weigh the established
healing potential due to nutritional or electrolyte deficiencies risk of an anastomotic leak with the potential increased
354 Ostomies
mortality and hospital stay from a stoma to help individualize Table 38.2 Risk factors for stoma-related complications
patient care. Furthermore, while a diverting loop ileostomy
Patient-related factors Surgical-related factors
may not change the anastomotic leak rate, it may decrease
the severity of an anastomotic leak, especially in low pelvic Obesity Emergent surgery
anastomosis having an impact on long-term function. Diabetes Stoma type
Nutritional status Stoma height
END STOMA Emergent surgery Surgeon specialty
Age >65
When a primary anastomosis is determined to be technically Gender
impossible, too high risk, or the patient’s function will be Steroid use
unacceptable, an end stoma is created. The colon or the small
intestine can be utilized for this purpose, and either would be
an option in our case study. A Hartmann procedure, resulting complication rates, with one study showing colorectal spe-
in an end colostomy with a rectal stump, may save operative cialists to have nearly half the number of stoma complica-
time as well as preserve intestinal length and colon function tions when compared with general surgeons (27) (Table
for this patient. However, in the setting of chronic obstruction, 38.2). In the upcoming section, we review the most common
some may find the proximal colon too dilated or unhealthy to complications associated with intestinal stomas, including
bring up as a colostomy, and therefore, an abdominal colec- their management and prevention.
tomy with an end ileostomy may be preferred.
In regard to stoma complications, similar rates exist DEHYDRATION
between end ileostomies and colostomies (22,23). However,
there are clear advantages and disadvantages to each that may More commonly seen after ileostomies than colostomies,
influence surgical choice. Common complications seen with dehydration remains a leading cause of postoperative read-
ileostomies include high output resulting in dehydration, skin missions following stoma creation. The magnitude of this
irritation, and requirement for nighttime emptying (22,23). complication cannot be overstated, as demonstrated in a
Patients with colostomies have problems with odor and para- study by Fish et al. (2017) in which postoperative readmis-
stomal hernias, especially in the setting of obesity (22,23). It is sions were monitored in 407 patients following ileostomy
important to note that approximately 50% of the time a tem- creation. A 60-day readmission rate of 28% was noted, with
porary stoma will end up becoming permanent (24). the most frequent reason for readmission being dehydration
(42%) (29). Additionally, patients readmitted with dehydra-
tion were noted to have increased morbidities, longer hospi-
tal lengths of stays, and repeat readmissions (29).
STOMA COMPLICATIONS Electrolyte imbalances after ileostomy creation, includ-
ing hyponatremia, hypokalemia, hypomagnesemia, and
The incidence of stoma complications varies widely in the hypocalcemia, are most commonly observed between
literature, depending on the definition of the complication postoperative days 3–8 (24,30). These imbalances can lead
within the study and the length of follow-up, but many esti- to additional complications such as renal failure and car-
mate an incidence of 50%–70% (15,21,25–27). These com- diac arrhythmias. Careful monitoring in the postoperative
plications can be divided into two categories: early and late period is critical to early recognition and management.
(Table 38.1). Early complications refer to those within the Patients may present with stoma output exceeding 1–2 L/
first 30 days of surgery, and late complications are defined day, or more generalized signs, such as increased weakness,
as those thereafter, often observed between 6 and 10 weeks malaise, nausea, and vomiting. Higher risks of dehydration
postoperatively (1). Risk factors associated with stoma are noted in the setting of proximal stomas, short gut, small
complications include emergent operation, age >65 years, bowel obstruction, sepsis, and diuretic use (31,32).
obesity, and diabetes (25). Stoma height and type can affect Implementation of “ileostomy pathways” that empha-
complications rates (28). The specialty of the surgeon cre- size patient education prior to stoma creation and active
ating the stoma has also been demonstrated to influence management and attention to intakes and outputs in the
postoperative period have demonstrated great appeal; Nagle
Table 38.1 Stoma complications et al. (2012) reported that readmission rates secondary to
dehydration dropped from 15.5% to 0% in their study popu-
Early complications Late complications lation following the implementation of such a pathway at
Dehydration Stenosis Best Israel Deaconess Medical Center. Central to this ileos-
Skin complications Parastomal hernia
tomy pathway is active participation by the patient in his or
her stoma care and output monitoring during the patient’s
Mucocutaneous separation Prolapse
hospital stay (33).
Retraction Stomal varices
Management of dehydration involves fluid resuscitation
Ischemia/necrosis
and correction of electrolyte abnormalities. Patients should
Stoma complications / Ischemia/necrosis 355
be counseled to consume electrolyte-enriched solutions,

such as sports drinks, as opposed to hypotonic fluids such
as water, fruit juices, and tea, which can augment hypona-
tremia (24). Dietary counseling regarding fiber supplemen-
tation and avoidance of foods that are likely to increase
stoma output should be provided. Addition of pharmaco-
logic therapies, such as loperamide, diphenoxylate, and
opioid derivatives, may be required if outputs remain per-
sistently high.
SKIN COMPLICATIONS
Cutaneous complications following stoma creation may
result from mechanical, chemical, allergic, or infectious
causes, and are more frequently observed in the setting of Figure 38.1 Pyodermia gangrenosum.
ileostomies, as a consequence of the high volume, liquid-
like, alkaline output associated with this type of stoma (24).
Obesity and diabetes have been identified as independent MUCOCUTANEOUS SEPARATION
risk factors for cutaneous complications (34). Poor siting,
incorrect appliance application, and other stoma-related Mucocutaneous separation refers to separation of the stoma
complications, such as prolapse and hernia, may further from the surrounding skin, and may occur as a result of
exacerbate local skin irritation, highlighting the impor- improper stoma maturation, excessive traction, or poor
tance of preoperative marking and postoperative educa- healing (24). Once identified, the extent of the detachment
tion. Appliances should generally be changed no more should be properly evaluated by an enterostomal nurse so
frequently than every 3 days, to prevent excessive skin that treatment can be initiated. Predisposing risk factors,
trauma. Following appliance removal, the skin should be including the presence of an immunocompromised state,
gently cleaned with water and dried. Wafers should be cut diabetes, and smoking, should be optimized whenever pos-
to the appropriate size, and barrier creams, powders, and sible. The space between the skin edge and stoma should be
pastes should be applied to protect the skin from stoma cleaned and filled with a paste or powder to promote heal-
effluent. ing. Long-term separation can result in stomal retraction
Allergic dermatitis can be treated with topical steroid and stenosis.
creams and antihistamines, while antifungal creams or
antibiotics may be required in the setting of infectious eti- STOMA RETRACTION
ologies. Candida albicans is the most common infectious
cause of peristomal skin irritation, as the warm, moist peri- Stoma retraction is most often seen in the acute postopera-
stomal skin provides the perfect fungal environment; treat- tive period and may ensue from inadequate mobilization
ment with a light dusting of miconazole nitrate 2% powder of the bowel and/or mesentery, leading to undue tension,
is sufficient in most cases (34). or from impaired wound healing. Retraction is most often
seen with ileostomies, with rates of up to 17% observed (34).
PYODERMA GANGRENOSUM Treatment frequently involves repeat laparotomy with fur-
ther mobilization of the bowel. Retracted ileostomies often
Peristomal pyoderma gangrenosum is a rare and debilitat- require additional mobilization, and in selected cases suture
ing cutaneous complication associated with inflammatory fixation of proximal bowel to the fascia. Retracted colosto-
bowel disease. The initial presentation is one of small pus- mies often require high ligation of mesenteric vessels, relax-
tules, which progress into large, painful, ulcers with necrotic ing incisions, flexure takedown, and mobilization along the
edges (Figure 38.1). The diagnosis is frequently made only peritoneal reflections. A final option is conversion to an
after other more common causes have been ruled out. end-loop stoma (24,34).
Treatment frequently begins with topical or intralesional
steroids, but response rates are variable, and other systemic ISCHEMIA/NECROSIS
treatments, including IV steroids, infliximab, dapsone, and
immunomodulators, have been tried also with variable suc- Necrosis of the stoma is an uncommon complication, with
cess. Conflicting reports exist with regard to the association incidence rates ranging between 1% and 34%, and is usually
between pyoderma gangrenosa and intestinal disease activ- noted in the immediate postoperative period (30). Ischemic
ity, but shorter healing times have been observed follow- changes should be assessed in terms of both extent and
ing the resection of actively diseased bowel (24,34). Stoma depth; while small, superficial patches of mucosal ischemia
relocation is generally not advised, as the disease frequently can frequently be treated conservatively, larger areas or full-
recurs at the new stoma site. thickness involvement may require stomal revision and/or
356 Ostomies
repeat laparotomy. It is important to differentiate whether

(c)
or not the ischemia extends below the level of the fascia.
In order to facilitate this assessment, a lubricated test tube (a)
or endoscope can be inserted into the stoma and transil- a
luminated so that the color of the mucosa can be evaluated.
1
Necrotic changes noted below the level of the fascia man-
date reexploration and resection of any dead bowel. In some b
cases, evaluation with a flexible scope can further assist in 2
evaluation. For ischemia/necrosis above the fascia, immedi- a
(b)
ate surgery is not necessary, and in some cases, later stoma 3 c
4 b 1
revision will be needed due to stenosis.
Predisposing factors to stomal ischemia and necro- d 2
sis include arterial insufficiency, venous congestion, and
mechanical factors, such as inadequate trephine size. Care 3
should be taken to assess for ischemia, after division of the 4
c
bowel and prior to the creation of the stoma. Excessive trim-
ming of the mesentery or de-fatting of the bowel wall should d
be avoided, so as to avoid injury to the collateral vessels that
are providing the blood supply to the stoma. When creating Figure 38.2 Z-plasty repair for stenosis: (a) the stoma is
separated from the skin and underlying subcutaneous
colostomies, at least a 1 cm segment of mesentery should be
tissue, then incisions are made in the skin and bowel as
preserved at the edge of the bowel so that the marginal artery depicted, (b) the bowel is realigned to fit within the new
is not compromised (24). An adequate trephine should be skin incisions, (c) stoma is matured to the skin.
created to avoid constriction of the mesenteric vessels as the
bowel is brought out through the abdominal wall. In the
setting of the obese patient, consideration should be given presence of end stomas. Although transrectus stoma cre-
to bringing out the stoma in the upper abdomen or creation is thought to reduce the incidence of parastomal her-
ation of an end-loop stoma, to avoid ischemic complications nias, these hernias can occur even when stomas are brought
secondary to undue tension (24). out through the rectus musculature. It has been suggested
that an increase in the stoma aperture size of even 1 mm
STENOSIS can lead to a 10% increase in the risk of parastomal hernia-
tion (36). While uncommon in the immediate postopera-
Stomal stenosis refers to a narrowing of the stomal lumen tive period, the incidence of these hernias correlates with
at the skin or fascial level. Symptomatic patients may pres- the duration of the stoma, and ranges from 14% to 40% (24).
ent with symptoms of partial obstruction or decreased stool Evaluation of patients suspected to have parastomal her-
caliber. Although this complication is typically seen as a nias should include examination in the supine and upright
consequence of prior ischemia or stomal retraction, the dif- positions, as well as during a Valsava maneuver. In some
ferential diagnosis should include Crohn disease and malig- cases, additional imaging studies, such as computed tomog-
nancy. Initial management consists of dietary modification raphy scans, may be required to further characterize the
(i.e., low-residue diet), fluid hydration, and gentle catheter extent of the defect. Treatment options vary with degree
irrigation, in the setting of colostomies. The utility of serial of symptomology. Minimally symptomatic hernias can be
dilation has been questioned, as repeated trauma can stimu- treated using stoma belts and modified pouching systems,
late fibrosis, thus worsening the underlying stenosis (1). If while more symptomatic hernias may require surgical
conservative measures fail, surgical options, including sto- intervention, ranging from primary and mesh repairs to
mal revision and Z- or W-plasty techniques (Figure 38.2), stoma re-siting. Consideration of stoma reversal should be
whereby the skin opening is enlarged using plastic surgery entertained before embarking upon repair, because rates of
methods, can be considered (1,35). A A-plasty technique is recurrence remain high, ranging from 7% to 100% depend-
particularly helpful when bowel length is limited, such as in ing on the technique of repair used (24,30,34).
continent ileostomies. Primary parastomal hernia repairs have largely been
abandoned secondary to high recurrence rates. This type
PARASTOMAL HERNIA of repair is best suited for patients in whom a repeat lapa-
rotomy may be contraindicated or deemed high risk. The
Parastomal hernias are incisional hernias occurring near technique involves dissection through the subcutaneous
the site of a stoma that can result in symptoms ranging tissues in the parastomal space, excision/reduction of the
from an unsightly bulge and difficulty with pouching to hernia sac, and primary repair of the fascial defect.
bowel ischemia, obstruction, and perforation. Risk factors Mesh repairs involve the placement of a synthetic or bio-
for parastomal hernias include obesity, pulmonary dis- logic mesh in the overlay, underlay, or sublay positions, and
ease, malnutrition, steroid use, emergency surgery, and the can be performed using a laparoscopic or open approach
Stoma complications / Parastomal hernia 357
(a)
(b) Subcutis Bowel passage

Anterior rectus Rectus muscle
sheath
Posterior rectus
sheath Prosthesis
(c)
Figure 38.4 Keyhole technique. (From Hansson BM et al.
Ann Surg 2012;255(4):685–95.)
two meshes (38). Using this method, only one recurrence was
noted in 47 patients, resulting in a much lower overall recur-
rence rate of 2.1% (37,38).
Stoma re-siting is another option for repair. This tech-
nique may require a repeat laparotomy. A new site should
be selected prior to the procedure and properly marked.
Figure 38.3 Mesh positions: (a) overlay, (b) under-
A site on the contralateral site of the abdomen is generally
lay, and (c) sublay. (From Hansson BM et al. Ann Surg
2012;255(4):685–95.)
advised, but a site on the same side (upper abdomen) is also
an option. Unfortunately, this procedure is associated with
a risk of incisional hernia formation at the site of fascial clo-
(Figure 38.3) (37). Overlay approaches refer to the place-
sure, as well as a risk of recurrent parastomal herniation,
ment of mesh above the level of the fascial closure, while
at the new stoma site. To avoid this problem, a technique
sublay or retrorectus repairs involve the placement of mesh
called translocation is considered.
between the rectus abdominus muscle and posterior sheath.
With translocation, the ostomy is mobilized locally. The
Recurrence rates have been observed to be higher with the
mucocutaneous junction is divided, and the stoma is mobi-
onlay approach in comparison to the retromuscular tech-
lized down and through the fascia in a manner similar to
nique (18.6% versus 6.9%) (37).
The two most common types of intraperitoneal repairs
are the keyhole and Sugerbaker techniques (Figures 38.4 and
38.5) (37). Both intraperitoneal repairs start with identifica-
tion and reduction of the hernia sac. In the keyhole technique,
the stoma is brought out through a 2–3 cm “keyhole” within a
mesh placed intraperitoneally, in such a manner that 4–5 cm
of fascial overlap is achieved (34,38). The “Sugarbaker” tech-
nique, as originally conceived by Paul Sugarbaker in 1958,
involves placement of mesh around the fascial defect and
positioning the bowel such that it travels laterally between the
mesh and abdominal wall for 5 cm, before it enters the perito-
neal cavity (38). In a meta-analysis of 270 laparoscopic repairs
involving the keyhole and Sugarbaker techniques, recurrence
rates of 20.8% were observed with the keyhole method, com-
pared to 11.6% with the Sugarbaker technique (37). A sand-
wich technique involving a combination of the keyhole and
Sugarbaker methods has been described by Berger et al., in
which one piece of mesh is fashioned as a keyhole around the
stoma, while a second, larger piece of mesh is placed around Figure 38.5 Sugarbaker technique. (From Hansson BM
the fascial defect, and the bowel is lateralized between the et al. Ann Surg 2012;255(4):685–95.)
358 Ostomies
closing a stoma. The presence of a parastomal hernia often

(a)
eases this dissection. With retractors, the abdominal wall
is elevated and through the ostomy opening and adhesions
between the abdominal wall between the old and new stoma
site are divided. With the surgeon’s hand inside the abdo-
men, the new ostomy site is created. A clamp is inserted
into the new site, and the end of the open bowel is grasped.
The bowel is then maneuvered to the new site. Some mobi-
lization of the bowel may be required to obtain adequate
length. The bowel is then matured at the new site, and the
old ostomy and hernia site is closed. This repair is often
reinforced with mesh (Figure 38.6).
Given the prevalence and morbidities associated with
parastomal hernias, increasing attention has been placed (b)
on the use of prophylactic mesh at the time of the original
stoma creation. In a meta-analysis involving 649 patients,
Cross et al. observed that prophylactic mesh placement
reduced the rate of parastomal hernia repair by 65%, without
an increase in mesh-related complications, including infec-
tion, stomal necrosis, and stenosis (39). Placement of sublay
mesh in the setting of open surgery and use of a Sugarbaker
technique with synthetic mesh in the laparoscopic setting
have been shown in multiple studies, including randomized
control trials, to prevent hernias (40–44). While this has
not been adopted as standard practice for many colorectal
surgeons, it should be considered, especially in patients at
increased risk for hernia formation.
PROLAPSE
Stomal prolapse is defined as a full-thickness protrusion of
the bowel wall through the stomal orifice. It can be char-
acterized as fixed, with a permanent length of protruding
(c)
bowel, or sliding, wherein a variable length of bowel is noted
to prolapse with increased abdominal pressure (Figure 38.7).
Prolapse is more commonly noted with loop stoma, as com-
pared to end stomas with transverse loop colostomies car-
rying a risk of prolapse of up to 30% (45), and frequently
involves the distal stomal limb. The reasons for this are
unclear but may be related to atrophy of the distal limb sec-
ondary to disuse or excessive mobility due to lack of fixation
(24,34). It has been suggested that distal, as compared with
proximal transverse loop stomas, may be associated with
decreased rates of prolapse, secondary to partial fixation of
the distal limb by the splenic flexure (24).
Management of stomal prolapse varies with symptom- Figure 38.6 Translocation of stoma: (a) The stoma is
atology. Often, stomal prolapses are asymptomatic but may mobilized off the abdominal wall, (b) abdominal wall
result in cosmetic concerns or difficulties with pouching. defect created at the new, pre-chosen stoma site, (c)
Intermittent prolapses can be manually reduced via appli- clamp is used to translocate the bowel from original site
cation of gentle pressure, following the application of table to the new location.
sugar to help decrease the edema content of the bowel.
Surgical intervention is merited in the setting of obstruction,
ischemia, or perforation. Surgical options include stoma STOMAL VARICES
reversal, excision of redundant bowel and refashioning of
the stoma, or Delorme or Altemeier-type procedures (1). In Stomal varices present a potentially life-threatening com-
other cases, conversion of a loop stoma to an end or end-loop plication in patients with portal hypertension. Making the
stoma or stoma re-siting may be required (34). diagnosis often requires a heightened level of suspicion; an
Prevention / Pre-op preparation 359
PREVENTION
Due to the high incidence of stomal complications and

substantial economic cost, a focus has been placed on pre-
vention. For colorectal surgery patients, multiple steps can
be taken in the perioperative and intraoperative setting to
help reduce the incidence of complications and ultimately
enhance the experience for the patient.
PRE-OP PREPARATION
Figure 38.7 Stomal prolapse. The management of a patient for colostomy or

ileostomy starts before the operation.
understanding of the state of the patient’s liver disease cou- B.N. Brooke (1954)
pled with a careful examination of the stoma itself is critical.
A bluish collection of vessels is seen in the subcutaneous tis- In the era of enhanced recovery after surgery (ERAS) path-
sue surrounding the ostomy (Figure 38.8). It is not uncom- ways, and the Enhanced Recovery Program (ERP) in the
mon for this diagnosis to be overlooked; in a case series United Kingdom, there has been an increased focus on
presented by Spier et al. (2008) involving eight patients, the pr-operative patient education. For the colorectal surgery
diagnosis was initially overlooked in six (46). If the stoma patient, a large portion of this education focuses on stoma
appears normal, and the level of clinical suspicion remains marking and education. The importance of this was rec-
high, evaluation should include ultrasound, computed ognized in the 2015 position statement (47), revised from
tomography, or magnetic resonance imaging studies. the original in 2007 (48), published jointly by the American
Local treatment options for management of stomal vari- Society of Colon and Rectal Surgeons and Wound, Ostomy,
ceal bleeding include compression, sclerotherapy, and suture and Continence Nurses Society stating that ostomy site
ligation. These are frequently temporizing measures, as the selection and education should be performed preoperatively
bleeding commonly recurs. Mucocutaneous separation, whenever possible.
stomal relocation, and percutaneous embolization are simi- There are multiple reasons for this. Preoperative site
larly associated with high recurrence rates. Pharmacologic selection has been shown to reduce stoma-related complica-
therapy with ß-blockade has been proposed, but the efficacy tions and result in improved independent care of the stoma
of this therapy has been questioned (46). Definitive treat- by the patient (21,49,51). The preoperative education allows
ment options include surgical and percutaneous shunting time for the patient to psychologically prepare for the pres-
(i.e., transjugular intrahepatic portosystemic shunt) and ence of a stoma, develop basic understanding of supplies
liver transplantation. In a literature review by Conte et al. used and general care, and answer common lifestyle-related
(1990), no cases of rebleeding were observed in patients with questions (51–54). In a randomized study, Chaudhri et al.
ileostomies and colostomies following shunting; addition- (2005) found that preoperative education resulted in sig-
ally, shunting was noted to have a low overall morbidity and nificant difference in stoma proficiency (5.5 versus 9 days),
was associated with improved patient mortality (47). decreased hospital stay (8 versus 10 days), and hospital anxi-
ety depression scale tended lower (51).
To ensure appropriate site selection, patients should be
evaluated standing and sitting, both fully clothed and with
skirt lifted. During this process, observations of belt line,
other stoma bags, scars, or skin creases can be made. The
patient is then moved into the supine position with shirt off.
By having the patient lift his or her head off the exam table,
the rectus muscle can be identified and marked. The stoma
site marking must be made overlying the rectus muscle and
should be at least 5 cm away from surgical incision, centered
in the quadrant of interest, and within the patient’s visual
field. Avoidance of bony prominences, scars, and skin folds
(which often are not visible with the patient in supine posi-
tion) is important. The most common stoma site is within
the “ostomy triangle” bordered by the anterior superior iliac
Figure 38.8 Parastomal varices. spine, pubic tubercle, and umbilicus (55) (Figure 38.9).
360 Ostomies
Faceplate
Infraumbilical
fat fold
Rectus abdominis m. Ant. rectus sheath

Post. rectus sheath Peritoneum
Figure 38.9 Stomal siting.
Special considerations should be made for patient with INTRAOPERATIVE TECHNICAL TIPS
contractures or impaired mobility. Obese patients are at
higher risk for stoma complications and can be more dif-
ficult to find an ideal stoma site. It is important to keep in The closest attention must therefore be paid to the
mind that there may be a significant shift in subcutaneous minutiae of technique and the subsequent care.
tissue when an obese patient is moved from sitting to stand-
ing. It is typically recommended that an obese patient be BN Brooke (1952) (56)
marked at a higher site on the abdominal wall, as the upper
quadrants are often more visible to the patient, thinner, and Stomas are created toward the end of what can typically be
closer to the intended bowel. Once marked, it is important considered difficult and somewhat time-consuming oper-
to have the patient reassess the site himself or herself to con- ations. As such, exquisite care must be taken to continue
firm visibility and access (Figure 38.10). to adhere to strong surgical principles and techniques, in
Step 1 Step 2
Look at the profile of the patient. Notice where the abdomen While patient is seated, look for skin folds and creases.
curves back under toward the body. The underside of the Note and avoid skin folds and creases.
abdomen is not visible to the patient. Avoid this area.
Identify
creases and
folds.
Line of sight
Avoid
Patient cannot
see below the
line of sight.
Step 3 Step 4
Identify and target the rectus abdominis muscle below Mark optimal stoma sites on the rectus abdominis, that
the ribs. are in patient’s line of sight, while avoiding creases and
skin folds.
Creases when seated

Below ribs
X X Mark
prefered site
Below line of sight
Rectus for stoma.
abdominis
muscle
Figure 38.10 Marking the obese patient. (From Salvadalena G et al. J Wound Ostomy Continence Nurse
2015;42(3):249–52.)
Prevention / Postoperative support 361
order to prevent surgical-related complications. The stan- (a)

dard of care for everting ileostomies was first established
by Brooks in 1952 (56) and continues today. When creating
an ileostomy, it is generally recommended that approxi-
mately 6–8 cm of small bowel be eviscerated for creation
of an end ileostomy, and 3 cm at the apex of small bowel
for an ileostomy. This allows for an eversion of at least
2 cm above the skin level. Persson et al. (2010) followed
180 patients with stomas over a 2-year period and found
that all ileostomies less than 2 cm in height were associ- (b)
ated with fewer complications (57). Hall (1995) has sug-
gested making the inferior edge of a stoma approximately
1 cm shorter so as to tip the bowel and help prevent skin
irritation (58).
Similar to Brooke’s technique for the small bowel, sur-
geons have found that everting a colostomy, coined “spouted
colostomy” by Marsh (2007), can result in decreased com- Figure 38.11 (a) Cross section demonstrating midline
plications (59). It is generally recommended that 2–4 cm of incision and areas of subcutaneous fat excision. (b) After
colon be eviscerated for creation of an ideal colostomy to removal of excess subcutaneous tissue, incision is closed,
allow for appropriate eversion. The appropriate height of an flaps attached to fascia, and stoma matured with ade-
everted colostomy has varied in research. All recommend quate eversion.
a colostomy height over 1 cm to be associated with fewer
complications (30,57), and some have suggested height over anastomosis), but care must be taken to preserve the mar-
1.5 cm (28). ginal artery. Scoring or creating a window in the mesentery
Thick abdominal wall and a shortened and fat mesen- may provide additional length. Excess subcutaneous fat can
tery make ideal stoma creation particularly difficult in the be removed as described previously. When these techniques
obese patient. As previously mentioned, stoma siting higher do not work, creation of a loop end or “pseudo-loop colos-
on the abdominal wall can help. If the patient has not been tomy,” in which only the antimesenteric border of the colon
marked preoperatively, consider intraoperative fluoros- is brought through the abdominal wall and matured to the
copy to correctly identify the anterior-superior iliac spine skin without eversion, can be considered (62) (Figure 38.12).
to guide location (55), as traditional landmarks such as the Similar techniques can be used for an ileostomy that has
umbilicus may be misleading. Once your site has been cho- difficulty reaching the skin surface, including mobilization
sen, using a wound protector at the stoma site can aid in up to the duodenum, sacrifice of the ileocolic artery, or cre-
pulling the bowel and mesentery up. All attempts should be ation of a pseudoloop without eversion (20).
made to preserve the marginal artery at or above the level of A colon distended from a distal obstruction is another
the skin to ensure adequate blood supply. common clinical scenario that can make stoma creation dif-
An additional option is abdominal wall modification ficult. In addition to being dilated, this bowel has a tendency
or contouring. The technique has been described by sev- toward ischemia and impaired mobility. The increased
eral authors and uses plastic surgery techniques to create abdominal wall defect required to bring the bowel up to the
an abdominal wall that supports a well-constructed stoma skin can increase the risk of parastomal hernia. The most
(54,60,61). In a manner similar to an abdominoplasty, the important intraoperative maneuver in this setting to make
skin and subcutaneous tissue are mobilized off the fascia colostomy creation easier is to decompress the bowel ideally
via a low transverse or midline incision. If a stoma is pres- prior to bringing up through the abdominal wall or alterna-
ent, it is mobilized from the skin and subcutaneous tissue tively once it is up.
to the fascia. Excess subcutaneous fat is then excised from
the skin (Figure 38.11). The flaps are not as thin as those POSTOPERATIVE SUPPORT
produced by plastic surgeons to reduce the potential for
flap loss. After removing the fat, the stoma is then brought The introduction of ERAS protocols has resulted in a
through the skin at its original site or through a new loca- reduction in readmissions due to complications in the
tion. If excess skin is present, it can be excised before the colorectal surgical population (63); however, complications
stoma is brought through the skin. Subcutaneous drains associated with an ileostomy remain the most likely cause
are usually placed to encourage adherence of the flaps and of readmission—emphasizing the importance of appropri-
to reduce seromas. ate postoperative support. During hospitalization, a wound
For a colostomy that will not reach the skin, it is impor- care nurse should follow the patient. Their role, during this
tant to complete full mobilization of the bowel, medial, lat- time, is to reinforce education introduced preoperatively,
eral, and around the flexure. The main blood supply can be provide hands-on teaching for stoma appliance changes and
taken to increase length (as one would for a low coloanal care, and provide initial supplies for the patient.
362 Ostomies
(a) (b)
(c)
Figure 38.12 Loop-end (pseudo-loop) colostomy. (a) After the bowel is divided a loop of colon is brought up through the
defect (b) An incision is made in the distal end of the loop and matured in standard brooke fashion over a stoma rod (c)
Final appearance of loop-end colostomy.
It is recommended that upon discharge a patient has approach is preferred, patients with a prior Hartmann pro-
established follow-up with a wound care/stoma specialist. cedure or those in which the distal remnant is not available
The frequency of postoperative visits will be individual- via a local approach are obviously forced to undergo a repeat
ized and dependent on patient needs. In addition to pouch laparotomy. Surprisingly, there is a paucity of recent data
changes, enterostomal nurses can cover topics related to that highlight the potential perils of this seemingly benign
physical, psychologic, and social issues that arise when operation. The most recent study, published in 2005, was a
living with a stoma. This type of support has been shown retrospective review of 533 patients undergoing stoma clo-
to improve a patient’s quality of life with a stoma (64). sure at the University Hospital of Vienna (64). The majority
Moreover, they can monitor for and begin early interven- of the patients (51%) underwent reversal of a colostomy, 44%
tion for any complications that may arise (57). had closure of an ileostomy, and 5% had combined reversals
of both a colostomy and an ileostomy. All patients under-
went a laparotomy using the intraperitoneal approach. Their
30-day mortality was 3% (15 patients) with rates similar
OSTOMY REVERSAL for either ileostomy or colostomy reversal. Causes of death
were multisystem organ failure after nonsurgical compli-
Reversal of temporary stomas should be undertaken as cations in nine patients, and anastomotic leakage, missed
soon as physiologically feasible to reestablish gastrointes- small bowel injury, and cecal perforation in the remaining
tinal continuity and for psychological improvement. This, six patients. Overall complications were 20%, with anasto-
of course, implies that the purpose of the stoma placement motic leakage (5%), ileus (4%), postoperative bleeding (2%),
has been met, and the patient is capable and a candidate for and wound infection (2%). When analyzing patient-related
another operative procedure. There are two main operative factors between survivors and nonsurvivors, only advanced
approaches to ostomy reversal: local or via a laparotomy. patient age was found to be statistically significant. This
While both approaches are associated with inadvertent study, which highlights the potential morbidity of stoma
enterotomies, bleeding, wound infections, and anastomotic reversal, also emphasizes the importance of meticulous sur-
complications, the biggest advantages of the laparotomy gical technique required in these challenging patients with
approach are improved exposure and the ability to reexplore reoperative abdomens.
the abdomen. Certainly, the type of ostomy is important Our approach to ostomy reversal begins with a thorough
to consider when planning the operative approach as loop preoperative evaluation, which includes interrogation of the
ileostomies are technically the least challenging to reverse distal colon with either a barium enema or endoscopy. The
and are often amenable to local reversal. Although a local primary reason for which sentinel procedure was performed
References 363
(b)
SUMMARY
(a)
Creation of intestinal stomas is a common procedure,
meant to ultimately enhance the patient’s recovery
and quality of life, but with complication rates of up
to 70% with reoperations required in up to 20%, this
can have significant impact both on patient’s life and
(c) health-care economics. For this reason, thoughtful
decision-making, meticulous surgical technique, and
thorough perioperative education are imperative.
REFERENCES
1. Husain SG, Cataldo TE. Clin Colon Rectal Surg. 2008;
21:31–40.
Figure 38.13 Mercedes or triangular closure. (a) Stoma 2. Lewis L. History and evolution of stomas and
site with fascia closed. (b) Initial approximation of skin and appliances. In: Taylor P (ed.) Stoma Care in the
subcutaneous fat. (c) Completed closure with small area in Community: A Clinical Resource for Practitioners.
center left open for drainage and secondary healing.
London: EMAP Healthcare. 1999, pp. 1–20.
3. Arezzo A et al. Surg Endosc. 2016 (epub).
is important to consider, since it may reveal if the purposes 4. Sasaki K et al. Dis Colon Rectum. 2012;55(1):72–8.
of the ostomy have been met and may potentially alter the 5. Lim JF et al. Dis Colon Rectum. 2005;48(2):205–9.
decision on reversal. An obvious but sometimes overlooked 6. Lee, YM et al. J Am Coll Surg. 2001;192:719–25.
step should also be the evaluation of the patient’s sphincter 7. Steinermann DC et al. Langenbecks Arch Surg. 2015;
tone and ability to control fecal stream once continuity has 400(5):609–16.
been restored. This may require not only clinical evaluation, 8. Constantinides VA et al. Dis Colon Rectum. 2006;
but formal documentation through anorectal physiology 49(7):966–81.
testing including manometry. Baseline poor sphincter tone 9. Di Saverio S et al. Surg Endosc. 2016;30(12):
or incontinence should be considered a contraindication for 5656–64.
ostomy reversal in all but the rarest of cases. Finally, addi- 10. Alizai PH et al. Int J Colorectal Dis. 2013;28(12):
tional patient factors that can be altered, such as nutritional 1681–8.
status, steroid use, and tobacco abuse, should be optimized 11. Oberkolfer CE et al. Ann Surg. 2012;256(5):819–26.
prior to surgery. When planning the operative approach for 12. Rullier E et al. World J Surg. 2001;25:274–7.
end colostomy reversals, additional factors to consider prior 13. Harris DA et al. Ann R Coll Surg Engl. 2005;87:
to embarking on the operation should include the expected 427–31.
amount of adhesive disease likely to be encountered or pre- 14. Hanna MH et al. Langengecks Arch Surg. 2015;400:
viously encountered (i.e., review prior operative notes), 145–52.
whether there is a history of prior abdominal or pelvic radi- 15. Ihnat P et al. Surg Endosc. 2016;30(11):4809–16.
ation, concomitant pathology such as the presence of inci- 16. Wu S et al. World J Gastroenterol. 2014;20(47):
sional hernias, and the type of ostomy. For instance, patients 18031–7.
with multiple prior surgeries and a history of radiation will 17. Gu W, Wu S. World J Surg Oncol. 2015;13(9).
most likely benefit from a laparotomy approach that 18. Jafari MD et al. Ann Surg. 2013;79(10):1034–9.
includes preoperative ureteral stent placement, while those 19. Gastringer I et al. Br J Surg. 2005;92(9):1137–42.
patients with loop ostomies without any other comorbidi- 20. Wong NY, Eu KW. Dis Colon Rectum. 2005;48(6):
ties can be managed with a local approach. Whether a sta- 2076–9.
pled or hand-sewn anastomosis is performed is up to the 21. Nastro P et al. Br J Surg. 2010;97(12):1885–9.
surgeon’s discretion. Key technical points in each method, 22. Robertson I et al. Colorectal Dis. 2005;7:279–85.
however, are to ensure adequate mobilization and visualiza- 23. Leenen LP, Kuypers JH. Dis Colon Rectum. 1989;32:
tion of the distal colonic or rectal stump with resection of 500–4.
the exteriorized bowel or end stump back to normal healthy 24. Kwiatt M, Kawata M. Clin Colon Rectal Surg. 2013;
bowel prior to the anastomosis. Finally, delayed primary 26:112–21.
closure or partial closure is performed for the area in which 25. Arumugam PJ et al. Colorectal Dis. 2003;5(1):49–52.
the stoma was placed, and drains are not routinely placed 26. Mahjoubi B et al. Colorecal Dis. 2005;7:582–7.
(Figure 38.13). 27. Saghir JH et al. Eur J Surg. 2001;167(7):531–4.
364 Ostomies
28. Cottam J et al. Colorectal Dis. 2007;9:834–8. 46. Conte JV et al. Dis Colon Rectum. 1990;33(4):308–14.
29. Fish DR et al. Ann Surg. 2017;265(2):379–87. 47. Salvadalena G et al. J Wound Ostomy Continence
30. Velasco M et al. Cir Esp. 2014;92:149–56. Nurs. 2015;42(3):249–52.
31. Hayden DM et al. J Gastrointest Surg. 2013;17(2): 48. American Society of Colon and Rectal Surgeons
298–303. Committee Members; Wound Ostomy Continence
32. Messaris E et al. Dis Colon Rectum. 2012;55(2): Nurses Society Committee Members. J Wound
175–80. Ostomy Continence Nurs. 2007;34(6):627–8.
33. Nagle D et al. Dis Colon Rectum. 2012;55(12): 49. Park JJ et al. Dis Colon Rectm. 1999;42:1575–80.
1266–72. 50. Person B et al. Dis Colon Rectum. 2012;55(7):783–7.
34. Bafford A, Irani J. Surg Clin North Am. 2013;93(1): 51. Chaudhri S et al. Dis Colon Rectum. 2005;48(3):504–9.
145–66. 52. Millan M et al. Colorectal Dis. 2010;12(7):e88–92.
35. Beraldo S et al. Colorectal Dis. 2006;8(8):715–6. 53. Gulbiniene J et al. Medicine (Kaunas.) 2004;40(11):
36. Pilgrim CH et al. Dis Colon Rectum. 2010;53(1):71–6. 1045–53.
37. Hansson BM et al. Ann Surg. 2012;255(4):685–95. 54. Beck SJ. Clin Colon Rectal Surg. 2011;24(4):259–62.
38. Aquina CT et al. Dig Surg. 2014;31(4–5):366–76. 55. Brooke BN. Lancet. 1952;2:102–4.
39. Cross AJ et al. Br J Surg. 2017;104(3):179–86. 56. Persson E et al. Colorectal Dis. 2010;12:971–6.
40. Lopez-Cano M et al. Hernia. 2017;21(2):177–89. 57. Hall C et al. Br J Surg. 1995;82:1385.
41. Janes A et al. World J Surg. 2009;33(1):118–21. 58. Marsh P, Clark JS. Ann R Coll Surg Eng. 2007;89(1):78.
42. Serra-Aracil X et al. Ann Surg. 2009;249(4):583–7. 59. Cataldo, P. Clin Colon Rectal Surg. 2008;21(1):17–22.
43. Lambrecht JR et al. Colorectal Dis. 2015;17(10): 60. Evans JP et al. Dis Colon Rectum. 2003;46(1):122–6.
O191–7. 61. Beck DE. Clin Colon Rectal Surg. 2008;21(1):71–5.
44. Shellito PC. Dis Colon Rectum. 1998;41(12):1562–72. 62. Shah PM et al. Dis Colon Rectum. 2017;60(2):219–27.
45. Spier BJ et al. Clin Gastroenterol Hepatol. 2008;6(3): 63. Karadag A et al. Int J Colorectal Dis. 2003;18:234–8.
346–52. 64. Pokorny H et al. Arch Surg. 2005;140:956–60.
39
Operative and nonoperative therapy
for chronic constipation
ROBERT J. SINNOTT, MICHELLE C. JULIEN, AND DANIEL E. SARMIENTO
A 24-year-old female presents with a 2-year history of Constipation is an extremely common complaint in North
constipation. She stated that the symptoms started America with a prevalence between 2% and 27% of the
after surgery for perforated appendicitis. Medical population, estimating that about 63 million people are suf-
management with fiber supplements, therapeutic fering from this diagnosis. Women report two- to threefold
enemas, and various forms of laxatives have failed. higher incidence in constipation than men. There is a higher
Her symptoms have had a significant impact in her incidence of constipation in non-Caucasians as well as indi-
daily activities. viduals with less education and lower income. Additionally,
age has played a role in constipation with multiple studies
showing that older age leads to a higher prevalence of con-
stipation (1). In an effort to standardize the definition of
CASE MANAGEMENT constipation to identify patients who might require further
evaluation and treatment, a system-based classification was
Colonoscopy revealed no stricture or stenosis. established by a panel of experts with specific parameters in
Sitzmarks study was positive and demonstrated 2006 by the Rome Committee and is termed the Rome III
markers distributed throughout her colon with a Criteria (Table 39.1) (2). For a diagnosis of functional con-
preponderance in the rectosigmoid junction. Anal stipation to be made, the listed criteria must be met for at
manometry had a positive rectoanal inhibitory reflex, least 3 months with the onset of symptoms initiating at least
excluding Hirschsprung disease. Conventional 6 months prior to diagnosis.
defecography showed a partial outlet obstruction. Normal defecation is a complex process that results from
Scintigraphy study showed decreased whole gut tran- stool formation, colon motility, and pelvic floor function. It
sit. Diagnosis of colonic inertia as well as pelvic outlet is a coordinated event that relies on the interaction of the
dysfunction was made. Because of her debilitating autonomic, enteric, and somatic nervous systems stimu-
symptoms, she initially underwent loop ileostomy. lating the contraction of smooth muscle in the colon wall,
Biofeedback therapy was started and continued for resulting in the forward movement of stool and the ultimate
over a year. Repeat defecography showed improve- relaxation of the anal sphincters allowing for evacuation.
ment. She underwent total abdominal colectomy with Disruption of any part of this pathway will lead to some
takedown of her ileostomy and ileorectal anastomosis. form of constipation. Contributing factors may include diet,
She had a prolonged postoperative ileus but was medications, neurologic or endocrine disorders, psychoso-
discharged without further complication. Her bowel cial issues, colonic disease, or pelvic floor abnormalities
function has markedly improved. She continues with (Tables 39.2 and 39.3) (1–3). Often patients may have consti-
biofeedback and is under the care of a psychologist pation with no identifiable cause.
for behavioral issues but views her quality of life as an Several subtypes of constipation have been identified,
improvement to her preoperative state and does not including slow transit constipation, pelvic outlet obstruc-
regret having surgery. tion or pelvic constipation, and combined slow transit con-
stipation with outlet obstruction. Slow-transit constipation,
365
366 Operative and nonoperative therapy for chronic constipation
Table 39.1 Rome III diagnostic criteria for constipation Table 39.2 Drugs associated with constipation
Criteria must be fulfilled for the last 3 months Anticholinergics Antidepressants

Symptom onset at least 6 months before diagnosis Antipsychotics
1. Must include two or more of the following: Cation- Aluminum (antacids and sucralfate)
a. Straining ≥25% of defecation containing Bismuth
b. Lumpy or hard bowel movements ≥25% of agents Calcium (antacids and supplements)
defecation Iron supplements
c. Sensation of incomplete evacuation ≥25% of Neurally active Opiates
defecation agents Antihypertensives
d. Sensation of anorectal obstruction ≥25% of
Ganglionic blockers
defecation
Vinca alkaloids
e. Manual maneuvers to facilitate bowel movement
Calcium channel blockers
≥25% of defecation
Others Antihistamines
f. Fewer than three defecations per week
2. Loose stools are rarely present without the use of Antiparkinsonian drugs
laxatives Diuretics
3. Insufficient criteria for irritable bowel syndrome Nonsteroidal anti-inflammatory drugs
also referred to as colonic inertia, is a motility disorder in of symptoms. In addition, questions specifically addressing
which stool moves through the colon at a slow rate. In some defecatory habits including excessive straining, bloating,
patients, only the colon is affected, and patients may not the ability to sense and completely evacuate, and maneu-
have bowel movements for days to weeks at a time. Pelvic vers such as perineal pressure or digitalization of the vagina
constipation, a second subtype of constipation, is the lack of used to aid in evacuation should be asked. A stool diary
coordination of the pelvic floor during defecation. Examples kept by the patient, which details stool form and frequency,
of this include rectocele, enterocele, sigmoidocele, rectal may also provide valuable information. Dietary intake, fluid
prolapse, and rectal intussusception. Pelvic constipation consumption, exercise habits, medications including both
results in excessive straining, the need for digital evacua- prescription and over-the-counter supplements, and family
tion of stool, and incomplete evacuation (4,5). history of colorectal cancer should be assessed. The number
and type of laxatives used as well as their effect on bowel
function should be documented. Finally, any associated or
undiagnosed psychiatric, neurologic, or endocrine disorder
EVALUATION OF CONSTIPATION such as diabetes or hypothyroidism or a history of sexual
abuse should be identified in the initial assessment.
The initial evaluation of a patient with constipation should Once a detailed history has been obtained, a complete
always begin with a thorough history and physical examina- physical examination directed at the abdominal and ano-
tion to help distinguish constipation subtypes. Information rectal regions is required. The abdominal examination in
collected during the history should include details regard- general will be unrevealing; however, any specific areas of
ing stool consistency, blood in the stool, caliber and fre- tenderness on palpation or evidence of distention result-
quency of bowel movements, as well as onset and duration ing from colonic dilatation should be noted. Anorectal
Table 39.3 Medical conditions causing constipation
Endocrine and Collagen vascular

metabolic Neurogenic and musculoskeletal
Chronic renal failure Autonomic neuropathy Amyloidosis
Diabetes mellitus Cerebrovascular disease Dermatomyositis
Hypothyroidism Dementia Myotonic dystrophy
Hypercalcemia Depression Systemic sclerosis
Hypokalemia Multiple sclerosis Scleroderma
Pregnancy Muscular dystrophy
Milk-alkali syndrome Parkinson disease
Porphyria Spinal cord lesions
Carcinomatosis Hirschsprung disease
Chagas disease
Functional evaluation and diagnostic studies / Scintigraphy 367
examination should include a direct inspection for any respect to cost and the convenience of being able to perform
associated anorectal pathology, such as hemorrhoids, fis- the test in the colorectal surgeon’s office. When utilizing
sures, fistulas, or prolapse. Digital rectal exam should be manometry, the surgeon is able to interpret the analysis and
performed to exclude a mass, stricture, anal hypertonia, or discuss the findings with the patient after the procedure is
fecal impaction that can be associated with constipation. In completed.
addition, a straining maneuver will assist in diagnosing a
rectocele, internal or full-thickness rectal prolapse, cysto- BALLOON EXPULSION TEST
cele, or pelvic floor descent.
The balloon expulsion test is an essential component of
physiologic testing to assess evacuatory function with
regard to constipation. There is no standardized method of
FUNCTIONAL EVALUATION AND performing the test. In general, the balloon is attached to
DIAGNOSTIC STUDIES a catheter and placed into the rectum and inflated. A typi-
cal balloon is inflated with 50 mL of warm water. In the
The initial diagnostic workup for constipation should seated position, the patient attempts to expel the balloon.
include an anatomic evaluation of the colon with colonos- In a normal person, expulsion time is usually within 1 min-
copy, barium enema, or virtual colonoscopy to exclude a ute (9). The time is recorded for the length of time required
stricture or mass. A colonoscopy should be recommended for expulsion. Variables in the technique include the type
if the patient meets guidelines for screening or if symptoms of balloon, the material used to fill, and the amount filled.
such as rectal bleeding, change in bowel habit, weight loss, Although this test has the convenience of being performed
or anemia are identified (6). A colonoscopy is considered in the office with specificity ranging from 80% to 90%, its
in other patients as it provides a picture of the large intes- sensitivity is low at around 50% (10).
tine that more readily identifies strictures and volvulus and
may provide better screening in patients who often achieve RADIOPAQUE MARKERS
adequate cleansing required for an adequate colonoscopy.
Blood tests used to identify anemia, hypothyroidism, diabe- The ingestion of radiopaque markers to assess constipa-
tes, and other medical conditions can be helpful in assess- tion has been utilized since the 1960s (11). In this test a
ing constipation. In patients with no other underlying cause capsule with multiple radiopaque rings is used, and serial
for constipation and in those who do not respond to basic radiographs are taken. There are several protocols for this
treatments including fiber supplements and laxatives, then type of study, but all require cessation of laxatives during
further physiological testing is indicated, as discussed in the the study. Contributing factors may include diet, medica-
following text. tions, neurologic or endocrine disorders, psychosocial
issues, colonic disease, or pelvic floor abnormalities (Tables
ANAL MANOMETRY 39.2 and 39.3) (1,3). Often patients may have constipation
with no identifiable cause. Accurate documentation of the
The study and recording of pressures in the anal canal has date and time of ingestion of the capsule is important so
been practiced for decades, offering valuable information that plain radiographs are scheduled to be taken on days 1,
regarding several aspects of anorectal function. It is one of 3, and 5 to monitor the passage of the markers. A normal
the most commonly utilized diagnostic tests in the evalua- study requires passage of 80% of the markers by day 5 (12).
tion of pelvic floor function. The basic equipment needed for The number and distribution of the markers are noted at
manometry consists of a probe to sense pressure, a record- the end of the study. The segmental transit times are then
ing device, a monitor, and a software system to aid in analy- calculated by the distribution of the markers. The diagnosis
sis. Manometry can assess the anal canal tone by measuring of colonic inertia can be made if greater than five markers
resting pressure (difference between the intrarectal pressure are noted throughout the colon on day 5. Outlet obstruction
and the anal canal pressure), anal canal length, and squeeze should be considered when multiple markers have accumu-
pressure (contribution of the external anal sphincter and lated in the region of the rectosigmoid junction and have
puborectalis muscles) (6). Rectoanal inhibitory reflex (RAIR) not progressed into the rectum or been evacuated on radio-
allows discrimination of solid, liquid, and gas. The presence graphic review (13). Radiopaque markers have the benefit
of this reflex confirms a functioning myenteric plexus and, of decreased cost; however, the patient is subjected to small
therefore, the absence of Hirschsprung disease (7). Rectal amounts of ionizing radiation.
sensation can also be assessed by incrementally inflating a
balloon in the rectum. The patient’s first sensation and urge SCINTIGRAPHY
to defecate and maximum tolerated volumes are recorded.
Alterations in rectal sensation are contributed to constipa- Scintigraphic techniques have already been used success-
tion. A systematic review has demonstrated that anorectal fully to characterize the propulsive or contractile activity,
manometry shows that 20%–75% of patients suffer from or both, of the esophagus, stomach, gallbladder, and small
anorectal dysfunction (8). Manometry is advantageous with intestine. Colonic transit is evaluated by assessing the
pattern of the tracer and percent excretion at 24, 48, and comparing echodefecography and conventional defecogra-
72 hours (14). The use of γ-emitting radionuclide mark- phy in the diagnosis of anal rectal disorders (13).
ers would seem to be ideal for the study of colonic motility Dynamic 3D transvaginal and transrectal ultrasonog-
because of low radiation exposure, noninvasive imaging, raphy (TTUS) using a biplane transducer has been shown
and patient comfort. Studies have shown the ability to delin- to be an acceptable alternative method in the evaluation of
eate segmental colonic transit. However, despite attempts to women with obstructed defecation syndrome. In a recent
simplify the techniques and make them more applicable for review, TTUS achieved high rates of concordance compared
clinical use, measurement of colonic transit time by scintig- with echodefecography in the diagnosis of anismus, recto-
raphy is not yet widely available (15). cele, enterocele/sigmoidocele, and intussusception (21). The
benefits of ultrasound techniques compared to other imag-
ing modalities such as defecography and dynamic magnetic
resonance imaging (MRI) include the ability to visualize the
ANATOMICAL EVALUATION AND anal sphincter looking for defects, atrophy, and scarring as
FUNCTIONAL STUDIES well as identifying dynamic dysfunction. In addition, it does
not expose patients to radiation, is readily available in the
DEFECOGRAPHY office making it cost effective, and involves less discomfort
compared to conventional imaging. In a study by Vitton
Conventional defecography, also referred to as evacuation et al., patient tolerance was significantly better for anorectal
proctography, is widely used to evaluate posterior pelvic endosonography compared to defecography and MRI, and
floor dysfunction related to obstructed defecation syn- the majority of patients would choose ultrasound over the
drome (16). It assesses dynamic changes in the rectal wall, other studies if a second imaging study was necessary (22).
anal canal, vagina, and pelvic floor anatomy during the def- Available techniques include 2D/3D/4D, which allow for
ecation process. Specific measurements of the anal rectal multiplanar reconstructions of the complex pelvic floor (23).
angle, extended perineal descent, and puborectalis length
can be calculated. The rectal contrast is thickened to simu- PELVIC MAGNETIC RESONANCE
late stool weight and consistency. Images are obtained when IMAGING
the patient is asked to squeeze and contract the external
anal sphincter and puborectalis at rest and with Valsalva. In recent years, MRI has been used more often due to stron-
Patients with pelvic dyssynergy related to a nonrelaxing ger gradients and shorter scanning time to help diagnosis
puborectalis muscle can be identified by failure of the anal pelvic pathology. Some of the pelvic pathologies that can be
rectal angle to open, persistence of the puborectalis com- explored using MRI include rectocele, rectoanal intussus-
pressing the rectum, and inadequate rectal emptying of the ception, measurement of the anorectal angle, and perineal
barium paste (17,18). If the patient digitates to facilitate or descent (24). In a study by Matsuoka and colleagues, pelvic
initiate evacuation, images during these maneuvers are per- MRI was compared to videoproctography (VP). They found
formed (19). One of the advantages of defecography is that that patients had no radiation exposure, and no rectal con-
this can be performed in the upright sitting position allow- trast was needed to evaluate the above-mentioned patholo-
ing for the natural physiologic influence of gravity, abdomi- gies, making it an attractive imaging modality for patients.
nal wall muscles, and weight of the intestines to exert their However, at the conclusion of their study, they found no
effects on the patient’s ability to evacuate. This type of difference between the use of MRI (which is 10 times more
imaging modality can also detect rectocele, rectal intussus- expensive) and VP. In fact MRI was not able to show small
ception, rectal prolapse, enterocele and sigmoidocele, and rectoceles and anorectal intussusception, and this was
descending perineum syndrome (19). Disadvantages of this attributed to patients having to be in the prone position and
method include the inability to evaluate soft tissues, provide not using rectal contrast (25).
information about sphincter defects, exposure to radiation,
and an awkward and embarrassing experience for patients.
MEDICAL TREATMENT OF
ULTRASOUND CONSTIPATION
Current advancement in imaging technologies and the
development of three-dimensional (3D) ultrasound equip- Constipation is a common problem, and unfortunately
ment have enhanced the ability to study pelvic floor anat- its treatment has been far from satisfactory. A 2002 meta-
omy and dysfunction in different planes. Echodefecography, analysis suggested that there was little credible evidence to
introduced by Murad-Regadas (20), is a dynamic 3D anal support many of the drugs that are commonly used, espe-
rectal ultrasonography technique using a 360° rotational cially the over-the-counter preparations (26). However, this
transducer placed in the rectum with automatic scans in the analysis grouped all agents into a single “laxative group,”
axial, sagittal, and oblique planes. In a multicenter study which may have obscured any benefits of individual medi-
of 86 women, a high degree of agreement was noted when cations (27).
Medical treatment of constipation 369
Patient education and dietary and lifestyle modifications categorized as soluble or insoluble. Examples include psyl-
play an essential role in the initial treatment of constipation, lium (Metamucil and Kosyl), methylcellulose (Citrucel), and
regardless of its etiology. Education should include proper calcium polycarbophil (Fibercon). Dietary additives such as
sitting positions and an explanation of normal physiologic bran may cause abdominal bloating and discomfort, which
bowel patterns (28). It is also wise to advise the patient that in turn may decrease patient compliance. Gradual dose
a daily bowel movement is not a requisite to good health, increase may minimize these symptoms. However, fiber may
and all providers should strive to decrease patient anxiety not be the most appropriate therapy for all causes of consti-
over the act of defecation. It is up to the physician to iden- pation. In a study published by Winfried et al., fiber was of
tify patients who need psychological support, since consti- limited value in 83% of patients with slow-transit constipa-
pation may be aggravated by stress or depression and should tion and 63% of patients with a disorder of defecation and
communicate to the patient that his or her symptoms will outlet obstruction. Conversely, 85% of patients without a
not improve overnight, and modifications of the treatment pathological finding or underlying motility disorder either
regimen may be required. improved or became asymptomatic with fiber therapy (35).
General measures such as adequate hydration and regu- These data suggest that a therapeutic trial of dietary fiber,
lar exercise may improve overall health; however, there is in the range of 25–35 g/day, should be considered as initial
no evidence to support success in the treatment of chronic treatment for patients with constipation, with the expecta-
constipation except in situations of dehydration (29,30). tion of a more effective response in normal-transit or fiber-
Nonetheless, diet modification to increase water and fiber deficiency constipation patients as opposed to those patients
consumption is considered an important, first-line compo- with slow-transit constipation or pelvic floor dysfunction.
nent and is typically recommended before technical inves- When nonpharmacologic management does not improve
tigations of pelvic floor function and colon motility are symptoms, laxative medications should be added for the
performed or medications are started (31–35). Dietary fiber management of constipation. A list of common medications
supplementation has been shown to allow discontinuation of used to treat constipation is shown in Table 39.4. There are
laxatives in 59%–80% of elderly patients with chronic idio- many options available, and the choice of therapy should be
pathic constipation while improving body weight and well- subject to a consensus between the treating physician and
being (36). patient preference (39).
Fibers are bulking laxatives known to increase fecal mass/ Osmotic laxatives are poorly absorbed or nonabsorbed
weight by absorbing and retaining fluid, which in turn stim- compounds that work by increasing water content within
ulates motility and accelerates colonic transit time, resulting the large bowel; they include milk of magnesia, magne-
in an increase in bowel frequency (37,38). Its treatment relies sium citrate, and sodium phosphate. These agents are not
heavily on the ingestion of an adequate amount of fluid to recommended in patients with cardiac and renal dysfunc-
reach efficacy. They are also reasonably inexpensive, making tion given that excessive absorption may lead to electro-
an effective therapeutic intervention for addressing consti- lyte abnormalities and volume overload. When ingested as
pation-related bowel dysfunction (32,35). Fiber is found in hypertonic solutions, rapid osmotic equilibration occurs,
grains, fruits, vegetables, nuts, seeds, and beans, and can be and overuse may result in significant dehydration (40).
Table 39.4 Medications commonly used for constipation
Category Example Mechanism of action

Bulking laxatives • Psyllium Increase fecal mass to stimulate peristalsis
• Methylcellulose
• Calcium polycarbophil
Osmotic laxatives • Lactulose Increase water content in large bowel
• Sorbitol Lactulose is fermented by colonic bacteria, which decrease
• Magnesium salts gut pH, resulting in fecal volume expansion and
• Polyethylene glycol (PEG) accelerated transit
Stimulant laxatives • Senna Increase intestinal motility
• Cascara
• Bisacodyl
Softeners • Mineral oil Change stool composition creating softer stools
• Docusate
• Glycerine
Prosecretory agents • Lubiprostone Activate chloride channels to increase intestinal chloride
• Linaclotide secretion
Prokinetics • Almovipan Μµ-opioid receptor antagonist
Serotonin receptor agonist • Prucalopride Selective 5-HT4 agonist
Lactulose and sorbitol are nonabsorbable disaccharides. Assessment of Constipation quality of life score, and all
Lactulose is readily fermented by the colonic bacterial flora subscales in the hisacodyl-treated patients, compared with
with the production of short-chain fatty acids and various those who received placebo (p ≤ 0.70), and treatment with
gases, which act as osmotic agents that stimulate intestinal bisacodyl was well tolerated (50).
motility and secretion (41). Lactulose has been shown to Senna is a member of the anthraquinone family of laxa-
increase stool frequency in chronically constipated patients tives that are common constituents of herbal and over-
(42); side effects may include abdominal bloating, discom- the-counter laxatives that are metabolized in the colon by
fort, and flatulence, which may decrease patient compliance. bacteria into their active forms. In a trial of elderly nurs-
Sorbitol is a poorly adsorbed sugar alcohol that produces ing home residents (n = 77), a senna and fiber combina-
similar effects. In a trial of constipated men over the age tion was reported to be better than lactulose in improving
of 65, sorbitol administered as a 70% syrup (10.5 g/15 mL; stool frequency, stool consistency, and ease of passage (51).
15–60 mL daily) was equivalent to lactulose in improving Furthermore, the senna and fiber combination was 40%
symptoms (43). Furthermore, it was cheaper and better tol- cheaper than lactulose therapy.
erated during a 4-week trial. Side effects of stimulant laxatives may include aller-
High molecular weight polyethylene glycol (PEG) is a gic reactions, electrolyte imbalances, melanosis coli, and
large polymer with substantial osmotic activity commonly “cathartic colon.” Melanosis coli is an abnormal pigmenta-
used along with a balanced electrolyte solution for colon tion of the colonic mucosa caused by the accumulation of
cleansing in preparation for colonoscopy and bowel surgery apoptotic epithelial cells being phagocytosed by macrophages
(44,45). A 2011 Cochrane Review evaluated relevant data (52). “Cathartic colon” is an alteration of the normal colon
to determine whether lactulose or polyethylene glycol was anatomy. It includes colonic dilatation, loss of haustra folds,
more effective in treating chronic constipation and fecal strictures, colonic redundancy, and wide gaping of the ileoce-
impaction. The meta-analysis included 10 randomized con- cal valve “fish mouth appearance” (53). Despite this common
trolled trials; their findings indicated that polyethylene gly- finding, current evidence supports the safety of currently
col is better than lactulose in outcomes of stool frequency available laxatives at recommended doses for long-term use.
per week, form of stool, relief of abdominal pain, and the Newer agents for constipation such as prosecretory
need for additional products (46). PEG is a reliable treat- agents (lubiprostone and linaclotide) and prokinetic agents
ment for chronic constipation because of its high efficacy, (alvimopan) may be considered when dietary modifica-
proven benefit when compared with other agents, and its tions, as well as osmotic and stimulant laxatives, have failed.
long-term effectiveness. Other agents such as MiraLAX Lubiprostone (Amitiza) is an oral bicyclic functional
have been effectively used as osmotic laxatives for the treat- fatty acid that activates the type 2 chloride channels located
ment of chronic constipation. MiraLAX does not contain on the intestinal epithelial cell leading to an active secre-
absorbable salts and generally does not alter measured levels tion of chloride into the intestinal lumen (54,55), This
of core electrolytes as well as calcium, glucose, blood urea drug has been shown to slow gastric emptying but accel-
nitrogen (BUN), creatinine, and serum osmolality (47–49). erate small bowel and colonic transit time at 24 hours in
Stimulant laxatives are not absorbed and exert their healthy volunteers (56). In a multicenter, 4-week, double-
effects at the mucosal level, causing a reduction in the blind, randomized, placebo-controled trial, lubiprostone
absorption of water and electrolytes and stimulating the produced a bowel movement in the majority of individu-
production of secretions, which creates a prokinetic effect als within 24–48 hours of initial dosing and significantly
in the colon. Abdominal discomfort and cramping are com- increased the number of spontaneous bowel movements
mon side effects of these agents. Bisacodyl produces defeca- within the study period (57). These findings are similar to
tion within 6–8 hours of taking the tablet, or 15–30 minutes other well-developed studies (55). Lubiprostone is currently
after suppository. Its mechanism of action is the conversion approved by the U.S. Food and Drug Administration for
into the active metabolite, bis-(p-hydroxyphenyl)-pyridyl- the treatment of chronic idiopathic constipation in adults,
2-methane (BHPM) which has a dual action as an antiab- opioid-induced constipation in adults with chronic non-
sorptive-secretory agent as well as inducing high-amplitude cancer pain, and irritable bowel syndrome with predomi-
propagated contractions of the bowel resulting in a direct nant constipation in women older than 18 years of age
prokinetic effect (50). In a randomized, double-blind, pla- (58). Common side effects include nausea, headaches, and
cebo-controlled, multicenter trial in the United Kingdom, diarrhea.
patients were randomly assigned, in a 2:1 ratio, to groups Linaclotide (Linzess) is a minimally absorbed peptide
that were given 10 mg of bisacodyl or placebo, once daily, agonist of the guanylate cyclase C receptor. It results in the
for 4 weeks. Patients used an electronic diary each day to generation of cyclic guanosine monophosphate (cGMP),
record information relating to their constipation. The num- which in turn, activates transmembrane channels inducing
ber of complete spontaneous bowel movements per week the secretion of chloride and bicarbonate into the intestinal
during the treatment period increased from 1 ± 0.1 in both lumen, increasing luminal fluid secretion and accelerating
groups to 5.2 ± 0.3 in the bisacodyl group and 1.9 ± 0.3 in intestinal transit. In two randomized, controlled, double-
the placebo group. Compared with baseline, there was a blind multicenter studies, Lembo et al. attempted to deter-
statistically significant improvement in the overall Patient mine the efficacy and safety of linaclotide in patients with
Biofeedback therapy 371
chronic constipation. After randomization, patients received patients with input regarding their performance during
either 145 micrograms or 290 micrograms of linaclotide ver- attempted defecation maneuvers.
sus placebo daily for 12 weeks. The primary efficacy endpoint The use of biofeedback therapy for the treatment of pelvic
was three or more complete spontaneous bowel movements floor dyssynergia has been well documented (70,71). These
(CSBMs) per week and an increase of one or more CSBMs studies report success rates ranging from 30% to 100% with
from baseline during at least 9 of the 12 weeks. This was over two-thirds of patients benefiting from biofeedback
reached in 16% and 21% of patients who received 145 micro- training. In a prospective randomized controlled trial, Rao
grams of linaclotide and by 19.4% and 21.3% of patients who et al. investigated the efficacy of biofeedback with either
received 290 micrograms versus 3.3% and 6% of patients who sham feedback therapy or standard therapy (diet, exercise,
received placebo (p < 0.01). The authors concluded that lina- and laxatives) in 77 subjects with chronic constipation and
clotide significantly reduced bowel and abdominal symp- dyssynergic defecation (68) at baseline and after 3 months of
toms of constipation, but long-term risk and benefits need treatment. Physiologic changes were assessed by anorectal
to be assessed (59). Currently, linaclotide is approved in the manometry, balloon expulsion, colonic transit study, and
United States for the treatment of irritable bowel syndrome symptomatic changes. Subjects in the biofeedback group
with constipation and chronic idiopathic constipation. were more likely to correct their dyssynergia, improve def-
Alvimopan, is a peripherally acting Μµ-opioid receptor ecation indexes, and decrease balloon expulsion time.
antagonist that does not cross the blood-brain barrier and, In regard to patients with outlet obstruction with asso-
therefore, does not inhibit the analgesic effect of opioids. ciated pelvic floor dyssynergia, a 2006 randomized, con-
Alvimopan can increase motility in opioid-induced colonic trolled trial compared patients with pelvic floor dyssynergia
constipation without compromising analgesia; however, treated with biofeedback versus laxative therapy with poly-
this does not seems to extend to patients with idiopathic ethylene glycol and education. The results favored the bio-
constipation transit in healthy subjects (60,61). This drug feedback-treated group, and benefits continued to be seen
has been shown to be effective in the treatment of acute after 2 years of initial treatment. The authors conclude that
postoperative ileus (62). biofeedback therapy should be the treatment of choice for
Other agents such as elobixibat (63,64), and procalopride this type of constipation (72).
(65), are currently under investigation and not formally Biofeedback therapy may rely on EMG monitoring of
approved by the U.S. Food and Drug Administration for muscle tone or anorectal pressures for patient training. A
the treatment of chronic idiopathic constipation. However, 2003 meta-analysis compared both methods and concluded
Plecanatide (66), which functions as a guanylate cyclase-C that patients with pressure biofeedback protocols showed
(GC-C) agonist and is taken orally once daily, acts locally significantly better outcomes (mean success rate of 78%) ver-
on the luminal surface of the intestinal epithelium resulting sus EMG biofeedback (mean success rate of 70%). Further
in increased intestinal fluid and accelerated transit has been analysis compared intraanal to perianal EMG biofeedback,
recently approved. and the results showed no significant difference between the
two subgroups (69% versus 72%, respectively). The overall
data showed a success rate ranging from 69% to 78%, regard-
less of which protocol or what instrumentation was used (69).
BIOFEEDBACK THERAPY The treatment sessions are typically performed by nurses
or physical therapists with advanced training and interest in
As described previously, a subtype of constipation identi- pelvic floor disorders. Outcomes depend on the affect and
fied as pelvic constipation may be the result of pelvic floor patience of the therapist and patient acceptance to the thera-
dyssynergia also known as anismus. It is characterized by pist’s technique. Other factors influencing the outcome of
a failure of the abdominal, rectal, and pelvic floor and anal biofeedback therapy have been studied. In a recent review,
sphincter muscles to coordinate and complete the act of def- additional predictors of successful outcome include the
ecation resulting in an impaired propulsion of stool from number of sessions attended (five or more) and whether the
the rectum, paradoxical anal contraction, inadequate anal completion of therapy was determined by the therapist (63%
relaxation, or a combination of these (67). This represents success rate) as opposed to the patient terminating treatment
about one-third of patients with chronic constipation seen prematurely (25% success rate) (73). Unfortunately, to date,
in tertiary care centers (68). no physiologic (manometry and balloon expulsion test),
Biofeedback therapy plays an important role in patients anatomic (rectocele, intussusception, or abnormal perineal
with constipation due to pelvic floor dyssynergia, especially descent), or demographic (age, gender, or duration of symp-
after failure of conservative management (69). Biofeedback toms) variable has been able to be identified that would influ-
therapy is based on operant conditioning to reinforce posi- ence treatment outcome. Nevertheless, many researchers
tive behavior; it uses electronically amplified recordings of suggest that psychopathology may play an important role (2).
pelvic floor muscle contractions (electromyography [EMG]) Despite the lack of a reproducible standardized tech-
or anorectal pressure tracings to teach patients how to relax nique, biofeedback therapy is a relatively inexpensive treat-
pelvic floor muscles and to strain more effectively when they ment option that has no related risk of side effects. Of note,
defecate (2). Visual or auditory feedback is used to provide the presence or absence of irritable bowel syndrome does
not appear to impact the success rates of biofeedback for Table 39.5 Results of total abdominal colectomy with
constipation (74). ileorectal anastomosis
There is insufficient data supporting the use of biofeed-
Number of Success Follow-up
back therapy for the treatment of slow-transit constipation
Study patients (%) (Mo)
(75,76).
Failure of biofeedback therapy poses a significant treat- Beck 14 100 14
ment problem. Botulinum toxin injection has been proposed Nylund 40 72.50 132
as an alternative therapeutic modality for patients with refrac- FitzHarris 75 80 —
tory pelvic floor dyssynergia. Injection is directed into the Fan 24 82 23
puborectalis muscle and external anal sphincter. Symptom Hassan 104 85 104
improvement is reported to be short lived (1–3 months) with Webster 55 89 12
incontinence reported in 25% of patients. Because of these Zutshi 64 92 128
results and expense of the drug, this treatment modality Pikarsky 50 100 106
should be reserved for those patients with severe symptomatic Wexner 16 94 15
pelvic dyssynergia who have failed all other therapies (77). Pinedo 20 95 25
Glia 14 86 60
Riss 12 50 84
SURGICAL OPTIONS Sohn 37 81.8 41
Sheng 68 94 40.7
The role of surgical intervention for patients who present with Source: Pinedo G et al. Surg Endosc. 2009;23:62–5; Pinedo G
et al. Surg Endosc. 2009;23:62–5; Sohn G et al. J Korean
constipation should only be considered when the classifica- Soc Coloproctol. 2011;27(4):180–7; Fan CW, Wang JY. Int
tion has been accurately identified as result of a thorough Surg. 2000;85:309–12; Hasson I et al. J Gastroinest Surg.
constipation workup, including colon transit studies and 2006;10:1330–7; Riss S et al. Colorectal Dis. 2009;11(3):
pelvic floor physiology testing. In addition, symptoms should 302–7.
be severe and incapacitating and are not responsive to nutri-
tional or pharmacological therapy. Lahr evaluated over 2,000 commonly performed (93,94). A report by Xu comparing
patients who presented with severe constipation and found laparoscopic versus open colectomy for slow-transit consti-
only 9.9% met criteria for surgical treatment (78). As with any pation confirms the established benefits of the minimally
operative procedure, patient selection is critical for success. A invasive techniques in colon surgery, including less blood
decision for surgery requires a detailed discussion regarding loss, earlier return of bowel function, better cosmesis, and
standard operative risk and postoperative complications as shorter hospital stays (95).
well as appropriate expectations as they relate to improvement Variations of the procedure have been developed, includ-
in bowel frequency and the resolution of symptoms including ing preservation of the cecum with cecorectal anastomosis
abdominal pain and bloating, which are often associated in or a portion of the sigmoid colon with ileosigmoid anas-
these patients and can persist after surgical intervention (79). tomosis. The intent is to reduce the side effects of diarrhea
For patients who have documented slow-transit consti- and electrolyte abnormalities that can be associated with
pation and normal pelvic floor physiology, total abdominal total abdominal colectomy. Sarli reported an antiperistal-
colectomy with ileorectal anastomosis is generally consid- tic anastomosis between the cecum and the rectum with-
ered to be the procedure of choice. Other options include out mesenteric rotation with good results, with only one
subtotal colectomy with ileosigmoid anastomosis, subto- case of recurrent constipation (96). Marchese reported 22
tal colectomy with cecorectal anastomosis, and segmental patients who underwent cecorectal anastomosis for slow-
colectomy. transit constipation. Over 88% of the patients were pleased
Abdominal colectomy was first reported by the British with the results and expressed a willingness to repeat the
surgeon Arbuthnot-Lane in 1908 for patients with constipa- procedure given the same preoperative conditions. They
tion (80). Overall success of total abdominal colectomy with initially concluded that cecorectal anastomosis did not
ileorectal anastomosis is exceedingly high with good rates appear to be inferior to subtotal colectomy with ileorectal
of clinical improvement and patient satisfaction ranging anastomosis in terms of therapeutic effectiveness, postop-
from 50% to 100% (81–86) (Table 39.5). One summarizes erative morbidity, and overall impact on quality of life (97).
many of the results of total abdominal colectomy with ileo- However, a subsequent review by the same author reported
rectal anastomosis (87–92). Morbidity after total abdominal three delayed surgical complications from cecorectal anas-
colectomy with ileorectal anastomosis can be significant, tomosis relating to a dysfunctional or mobile cecal stump
including anastomotic leak seen in up to 10% of patients, requiring surgical revision. Additional reports support the
small bowel obstruction with a reported incidence up to concept that maintenance of the cecal reservoir can result
30%, and postoperative ileus seen in up to 25% of patients in dilatation and recurrence of constipation symptoms (98).
(83,85). Minimally invasive approaches have been success- In an attempt to improve functional results with this tech-
fully applied to this procedure with good results and are nique, Wei concluded that postoperative outcomes can be
Surgical options 373
optimized by shortening the length of the ascending colon excision of the pouches because of poor functional results
reservoir above the ileocecal junction to 2–3 cm as opposed (108). Proctocolectomy with ileoanal pouch creation
to 10–15 cm (99). should be chosen carefully as a surgical option for patients
Sigmoid preservation does not have any major advan- with refractory constipation, regardless of the circum-
tage compared with other procedures and can predispose stance, and should be considered only when a stoma is the
to postoperative constipation. Pemberton reported a 50% only remaining option. The available literature does not
conversion rate from ileosigmoid anastomosis to ileorectal justify the risk of complications from this procedure for
anastomosis (100). A retrospective comparison of cecorectal this indication, and patients should be extensively coun-
anastomosis versus ileosigmoid anastomosis demonstrated seled regarding functional expectations (109).
that the cecorectal anastomosis was more often associated There is a subset of patients who may benefit from fecal
with persistent constipation and lower patient satisfaction diversion, albeit an extreme measure, as a reasonable alter-
(73% versus 93%) (101). Despite the theoretical advantages native for the management of constipation and should be
of these procedures, preservation of the cecal reservoir and reserved for those who have failed other available treat-
sigmoid colon can result in the recurrence of constipation ments. In addition, this procedure can be considered in
symptoms. symptomatic patients who may have global intestinal motil-
Segmental colectomy has been advocated by some for the ity disorder, combined slow-transit constipation, as well
treatment of slow-transit constipation with the theoretical as a pelvic outlet problem, in which case it is uncertain as
advantages being a reduction in the incidence of diarrhea to whether the patient will benefit from colonic resection,
that can be seen with total abdominal colectomy (102,103). fecal incontinence, or who may not tolerate a colon resec-
Modifications of colon transit studies have been used to tion. Ileostomy is generally the preferred stoma, and relief of
determine segmental colon transit impairment, which symptoms must be weighed against potential complications
would subsequently influence a decision to perform a seg- associated with an ileostomy, including electrolyte abnor-
mental resection, although the reliability of these studies malities, dehydration, and other stomal problems includ-
has been questioned (103,104). You et al. reported a series ing hernia and retraction. Scarpa reviewed outcomes in 24
of patients who underwent different types of colectomies patients who underwent ileostomy for constipation with
including left and right colectomy according to the distri- a 96% success rate in alleviating symptoms. Four patients
bution or accumulation of markers in the colon. Follow-up subsequently underwent ileostomy closure, and 50% of
in 2 years revealed that all but three patients had satisfac- them developed recurrent constipation (110).
tory bowel movements without worsening constipation. Another less invasive option in the treatment of consti-
Three patients eventually went on to have subtotal colec- pation is the use of antegrade colonic enema (ACE), which
tomy with ileorectal anastomosis (105). In addition, Lundin was initially described by Malone for the treatment of fecal
reported on 28 patients who underwent segmental resec- incontinence in children, utilizing enemas delivered to the
tions based on impaired transit in one segment by radio- cecum to wash out the colon (111). The technique has also
graphic imaging. While 23 patients were satisfied with the been described in adults with neurogenic constipation sec-
outcome, 5 subsequently required additional surgery (106). ondary to spinal cord injury, as well as slow-transit consti-
It must be emphasized that interpretation of radial opaque pation or obstructive defecation (112–116). Over time, there
marker studies has limitations, and the decision to proceed have been modifications to the original technique with the
with segmental resection should be approached with cau- creation of catheterizable conduit most commonly using
tion since controlled data are lacking, and failure rates can the appendix left in situ to administer the enema, although
be high (106). a tube cecostomy and ileostomy have been described (117).
The role of proctocolectomy as well as completion proc- The benefits of this approach are the avoidance of a colec-
tectomy after total abdominal colectomy and ileorectal tomy and functioning stoma. The main disadvantage is a
anastomosis with ileo pouch anal anastomosis has been high incidence of surgical complications, including stenosis
described for slow-transit constipation. Although the data of the conduit occurring in up 23%–100% of patients and
are limited, Hosie reported on 13 patients with functional leakage (112,117). The placement of an indwelling cath-
bowel disorders including 8 who had recurrent constipa- eter has been shown to decrease the incidence of stenosis,
tion after colectomy for slow-transit constipation and 5 although wound infection and catheter dislodgement are
who had constipation associated with megarectum and frequent (118). Two series of just over 100 patients reported
megacolon. Mean frequency of defecation was improved, the use of the ACE procedure for constipation and reported
and 85% of the patients surveyed felt the operation had satisfactory function in 42%–47% of patients (112). Success
been worthwhile with an improvement in their quality of is generally reported as technical success, with less reli-
life. Two patients had the pouch converted to an ileostomy able data on the improvement of constipation symptoms
due to persistent complications and a poor functional and no comparison with other therapies for constipation.
result (107). Keighley reported his series of patients who Hirst reported that 65% of patients experienced subjective
had previously undergone total colectomy with ileorectal improvement in defecation, but many patients were lost to
anastomosis who had recurrent difficulty with evacuation follow-up (113). Patient compliance is an important factor
of stool. Four of these patients subsequently underwent in judging the success of this approach, and most series
report only 50% of patients use the conduit for irrigations from a rectocele, and size of a rectocele being greater than
in the long term (112,113,119). Selection of this procedure 4 cm (126). The basic premise of the operation is to recreate
should be reserved for highly motivated patients whose con- the rectovaginal septum to eliminate the bulge into the pos-
stipation is secondary to neurogenic causes and who under- terior wall of the vagina.
stand the daily long-term commitment necessary to achieve Theoretical benefits of the transvaginal repair include
success. better visualization and access of the endopelvic fascia
and the levator muscles as well as maintenance of the rec-
tal mucosa integrity that may reduce infection and fistula
complications (109). Transrectal repair has a theoretical
SURGERY FOR COMBINED DISORDERS advantage of less sexual and defecatory dysfunction as well
as addresses any other anorectal pathology. A transperineal
In patients with mixed disorders of slow-transit constipa- approach is appealing for patients with both a symptomatic
tion and obstructed defecation, treatment is complex and rectocele and fecal incontinence as a result of a sphincter
requires a multidisciplinary approach. The role of total defect, which would allow for a concomitant sphinctero-
abdominal colectomy with ileorectal anastomosis remains plasty to be performed.
somewhat controversial, since a favorable outcome is less There have only been a few prospective studies done to
predictable than for patients with pure slow-transit consti- assess the efficacy of these different approaches. Arnold
pation (82). Ragg evaluated 541 patients with chronic consti- compared transrectal versus transvaginal approaches,
pation and identified a 29% incidence of outlet obstruction which demonstrated an equal complication rate in both
in conjunction with colonic inertia (120). Hedrick created a groups. Over 50% of the patients had postoperative consti-
classification scheme in patients with obstructed defecation pation, and up to 30% had some component of gas or liquid
based on functional and mechanical etiologies. Functional stool incontinence (127). Nieminen examined transvaginal
causes include short segment Hirschsprung disease, pelvic versus transrectal repair in 30 patients and demonstrated
floor dyssynergia, and neuropathy which would include improvement of the outlet obstructive symptoms in over
spinal cord lesions and neurologic disease such as multiple 70% of patients, with less recurrence in the transvaginal
sclerosis. Mechanical causes include internal intussuscep- group 12 months after surgery (128).
tion, enterocele, sigmoidocele, rectocele, and rectal prolapse Surgical repair of rectal intussusception can be considered
(121). A comprehensive workup is critical in the assessment in patients with severe symptoms of obstructive defecation
of these patients so that an appropriate treatment strategy only after failing nonoperative treatments, since in almost
can be formulated. half of the cases, functional improvement can be achieved
In general, patients with slow-transit constipation and without the use of surgery (129). Several approaches have
pelvic floor dyssynergia should be treated with biofeedback been described, including different rectopexy techniques,
before colectomy, because total abdominal colectomy with stapled transanal resection, and the Delorme procedure.
ileorectal anastomosis in this group of patients is associated Although these procedures may resolve anatomic issues
with higher rates of recurrent constipation and lower rates like rectal ulcers and document improvement on repeat
of patient satisfaction (82). Biofeedback has been shown defecography, caution still needs to be applied, because they
to help patients with constipation and obstructed defeca- may not improve and can potentially worsen functional
tion. A prospective randomized trial investigating the effi- outcomes for patients (130,131).
cacy of biofeedback concluded that the use of biofeedback Ventral mesh rectopexy is a relatively new and promising
improved defecation indexes and decreased balloon expul- technique to correct rectal intussusception. A theoretical
sion time, resulting in an improvement in physiologic bowel advantage of this approach is that it avoids posterior lateral
function in patients with obstructed defecation (122). The rectal mobilization and thereby minimizes the risk of post-
timing of biofeedback therapy remains controversial with operative constipation. Several studies have shown improve-
some authors advocating its use in the preoperative period, ment in constipation in 80%–95% of patients with minimal
while others have reported efficacy in the postoperative new-onset constipation (132,133). Portier reported a series
period (123,124). of 40 patients who underwent ventral mesh rectopexy with-
If the slow-transit constipation is associated with rectal out sigmoid resection for rectal intussusception and found
intussusception or a nonemptying rectocele/enterocele on that 65% of patients on self-assessment responded as being
defecography, repair of the outlet obstruction is recom- cured, with another 33% showing improvement (133).
mended before or concomitant with total abdominal colec- Laparoscopic and robotic-assisted ventral mesh rectopexy
tomy with ileorectal anastomosis (109,125). is now being performed with acceptable recurrence rates,
Rectocele repair has been described using transvaginal, good functional results, and low mesh-related morbidity
transrectal, or transperineal approaches with and without (134,135).
mesh with good results in patients with symptoms of outlet Stapled transanal rectal resection, initially reported
obstruction. Indications for surgery include manual manip- by Longo (136), has been performed for the treatment of
ulation of the vaginal wall or rectum to evacuate, findings of obstructed defecation resulting from rectocele and intus-
an abnormal defecography with lack of emptying of contrast susception. The procedure utilizes a stapler for endorectal
Complications 375
resection to remove the redundancy of the distal rectum, a year and have failed medical therapy (148). It is currently
allowing for a less obstructed pathway to defecation. The lit- not approved by the U.S. Food and Drug Administration for
erature is difficult to interpret because of the diverse patient this condition in the United States.
population as well as a lack of prospective randomized The reported success rates range from 42% to greater
studies. In addition, there has been inconsistent assessment than 90%, and several prospective studies have shown
of long-term outcomes. Despite this, several studies have improvement in the Cleveland Clinic constipation score
attested to the successful results of this procedure in treat- (145,151,152). Thomas reviewed 13 studies demonstrat-
ing obstructed defecation with satisfaction rates ranging ing the use of SNS for the treatment of constipation. In
from 64% to 86% (137–139). In a review of 344 patients, 81% patients who proceeded to permanent SNS, up to 87%
of the patients were highly satisfied, and rectal urgency was showed improvement in symptoms based on quality of life
resolved in all patients at a median follow-up of 81 months and patient satisfaction scores at a medium follow-up of 28
(140). Concerns about long-term outcomes and serious post- months (146). Some studies do not report efficacy separat-
operative complications including pain, urgency, inconti- ing patients with slow-transit versus outlet obstruction, but
nence, constipation, rectal diverticulum, retroperitoneal Ratto reported that improvement was better in patients with
emphysema, bleeding, the development of rectovaginal obstructed defecation (147).
fistula, and long-term proctalgia associated with inflamma- Dinning completed a randomized, double-blind pla-
tion from retained staplers have deterred rapid acceptance cebo-controlled two-phase crossover study for SNS in slow-
of this procedure in the United States (141,142). Careful transit constipation. The short-term trial compared sham
patient selection, detailed informed consent outlining both stimulation with SNS with the primary outcome being days
benefits and potential risks, and surgeon experience in per- per week with the feeling of complete evacuation. There was
forming this procedure are vital to maximize the benefits no significant difference in the response rate between the
and minimize potential complications. SNS group (30%) versus the sham stimulation group (21%)
The optimal surgical treatment for slow-transit consti- (153). A longer follow-up at 2 years concluded that SNS was
pation associated with concomitant obstructive defeca- not effective treatment for slow-transit constipation. One
tion remains controversial. Based on the pathophysiologic reason for such a low success rate in this study was that all
changes of mixed constipation, a novel surgery entitled patients proceeded to permanent implantation regardless of
the Jinling procedure aims to solve the coexistence of their response to peripheral nerve evaluation (PNE) (154).
obstructed defecation and delayed colon transit in one oper- A recent Cochrane Review for SNS for constipation in
ation. This procedure adds a new side-to-side anastomosis adults concluded that SNS did not seem to improve symp-
to the low colorectal posterior anastomosis after subtotal toms in people with constipation, but there were only two
colectomy, which fixates the right colon mesentery with the randomized trials providing evidence, of which one trial
rectal stump acting as a rectopexy resulting in elevation of had only two participants (155).
the pelvic floor musculature and relieving the symptoms of While there is prospective case series data available sup-
obstructed defecation (143). In a review of 117 patients who porting the beneficial effects SNS for chronic constipation,
underwent this procedure, there was a reduction in validated the majority of reports were uncontrolled and had no com-
constipation scores that was maintained for over 4 years, as parison with any other treatment modalities. In addition,
well as improvement in the postoperative GI quality of life there was no consistent definition of constipation or uni-
with high patient satisfaction rates (110). In addition, Reshef form methods to measure response and improvement in
studied the outcomes for patients with pure slow-transit these studies. Additional studies are necessary to assess the
constipation (102) compared to those with slow transit and potential efficacy of this treatment option and provide evi-
features of obstructed defecation (41) who underwent total dence to determine selection criteria for its use.
abdominal colectomy with ileorectal anastomosis. They
concluded that total abdominal colectomy can be offered
to a highly select group of patients with a combined disor-
der, with the expectation of achieving equivalent long-term
COMPLICATIONS
results and high patient satisfaction rates (144).
Sacral nerve stimulation (SNS) has been used success- The success rate for total colectomy with an ileorectal anas-
fully for pelvic floor disorders including urinary incon- tomosis for slow-transit constipation has been shown to
tinence, overactive bladder, and fecal incontinence. The be around 90% overall; however, morbidity is high, which
mechanism of action is poorly understood, but it is thought can be attributed to those risks directly related to any colon
to alter pelvic afferent pathways and central mechanisms resection as well as long-term functional problems. Short-
(145). Neurostimulation has now emerged as a potential term complications that can occur after any bowel operation
treatment for chronic constipation from slow-transit or include bleeding, urinary tract infections, thromboem-
outlet dysfunction, primarily in institutions outside of the bolic events, adverse reaction to anesthesia, and surgical
United States (145–150). A recent European consensus state- site infections, which can be superficial or intraabdominal
ment was in agreement that SNS could be considered for secondary to an anastomotic leak, which has been reported
patients who have constipation symptoms for greater than up to 11% of patients (85). Postoperative ileus and small
bowel obstruction are the most frequent complications seen noninflammatory reasons, Reshef noted a significantly
after total abdominal colectomy for constipation. Long- greater overall morbidity in the colonic inertia group. There
term postoperative functional complaints include chronic were higher rates of postoperative abscess, more urinary
abdominal pain, bloating, diarrhea, and incontinence. tract infections, higher rates of wound infection, and higher
In a review of 144 patients who underwent total col- rates of postoperative ileus and increased length of stay. In
ectomy for slow-transit constipation, postoperative ileus addition, the 30-day readmission rate was also significantly
was the most common complication. And 10% of patients higher (159).
underwent reoperation in the immediate postoperative In an effort to improve outcomes specifically aimed at
period, including 9 for anastomotic leak, 5 for postoperative decreasing ileus rates and lowering the incidence and sur-
bleeding, and 1 for a missed enterotomy. Wound infection gical treatment of small bowel obstruction, the implemen-
was seen in 14% of patients, and urinary tract infection was tation of enhanced recovery after surgery protocols and
seen in 12% (144). minimally invasive colorectal techniques has demonstrated
Knowles reviewed 32 series looking at the outcome of a reduction in morbidity, decrease in overall length of hos-
colectomy for slow-transit constipation from 1991 to 1998. pital stay, earlier return of bowel function, as well as a low
The mortality rate varied from 0% to 6%. The most common readmission rate in colectomy patients (160–164). In addi-
postoperative complication was small bowel obstruction, tion, the placement of an antiadhesive agent at the time of
which generally occurs from adhesions created at the time surgery has been shown to reduce the incidence of adhesions
of colectomy, although others have reported small bowel that can cause small bowel obstruction (165). In comparing
pseudo-obstruction secondary to a neuropathic disorder of laparoscopic versus open procedures in colorectal surgery,
the myenteric plexus limiting bowel motility (156). In his the need for operative intervention for small bowel obstruc-
review, the median incidence was 18% with a reoperation tion has been shown to significantly lower after laparoscopic
rate of 14% (82). Others have reported an incidence of small operations as compared to open procedures (166). Utilizing
bowel obstruction in up to 25% of patients with more than these protocols and surgical approaches achieves the goals
half requiring surgical intervention for lysis of adhesions of improving the quality of patient care and satisfaction,
(83,85,157). Other significant complications include diar- while reducing health care–related cost.
rhea, which is generally a short-term complication resulting
from the absence of the ability of the colon to absorb water,
which has been reported in 14%–46% of patients. Medical
treatment including fiber supplements, antimotility agents,
RECURRENCE OF CONSTIPATION/
and binders, as well as time to allow for intestinal adaptation
OUTCOMES
may reduce bowel frequency and improve the consistency
of stool. Some degree of incontinence has been reported in Total abdominal colectomy with ileorectal anastomosis is
up to 50% of patients with a median 14% (79,82). Chronic an effective method of treatment for medically resistant
abdominal pain can be a frequent complaint in this group colonic inertia. After an appropriate preoperative workup
of patients, with an incidence in up to 41% of patients. The including colon transit and pelvic floor studies, successful
need for permanent ileostomy as a result of these complica- outcomes have been reported to be greater than 90%.
tions can be seen in up to 5% of patients (79,82,85). In 26 studies involving 1,047 patients with a mean follow-
FitzHarris noted that despite the statistically significant up of 44.8 months, the rate of bowel movements reported by
negative impact the long-term consequences of chronic the patients was 2.8 times per day, whereas recurrent con-
abdominal pain, diarrhea, and incontinence had on gas- stipation was reported in up to one-third of patients with a
trointestinal quality of life index scores, the vast majority mean of 9% (0%–33%) of 683 patients in 17 series (167).
of patients (over 90%) stated they would undergo subtotal It must be noted that definitive conclusions regarding the
colectomy again if given a second chance (79). effectiveness of surgery cannot be universally decided. The
The unrecognized presence of gastrointestinal dysmotil- methods used to assess the outcome of surgery and patient
ity may explain the high rate of ileus and postoperative small satisfaction vary greatly. Some studies use patient satisfac-
bowel obstruction. Glia looked at outcomes of colectomy for tion as a criterion for success, which can be very subjective
slow-transit constipation in relation to the presence of small and inaccurate. The ideal way to measure success should
bowel dysmotility evidenced by abnormal antral duodenal include standardized outcome measures such as question-
manometry. They found a trend toward better long-term naire-based protocols that assess quality of life, postop-
results after surgery for slow-transit constipation in patients erative complications, and functional outcome as it relates
with normal manometry before the operation, specifically specifically to bowel function. The process of distributing
in the incidence of constipation, diarrhea, chronic abdomi- and collecting questionnaires as well as patient compliance
nal pain, and abdominal bloating (158). rates remains an obstacle and presents a challenge in accu-
In addition, the indication for colectomy has been mulating outcome data when assessing the results of sur-
shown to influence postoperative morbidity. In comparing gery for constipation. However, the overall rate of success
total abdominal colectomy with ileorectal anastomosis in or satisfaction documented in 39 studies involving 1,423
patients undergoing surgery for colonic inertia versus other patients was 86% (39%–100%) (167).
References 377
Psychiatric history and mental well-being are also

important factors to consider in the selection of patients for rehabilitative treatment. Of all the patients who pres-
surgery and evaluating outcomes. Pluta et al. found supe- ent with chronic constipation, only a small percentage
rior results in patients with no psychiatric history compared will be refractory to medical management. Patients
to those with a psychiatric history, while others noted poor who do not respond to these forms of treatment may
outcomes with higher permanent stoma rates in patients be candidates for surgical intervention. The goals of
with psychologic conditions (168). In contrast, additional surgical management for constipation are to increase
series reported by Nylund and Reshef did not find psychi- the number of bowel movements and improve the
atric history to be a predictor of poor outcomes (166,169). quality of life for patients. A precise and thorough
In a more detailed study by Heyman et al. utilizing the evaluation of the morphology and function of the
Minnesota Multiphasic Personality Inventory in a series of bowel and pelvic floor is essential to accurately catego-
patients with severe constipation, it was noted that the inci- rize the classifications of chronic constipation, which
dences of hypochondriasis, depression, and hysteria were include slow-transit constipation, obstructed defeca-
significantly increased (170). Furthermore, O’Brien et al. tion, and combined disorders. This fundamental prin-
noted that a history of sexual abuse and not psychiatric his- ciple will guide clinicians in their decisions to select
tory was associated with persistent abdominal symptoms the appropriate therapy necessary to achieve a suc-
after total abdominal colectomy (171). cessful outcome, satisfy expectations associated with
Considerations for the recurrence or persistence of con- treatment, and minimize the recurrence of symptoms.
stipation after colectomy include the unrecognized presence Future developments in the field of neurostimula-
of pelvic floor dysfunction and the presence of a neuro- tion and understanding its effect on the properties of
pathic disorder that may extend proximally into the small nerves in the gastrointestinal tract may provide a less
bowel or even the entire gastrointestinal tract, resulting in a invasive treatment with better outcomes in the man-
global gastrointestinal motility disorder or the direct result agement of patients with chronic constipation.
of incomplete colonic resection (2).
Evaluation of these patients requires an initial assess-
ment for concomitant outlet obstruction with pelvic floor REFERENCES
physiology testing, since the presence of pelvic floor dys-
function has been shown to significantly decrease success- 1. Brandt LJ et al. Am J Gastroenterol. 2005;100(Suppl.
ful outcomes after surgery. If pelvic dyssynergy is identified, 1):S5–21.
then biofeedback therapy can be considered to improve 2. Papaconstantinou HT. Operative and nonopera-
symptoms and outcome (84). tive therapy for chronic constipation. In Whitlow CB
Radiographic imaging with contrast enemas and endo- et al. (eds). Improved Outcomes in Colon and Rectal
scopic visualization to identify anatomical abnormali- Surgery. London: Informa, 2010, pp. 361–374.
ties related to prior surgery and the remaining rectum is 3. Sandler RS et al. Am J Public Health. 1990;80:185–9.
essential to determine if recurrence of constipation is the 4. Ribas Y et al. Dis Colon Rectum. 2011;54:1560–9.
direct result of a dysfunctional or mobile cecal stump from 5. Knowles CH et al. Dis Colon Rectum. 2003;46:
a cecorectal anastomosis or a dysmotile portion of colon as a 1716–17.
result of a segmental resection or ileosigmoid anastomosis. 6. Wexner SD et al. Dis Colon Rectum. 1991;34:851–6.
All of these technical variants have been shown to be asso- 7. Tobon F et al. N Engl J Med. 1968;278:188–93.
ciated with a higher incidence of constipation recurrence, 8. Rao SS et al. Am J Gastroenterol. 2005;100:1605–15.
and reoperation with completion colectomy and conversion 9. Ternent CA et al. Dis Colon Rectum. 2007;50:
to ileorectal anastomosis may be necessary to correct these 2013–22.
types of surgical failure (97,98,100). 10. Beck DE. Ochsner J. 2008;8:25–31.
If the above testing is inconclusive and does not provide 11. Hinton JM et al. Gut. 1969;10:842–7.
sufficient evidence to proceed with either biofeedback or 12. Nam YS et al. Dis Colon Rectum. 2001;44:86–92.
reoperative surgery, then further testing to assess whole gas- 13. Videlock EJ et al. Neurogastroenterol Motil. 2013;25:
trointestinal transit should be performed to identify patients 509–20.
with a neuropathic disorder causing pseudo-obstruction 14. Metcalf AM et al. Gastroenterology. 1987;92:40–7.
who may benefit from ileostomy. 15. Cowlam S et al. Colorectal Dis. 2008;10:818–22.
16. Felt-Bersma RJ et al. Dis Colon Rectum. 1990;33:
277–84.
SUMMARY 17. Karlbom U et al. Int J Colorectal Dis. 1998;13:141–7.
18. Jones PN et al. Dis Colon Rectum. 1987;30:667–70.
Chronic constipation is a common disease that is mul- 19. Bremmer S et al. Dis Colon Rectum. 1995;38:969–73.
tifactorial in origin. In most cases, it can be success- 20. Murad-Regadas SM et al. Surg Endosc. 2008;22:974–9.
fully managed by diet modification, medication, or 21. Murad-Regadas SM et al. Dis Colon Rectum. 2014;
57:228–36.
22. Vitton V et al. Dis Colon Rectum. 2011;54:1398–404. 58. Gras-Miralles B, Cremonini F. Clin Interv Aging. 2013;
23. Bozkurt MA et al. Ulus Cerrahi Derg. 2014;30:183–5. 8:191–200.
24. Wieczorek AP et al. World J Urol. 2011;29:615–23. 59. Lembo AJ et al. N Engl J Med. 2011;365:527–36.
25. Varma MG et al. Dis Colon Rectum. 2008;51:162–72. 60. Gonenne J et al. Clin Gastroenterol Hepatol. 2005;
26. Jones MP et al. Dig Dis Sci. 2002;47:2222–30. 3:784–91.
27. Ramkumar D, Rao SS. Am J Gastroenterol. 2005;100: 61. Webster L et al. Pain. 2008;137:428–40.
936–71. 62. Camilleri M. Neurogastroenterol Motil. 2005;17:
28. Siegel JD, Di Palma JA. Clin Colon Rectal Surg. 2005; 157–65.
18:76–80. 63. Acosta A, Camilleri M. Therap Adv Gastroenterol.
29. Meshkinpour H et al. Dig Dis Sci. 1998;43:2379–83. 2014;7:167–75.
30. Young RJ et al. Gastroenterol Nurs. 1998;21:156–61. 64. Wong BS, Camilleri M. Expert Opn Investig Drugs.
31. Whitehead WE et al. J Am Geriatr Soc. 1989;37: 2013;22:277–84.
423–9. 65. Woodward S. Br J Nurs. 2012;21:982, 984–6.
32. Anti M et al. Hepatogastroenterology. 1998;45: 66. Shailubhai K et al. Dig Dis Sci. 2013;58:2580–6.
727–32. 67. Rao SS et al. Am J Gastroenterol. 1998;93:1042–50.
33. Ashraf W et al. Aliment Pharmacol Ther. 1995;9: 68. Rao SS et al. Clin Gastroenterol Hepatol. 2007;5:
639–47. 331–8.
34. Rodrigues-Fisher L et al. Clin Nurs Res. 1993;2: 69. Heymen S et al. Dis Colon Rectum. 2003;46:1208–17.
464–77. 70. Enck P. Dig Dis Sci. 1993;38:1953–60.
35. Voderholzer WA et al. Am J Gastroenterol. 1997;92: 71. Rao SS et al. Dig Dis Sci. 1997;15(Suppl. 1):78–92.
95–8. 72. Chiarioni G et al. Gastroenterology. 2006;130:
36. Sturtzel B et al. J Nutr Health Aging. 2009;13:136–9. 657–64.
37. Burkitt DP et al. Lancet. 1972;2:1408–12. 73. Gilliland R et al. Br J Surg. 1997;84:1123–6.
38. Schiller LR. Aliment Pharmacol Ther. 2001;15:749–63. 74. Ahadi T et al. J Res Med Sci. 2014;19:950–5.
39. Madoff RD, Fleshman JW. Gastroenterology. 2003; 75. Brown SR et al. Dis Colon Rectum. 2001;44:737–9.
124:235–45. 76. Chiarioni G et al. Gastroenterology. 2005;129:86–97.
40. Lembo A, Camilleri M. N Engl J Med. 2003;349: 77. Maria G et al. Dis Colon Rectum. 2000;43:376–80.
1360–8. 78. Lahr SJ. Am Surg. 1999;65:1117–23.
41. DiPalma JA. Rev Gastroenterol Disord. 2004; 79. FitzHarris GP et al. Dis Colon Rectm. 2003;46:
4(Suppl. 2):S34–42. 433–40.
42. Bass P, Dennis S. J Clin Gastroenterol. 1981;3(Suppl.): 80. Lane WA. Br Med J. 1908;1:126–30.
23–8. 81. Wexner SD et al. Dis Colon Rectum. 1991;43:851–6.
43. Lederle FA et al. Am J Med. 1990;89:597–601. 82. Knowles CH et al. Ann Surg. 1999;230:627–38.
44. Schiller LR et al. Gastroenterology. 1988;94:933–41. 83. Webster C, Dayton M. Am J Surg. 2001;182:639–44.
45. Toledo TK, Di Palma JA. Aliment Pharmacol Ther. 84. Redmond JM et al. Am J Gastroenterol. 1995;90:
2001;15:605–11. 748–53.
46. Lee-Robichaud H et al. Cochrane Database Syst Rev. 85. Pikarsky A et al. Dis Colon Rectum. 2001;44:1898–9.
2010;(7):CD007570. 86. Ripetti V et al. Surgery. 2006;140:435–40.
47. Di Palma JA et al. Am J Gastroenterol. 2002;97: 87. Pinedo G et al. Surg Endosc. 2009;23:62–5.
1776–9. 88. Sheng QS et al. J Dig Dis. 2014;15:419–24.
48. DiPalma JA et al. Am J Gastroenterol. 2000;95: 89. Sohn G et al. J Korean Soc Coloproctol. 2011;27(4):
446–50. 180–7.
49. Corazziari E et al. Gut. 2000;46:522–6. 90. Fan CW, Wang JY. Int Surg. 2000;85:309–12.
50. Kamm MA et al. Clin Gastroenterol Hepatol. 2011; 91. Hasson I et al. J Gastroinest Surg. 2006;10:1330–7.
9:577–83. 92. Riss S et al. Colorectal Dis. 2009;11(3):302–7.
51. Passmore AP et al. Pharmacology. 1993;47(Suppl. 1): 93. Hsiao KC et al. Int J Colorectal Dis. 2008;23:419–24.
249–52. 94. Athanasakis H et al. Surg Endosc. 2001;15(10):
52. Oster JR et al. Am J Gastroenterol. 1980;74:451–8. 1090–2.
53. Urso FP et al. Radiology. 1975;116:557–9. 95. Xu LS, Liu WS. Am Surgeon. 2012;78:495–6.
54. Cupoletti J et al. Am J Physiol Cell Physiol. 2004;297: 96. Sarli L et al. Dis Colon Rectum. 2001;44(10):1514–9.
C1173–83. 97. Marchesi F et al. World J Surg. 2007;31:1658–64.
55. Johanson JF, Ueno R. Aliment Pharmacol Ther. 2007; 98. Fasth S et al. Acta Chir Scand. 1983;149:623–7.
25:1351–61. 99. Wei D et al. BMC Gastroenterol. 2015;15:30.
56. Camilleri M et al. Am J Physiol Gastrointes Liver 100. Pemberton JH et al. Ann Surg. 1991;214:403–11.
Physiol. 2006;290:G942–7. 101. Feng Y, Jianjiang L. Am J Surg. 2008;195:73–7.
57. Johanson JF et al. Am J Gastroenterol. 2008;103: 102. Kamm MA et al. Int J Colorectal Dis. 1991;6(1):49–51.
170–7. 103. De Graff EJ et al. Br J Srg. 1996;83(5):648–51.
References 379
104. Ehrenpreis ED et al. Gastroenterology. 1997;110A: 139. Schwandner O, Fürst A, German StaRR Registry
728. study Group. Langenbecks Arch Surg. 2010;395:
105. You YT et al. Am Surg. 1998;64(8):775–7. 505–13.
106. Lundin E et al. Br J Surg. 2002;89(10):1270–4. 140. Goede AC et al. Colorectal Dis. 2011;13:1052–7.
107. Hosie KB et al. Br J Surg. 1990;77:801–2. 141. De Nardi P et al. Tech Coloproctol. 2007;11:353–6.
108. Keighley MR et al. Gut. 1993;34(5):680–4. 142. Pescatori M et al. Int J Colorectal Dis. 2005;20:83–5.
109. Paquete IM et al. Dis Colon Rectum. 2016;59: 143. Ding W et al. Dis Colon Rectum. 2015;58:91–6.
479–92. 144. Reshef A et al. Int J Colorectal Dis. 2013;28:841–7.
110. Scarpa M et al. Colorectal Dis. 2005;7(3):224–7. 145. Carrington EV et al. Neurogastroenterol Motil. 2014;
111. Malone PS et al. Lancet. 1990;336(8725):1217–8. 26:1222–37.
112. Lees NP et al. Colorectal Dis. 2004;6:362–8. 146. Thomas GP et al. Br J Surg. 2013;100:174–81.
113. Hirst GR et al. Tech Coloproctol. 2005;9:217–21. 147. Ratto C et al. Colorectal Dis. 2015;17:320–8.
114. Poirier M et al. Dis Colon Rectum. 2007;50:22–8. 148. Maeda Y et al. Colorectal Dis. 2015;17:o74–87.
115. Buntzen S, Laurberg S. Dis Colon Rectum. 2008;51: 149. Graf W et al. Neurogastroenterol Motil. May 2015;
1523–8. 27(5):734–9.
116. Meurette G et al. Gastroenterol Clin Biol. 2010;34(3): 150. Ortiz H et al. Dis Colon Rectum. 2012;55:876–80.
209–12. 151. Pinto RA, Sands DR. Gastrointest Endoscopy Clin N
117. Biyani D et al. Colorectal Dis. 2007;9:373–6. Am. 2009;19:83–116.
118. Patton V, Lubowski DZ. Dis Colon Rectum. 2015;58: 152. Kamm ma et al. Gut. 2010;59:333–40.
457–65. 153. Dinning PG et al. Am J Gastroenterol. 2015;110:
119. Worsøe J et al. Dis Colon Rectum. 2008;51:1523–8. 733–40.
120. Raag J et al. Colorectal Dis 2011;13:1299–302. 154. Patton V et al. Dis Colon Rectum. 2016;59:878–85.
121. Hedrick TL, Friel CM. Gastroenterol Clin N Am. 2013; 155. Thaha MA et al. Sacral Nerve Stimulation for Fecal
42:863–76. Incontinence and Constipation in Adults (Review).
122. Talley NJ et al. Am J Gastroenterol. 1996;91:19–25. Cochrane Database of Systematic Reviews 2015,
123. Nyam DC et al. Dis Colon Rectum. 1997;40(3):273–9. Issue 8. Art. No.: CD004464. DOI: 10.1002/14651858.
124. Bernini A et al. Dis Colon Rectum. 1998;41(11): CD004464.pub3.
1363–6. 156. Krishnamurthy S et al. Gastroenterology. 1985;88:
125. Zenilman ME et al. Arch Surg. 1989;124:947–51. 26–34.
126. Mellgren A et al. Dis Colon Rectum. 1995;38:7–13. 157. Zutshi M et al. Int J Colorectal Dis. 2007;22:265–9.
127. Arnold MW et al. Dis Colon Rectum. 1990;33:684–7. 158. Glia A et al. Dis Colon Rectum. 2004;47:96–102.
128. Nieminen K et al. Dis Colon Rectum. 2004;47: 159. Reshef A et al. Colorectal Dis. 2012;15:481–6.
1636–42. 160. Senagore AJ et al. Dis Colon Rectum. 2009;52(2):
129. Murad-Regadas SM et al. Arq Gastroenterol. 2012;49: 183–6.
135–42. 161. Delaney CP et al. Ann Surg. 2008;247(5):819–24.
130. Van Tets WF, Kuijpers JH. Dis Colon Rectum. 1995;38: 162. Bona S et al. World J Gastroenterol. 2014;20(46):
1080–3. 17578–87.
131. Christiansen J et al. Dis Colon Rectum. 1992;35: 163. Feldman LS, Delaney CP. Surg Endosc. 2014;28(5):
1026–8. 1403–6.
132. Slawik S et al. Colorectal Dis. 2008;10:138–43. 164. Sample C et al. J Gastrointest Surg. 2005;9:803–8.
133. Portier G et al. Colorectal Dis. 2011;13:914–7. 165. Becker JM et al. J Am Coll Surg. 1996;183:297–306.
134. Owais AE et al. Colorectal Dis. 2014;16:995–1000. 166. Reshef A et al. Surg Endosc. 2013;27:1717–20.
135. Van Iersel JJ et al. World J Gastroenterol. 2016; 167. Bove A et al. World J Gastroenterol. 2012;18(36):
22(21):4977–87. 4994–5013.
136. Longo A. Annual Cleveland Clinic Florida Colorectal 168. Pluta H et al. Dis Colon Rectum. 1996;39(2):160–6.
Disease Symposium. 2004. 169. Nylund G et al. Colorectal Dis. 2001;3(4):253–8.
137. Reboa G et al. Dis Colon Rectum. 2009;52: 170. Heyman S et al. Dis Colon Rectum. 1993;36:593–6.
1598–604. 171. O’Brien S et al. Dis Colon Rectum. 2009;52(11):
138. Stuto A et al. Surg Innov. 2011;18:248–53. 1844–7.
40
Colorectal trauma
ALISON ALTHANS AND SCOTT R. STEELE
CHALLENGING CASE look. His bowel appeared healthy and in-continuity

restoration was performed for both the small and large
A 19-year-old male was involved in a blast injury with bowel. Following a negative air-leak test on the colon,
multiple shrapnel wounds to the abdomen. He was he was closed, extubated, and recovered uneventfully.
initially alert and talking on presentation and slightly
tachycardic (HR 118) and mildly hypotensive (systolic
blood pressure 95 mm HG), but he responded to INTRODUCTION
fluid resuscitation. Chest x-ray demonstrated a small
left-sided pneumothorax and abdominal plain film
Since the experience of military surgeons in managing colon
showed multiple metallic fragments. Physical exami-
and rectal traumas has been crucial in shaping the way civil-
nation was significant for peritonitis on abdominal
ian colorectal injuries are now managed, it is important to
examination. After he was given cefazolin and metro-
consider these traumas in the context of wartime history.
nidazole, he was taken to the operating room. He was
During the American Civil War, the risk of infection and
found to have multiple small bowel, descending and
sepsis secondary to any surgical intervention for abdominal
sigmoid colon through-and-through injuries (Figure
trauma made expectant management the mainstay (1). At
40.1) and a moderate amount of intraabdominal con-
the beginning of World War I, most abdominal wounds were
tamination of stool.
managed nonoperatively. However, as the war continued on,
Cuthbert Wallace and other defiant British surgeons insisted
on laparotomy, arguing that hemorrhage was the chief cause
CASE MANAGEMENT of early mortalities (2,3). With the implementation of expe-
dited evacuation of the injured from the battlefield and sub-
After initially packing the abdomen and continuing sequent operation, mortality from abdominal wounds began
resuscitation, hemostasis was achieved on two small to fall (4). Throughout World War II and the first half of the
bowel mesenteric bleeding vessels. Further stool twentieth century, the procedure of choice was largely exte-
contamination was controlled by stapling off and riorization of the affected bowel—taking the portion of the
resecting the bowel with significant injury and loss of bowel wall that was injured and bringing it to the skin as a
wall integrity and oversewing smaller areas. The colon stoma. The most important work shaping these guidelines
injuries were resected due to the >50% loss of colon was Ogilvie’s “Abdominal Wounds in the Western Desert,”
wall. After washing the abdomen out with several liters in which he advocated for exteriorization or proximal colos-
of warm normal saline, the patient remained mildly tomy—even though his data did not clearly support this
hypotensive despite appropriate aggressive resuscita- recommendation (5,6). Despite the fact that mortality rates
tion. The bowel was left in discontinuity, the abdomen were similar between primary repair and diversion/exte-
was packed and left open utilizing a sterile negative riorization groups in Ogilvie’s series, the U.S. Office of the
therapy wound device, and the patient was taken to Surgeon General mandated against primary repair (6,7).
the intensive care unit for continued resuscitation. The Management of rectal injuries also improved during this
following day after his hemodynamics and laboratory time as presacral drainage and distal rectal irrigation were
values were normalized and he was normothermic, he implemented (8). Along the way, mortality rates following
was taken back to the operating room for a second colon and rectal traumas had dropped from the nearly 90%
initially seen in the Civil War era to less than 40% (9).
380
Preoperative assessment 381
have an incidence of somewhere between 1% and 3% in

civilian trauma centers, and 5%–10% of those injured in
combat (17). In both cohorts, penetrating injuries are most
common (18,19). However, in war injuries, there is much
greater concern for high-energy and blast mechanisms.
These are far more important to consider, as they are often
associated with concomitant damage to adjacent organs. In
addition, the thermal injury that occurs with this type of
mechanism often leads to a delay in the injury presentation,
as it may evolve over time. What initially appears to be a
normal appearing section of bowel adjacent to the wound
may progress to more severe injury that warrants resec-
tion. Failure to recognize this injury pattern may contrib-
ute to subsequent complications. A recent series describing
the epidemiology of colorectal trauma in Operation Iraqi
Freedom and Operation Enduring Freedom found that 35%
Figure 40.1 Shrapnel injury to the small bowel with a hex
of colorectal injuries among combatants could be attributed
nut. Similar injuries were present to the small and large
bowel. to explosions or blasts, consistent with the widespread use
of improvised explosive devices (16). While this mechanism
of injury may also be seen in the civilian setting, especially
The transition to surgical management was also accom-
in light of the more recent world events, the vast majority
panied by simultaneous advances in antibiotic therapy, fluid
in the civilian sector are due to gunshot and stab wounds
resuscitation, blood banking, surgical techniques, and equip-
(15). Blunt trauma usually occurs in the context of motor
ment. Continued experience with diversion and exterioriza-
vehicle crashes, and incidences are as infrequent as 0.02%
tion procedures used during the Korean and Vietnam Wars,
in the civilian setting and 5% in war (20,21). Rectal injuries
in alignment with the previously established doctrine against
may also include accidental or intentional impalement, iat-
primary anastomosis, led to improved outcomes and dropped
rogenic injury, secondary damage from bony fragments in
the mortality rate even further to ∼10% (10). Although there
a pelvic fracture, and insertion of foreign bodies, or with
was some experimentation with primary repair in some
extensive perineal trauma (22) (Figure 40.2).
right-sided colon injuries during this time (11), colostomy
remained standard practice until the 1970s. Civilian prac-
tices largely paralleled those of the combat setting, until the
late 1970s at which point surgeons questioned the differences PREOPERATIVE ASSESSMENT
in militant and civilian traumas, the possible compounding
factors that may have contributed to the success of diver-
As with all trauma patients, those with colorectal trauma are
sion in military trauma, and the potential benefit of primary
first evaluated for airway, breathing, and circulation issues,
anastomosis in carefully selected and optimized cases (12,13).
and assessed according to the Advanced Trauma Life Support
These questions were addressed by Stone and Fabian’s ran-
primary and secondary surveys. Fluids and blood products
domized study that demonstrated better outcomes for repair
are administered as needed. Additionally, patients with
rather than diversion, although they did exclude patients
with extensive blood loss, organ damage, fecal contamina-
tion, or delay to operation (13). This study and the subsequent
prospective randomized studies that supported these data
led to increasing support for a single-staged procedure with
primary anastomosis that continues today (14,15). Especially
in the context of the recent military conflicts in Iraq and
Afghanistan, however, management of colorectal trauma is
still heavily debated and warrants continued appraisal and
application of case-by-case decision-making (16). In this
chapter, we focus on the management and controversies of
the care of the patient with colorectal trauma.
ETIOLOGY
The incidence, mechanism, and type of colorectal injury

vary between civilian and combat settings. These traumas Figure 40.2 Perineal injury.
382 Colorectal trauma
life-threatening injuries (especially those with features of the steps in diagnosis, most colorectal injuries are identified
“lethal triad” of acidosis, hypothermia, and coagulopathy) intraoperatively or by other diagnostic techniques (27).
may be considered for urgent damage control laparotomy
(DCL). Under this protocol, the patient undergoes a staged DIAGNOSTIC PERITONEAL LAVAGE
process of abbreviated laparotomy with control of hemostasis
and fecal soilage, temporary closure, aggressive resuscitation Diagnostic peritoneal lavage (DPL) was first introduced by
in the intensive care unit, and a later operation for defini- Root and colleagues in 1965 and is particularly applicable
tive repair or diversion after the patient has been stabilized for the evaluation of hemodynamically unstable patients
(23). In the context of colorectal surgery, this approach has that are suspected to have blunt intraperitoneal colorectal
been particularly useful in patients with presacral bleeding trauma. While not performed as commonly as in the past,
and colorectal perforations (24), even in the setting of elective it still is a useful adjunct to obtain urgent information and
surgery. Patients not necessitating urgent laparotomy may stratify patients. Before proceeding with DPL, an attempt
undergo any of several diagnostic methods described in the is made to aspirate free intraperitoneal fluid (the diagnostic
following sections and summarized in Table 40.1. peritoneal aspirate). If 10 mL or more of gross blood is aspi-
rated, no further steps are taken because an injury is likely
and exploration is often warranted. If not, the peritoneal
cavity is lavaged with 1 L of either normal saline or lactated
DIAGNOSIS OF COLON INJURIES Ringer solution, and then the effluent fluid is evaluated. The
lavage is considered positive if more than 100,000 red blood
PHYSICAL EXAMINATION cells/mm3 or 500 white blood cells/mm3, bile, amylase, bac-
teria, or particulate matter are present on microscopic anal-
Depending on the severity and context of the injury at hand, ysis. These patients necessitate laparotomy.
conducting a physical examination may be impractical or DPL is more sensitive than physical examination and is
have limited utility in establishing a definitive diagnosis, also rapid and inexpensive (27,28). The extremely high sen-
especially in those with blunt trauma (25). Examination sitivity, however, comes at the cost of false-positive results
is especially difficult in patients who are intoxicated, have that may occur in the context of solid organ lacerations, ret-
neurological impairments, or have multisystem trauma. roperitoneal hematoma, or iatrogenic injury that results in
Potential pertinent positives may include evidence of pen- nontherapeutic laparotomy (29). Additionally, because DPL
etrating wounds, peritoneal signs associated with hollow is an invasive procedure, it may result in complications,
viscous perforation, “seat belt sign” ecchymosis in blunt such as intraabdominal organ or neurovascular damage
injuries, signs of chance fracture, and gross blood on digi- (30). DPL is often contraindicated when the patient is preg-
tal rectal exam (DRE) (7,26). While physical examination nant. Care should be taken in the morbidly obese patient
and thorough history when possible remain important first and in those with a prior laparotomy.
Table 40.1 Comparison of diagnostic techniques for colorectal injury
Accuracy Sensitivity Specificity Advantages Disadvantages

DPL 92%–98% 87%–100% 84%–98% Inexpensive Invasive
Rapid Misses retroperitoneal
Mobile injury
Very sensitive Relatively low specificity
No radiation
Abdominal CT 82%–99% 92%–98% 97%–99% Noninvasive Expensive
Very sensitive and specific Timely
Identification of specific injury Radiation exposure
Identification of concurrent Requires transport
extraabdominal injury
Not operator dependent
FAST 96%–98% 73%–88% 98%–100% Inexpensive Interobserver variability
Rapid Can be hampered by
Mobile obstruction of view (e.g.,
Noninvasive obesity, fractures,
Highly specific intraabdominal air)
Allows for concurrent resuscitation Relatively low sensitivity
Can be repeated
No radiation
Diagnosis of colon injuries / Diagnostic laparoscopy 383
FOCUSED ABDOMINAL SONOGRAPHY

FOR TRAUMA
As an alternative to DPL, a focused assessment with sonog-
raphy for trauma (FAST) may be utilized for initial evalua-
tion of a hemodynamically unstable blunt trauma patient.
This technique may also be used for stable patients with both
blunt and penetrating injuries, and involves ultrasound
views of Morison’s pouch (hepatorenal recess), the pouch of
Douglas (retrovesical intraperitoneal cavity), the splenore-
nal space, and the pericardium. If an extended examination
(E-FAST) is performed, bilateral hemithoracic and upper
anterior chest wall views are also obtained. The ultrasound
is positive if hemoperitoneum is observed. Similar to DPL,
a positive FAST often requires laparotomy in the unstable
patient. In the stable patient, the free fluid can be followed
with other diagnostic imaging such as a computed tomog-
raphy (CT) scan to look for solid organ injury that may not
require a laparotomy.
FAST is easily accessible, can be performed while Figure 40.3 CT scan demonstrating bullet tract through
resuscitation is ongoing, is noninvasive, and is specific the retroperitoneum (white arrows).
(27,31,32). It is safe for pregnant patients and children
and may be repeated in serial investigations to potentially
avoid DPL, CT, and/or laparotomy. However, because DIAGNOSTIC LAPAROSCOPY
a positive exam relies on the presence of intraperitoneal
fluid (for most operators, about 200 mL), sensitivity is Laparoscopy is unique in that it may be used in both the
decreased, and a CT scan should be obtained before a diagnosis and minimally invasive treatment of colorectal
course of nonoperative management is begun (33,34). traumas (Figure 40.4). Laparoscopic exploration follow-
Finally, care should be taken for the patient with retroperi- ing penetrating abdominal wounds is especially useful in
toneal fluid that may not be picked up as readily on FAST detecting injuries missed by standard imaging (especially
examination. diaphragmatic), and also in minimizing rates of nonthera-
peutic laparotomy (39). A prospective study conducted by
Lin et al. in 2010 found that the implementation of diagnos-
COMPUTERIZED TOMOGRAPHY
tic laparoscopy decreased the nontherapeutic laparotomy
CT is the imaging modality of choice for colorectal injury, rate from 58% to 0% in hemodynamically stable patients
typically used in the evaluation of hemodynamically stable with abdominal stab wounds, with an accuracy of 100% (40).
patients who either had a positive FAST examination or Similarly, a 2014 retrospective study by Lee et al. showed
went straight to CT. Views of the head, chest, abdomen, and that 16% of their blunt abdominal trauma patients avoided
pelvis with IV contrast are typically obtained in order to unnecessary laparotomy when initially evaluated with lapa-
scan for multiple injuries. roscopy (41). In select cases, surgeons who are experienced
The high levels of sensitivity, specificity, and accu- in advanced laparoscopic techniques may take the oppor-
racy in abdominal CT for blunt abdominal trauma have tunity to address the injuries identified on diagnostic lapa-
been well established in the literature (35–37). These roscopy in a minimally invasive manner. Laparoscopy does
values are somewhat decreased in the context of mes-
enteric injury and potentially in hollow viscous, where
DPL may be more appropriate (27). Though CT requires
more time and patient transport than FAST and DPL, it
has the unique ability to detect occult injury and identify
the specific lesion. Improvements in radiologic technol-
ogy over time have decreased scan time and allowed for
detection of more subtle signs of bowel injury. Findings
that may indicate injury include thickening of the bowel
wall, tearing of the bowel wall, abdominal free fluid or air,
and mesenteric fat stranding (38). In addition, due to the
availability of rapid thin-slice CT imaging, trajectory of
projectile fragments or bullets can often be identified on Figure 40.4 Rectal injury identified on laparoscopy
CT (Figure 40.3). (white arrow).
come with the small risk for inducing tension pneumotho- Table 40.2 AAST colon and rectal injury scales
rax, hypotension, or gas embolism (42). The role of laparos-
Grade Injury description
copy in the diagnosis and treatment of colorectal trauma
will likely evolve as the technology progresses and more Colon
surgeons gain expertise in the involved techniques. I (a) Contusion or hematoma without
devascularization
Diagnosis of rectal injuries (b) Partial-thickness laceration
II Laceration ≤50% of circumference
Diagnosis of rectal injuries is particularly complicated by III Laceration >50% of circumference
the fact that they may occur in the context of trauma to the
IV Transection of the colon
pelvic bone, which may need to be concomitantly addressed.
V Transection of the colon with segmental
DRE has classically been used in secondary survey to assess
tissue loss
for sphincter tone, bleeding, or palpable injury to the rectal
wall, but its use has become somewhat controversial as it has Rectum
varying rates of sensitivity in detecting rectal injury (43,44). I (a) Contusion or hematoma without
More accurately, the finding of gross blood, bone segments, devascularization
or a palpable defect should indicate the need for other tests to (b) Partial-thickness laceration
definitively diagnose the injury. A negative DRE in the set- II Laceration ≤50% of circumference
ting of other information that is potentially compatible with III Laceration >50% of circumference
a rectal injury warrants further investigation. Comparing
IV Full-thickness laceration with extension
DRE to rigid sigmoidoscopy, sigmoidoscopy has been dem-
into the perineum
onstrated to be more specific, especially in the context of
V Devascularized segment
extraperitoneal rectal lesions (45). Proctosigmoidoscopy,
however, may have limited utility without a bowel prepara-
tion. Because there is little downside to DRE, the authors choice for the majority of colon injuries. This was set into
would still recommend DRE when possible for identifica- motion by the aforementioned Stone and Fabian’s clas-
tion and treatment of injuries that may not be detected on sic randomized trial that resulted in equivalent mortality
imaging. Finally, because up to one-third of rectal injuries between the primary repair and colostomy groups (1.5%
are accompanied by a genitourinary injury, a CT scan with versus 1.4%, p > 0.05), as well as increased morbidity and
both rectal and bladder contrast is often recommended in length of stay for patients in the colostomy group (13).
hemodynamically stable patients (21,46). Several prospective series since this study have demon-
strated the favorable use of primary repair, resulting in safe
Injury scales and successful treatment of ∼73%–85% of civilian injuries
and a more uncertain range in the military setting (50–54).
As a means of standardizing the assessment of traumatic These studies have proven primary repair to have equiva-
colorectal injury, several injury scales have been developed. lent if not superior outcomes to diversion in most clinical
The first was established by Flint and associates in 1981 and scenarios.
was used to classify which levels of injury could safely undergo Stone and Fabian’s study, however, required manda-
primary repair versus colostomy (47). The Penetrating tory colostomy for patients with extensive blood loss,
Abdominal Trauma Index (PATI) was developed the same organ damage, fecal contamination, or delay to operation
year by Moore et al., and takes into account all injured in approximately half their cohort—thus eliminating the
intraabdominal organs and the severity of each injury (48). sickest of patients and preventing assessment of how these
Most recently, the American Association for the Surgery of patients fare with primary repair. This issue was addressed
Trauma (AAST) developed a similar scale for all abdominal by Chappuis et al., who only excluded patients who had
organs with the goal of further standardizing assessment for a rectal injury rather than a colonic one. They found that
unifying research and management efforts (49). The AAST complication rates were similar between the two groups,
scales for colon and rectal injury can be seen in Table 40.2. and concluded that primary repair could safely be per-
formed in all civilian penetrating colon traumas; however,
they did not report any statistics on their sample size of only
56 patients (55).
MANAGEMENT OF COLON INJURY Since Chappuis and colleagues’ study, there have been
three major prospective randomized trials comparing
THE CONTROVERSY OF PRIMARY REPAIR primary repair and diversion while only excluding rectal
injuries (56–58). Gonzalez et al. also published the results
We have come a long way since the time of Ogilvie when of the first 109 patients in their study 4 years prior but con-
exteriorization was standard and primary repair was man- tinued the study in an attempt to assess outcomes of repair
dated against, and primary repair is now the procedure of of destructive injuries (57,59). The results of the subsequent
Management of colon injury / The role of resection with anastomosis or diversion 385
Table 40.3 Prospective randomized trials comparing primary repair and diversion for colon injury without exclusion based
on risk factors
Number of Morbidity of Morbidity of

Author (year) patients primary repair, % diversion, % Other conclusions Study weaknesses
Sasaki (1995) (56) 71 19 36 Broke primary repair group Small study
into resection and sample; only 28
nonresection groups and patients diverted
found no differences
Gonzalez (2000) (57) 176 18 21 Patients with PATI score >25 Only 9% of injuries
have similar complication were destructive
rates regardless of
treatment modality
Kamwendo (2002) (58) 240 26 18 Delay from time of Majority of injuries
penetrating colon injury is (80%) were due
not a contraindication to to low-energy
primary repair; primary gunshot
repair is associated with
shorter operation time
and length of stay
study published in 2000 and the other two major random- as a “primary repair,” without distinguishing a direct anas-
ized trials are summarized and compared in Table 40.3. tomosis from one requiring resection before restoring con-
Each of these studies showed similar complication rates tinuity. The Sasaki et al. study did conduct a subanalysis
between the two groups, with Sasaki and Gonzalez dem- where the primary repair group was separated by resection
onstrating lower rates in the primary repair group, and versus no resection, and found no differences between the
Kamwendo showing a statistically insignificant (p = 0.21) two groups (56). However, there were only 29 patients in the
higher rate in the primary repair group. All three studies resection group and 24 patients in the nonresection group.
concluded in support of primary repair as the standard of As suggested by the development of injury scales, stratify-
care for penetrating colon injury. ing colon injuries by the extent of damage may be helpful
Despite these data, uncertainty regarding the safety of a in determining whether primary repair with no resection,
single-staged operation with primary anastomosis persisted. resection with primary anastomosis, or resection with
In 2003, a Cochrane meta-analysis of the previously men- diversion is the most suitable management for the patient at
tioned randomized studies (plus a 1992 study by Falcone hand (see Colon Algorithm, Figure 40.5).
et al. that had stringent exclusion criteria similar to Stone and First, it is helpful to distinguish whether the colon injury
Fabian) was compiled to conduct a review of 361 patients in the is nondestructive or destructive. A nondestructive injury is
primary repair group and 344 in the diversion group (60,61). one that would be classified as grades I–III on the AAST
This analysis found no difference in mortality, total infec- colon injury scale (63). Based on the findings of the random-
tious complications (odds ratio [OR] = 0.44; 95% confidence ized trials and reviews of these studies conducted through-
interval [CI], 0.17–1.1), abdominal infections including dehis- out the years, ample evidence supports the safety and
cence (OR = 0.67; 95% CI, 0.35–1.3), or abdominal infections efficacy of primary repair without resection or diversion for
excluding dehiscence (OR = 0.69; 95% CI, 0.34–0.9) between nondestructive trauma to the colon. This was further sup-
primary repair versus diversion (61). Total complications ported by a 2003 Maxwell and Fabian review of prospective
(OR = 0.54; 95% CI, 0.39–0.76), and wound complications and retrospective studies of civilian colon injury that sepa-
excluding dehiscence (OR = 0.43; 95% CI, 0.24–0.77) all sig- rately analyzed nondestructive and destructive wounds,
nificantly favored primary repair (61). Another meta-analysis showing that primary repair with minimal amounts of
conducted the year before the Cochrane Review that did not debridement was clearly the treatment of choice for nonde-
include the Kamwendo study, which was found to contribute structive injuries (64). For these patients, the overall com-
heterogeneity to the Cochrane review, yielded ORs in favor of plication rate was 14% in the primary repair group and 30%
primary repair for each condition (62). in the colostomy group, and intraabdominal abscess rates
were 4.9% and 12%, respectively (64).
THE ROLE OF RESECTION WITH Destructive wounds, on the other hand, typically require
ANASTOMOSIS OR DIVERSION some degree of resection, as a result of transection of the
bowel or devascularization due to mesenteric injury. Because
All of the randomized trials describe thus far classified all most of the prospective studies conducted through the early
surgeries that occurred in one operation with anastomosis 2000s did not include a substantial amount of destructive
Suspected colon injury
Hemodynamically stable?
No
Yes
Acidotic, hypothermic,
and/or coagulopathic?
Yes No
Obtain labs, plain films, CT, +/–
Damage control laparotomy FAST, DPL diagnostic laparoscopy
with temporary closure
Negative Positive
Positive Negative
Rewarm and Consider extra-
resuscitate in ICU colonic injury
Colon injury Observe
present
Return to OR for definitive
treatment after stabilization
Destructive injury?
Yes No
Resection Primary repair
Patient in shock, received > 4 units

pRBCs, or has significant co-morbidities?
Yes No
Consider Hartmann Primary anastomosis

or
Anastomosis with proximal diversion
Figure 40.5 Colon algorithm.
injuries, it was less clear if primary repair was just as safe controlling for the traditional risk factors described in the
and effective for these patients as for those with nondestruc- literature, such as shock, transfusion >6 units of packed
tive injuries. In 2001, Demetriades and colleagues published red blood cells, PATI score >25, and delay to operation >6
a landmark study that, though not randomized, prospec- hours. Thirteen patients in the primary repair cohort expe-
tively evaluated the outcomes of resection with primary rienced an anastomotic leak, and one patient in the diversion
anastomosis versus diversion for 297 patients with destruc- group had a leak from the Hartmann pouch. No patients
tive colon injuries (15). Treatment was determined at the died from leak, and no risk factors were identified for leak.
surgeon’s discretion, resulting in two-thirds of patients The authors of this study concluded that all destructive
treated with primary anastomosis and one-third with colon injuries requiring resection should be managed with
diversion. There was lower colon-related mortality in the primary repair regardless of risk factors present.
primary anastomosis group (0% versus 4%, p = 0.012), and There remain adamant supporters on either side of the
similar rates of abdominal complications (22% versus 27%, argument for primary anastomosis versus either proximal
p = 0.373). Univariate analysis demonstrated that severe diversion or Hartmann, especially for destructive colon
fecal contamination, transfusion requiring greater than 4 injuries in patients with the traditional risk factors. Rather
units of packed red blood cells within the first 24 hours, and than universally implementing primary repair regardless
single-agent antibiotic prophylaxis were all independently of risk factors, some have attempted to standardize a clini-
associated with increased complications. However, when cal algorithm based on the presence of potential risk factors
controlling for these risk factors on multivariate analysis for anastomotic leak and/or mortality related to suture line
comparing primary anastomosis with diversion, there was failure. Stewart et al. found that patients with destructive
no significant difference in risk for complications. They also wounds that also received >6 units of packed red blood cells
found no differences between the two treatment arms when and/or had significant comorbid conditions were had a 42%
Management of colon injury / Colostomy closure 387
risk for leak, while their healthy counterparts who did not The literature describing long-term outcomes of patients
require massive transfusion had a leak rate of just 3% (65). undergoing repair of colon injury after damage control sur-
Furthermore, one-third of patients suffering a leak died. gery is sparse, lacking randomized studies. The current set
Considering these concerning data, this institution imple- of reviews of DCL patients yields conflicting data, with some
mented a clinical pathway, whereby patients with destructive studies suggesting that restoration of continuity is unsafe
injuries and either significant comorbid conditions or >6 after DCL, and others that repair after DCL is equivalent to a
unit transfusion were treated with end colostomy and those single operation. It is important to note that patients undergo-
with destructive injury but without either of these conditions ing DCL have been shown to have higher leak rates, increased
were treated with resection and primary repair (50). By fol- risk of ventral hernia, and more colon-related complications
lowing this algorithm, the authors were able to reduce their than those who undergo repair by a single laparotomy (both
overall leak rate from 14% to 3%, as well as decrease their primary repair or resection with anastomosis or diversion)
abscess and mortality rates. Despite the clear improvements (70,71). Another study also showed higher leak rates in the
made by this series, conflicting data continue to emerge that DCL group, but still concluded that it was a safe option (72).
still support routine primary repair regardless of any patient Miller and associates found no leaks in the DCL group and
risk factors (66). The authors believe that diversion is still one in the primary repair group, and no differences in abscess
useful in select cases, and that the judgement of an experi- rates and overall survival between the patients who under-
enced surgeon is required on a case-by-case basis to choose went anastomosis versus diversion after DCL (73). A more
between primary repair and diversion. The risk factors recent study concluded that anastomosis in DCL patients is
shown in some of the literature, technical issues encountered feasible if completed during the first operation post-DCL,
during the case, and circumstances unique to each patient but anastomoses performed beyond this point are associated
must be considered when weighing treatment options. with an eight times greater risk of leak compared to those in
Of note, there are important differences between civilian single laparotomy and first operation post-DCL patients (74).
and military colon traumas, but primary repair is still appli- There are even fewer studies detailing DCL in the mili-
cable to combat injuries. As stated, these patients are more tary setting, but it has been shown to minimize compli-
likely to have trauma involving high-energy mechanisms, cation rates for severely injured patients (19). Despite the
have multiple associated injuries, and have burns, and care varying data that exist, damage control laparotomy remains
should be taken to consider the evolution of the injury over a useful strategy for improving outcomes of rapidly dete-
time. The limitations of treatment in some combat settings riorating patients and delaying decision-making regarding
also pose challenges as compared to civilian trauma centers, definitive repair of the colon. This may be particularly rel-
with limited supplies and prolonged transport time often evant for treating some of the high-energy and blast injuries
required. Considering these factors, primary anastomosis encountered especially in the battlefield (Figure 40.6).
has thus far been utilized less frequently in the military set-
ting than in civilian traumas and has been encouraged to be COLOSTOMY CLOSURE
approached with increased caution (67,68). However, when
implemented appropriately, similar results may be obtained It has classically been thought that diversion is a “safer”
to those in the civilian literature. A 2007 study of Operation option as compared to primary anastomosis, especially in
Iraqi Freedom colorectal trauma patients from 2003 to 2004 hemodynamically unstable patients with increased comor-
showed that diversion was associated with lower leak rates bidities. However, colostomy creation and takedown are
compared to primary anastomosis, but rates of sepsis and
mortality did not differ between the two groups (54).
DAMAGE CONTROL
The concept of damage control has been useful in prevent-
ing the rapid decline of patients that are acidotic, hypother-
mic, and coagulopathic in order to avoid the lethal triad.
After an abbreviated procedure for control of hemorrhage
and prevention of further intraabdominal contamination,
the patient is taken to the ICU for stabilization, re-warm-
ing, and resuscitation. Not only does this save the patient
from deteriorating, but it also permits time for the surgeon
to decide if an anastomosis or diversion will be performed
in the subsequent operation. This protocol was first elabo-
rated by Rotondo et al. in 1993, based on the experience of
the Navy with damaged ships, and proved to be a promis-
ing methodology to improve morbidity and mortality in an Figure 40.6 Damage control laparotomy in a combat
initially unstable state (69). environment.
procedures that are not without risks of their own. In a suspicion for concurrent injuries to these organs and to the
2000 review of 311 patients admitted for penetrating trauma surrounding bony pelvis and major vasculature as well. This
to the colon at a level I urban trauma center, ostomy cre- makes the principles of damage control laparotomy espe-
ation was found to be one of the most significant risk fac- cially pertinent to rectal injury (82).
tors for postoperative complications on both univariate
(p < 0.0001) and multivariate (p = 0.004) analysis (75).
Patients who have a complicated initial procedure have THE FOUR D’S
been shown to also be more likely to experience complica-
While the management of rectal injuries has evolved on a
tions after colostomy closure, especially if performed within
timeline similar to colon injuries, the adoption of primary
3 months of the trauma (76). The rates of complications
repair has been somewhat slowed and wary due to the pau-
associated with colostomy closure reported in the literature
city of class I data evaluating the outcomes of management
vary greatly, anywhere from 5% to 55% but most somewhere
of rectal trauma. Classically, the surgical treatment of rectal
around 20%–30% and some studies showing lower compli-
injury has been based on “the four D’s”: diversion, drain-
cation rates with loop compared to end colostomies (77–80).
age, distal washout, and direct repair when feasible. The use
Pertinent complications include enterotomy, wound infec-
of these four pillars was supported by reduced mortality in
tion and dehiscence, anastomotic leakage and stricture,
a study of rectal injuries during the Vietnam War, but the
enterocutaneous fistula, ileus, and ventral hernia. The fact
study included just 29 patients considering the rare nature
that an additional operation and hospital stay are required
of the injury (83). Each of these components, which have
for reversal also increases costs associated with colostomy.
been considered central dogma for the treatment of rectal
Although the timing of reversal remains controversial,
trauma, has been challenged in the subsequent literature,
same-admission reversals may be an effective strategy to
reduce costs while also minimizing complications (81). In but small sample sizes, lack of randomized control trials,
summary, while there is certainly a role for diversion, the and conflicting results have made coming to a uniform
morbidity associated with not only colostomy creation consensus quite difficult. A summary of the data reviewed
for the 2016 Eastern Association for the Surgery of Trauma
but also takedown must also be considered when deciding
(EAST) guidelines on the management of penetrating extra-
against primary anastomosis.
peritoneal rectal injury can be found in Table 40.4.
SUMMARY: COLON INJURIES

Diversion
Patients with traumatic colon injuries can often be managed
by in-continuity methods (i.e., primary resection, resection Similar to decision-making for the approach to colon
with primary anastomosis). Patients should first be evalu- management, considering the extent and location of the
ated for hemodynamic stability, and if needed, undergo a rectal damage as well as the overall stability of the patient
damage control laparotomy followed by resuscitation and can be used to determine when diversion versus primary
secondary laparotomy when stabilized. Complications may anastomosis is most appropriate (see Rectal Algorithm,
still occur, but sound technique and good surgical judgment Figure 40.7). As with colon injury, primary repair should be
are required when confronted with colon injuries. considered for nondestructive intraperitoneal rectal injury,
but resection is necessary with destructive injury. Primary
anastomosis is an option for the otherwise healthy and
hemodynamically stable patient, but those with comorbidi-
MANAGEMENT OF RECTAL INJURY ties or hemodynamic instability should be diverted (84,85).
In general, at least some form of diversion is considered
COMPARING TO COLON INJURY necessary in the context of the majority of extraperitoneal
rectal injuries in both civilian and military cases. This is
In general, the diagnosis and management of rectal injury largely because diversion has been confirmed to be safe and
are more complex than that of colon injury for several rea- effective for rectal injury, while studies describing repair
sons. Because the rectum is largely protected by the pelvic without any diversion are few, and conclusions are based
bones and surrounding soft tissue, it is less likely to sus- mainly on class III data. A 2006 prospective study conducted
tain injury; when injury does occur, however, these barri- by Gonzalez et al. managed 14 patients with nondestructive
ers make exposure and treatment of the affected area much extraperitoneal rectal injuries without fecal stream diver-
more difficult. Additionally, the rectum has both intraperi- sion, with no infectious complications related to the rectal
toneal and extraperitoneal portions. In general, the same defect (86). This management style is yet to become com-
principles for management of colon injuries discussed pre- mon practice as these results have not been replicated in
viously can be applied to intraperitoneal rectal injuries, but any substantial sample size or in a randomized study, while
lower rectal injuries have different diagnostic and manage- several larger studies have demonstrated consistent sepsis
ment implications. Finally, the close proximity of the rec- rates <33% and mortality rates <7% in the context of stoma
tum to the genitourinary system requires a high index of placement (67,87–91). Colostomy creation may be even more
Management of rectal injury / The four D’s 389
Table 40.4 Data reviewed in the 2016 EAST guidelines for penetrating extraperitoneal rectal injury
Number of Number of Infectious

studies patients Mortality, % complications, %
Diversion versus repair 14 549 Proximal diversion
Diversion 523 1.7 8.8
Primary repair + no 26 0 18.2
diversion
Drainage versus no 17 653 No drainage
drainage
Drainage 388 1.8 9.3
No drainage 265 0.75 6.1
Washout versus no washout 13 487 No washout
Washout 200 1 11
No washout 287 1.4 11.5
Suspected rectal injury
Hemodynamically stable?
No
Yes
Acidotic, hypothermic,
and/or coagulopathic?
Yes No
DRE, proctoscopy, obtain labs, plain
Damage control laparotomy DRE, proctoscopy films, CT, +/– diagnostic laparoscopy
with temporary closure
Negative Positive
Positive Negative
Rewarm and Consider extra-
resuscitate in ICU rectal injury
Rectal injury Observe
present
Return to OR for definitive
treatment after stabilization
Location of injury
Intraperitoneal Extraperitoneal Combined injury
Destructive injury? Destructive injury? Destructive injury?

Yes No Yes No Yes No
Resection Primary repair Consider Hartmann Proximal diversion Consider Hartmann Proximal diversion
rather than proximal rather than proximal
diversion diversion
Patient in shock, received > 4 units

pRBCs, or has significant co-morbidities?
Direct repair, presacral drainage, and distal
Direct repair of
Yes No washout may be utilized in the management of
intraperitoneal
extraperitoneal rectal injuries. However, there is
lesion if feasible
no strong evidence demonstrating significant
Consider Hartmann
Primary anastomosis benefit of any of these practices so they cannot
or
be recommended for routine use, rather, should
Anastomosis with
be used at the surgeon’s discretion.
proximal diversion
Figure 40.7 Rectal algorithm.

crucial in the military setting, where injury severity is often train of thought for the role of direct repair, drainage is even
escalated, and multiple traumas are sustained simultane- less useful in the context of a stoma.
ously (16,67). The largest retrospective study conducted thus far
The type of diversion performed is dependent on both reviewed 92 patients with rectal injuries, 86 of which
the extent of the injury, the skill and experience of the sur- received a proximal colostomy. None of these patients had
geon, and whether or not the stoma is to be reversed in the a drain placed, the infection rate was just 10%, and there
future. Considering the increased morbidity and mortal- was no incidence of sepsis (99). Similar results have been
ity associated with reversal of colostomy after a Hartmann obtained by other reviews (100,101). Additionally, a ran-
procedure, loop ostomy diversion is preferable when fea- domized control trial was conducted by Gonzalez et al. in
sible (92–94). However, if there is significant damage to the which all 48 patients were diverted but 23 received a drain
sigmoid and/or rectum, a Hartmann procedure is often and 25 did not. Infectious complications occurred in just
necessitated (89). 4% of the no-drain group and 8% of the drain group, lead-
ing the authors to conclude in favor of omission of presacral
Direct repair drainage in most cases (102). This practice may still be use-
ful in the setting of large destructive wounds with increased
It is generally accepted that rectal wounds that are intra- serosanguinous fluid drainage into the pelvis.
peritoneal should be repaired, because they can easily be
accessed without major dissection. Furthermore, intra- Distal washout
peritoneal wounds can be treated similar to colon injuries
as previously discussed. Extraperitoneal lesions, however, Distal washout involves either a washout from below or
are less amenable to repair and require significant opera- on-table lavage through a separate injury, colotomy, or
tive time. Therefore, the potential benefits of repair are appendectomy opening (Figure 40.8). Although distal rec-
often outweighed by the risks, and it is generally accepted tal irrigation was initially implemented for fear of leaving
that extraperitoneal injury repair should not be attempted remaining stool to act as a substrate for sepsis, opponents
unless the wound is easily visualized and requires minimal of this practice argue that forcing fluid into the rectum may
dissection (95–97). actually facilitate spread of feces and bacteria into previously
Whether or not a diversion is performed may also dic- uncontaminated planes. With little benefit attributable to
tate whether or not direct repair of the rectal wound is nec- washout, most surgeons have abandoned this component of
essary. When the fecal stream is diverted away from the management. The majority of data evaluating the outcomes
rectum, the risk of infectious complications is minimized, of washout are retrospective in nature, but most studies in
and foregoing direct repair has been shown to have no cor- the last two decades have shown no significant decrease in
relation with infection (88,97). This was demonstrated in morbidity associated with distal washout, and EAST simi-
a study by Weinberg et al. that followed an algorithm by larly recommends against it (91,97,103).
which intraperitoneal rectal injuries treated with primary
repair, extraperitoneal injury to the proximal two-thirds SUMMARY: RECTAL INJURY
or accessible distal segments of the rectum were managed
with direct repair and diversion, and inaccessible injuries to Rectal injuries need to be stratified by the location—intra-
the distal one-third were managed with proximal diversion versus extraperitoneal. Intraperitoneal injuries can often
and presacral drainage without direct repair (85). Following be treated as colon injuries, with the same issues of need
this clinical pathway, they were able to reduce the incidence for primary repair, resection, or diversion, along with the
of infectious complications by 50%. Others have attempted DCL. Extraperitoneal injuries often still require diversion.
primary repair of an accessible rectal injury without proxi- Drainage, direct repair, and distal washout are more con-
mal diversion, demonstrating no associated morbidity but troversial. The senior author prefers to avoid washout from
only in very small sample sizes (96). above and simply remove the gross stool where possible;
direct repair only when easy and directly amenable (fore-
Drainage going extensive dissection to get at the site of injury), and
drainage only when the area is open (avoiding making new
Presacral drainage is one of the other classic pillars of extra- planes to place drains).
peritoneal rectal injury management, but this practice has
also been supported by only weak evidence and has been OTHER STRATEGIES TO MINIMIZE
proven futile in other studies. Drains allow for extraction INFECTION IN COLORECTAL TRAUMA
of contaminated fluid or blood but also involve the invasive
nature of drain placement. While older studies of combat Patients with penetrating abdominal trauma, especially
trauma demonstrated decreased rates of pelvic infection when the colon and rectum are affected, are almost the per-
with drain placement, more recent studies have concluded fect hosts for infection. Discontinuity of the gastrointesti-
that drainage does not significantly reduce infection rates nal tract allows for microbial translocation, ischemic tissue
and is not as imperative as once thought (98). Similar to the and retained fragments allow for seeding and growth of
Anorectal foreign bodies 391
Figure 40.8 Rectal washout.
organisms, and a patient with sustained blood loss in shock 24 hours has not been shown to add any benefit, even in the
will have an impaired immune response (104). Therefore, context of DCL or high-risk patients (111–113). It is impor-
extra precaution must be taken to prevent infectious com- tant to not prolong antibiotic use so as to avoid promotion
plications in these trauma patients. There are several strate- of antibiotic-resistant strains (114).
gies that are commonly described in the literature and are
typically recommended in these cases.
Aggressive wound debridement should be undertaken
at the time of surgery. The goal is to remove all of the ANORECTAL FOREIGN BODIES
necrotic and nonviable tissue, and then clean the wound
with irrigation using normal saline or sterile water (105). In this section, we cover anorectal foreign bodies. The
At this point in time, any retained fragments from the reader is encouraged to review other extensive reviews on
penetrating weapon that are visualized should also be this topic by the senior author including an UpToDate dis-
carefully removed. These foreign objects may serve as a cussion on management (available at www.UpToDate.com).
foundation for abscess formation, thus increasing the risk Retained foreign bodies in the rectum are encountered rela-
for sepsis if left in the abdominal cavity (106). After these tively infrequently, but diagnosis and management are com-
procedures, the abdominal skin wound is usually left open plicated by the vast array of objects that may be found and
to heal by secondary intent in order to prevent surgical site the potential damage that each object can cause. Examples
infection (107). of items recovered are drugs, torches, broomsticks, aerosol
The use of prophylactic antibiotics in this setting has been cans, pipes, vibrators, a variety of foods, light bulbs, knives,
well described and supported by sufficient class I and class enema tips, and more. Most of the literature represents sin-
II data. The cornerstone for antibiotic therapy in colorectal gle-center case studies, but a recent Nationwide Inpatient
trauma is preoperative administration of broad-spectrum Sample (NIS) database review identified 3,359 cases with
IV antibiotics with anaerobic coverage continued for up to a primary diagnosis of rectal foreign body between 2009
24 hours. The accumulation of high-quality data supporting and 2011 (115). The majority of patients presenting in these
this tenant has led to recommendations in both the civil- cases are men, with published studies showing between 65%
ian and military sectors that are widely accepted (105,108). and 100% of their sample to be men (116). Most patients are
Following these guidelines can minimize the rate of post- between ages 30 and 40, but patients as old as 90 have been
operative infection to just 11% (109). A single-drug regimen reported (117–119).
is typically adequate for nondestructive injuries, but more Placement of the foreign object is typically classified as
extensive wounds may benefit from combination therapy voluntary versus involuntary, and/or sexual versus non-
(15,110). The administration of antibiotics for longer than sexual. Examples of scenarios and objects typical for each
category can be found in Table 40.5. A 2010 systematic pneumoperitoneum. Pneumoperitoneum, hypotension,
review of colorectal foreign bodies in 193 patients found that fever, tachycardia, and peritoneal signs indicate perforation.
sexual activity was implicated in 49% of cases, iatrogenic In this case, the patient should be immediately resuscitated
injury following self-treatment for anorectal pathology in and broad-spectrum antibiotics should be administered. If
25% of cases, self-insertion in the mentally ill in 5%, and the patient is stable enough for CT, this may be useful in
accidents or assaults in the remaining 21% of cases (120). locating the site of rectal perforation. Patients with an intra-
The injury may also be classified by the AAST rectal injury peritoneal perforation of the rectum or distal colon require
scale (Table 40.2), but the majority of foreign body–related laparotomy, and the approach to repair follows the same
injuries are grade I or less commonly grade II (116,118). guidelines as the trauma management algorithms discussed
Obtaining an accurate history is often challenging in previously (121). Extraperitoneal perforations may be man-
these instances, as it is not uncommon for patients to feel aged more conservatively (122).
embarrassed and reluctant to disclose all relevant informa- In most cases, the patient is stable without perforation
tion. The physician must also be prepared for the possibility and may be managed nonoperatively. The majority of foreign
of providing emotional support and a safe environment for bodies can be removed transanally at the bedside (120,121).
rape and assault victims. Most patients will present on the Patient relaxation can be encouraged with the use of a peri-
day of or several days after insertion of the object, but some anal nerve block or conscious sedation. Additionally, Foley
will wait months, making careful examination and removal catheter inflation, clamps, obstetric vacuum extractors,
even more imperative (120). Patients may report constipa- and endoscopy may be utilized to help retrieve the object
tion, anorectal or abdominal pain, bright red blood per if manual manipulation is not successful. If this becomes
rectum, or anal mucous discharge. Physical examination challenging, the patient may be asked to perform a Valsalva
findings will vary depending on the location of the object maneuver or can be admitted and observed to allow the
and the degree of injury caused. The item may be palpated object to descend distally. When none of these techniques
in the lower quadrants of the abdomen if located more prox- facilitate removal of the object, then transfer to the operat-
imally, or may be visualized or palpable upon DRE if more ing room and use of general anesthesia or spinal anesthetic
distal. Gross blood may be found, and in cases of sphincter may be required. The goal is to milk the object distally so
injury, tone will be decreased on DRE. If the rectum has that it may be accessed transanally. Some surgeons have uti-
been perforated above the peritoneal reflection, there may lized laparoscopy to achieve this (123,124). Finally, if none
be peritoneal signs. When foreign-body placement is sus- of these approaches are successful, then colotomy must be
pected but has not been disclosed by the patient, the phy- performed to remove the object.
sician should directly inquire about this possibility in a In general, patients with anorectal foreign bodies recover
nonconfrontational and nonjudgmental manner—empha- without major complications. However, there have been
sizing the importance of obtaining details regarding the cases of sepsis and even death reported in severe cases with
object, timing, and circumstances of the insertion in order perforation (125). Patients with perforation may require
to best proceed with management. diversion, which facilitates subsequent operation for rever-
Diagnosis is typically established using plain sal. Finally, injury to the sphincters can result in long-term
film radiographs to identify the object and rule out fecal incontinence or require sphincterotomy (126–128).
Table 40.5 Classification of anorectal foreign bodies and

common presentations of each ANAL SPHINCTER AND PERINEAL
Common patient profile/
INJURY
objects
Anal sphincter injury occurs most commonly in the setting
Sexual + voluntary Vibrators and dildos most
of obstetric trauma but may also result from anorectal for-
commonly, but many other
eign-body placement or iatrogenic during anorectal opera-
objects have been implicated
tions (129). The extent of the injury determines management
in these scenarios
strategy, as described by Hellinger (130). Superficial inju-
Sexual + involuntary Rape or assault victims, sodomy
ries may be debrided, and if the extent of injury is limited
with a variety of objects
to minimal damage to the internal sphincter, it may be left
Nonsexual + voluntary Body packing with illicit drugs
unrepaired. Destructive injuries often mandate proximal
or illegal paraphernalia
diversion and repair of the sphincter damage. This was clas-
Nonsexual + involuntary Children following ingestion of sically achieved with end-to-end direct apposition of the anal
plastic objects or utensils
sphincters, but more recently the practice of overlapping
Ingestion of bones from food
sphincteroplasty has been introduced (129). Surgeon prefer-
Mentally ill patients
ences and data literature on method of repair vary, but end-
Iatrogenic injury, retained
to-end apposition may be more effective in acute scenarios,
thermometers or enema tips
while overlap may be suitable for delayed repairs (131). More
References 393
advanced techniques include sphincter reconstruction with 18. Hatch Q et al. Surgery. 2013;154(2):397–403.
the gluteal and gracilis muscles, as well as construction of 19. Cho SD et al. Dis Colon Rectum. 2010;53(5):
artificial sphincter systems (132–134). These methods are 728–34.
complex and have variable success and fecal incontinence 20. Causey M et al. Clin Colon Rectal Surg. 2012;25(4):
rates, and they should only be attempted by surgeons in a 189–99.
delayed setting with extensive training and experience with 21. Perry WB. Trauma of the colon, rectum, and anus. In:
these procedures. Ultimately, up to 10% of sphincter repairs Steele SR, Hull TL, Read TE, Saclarides TJ, Senagore
may degenerate in the long term, and fecal incontinence AJ, Whitlow CB (eds.) The ASCRS Textbook of Colon
rates vary from study to study and may be underestimated, and Rectal Surgery. Cham, Switzerland: Springer
as this is a difficult endpoint to ascertain from patient inter- International, 2016, pp. 735–47.
views (135,136). The timing of this procedure is variable and 22. Herzig D. Clin Colon Rectal Surg. 2012;25(4):210–3.
somewhat controversial. Often the injury and surrounding 23. Shapiro MB et al. J Trauma. 2000;49(5):969–78.
tissues are not amenable to repair at the time of the surgery 24. McPartland KJ, Hyman NH. Dis Colon Rectum. 2003;
and are best left alone to attempt definitive repair at a later 46(7):981–6.
date. To divert at the time of the initial injury also is variable, 25. Schurink GW et al. Injury. 1997;28(4):261–5.
but discussion with the patient should be performed ahead 26. Chandler CF et al. Am Surg. 1997;63(10):885–8.
of time when possible. 27. Hoff WS et al. J Trauma. 2002;53(3):602–15.
Complex perineal injuries may be associated with more 28. Blow O et al. J Trauma Acute Care Surg. 1998;44(2):
life-threatening conditions such as hemorrhage and pel- 287–90.
vic fracture. Management of these issues takes priority 29. Liu M et al. J Trauma. 1993;35(2):267–70.
over repair of any sphincter damage, and the principles of 30. Davis JW et al. J Trauma. 1990;30(12):1506–9.
Advanced Trauma Life Support as well as damage control 31. Smith IM et al. Ann Surg. 2015;262(2):389–96.
surgery apply when needed. A review of these rare, com- 32. Rozycki GS et al. J Trauma. 1995;39(3):492–500.
plex clinical scenarios was published by Kudsk and Hannah 33. Branney SW et al. J Trauma. 1995;39(2):375–80.
in 2003, demonstrating an overall mortality rate of 32%, 34. Udobi KF et al. J Trauma. 2001;50(3):475–9.
and 21% pelvic sepsis rate (137). The initial surgical goals 35. Malhotra A et al. J Trauma. 2000;48(6):991–1000.
should be pelvic fixation and control of hemorrhage, and 36. Butela ST et al. Am J Roentgenol. 2001;176(1):
fecal diversion with wound debridement and irrigation can 129–35.
be addressed during later return to the operating room after 37. Shanmuganathan K et al. Am J Roentgenol. 2001;
stabilization. Wound coverage and restoration of sphincter 177(6):1247–56.
function are long-term goals. 38. Killeen KL et al. J Trauma. 2001;51(1):26–36.
39. Powell BS et al. Injury. 2008;39(5):530–4.
REFERENCES 40. Lin HF et al. World J Surg. 2010;34(7):1653–62.
41. Lee PC et al. Surg Innov. 2014;21(2):155–65.
1. Welling DR, Duncan JE. Clin Colon Rectal Surg. 2008; 42. Villavicencio RT, Aucar JA. J Am Coll Surg. 1999;
21(1):45–52. 189(1):11–20.
2. Bennett J. J R Soc Med. 1991;84:554–7. 43. Shlamovitz GZ et al. Ann Emerg Med. 2007;50(1):
3. Wallace C. Lancet. 1917;(ii):561–8. 25–33.
4. Bowlby A. Br Med J. 1917;I:705–21. 44. Esposito T et al. Int Braz J Urol. 2006;32(1):107–9.
5. Perry WB et al. Semin Colon Rectal Surg. 45. Hargraves MB et al. Am Surg. 2009;75(11):1069–72.
2005;15(2):70–9. 46. Anderson SW, Soto JA. Semin Ultrasound, CT MRI.
6. Ogilvie W. Surg Gyn Obs. 1944;78(2):225–38. 2008;29(6):472–82.
7. Steele SR et al. Dis Colon Rectum. 47. Flint LM et al. Ann Surg. 1981;193(5):619–23.
2011;54(9):1184–201. 48. Moore E et al. J Trauma. 1981;21(6):439–45.
8. Hughes L. Br J Surg. 1969;56(3):169–72. 49. Moore EE et al. J Trauma. 1990;30(11):1427–9.
9. MacFarlane C et al. J R Army Med Corps. 50. Miller PR et al. Ann Surg. 2002;235(6):775–81.
2002;148(1):27–31. 51. Baako BN. West Afr J Med. 1998;17(2):109–12.
10. Ganchrow MI et al. Arch Surg. 1970;100(4):515–20. 52. Hudolin T, Hudolin I. Br J Surg. 2005;92(5):643–7.
11. Aldrete JS et al. Ann Surg. 1970;172(6):1007–14. 53. Uravic M et al. Mil Med. 2000;165(3):186–8.
12. LoCicero J et al. J Trauma. 1975;15(7):575–9. 54. Steele SR et al. Dis Colon Rectum. 2007;50(6):870–7.
13. Stone HH, Fabian TC. Ann Surg. 1979;190(4):430–6. 55. Chappuis CW et al. Ann Surg. 1991;213(5):492–8.
14. Morken JJ et al. Surgery. 1999;126(4):693–700. 56. Sasaki LS et al. J Trauma. 1995;39(5):895–901.
15. Demetriades D et al. J Trauma. 2001;50(5):765–75. 57. Gonzalez RP et al. Am Surg. 2000;66(4):342–7.
16. Glasgow SC et al. J Trauma Acute Care Surg. 58. Kamwendo NY et al. Br J Surg. 2002;89(8):993–8.
2012;73(6 Suppl 5):S503–8. 59. Gonzalez RP et al. J Trauma. 1996;41(2):271–5.
17. Johnson EK, Steele SR. J Gastrointest Surg. 60. Falcone RE et al. Dis Colon Rectum.
2013;17(9):1712–9. 1992;35(10):957–63.
61. Nelson RL, Singer M. Cochrane Database Syst Rev. 105. Hospenthal DR et al. J Trauma. 2008;64(3 Suppl):
2003;(3):CD002247. S211–20.
62. Singer MA, Nelson RL. Dis Colon Rectum. 2002;45(12): 106. Poret HA3rd et al. J Trauma. 1991;31(8):1085–8.
1579–87. 107. Velmahos GC et al. Am Surg. 2002;68(9):795–801.
63. Choi WJ. J Korean Soc Coloproctol. 2011;27(4): 108. Goldberg SR et al. J Trauma Acute Care Surg. 2012;
166–72. 73(5 Suppl 4):S321–5.
64. Maxwell RA, Fabian TC. World J Surg. 2003;27(6): 109. Fullen W et al. J Trauma. 1972;12(282):289.
632–9. 110. Greer CLT et al. Am Surg. 2013;79(2):119–27.
65. Stewart RM et al. Am J Surg. 1994;168(4):316–9. 111. Fabian TC et al. Surgery. 1992;112(4):788–95.
66. Dubose J. J Gastrointest Surg. 2009;13(3):403–4. 112. Cornwell EE et al. J Gastrointest Surg. 1999;3(6):648.
67. Duncan JE et al. J Trauma. 2008;64(4):1043–7. 113. Kirton OC et al. J Trauma. 2000;49(5):822–32.
68. Sambasivan CN et al. Am Surg. 2011;77(12):1685–91. 114. Velmahos GC et al. Arch Surg. 2002;137(5):537–42.
69. Rotondo MF et al. J Trauma. 1993;35(3):375–83. 115. Pathak R et al. N Am J Med Sci. 2016;8(4):191.
70. Weinberg J et al. J Trauma. 2009;67(5):929–35. 116. Goldberg JE, Steele SR. Surg Clin North Am. 2010;
71. Brenner M et al. Arch Surg. 2011;146(4):395–9. 90(1):173–84.
72. Kashuk JL et al. Surgery. 2009;146(4):663–70. 117. Ooi BS et al. Aust N Z J Surg. 1998;68(12):852–5.
73. Miller PR et al. Am Surg. 2007;73(6):606–9. 118. Rodríguez-hermosa JI et al. Color Dis. 2007;9(6):
74. Anjaria DJ et al. J Trauma Acute Care Surg. 2014; 543–8.
76(3):594–600. 119. Clarke D et al. Color Dis. 2010;7(1):98–103.
75. Dente CJ et al. J Trauma. 2000;49(4):628–37. 120. Kurer MA et al. Color Dis. 2010;12(9):851–61.
76. Sola JE et al. Injury. 1993;24(7):438–40. 121. Lake JP et al. Dis Colon Rectum. 2004;47(10):1694–8.
77. Parks SE, Hastings PR. Am J Surg. 1985;149(5):672–5. 122. Fry RD et al. Dis Colon Rectum. 1989;32(9):759–64.
78. Pachter HL et al. J Trauma. 1990;30(12):1510–3. 123. Rispoli G et al. Dis Colon Rectum. 2000;43(11):
79. Berne JD et al. Surgery. 1998;123(2):157–64. 1632–4.
80. Pokorny H et al. Arch Surg. 2005;140(10):956–60. 124. Berghoff KR, Franklin ME. Dis Colon Rectum. 2005;
81. Khalid MS et al. Surgeon. 2005;3(1):11–4. 48(10):1975–7.
82. Arthurs Z et al. Am J Surg. 2006;191(5):604–9. 125. Waraich NG et al. N Z Med J. 2007;120(1260):U2685.
83. Lavenson GS, Cohen A. Am J Surg. 1971;122(2): 126. Madiba TE, Moodley MM. East Afr Med J. 2003;
226–30. 80(11):585–8.
84. Herr MW, Gagliano RA. Curr Surg. 2005;62(2):187–92. 127. Fernando R et al. Cochrane Database Syst Rev. 2006;
85. Weinberg J et al. J Trauma. 2006;60(3):508–14. 3(12):CD002866.
86. Gonzalez RP et al. J Trauma. 2006;61(4):815–9. 128. Fitzpatrick M et al. Am J Obstet Gynecol. 2000;183:
87. Ivatury RR et al. Am Surg. 1991;57(1):50–5. 1220–4.
88. Velmahos GC et al. World J Surg. 2000;24(1):114–8. 129. Brill SA, Margolin DA. Semin Colon Rectal Surg. 2005;
89. Merlino JI, Reynolds HL. Semin Colon Rectal Surg. 15(2):90–4.
2005;15(2):95–104. 130. Hellinger MD. Surg Clin North Am. 2002;82(6):
90. Bostick PJ et al. J Natl Med Assoc. 1993;85(6):460–3. 1253–60.
91. Burch JM et al. Ann Surg. 1989;209(5):600–11. 131. Cherry D, Greenwald M. Anal incontinence. In:
92. Sharma A et al. Color Dis. 2013;15(4):458–62. Beck D, Wexner S. (eds.) Fundamentals of Anorectal
93. Ince M et al. Color Dis. 2012;14(8):492–8. Surgery. New York, NY: McGraw-Hill, 1992,
94. Mosdell DM, Doberneck RC. Am J Surg. 1991;162(6): pp. 104–30.
633–7. 132. Wexner SD et al. Dis Colon Rectum. 2002;45(6):
95. Tuggle D, Huber PJ. Am J Surg. 1984;148(6):806–8. 809–18.
96. Levine JH et al. Am J Surg. 1996;172(5):575–9. 133. Madoff RD et al. Gastroenterology. 1999;116:549–56.
97. Mcgrath V et al. Am Surg. 1998;64(12):1136–41. 134. Wong WD et al. Dis Colon Rectum. 2002;45(9):
98. Armstrong R et al. Surgery. 1973;74:570–4. 1139–53.
99. Navsaria PH et al. World J Surg. 2007;31(6):1345–51. 135. Halverson AL, Hull TL. Dis Colon Rectum. 2002;45(3):
100. Levy R et al. J Trauma. 1995;38:273–7. 345–8.
101. Steinig JP, Boyd CR. Am Surg. 1996;62(9):765–7. 136. Jorge JM, Wexner SD. Dis Colon Rectum. 1993;36(1):
102. Gonzalez RP et al. J Trauma. 1998;45(4):656–61. 77–97.
103. Shannon F et al. J Trauma. 1988;28(7):989–94. 137. Kudsk KA, Hanna MK. World J Surg. 2003;27(8):
104. Fabian TC. Am J Surg. 1993;165(2A Suppl.):14S–9S. 895–900.
41
Prevention and management of urologic
complications after colorectal surgery
JACOB A. McCOY, J. CHRISTIAN WINTERS, AND SCOTT E. DELACROIX JR.
inflation of the retention balloon within the prostatic ure-

CHALLENGING CASE thra. A standard 16-French Foley catheter is an appropri-
ate choice for the initial attempt for most patients. It is not
An elderly man presents with gross hematuria. necessary to “test” inflate the balloon prior to attempted
Cystoscopic examination reveals a mass invading placement, as this will increase the size and decrease the
the left lateral bladder wall. Computed tomography rigidity of the distal aspect of the catheter. In men with
(CT) scan of the abdomen and pelvis demonstrates a a known history of benign prostatic hyperplasia, a 16- or
6.6 cm pelvic mass originating from the sigmoid colon 18-French coude-tipped catheter may be helpful as an initial
with bladder invasion and moderate left hydroure- intervention.
teronephrosis; colonoscopy with biopsy confirms It is essential to adequately lubricate the catheter and
sigmoid colon adenocarcinoma. insert the catheter past the point at which urine is returned
into catheter tubing prior to inflating the balloon. In men, it
is recommended to grasp the penis firmly, hold it on stretch
CASE MANAGEMENT (toward the ceiling), and advance the Foley to the point
at which the balloon inflation port projects off the side of
The patient undergoes en bloc resection of the tumor the catheter (“hubbed”). This ensures that the balloon of the
via open sigmoid colectomy with distal ureterectomy catheter has passed the prostatic urethra and entered the
and partial cystectomy. Four centimeters of left bladder prior to balloon inflation. Men with a tightly phi-
ureter is removed along with the left lateral bladder motic foreskin not allowing visualization of the glans may
wall. Reconstruction of the left ureter and bladder is require gentle dilation of the phimotic band with a hemostat,
accomplished with a Boari flap, psoas hitch, and ure- or urologic consultation. It may be difficult to locate the ure-
teral reimplantation over a JJ stent. A drain is placed thral meatus of the obese female or in a woman with a poste-
in the region of the ureteroneocystotomy. riorly located meatus (postradiation for anal squamous cell
The patient progresses well postoperatively with carcinoma [SCC]). An often-employed method of dealing
low drain output and is removed prior to discharge. A with difficult female anatomy is to block the anterior vaginal
cystogram performed 10 days postoperatively shows introitus with a finger, guiding the catheter ventrally toward
no evidence of leak from the bladder repair; his Foley the meatus. Intravesical placement of the catheter is con-
catheter is removed at this time, and 3 weeks later firmed with return of urine; if necessary, a Crede maneuver
the JJ stent is removed. Six months postoperatively, a (manual pressure on the lower abdomen) can be performed.
renal ultrasound shows no evidence of hydronephro- Patient dehydration secondary to nothing-by-mouth sta-
sis, and his creatinine is at his preoperative baseline. tus or bowel preparation can make it difficult to deter-
mine proper placement if one only looks for urine return.
If the entire catheter has been inserted and no urine has
been returned, gentle irrigation of the catheter with a 60 cc
URETHRAL INJURIES catheter-tipped syringe prior to balloon inflation is recom-
mended. If the catheter is in appropriate position, the cath-
The most common urologic injury in surgery is the trau- eter should easily irrigate and return fluid with aspiration. If
matic Foley catheter placement, often caused by premature irrigant goes in easily but does not return, a second syringe
395
396 Prevention and management of urologic complications after colorectal surgery
of saline may be injected and aspirated, as the bladder may Sequelae from treatment and/or intraoperative injury can
be inadequately distended to allow for aspiration. If aspira- include delayed urethral stricture with difficult voiding
tion is not easily accomplished after 120 cc has been injected, and bladder outlet obstruction. A retrograde urethrogram
the balloon should not be inflated. Intraoperative urologic will confirm the presence of a urethral stricture but must
consultation for evaluation and Foley catheter placement be done in the bilateral oblique as well as anterior-posterior
can be performed if the above measures fail. views. A retrograde urethrogram (RUG) can be performed
A common algorithm employed by urologists for dif- by affixing a 14-gauge angiocatheter to a 60 cc syringe filled
ficult Foley catheter placement includes first attempting a with standard water-soluble contrast. A RUG should be per-
standard 16-French Foley. If the patient is male and there formed around an indwelling catheter if already in place. If
is suspicion for prostatic hypertrophy, an 18-French coude- a radiographic enema is performed, water-soluble contrast
tipped catheter is often employed next. For nonurologists, is preferred, as it does not form concretions in the blad-
this may seem to be counterintuitive. The larger catheter der. Spontaneous closure of a urinary fistula is rare, but a
provides more rigidity to pass the enlarged prostate. trial of conservative urinary diversion (Foley catheter) for
Cystoscopy can aid in catheter placement over a wire or ure- 4–6 weeks for low-grade fistulas is recommended.
thral dilation, if necessary. If cystoscopy is unsuccessful, a Urinary fistulas are staged according to location, size,
suprapubic tube may be necessary and can be accomplished and patient’s history (1):
open or percutaneously.
Artificial urinary sphincters (AUSs) must be deactivated ●● Stage 1: Low (less than 4 cm from the anal verge and
prior to insertion of a Foley catheter. Deactivation is a dif- nonirradiated)
ferent mechanism than the normal operating “on” and “off” ●● Stage 2: High (greater than 4 cm for the anal verge and
cycling. It is the author’s experience that most patients do nonirradiated)
not know how to deactivate their AUS beyond the normal ●● Stage 3: Small (less than 2 cm irradiated fistula)
cycling mode. This deactivation must be performed prior ●● Stage 4: Large (more than 2 cm irradiated fistula)
to placement of a Foley catheter. Either a device represen- ●● Stage 5: Large (ischial decubitus fistula)
tative or a urologist can deactivate the sphincter preopera-
tively as an outpatient or on the day of surgery. A 12-French Enteric diversion by means of a diverting colostomy or
Foley catheter can then be placed with lubrication and care ileostomy is recommended for stages 3–5. The choices for
to ensure placement in the bladder prior to inflation of the repair are diverse and depend on local tissue integrity and
balloon. Failure to deactivate the AUS can result in ero- staging. A suprapubic catheter can be placed at the time of
sion of the urinary sphincter by means of pressure necro- repair in addition to a Foley catheter for maximal drain-
sis between the Foley catheter and the sphincter device. age (2). Transanal rectal flap advancement can be used for
Removal of the catheter should be done in standard fash- stage 1 fistulae or in combination with other techniques
ion postoperatively. The patient should be able to reactivate for higher-stage fistulae (3). Other techniques described
their AUS by cycling it as they normally would. If unable include:
to obtain urologic consultation prior to or intraoperatively,
a suprapubic catheter can be placed either by open or per- ●● Transanal-transsphincteric approach (dorsal lithotomy
cutaneous methods (preferably under ultrasound guidance anterior sphincterotomy) (4)
to avoid components of AUS). Care must be taken to avoid ●● York Mason/transsphincteric with rectal advancement
intraabdominal components, which are normally placed in flap (2,5,6) (jackknife posterior sphincterotomy)
the retropubic space. An inflatable penile prosthesis (IPP) ●● Perineal approach (jackknife or dorsal lithotomy) (7,8)
should not pose any additional difficulty in placing a ure- ●● Gracilis and rectus abdominus flaps (9,10)
thral catheter if lubrication and the aforementioned guide-
lines are adhered to. Surgical selection is based on fistula stage and the expe-
Urethral injuries are associated with extensive rectal rience of the reconstructive surgeon. Higher-stage fistulas
neoplasm or any inflammatory processes that alter surgi- and recurrences normally require regional flaps and pos-
cal planes, including pelvic radiation. Urethral injuries are sibly even urinary diversion (11). Outcomes for surgically
usually identified at the time of surgery secondary to iden- repaired rectourethral fistulas are overall favorable, with
tification of the indwelling Foley catheter. Repair of a small recurrences mostly dependent on stage and appropriate
urethral laceration can be performed with absorbable 3-0 or choice in initial surgical treatment. Success rates vary from
4-0 synthetic absorbable suture (SAS) on a tapered needle greater than 90% for low-grade fistulas to 70% for higher-
and delayed Foley catheter removal. If the patient has had grade fistulas (1–11). A retrograde urethrogram around
prior radiation or there is poor tissue composition, place- a Foley catheter (“pericatheter RUG”) at 4–6 weeks post-
ment of either an omental flap or local tissue flap to support operatively can be performed prior to urethral catheter
coverage of the repair is recommended. Injuries not identi- removal. Alternatively, a voiding cystourethrogram can be
fied at surgery can present postoperatively as urine drainage performed through a suprapubic tube after Foley catheter
per rectum, pneumaturia, or fecaluria if fistula is present. removal.
Bladder injuries 397
The injury can present as drainage from surgical incision,

BLADDER INJURIES increased output from surgical drain, vaginal leakage, ileus,
apparent oliguria, urinary ascites (with increasing potassium,
The location of the bladder within the pelvis and its close blood urea nitrogen, and serum creatinine secondary to reab-
proximity to the sigmoid colon and rectum predisposes sorption of urine through parietal peritoneum in the case
the bladder to injury during surgery of the colon and rec- of an unrecognized intraperitoneal injury), pneumaturia, or
tum. Iatrogenic injuries to the bladder can be staged as fecaluria in the cases of an enterovesical or colovesical fistula.
f ollows (12,14): Delayed urine leaks can be diagnosed radiographically by flu-
oroscopic cystogram or the CT cystogram (15). Traditional
Grade 1: Contusion, intramural hematoma, or partial- cystogram is performed by obtaining a prefilling film, plac-
thickness laceration ing a catheter, and distending the bladder with 200–300 cc of
Grade 2: Extraperitoneal bladder wall laceration <2 cm contrast, obtaining anterior-posterior and lateral films with
Grade 3: Extraperitoneal >2 cm or intraperitoneal <2 cm the bladder distended, and obtaining a postdrainage film. A
bladder laceration CT cystogram should similarly be performed with active ret-
Grade 4: Intraperitoneal bladder wall laceration >2 cm rograde filling of the bladder and adequate distention of the
Grade 5: Intra- or extraperitoneal bladder wall lacera- bladder. Passive opacification from IV contrast or inadequate
tion extending into the bladder neck or trigone (near retrograde distention of the bladder is usually inadequate to
ureteral orifice) identify smaller bladder injuries.
The development of a colovesical or enterovesical fis-
Risk factors for bladder injury include any process that tula is a delayed complication of cystotomy and/or pel-
distorts tissue planes and reduces surgical exposure. This vic radiation and/or inflammatory bowel disease (16,17).
includes adhesions or scarring from prior surgery, radia- Abdominal-pelvic CT scan with oral and/or rectal water-
tion, malignant infiltration, chronic inflammation, or soluble contrast has a greater sensitivity than cystoscopy in
infection. Injuries can be apparent intraoperatively or pres- diagnosing an enterovesical fistula (Figure 41.1). The most
ent in a delayed fashion. Intraoperative identification of the sensitive test to diagnose an enterovesical or colovesical fis-
injury allows for immediate cystorrhaphy usually in a two- tula is the poppy seed test (17). A 1.25-ounce container of
layer fashion (13). In open or robotic surgery, the mucosa poppy seed is mixed into a 12-ounce beverage or a 6-ounce
is closed in a running fashion using a 3-0 SAS followed by serving of yogurt and orally ingested by the patient. Urine is
a seromuscular running 2-0 SAS. The bladder can then be visually inspected for 48 hours, during which time identifi-
irrigated to ensure a watertight closure. In the laparoscopic cation of poppy seeds in the urine is a positive confirmatory
setting, a running one-layer closure is performed using a test for gastrointestinal fistula to the urinary tract. The sen-
2-0 SAS to close all three layers of the bladder. Care must sitivity and specificity in prior studies were 100% (17). This
be taken to ensure closure of the mucosal layer in the lapa- test does not provide anatomical information as in the case
roscopic one-layer technique. Again, the bladder should be of the abdominal-pelvic CT scan, but it is a much more cost-
irrigated to ensure a watertight closure. effective screening test in patients with equivocal symptoms
Repair can also differ depending on the location of the
injury. Anterior and dome injuries can be repaired primarily
as above. Posterior injuries involving the trigone or those near
the ureteral orifices (possible grade 5) dictate a more thor-
ough inspection of the bladder and an assurance of ureteral
integrity prior to closure. This is done through an anterior
cystotomy in the sagittal plane extending down toward the
pubic symphysis. This will allow placement of a Balfour or
Bookwalter self-retaining retractor and placement of bilateral
ureteral open-ended catheters. Giving the patient indigo car-
mine or methylene blue with Lasix can aid in identification
of the ureteral orifices by visualization of blue efflux. Closure
of the posterior bladder injury can then be done from the
bladder lumen, closing the muscular layer first using 2-0 SAS
followed by closure of the mucosal layer using 3-0 SAS. The
anterior cystotomy is then closed as described previously.
In cases where neoadjuvant radiotherapy has been used, an
interposition of omentum or perivesical fascia/peritoneal flap
is prudent to decrease the risk of fistula formation.
A bladder injury will usually manifest in the early
postoperative period, especially after removal of Foley catheter. Figure 41.1 Enterovesical fistula (arrow).
($5 versus more than $600) (17). When using barium con- (24). Of greatest importance to the surgeon is that arterial
trast, it is the authors’ recommendation to empty the blad- branches to the abdominal ureter approach from the medial
der after a fistula is diagnosed, as there have been reports of direction, whereas arterial branches to the pelvic ureter
barium concretions within the bladder. approach from the lateral direction (24). For the abdomi-
nal ureter, these branches originate from the renal artery,
gonadal artery, abdominal aorta, and common iliac artery.
After entering the pelvis, additional small arterial branches
URETERAL INJURIES may arise from the internal iliac artery or its branches, and
also from the middle rectal and vaginal arteries (24).
Injury to the ureter is one of the most common intraopera- The ureter will tend to adhere to the peritoneum dur-
tive urologic injuries in colorectal surgery. The incidence of ing its reflection rather than staying adherent to the psoas
iatrogenic injury to the ureter is reportedly from 1% to 10% muscle and underlying tissue. The ureter can be identified
(18–22). Iatrogenic ureteral injuries are of four types: lac- by visualization and by its peristaltic activity. Gentle pres-
eration, ligation, devascularization, and thermal or energy sure applied to the ureter will frequently cause p eristalsis—
related. Optimal treatment is early recognition and repair of termed the Kelly sign. The right ureter is adjacent to the
any ureteral injury. cecum, terminal ileum, and appendix. The left ureter is
related to the descending and sigmoid colon and their
ANATOMY mesenteries.
Iatrogenic ureteral injuries in colorectal surgery usually PREVENTION

occur in three distinct locations: at the takeoff of the infe-
rior mesenteric artery, where the infundibulopelvic liga- Ureteral catheterization is used to aid in identification of
ment/uterine vessels cross the pelvic brim, and between the ureters and to help identify ureteral injury, but catheters
the lateral rectal ligaments (Figure 41.2) (23). The course of do not prevent ureteral injury. The clinical value of prophy-
the ureter begins posterior to the renal artery around the lactic ureteral catheter placement prior to 162 laparoscopic
vertebral level L3 and continues along the anterior edge segmental left and right colectomies was assessed by Nam
of the psoas muscle. The gonadal vessels cross the ureter et al. There were no complications from placement of ure-
from lateral to medial in this region. The ureter next passes teral catheters (18). Postoperative urinary tract infection was
over the iliac vessels, generally marking the bifurcation of not increased. Total operative time was increased by 11.3
the common iliac into internal and external iliac arteries minutes. The ureteral catheter group included more difficult
Renal
artery
Ovarian artery
Inferior mesenteric artery
Common iliac artery
Internal iliac artery

Lateral rectal ligament
Superior vesical artery
Figure 41.2 Anatomy of the ureter.

Ureteral injuries / Types of injury 399
cases, including patients with Crohn disease and diverticuli- (a) (b)
tis. There were no ureteral injuries in any of the 162 patients
(18). An earlier study deemed ureteral catheterization nec-
essary in 27.5% of patients when assessed in a standardized
retrospective fashion (22). There were four complications,
presumably due to ureteral catheterization, which included
renal colic, oliguria, and one case of anuria attributed to ure-
teral edema after removal of the ureteral catheters (20,25).
Chahin et al. studied lighted ureteral stents/catheters placed
prior to laparoscopic colectomy in 66 patients (20). The most
common complication was self-limiting hematuria in 98.4%
of patients with an average duration of 2.5 days for unilateral (c)
stenting and 3.3 days with bilateral stenting.
It is the authors’ opinion that the choice for ureteral stent-
ing is a surgeon preference and depends on multiple variables,
including complexity of case, anatomy, and experience—
especially with the laparoscopic approach in a hostile abdo-
men. With greater operative experience, iatrogenic ureteral
injury decreases. In a study by Larach et al., the incidence of
conversions due to iatrogenic injuries showed a decline from
7.3% in the early group to 1.4% in the later experience group
(26). Once again, ureteral catheters have not been shown
to decrease ureteral injuries but do aid in identification of
the ureters, and most importantly, of an iatrogenic ureteral
injury. Ureteral catheters can be used to aid in diagnosis of
ureteral injury by retrograde injection of methylene blue Figure 41.3 Ureteroureterostomy. (a) Spatulation of
ureteral margins and placement of running locked sutures.
through the ureteral catheter or visualization of the catheter
Preferred technique. (b) Oblique anastomosis. (c) Spiral
itself. They can also be used to place a retrograde wire under anastomosis.
fluoroscopic guidance for placement of an indwelling ure-
teral double-J stent after identification of an injury.
may appear discolored, lack peristalsis, and may not bleed
at a transected site. The irradiated ureter is especially sus-
TYPES OF INJURY ceptible to this type of injury, as the normal healthy ureter
has numerous collaterals and is very resistant to devascular-
Laceration/transection ization, even with extensive dissection. The anatomy of the
A laceration or transection of the proximal or mid-ureter blood supply to the ureter (as previously described) should
can usually be repaired with primary anastomosis (ure- be known as the surgeon is carrying his or her dissection
teroureterostomy with spatulated ends), ureteral stent, and over the pelvic brim. Postradiation, many delayed ureteral
placement of a closed suction drain in the area of the repair strictures are not necessarily due to an iatrogenic injury but
(Figure 41.3). rather due to “treatment effect” in the surgical field.
Ligation Thermal
If a ligation injury is apparent intraoperatively, the clamp Thermal injuries will usually present in the early postopera-
or tie can be removed followed by ureteral stent placement tive period with either fistula or stricture formation. These
for up to 1 month. The patient should undergo repeat renal injuries are repaired in the same fashion as above depending
imaging at 3 months to assess for hydronephrosis to ensure on the location of the injury. Many laparoscopic surgeons
a ureteral stricture has not developed. If the injury is not use alternatives to monopolar dissectors because of the risk
identified until postoperatively, a retrograde ureterogram of thermal injury and delayed presentation of injuries. Even
and stent placement or percutaneous nephrostomy tube with these newer technologies, collateral tissue damage can
placement may be needed prior to surgical correction. be produced depending on the energy level and duration of
exposure. In animal models, use of the ultrasonic dissector
Devascularization (Ethicon or USSC) at a level of 3 for less than 10 seconds per
burst resulted in little to no collateral tissue damage (27).
A devascularization injury will not be evident intraopera- When using an ultrasonic dissector at levels of 4 or 5, energy
tively and results from the sacrifice of the segmental ure- time should be reduced to less than 5 seconds to prevent col-
teral blood supply. Intraoperatively a devascularized ureter lateral damage due to spread of thermal energy (27).
Location-dependent repair of the iatrogenic

ureteral injury
Repair of the injured ureter does not necessitate open con-
version if a urologist is available with advanced laparoscopic
skills. The basic principles of a ureteral anastomosis are as
follows: a tension-free anastomosis; well-vascularized spat-
ulated ends anastomosed over an indwelling ureteral stent;
use of an absorbable suture material 4-0 or 5-0; and place-
ment of a closed drain near the area of the repair. Do not use
nonabsorbable suture, as stone formation is inherent with
these nonabsorbable materials.
Proximal one-third
The boundaries of the proximal one-third ureter are the ure-
teropelvic junction (level of the kidney) and the pelvic brim
(sacroiliac joint on pelvic X-ray). Repairs of injuries to the
proximal ureter depend on the length of the damaged seg-
ment. Simple spatulated ureteroureterostomy (“U-U”) with
ureteral stent placement is the preferred method of repair if
there is significant length of the uninjured ureter to allow for
a tension-free anastomosis. A nephropexy can be performed
to bring the kidney caudad to decrease tension, if needed. In
cases with long segments of damaged ureters, a bowel inter-
position with tapered ileum or an appendiceal interposition
can be used (Figure 41.4). At specialized centers, autotrans-
plantation with reanastamosis to the iliac vessels and native
more distal ureter can be performed. Figure 41.4 Ureteral replacement by ileum (“ileal ureter”).
Left colon retracted medially. Ileum brought through a
Middle one-third hiatus in the colonic mesentery. Ileal ureter is in retroperi-
toneal position.
The preferred technique for midureteral repair is uretero-
ureterostomy, either laparoscopically or through the open midureter. A flap of the anterior bladder wall is raised in
technique. Care must be taken to maintain a tension-free a rectangular fashion and affixed to the psoas muscle in
anastomosis. This can usually be accomplished with a psoas the same fashion as a psoas hitch. The ureter is tunneled
hitch (Figure 41.5). The bladder is mobilized by ligating through the most proximal portion of the flap, and a neo-
the superior vesical pedicle on the contralateral side of the orifice is created as previously described. The bladder flap
injury. It is prudent to locate the contralateral ureter and is then tubularized and closed in a two-layer fashion using
ensure its integrity prior to this maneuver. The bladder can running 3-0 SAS to close the mucosa, followed by closure
then be opened through an anterior cystotomy and then of the seromuscular layer using 2-0 SAS (Figure 41.6).
secured to the psoas muscle and tendon using several 0-0 Prolonged bladder drainage with a Foley catheter for 7–14
SAS sutures through the seromuscular layer of the bladder. days is warranted.
Care must be taken not to include the genitofemoral nerve, The final option is the transureteroureterostomy. The
which is located within the belly of the psoas muscle. Suture surgeon tunnels the injured ureter under the posterior
should be placed in a linear fashion in line with the fascicles peritoneum overlying the great vessels. This allows a spatu-
of the muscle. The ureter can then be tunneled by passing a lated end-to-side anastomosis of the injured ureter to the
clamp from the lumen of the bladder through all layers of patient’s contralateral native uninjured ureter (Figure 41.7).
the bladder and then withdrawn with the distal aspect of the Contraindications include distal ureteral obstruction in
proximal salvaged ureter. The ureter should then be widely the uninjured ureter or a history of nephrolithiasis (relative
spatulated, and interrupted mucosal stitches (4-0 SAS) contraindication) (48).
should be used circumferentially to create the neo-orifice. A
ureteral stent can also be placed. The anterior cystotomy is Distal one-third
then closed as previously described. A closed suction drain
and Foley catheter are then left in place. The procedure of choice for the lower one-third ureteral
The Boari flap is another effective yet more complex injury is the ureteroneocystotomy (ureteral reimplant).
method for replacing an extensive loss of the distal and This may be accomplished primarily for very distal ureteral
Renal injuries 401
injuries or may require a psoas hitch or Boari flap for

patients with small capacity bladders and injuries near the
iliac vessels (24).
Delayed recognition
Ureteral injury may not be recognized at the time of sur-
gery, especially if ureteral catheters were not placed. Urine
leaks often present as urinoma, azotemia, ileus, sepsis, high
drain output, or flank pain. Cross-sectional imaging, espe-
cially with a delayed contrast phase, may be helpful in iden-
tifying a urine leak or ureteral obstruction. If a urine leak
is suspected due to high drain output, the drain fluid may
be sent for a creatinine level; results two to five times that
of serum (or higher) are indicative of the presence of urine.
Cystoscopy with retrograde urography is the most sensi-
tive method of identifying a ureteral injury and demon-
strates the exact location of a ureteral injury and also allows
attempted retrograde stent placement. If a stent cannot
successfully be placed retrograde, either due to stricture/
ligation or a significant ureteral disruption, a percutaneous
nephrostomy tube can be placed. Repair of an injury may be
performed at any point once the patient is stable for a return
to the operating room.
RENAL INJURIES
Direct renal injury is a rare occurrence in colorectal surgery.

McAnich et al. have reported that 90% of renal injuries can
be managed without nephrectomy (28). Though this work
Figure 41.5 Psoas bladder hitch. Mobilized bladder being does not address iatrogenic injuries, the principle of renal
anchored to psoas muscle, and the ureter is reimplanted. salvage should be adhered to. Every attempt to evaluate the
(a) (c)
(b)
Figure 41.6 Boari or bladder flap procedure. (a) Creation of tapered bladder flap, based posteriorly. (b) Submucosal ure-
teral reimplantation. (c) Closure of bladder flap.
and the ensuing blood loss is able to be managed by pressure

and hemostatic agents alone, a ureteral stent and Foley catheter
can be placed from below and the area drained with a closed
suction to prevent urinoma formation. Conservative manage-
ment is optimal as renorrhaphy, and exploration can lead to
unnecessary nephrectomy.
The left renal vessels are located posterior to the mes-
entery between the inferior mesenteric vein and aorta. If
a major vascular injury occurs and the patient’s intraop-
erative condition permits, every attempt should be made to
reestablish vascular integrity.
BLADDER DYSFUNCTION
The reported incidence of difficulty in reestablishing mic-

turition ranges from 15% to 25% after low anterior resec-
tion and up to 50% after abdominoperineal resection
(31). A thorough understanding of the neuroanatomy of
the pelvis and the technique of total mesorectal excision
(TME) and autonomic nerve preservation (ANP) can enable
both local tumor control and preservation of autonomic
nerve structures, thus reducing the risk of urogenital dys-
function (34,35). Favorable oncologic outcomes have been
reported for these nerve-sparing techniques (35–39). APR,
when performed in accordance with the principles of TME
and ANP, ensures the greatest likelihood of resecting all
regional disease while preserving both urinary and sexual
function (39). Locally advanced tumors and preoperative
chemotherapy and radiation can make identification of the
autonomic nerves and plexus more difficult and sometimes
Figure 41.7 Transureteroureterostomy. Right-to-left,
showing retroperitoneal tunnel anterior to the great impossible (34). The most common sequela from autonomic
vessels. nerve damage during surgery of the colon and rectum is
detrusor denervation and areflexia. This normally requires
clean intermittent catheterization, Foley catheter place-
extent of the injury as well as assess the entire genitourinary ment, or suprapubic tube placement depending on the over-
tract should be done prior to undertaking repair. A one- all dexterity and functional status of the patient. Damage to
shot intravenous pyelogram (IVP) can confirm contralat- the pudendal nerve or its branches from Alcock canal can
eral renal function. This can be done by giving the patient result in weakening of the striated urinary sphincter with
2 mL of contrast per kg IV up to a maximum of 150 mL. An resultant stress urinary incontinence and intrinsic sphinc-
on-the-table KUB is then done 10 minutes later. Simple pal- ter deficiency (rare).
pation of the contralateral kidney does not ensure function. Detrusor function (bladder contractility) is predomi-
The literature is full of anomalous solitary kidneys that were nantly mediated by the parasympathetic nervous system,
removed, necessitating dialysis or transplantation (29,30). namely, the pelvic nerve (33). These parasympathetic fibers
Pelvic kidneys have an anomalous blood supply generally originate from the spinal cord at the S2-S4 level. Pelvic nerve
arising from multiple arteries along the aorta and iliac ves- branches are redundant within the pelvis. The main trunks
sels. Ten percent are solitary and may easily be taken for to the bladder and proximal urethra course in the visceral
a pelvic mass, as they are not reniform and have a discoid pelvic fascia, also called the posterior endopelvic fascia (33).
shape (23). If calyceal or renal pelvis injury is suspected, a These preganglionic autonomic fibers course alongside the
retrograde pyelogram can be performed. superior vesical vasculature to synapse with postganglionic
Once the injury is well defined, repair can be decided. autonomic fibers within the bladder wall. Multiple pelvic
Minor renal lacerations or penetrating injuries may be repaired preganglionic nerves pass laterally from the pelvic floor
primarily with absorbable sutures and re-retroperitonealized over the rectal fascia investments en route medially to the
with perinephric fat, omentum, or hemostatic materials. Hilar bladder (Figure 41.8) (33).
control is paramount if an attempt at repair is to be performed. Sympathetic innervation to the bladder arises at the
If the injury is to the collecting system or renal parenchyma level of L2-L4 with a presynaptic fiber to the sympathetic
Bladder dysfunction 403
8
9
Afferent 10
Efferent 11
12
1
2
3
4
5 Parasympathetic
1
Hypogastric nerve nucleus
2
3
4
Pelvic nerve 5 Pudendal
nucleus
External sphincter Pudendal nerve
Figure 41.8 Innervation of lower urinary tract.
ganglion adjacent to the spinal cord. Synapse occurs in the Management of the postoperative patient with urinary
ganglion, and a long postganglionic fiber travels through retention varies widely between urologists. Of paramount
the pelvis to innervate the bladder. Through different end importance in treating urinary retention is accurate docu-
receptors located within the bladder, the sympathetic com- mentation of voided volumes and postvoid residuals, deter-
ponent of the autonomic nervous system helps to cause mined either by ultrasound (bladder scan) or initial urine
relaxation of the bladder body (compliance for storage) and return after catheterization. A lack of voiding may be due
contraction of the trigone and bladder neck at resting/stor- to oliguria rather than retention. After a patient’s Foley
age states. catheter is removed, a period of 6–8 hours is usually given
Somatic motor innervation to the striated pelvic floor to the patient to allow for spontaneous voiding. A bladder
musculature and sphincter arises from the S2-S4 level and scan may be performed if the patient has not urinated after
travels via the pudendal nerve through Alcock canal. The this trial, or sooner if the patient feels an urge to void and is
perineal branches of the pudendal nerve follow the perineal unable to. It is the authors’ practice to leave a Foley catheter
artery into the superficial pouch to supply the ischiocaver- indwelling for a period of 48–72 hours after a failed voiding
nosus, bulbospongiosus, and transverse perinei muscles. trial with a volume greater than 400 mL to allow the bladder
Some branches continue anteriorly to supply sensation to to decompress and recover from distention. A voiding trial
the posterior scrotum and perineum. Additional perineal may again be attempted in the same fashion afterward.
branches pass deep to the perineal membrane to supply the Repeated voiding trial failure may be an indication of
levator ani and striated urethral sphincter (40). bladder dysfunction. Management includes teaching clean
In the study by Junginger on TME, identification of the intermittent catheterization (CIC) or leaving an indwelling
pelvic autonomic nerves was complete in 72%, partial iden- Foley and having the patient return for full urodynamic eval-
tification in 10.7%, and not at all in 17.3% of patients (34). uation around 2–3 months postoperatively. Urodynamics
Univariate analysis showed that the case number (experi- can be a combination of pressure/flow studies with electro-
ence), gender (males greater than females), and T-stage myography tracings under fluoroscopic observation and,
(T1-2 versus T3-4) exerted an independent influence on sometimes, urethral pressure profiling. It may take up to 6
the achievement of complete pelvic nerve identification. months for bladder function to return to its new baseline, and
In this series of 150 patients with adenocarcinoma of the CIC may be a lifelong therapy. CIC is performed with a 12-
rectum, identification and preservation of the autonomic to 14-French low-friction catheter every 4–6 hours, and the
nerves were achieved in a majority of patients and led to duration can be adjusted based on the storage pressures and
the prevention of urinary dysfunction (4.5% versus 38.5%, bladder capacity at the time of urodynamic evaluation. There
p < 0.001) (34). are no drugs (including Bethanechol) with acceptable phar-
A Swiss group investigated the incidence and risk fac- macokinetics and side-effect profiles that have been shown to
tors for postoperative urinary retention in 513 consecu- clinically increase/improve contractility in the bladder.
tive patients undergoing colorectal surgery with early In a meta-analysis, Branagan et al. reviewed the colorec-
Foley removal due to an enhanced recovery pathway (49). tal surgery literature on suprapubic catheter placement fol-
Fourteen percent of these patients experienced urinary lowed by voiding trial versus urethral catheter placement
retention postoperatively; male gender and postoperative and standard trial of voiding postoperatively (31). They
thoracic epidural anesthesia were independent risk factors found favorable results for the suprapubic catheter in terms
for retention. Complications of urinary retention included of incidence of urinary tract infection, and a shorter mag-
decreased mobilization, increased pain, and longer dura- nitude and duration of pain and discomfort. The ability to
tion of IV hydration. simply clamp and unclamp the suprapubic catheter makes
management and voiding trials relatively simple, especially and were seldom referred or treated for symptoms postop-
in patients unable to perform CIC or those at especially eratively. Sexual dysfunction should be discussed with rec-
high risk for postoperative bladder dysfunction. Suprapubic tal cancer patients, and when appropriate, efforts to prevent
catheters are particularly useful if autonomic nerves have to and treat sexual dysfunction should be instituted (42).
be removed during radical pelvic surgery, because normal In a study by Nam et al. of patients undergoing TME
voiding may be difficult to reestablish and may take several and ANP for rectal carcinoma, factors that most affected
months to recover. In the select patient with voiding dys- postoperative sexual dysfunction were age older than 60
function and delayed recovery, suprapubic catheter place- (sexual desire, p = 0.019), time period within 6 months
ment results in less morbidity and patient discomfort than of surgery (erectile function, p = 0.04), and lower rectal
urethral catheterization (32). cancer (erectile function p = 0.02) (43). In the urologic lit-
erature, penile “rehabilitation” is started at approximately
1 month postoperatively with evidence suggesting that lack
of natural erections during this period of time produces
SEXUAL DYSFUNCTION cavernosal hypoxia (44). Prolonged periods of cavernosal
hypoxia induce fibrosis, which later increases the incidence
In the urologic community, an emphasis on postoperative of venous leak and thus potentiates long-term or permanent
sexual function has arisen from studies by Walsh on the erectile dysfunction.
anatomic retropubic prostatectomy with preservation of In consultation with a urologist, sexual dysfunction in
the neurovascular bundles that contribute to erectile func- the man can be treated with many different modalities.
tion (41). Most recently, postoperative penile rehabilitation For erectile dysfunction, oral phosphodiesterase inhibi-
is being performed in multiple settings with a theoretical tors, intraurethral vasoactive suppositories, intracaverno-
benefit of reducing the time of neuropraxia to the penis sal injections, vacuum erection devices, and implantable
and prevention of apoptosis-induced atrophy. Although no devices are all options. For ejaculatory dysfunction in
standardization exists with these rehabilitation programs, a patient desiring pregnancy, semen may be collected
patients are very interested, and at the authors’ institution from the bladder in the case of retrograde ejaculation.
this is discussed preoperatively. Sexual dysfunction has long Sympathomimetic agents may also be used. For refractory
been associated with rectal surgery in both male and female cases, electrovibratory ejaculation can be performed at spe-
patients. In male patients, erectile dysfunction is reported in cialized centers. It is important to discuss sexual function
5%–65% of patients, and ejaculatory dysfunction is reported with the patient both pre- and postoperatively, as there are
in 14%–69% (43). Damage to the sacral splanchnic nerve many therapeutic options that have been shown to be very
(parasympathetic) or the hypogastric nerve (sympathetic) satisfactory for both partners.
during surgery is the proposed mechanism of injury (43).
Sexual dysfunction is a broad term that encompasses fail-
ure of arousal, erection, orgasm, ejaculation, and emission.
Complaints from patients after radical pelvic surgery are usu- ARTIFICIAL DEVICES
ally mixed. Erection is parasympathetically mediated and is
governed by impulses traveling along the nervi ergentes (S2- Thousands of AUS and IPP have been implanted worldwide
S4) (41). The pelvic plexus is located retroperitoneally on the for the treatment of stress urinary incontinence and erectile
lateral surface of the rectum 5–11 cm from the anal verge dysfunction, respectively (Figure 41.9). The IPP has one to
with its midpoint located at the tip of the seminal vesicles. three components, while the AUS has three components.
The preganglionic fibers from the nervi ergentes coalesce The three-component systems will have a reservoir, pump,
on the pelvic wall with contributions from the sympathetic and cuff or prosthesis that is interconnected with reinforced
fibers from the hypogastric plexus (T10-L4). Damage to the tubing. These devices are silicone but develop a capsule
sympathetic plexus will result in problems with ejaculation, around them after implantation. The reservoir is typically
including retrograde ejaculation or anejaculation. found suprapubically in the space of Retzius. One should
In a study by Henderson et al., 81 women and 99 men make every attempt to refrain from entering this capsule
who had undergone curative rectal cancer surgery were and to prevent contamination of these silicone devices. If
given a validated sexual function questionnaire (42). Thirty- contamination occurs, either device removal or salvage
two percent of women and 50% of men were sexually active therapy with copious antibiotic irrigation is recommended.
compared with 61% and 91% preoperatively. Twenty-nine A urologist should be consulted if available. The risk of
percent of women and 49% of men reported that “surgery device contamination, postoperative infection, and damage
made their sexual lives worse.” Specific sexual problems to the tubing necessitating device removal or reoperation
in women were libido 41%, arousal 29%, lubrication 56%, should be discussed with the patient preoperatively. It is the
orgasm 35%, and dyspareunia 46%. In men complaints were authors’ practice to be very conservative in patients with
impotence/erectile dysfunction 84%, libido 47%, orgasm AUS, and we recommend all patients have their device deac-
difficulty 41%, and ejaculation difficulties 43%. Patients tivated by a urologist familiar with the AUS prior to place-
seldom remembered discussing sexual risks preoperatively ment of a urethral catheter. There are numerous reports of
References 405
6. Crippa A et al. Clinics 2007;62:699–704.

7. Yousseff AH et al. J Urol 1999;161:1498–500.
8. Visser BC et al. JACS 2006;195:138–43.
9. Zmora O et al. Dis Colon Rectum. 2006;49(9):1316–21.
10. Bruce RG et al. J Urol 2000;163:1212–5.
11. Elliott SP et al. J Urol 2006;176:2508–13.
12. Moore EE et al. J Trauma 1992;33:337–9.
a
13. Armenakas NA et al. JACS 2004;198:78–82.
14. Van Goor H. Colorectal Dis 2007;9(Suppl 2):25–34.
15. Deck AJ et al. J Urol 2000;164:43–6.
16. Jarrett TW, Vaughan Jr ED. J Urol 1995;153:44–6.
17. Kwon EO et al. J Urol 2008;179:1425–7.
18. Nam YS, Wexner SD. J Korean Med Sci 2002;17:633–5.
19. Larach SW, Gallagher JT. Semin Surg Oncol 2000;18:
265–8.
b 20. Chahin F et al. JSLS 2002;6:49–52.
21. Scala A et al. Colorectal Dis 2007;9:701–5.
22. Fry et al. Arch Surg. 1983;118:454–7.
23. Perlmutter AD et al. Anomalies of the upper urinary
tract. In Harrison JH, Gittees RF, Perlmutter AD et al.
(eds.) Campbell’s Urology, 4th Edition. Philadelphia,
c
PA: WB Saunders, 1979, pp. 1309–98.
24. Anderson KJ et al. Surgical anatomy of the retroperi-
toneum, adrenals, kidneys, and ureters. In Campbell-
Figure 41.9 Artificial urinary sphincter (AVS-800); Walsh Urology, 9th Edition. New York, NY: Elsevier,
American Medical Systems Inc, Minnetonka, Minnesota. 2007, pp. 34–7.
(a) Reservoir. (b) Cuff. (c) Pump. 25. Kyzer S, Gordon PH. Am Surg 1994;60:212–6.
26. Larach SW et al. Dis Colon Rectum 1997;40:592–6.
patients “turning off” their own AUS, when in reality, they 27. Emam TA, Cuschieri A. Ann Surg 2003;237:186–91.
only cycle them followed by urethral catheterization at the 28. McAninch JW et al. J Urol 1991;145:932–7.
time of surgery and resulting device erosion. This is a medi- 29. Granat M et al. Am J Obstet Gynecol 1980;138:233–5.
colegal issue that usually can be averted with a preoperative 30. Zusmer NR et al. Rev Interam Radiol 1980;5:95–6.
consultation with a urologist. 31. Branagan GW, Moran BJ. Dis Colon Rectum 2002;
The U.S. Food and Drug Administration approved sacral 45:1104–8.
neuromodulator is the Interstim device manufactured by 32. Chaudhri S et al. Dis Colon Rectum 2006;49:1066–70.
Medtronic Corp. (45). It is approved for use in patients with 33. Hollabaugh Jr RS et al. Dis Colon Rectum 2000;43:
refractory urgency and frequency or nonobstructive, non- 1390–7.
neurogenic urinary retention. A tined lead is placed through 34. Junginger T et al. Dis Colon Rectum 2003;46:621–8.
the S3 foramen, and an implanted generator is placed in a 35. Shirouzu K et al. Dis Colon Rectum 2004;47:1442–7.
pocket created in the gluteal area/upper hip. The manufac- 36. Saito N et al. World J Surg 1999;23:1062–8.
turer recommends against using electrocautery near the 37. Yamakoshi H et al. Dis Colon Rectum 1997;40:1079–84.
generator and not performing magnetic resonance imag- 38. Moriya Y et al. World J Surg 1997;21:728–32.
ing (MRI) on any patients with the Interstim device. It is 39. Enker WE et al. World J Surg 1997;21:715–20.
the authors’ practice to turn off the device with a Medtronic 40. Brooks. Anatomy for the Lower Urinary Tract and
supplied magnet prior to any radical pelvic operation. In Male Genitalia. 2;p69. Campbell-Walsh Urology,
thin patients, appropriate padding must be applied to the 9th Edition. New York, NY: Elsevier, 2007.
area of the implanted generator. MRI is contraindicated, 41. Walsh PC, Schlegel PN. Ann Surg 1988;208:391–400.
although there has been at least one study to show deactiva- 42. Hendren SK et al. Ann Surg 2005;242:212–23.
tion of the device prior to MRI to be safe (46,47). 43. Kim NK et al. Dis Colon Rectum 2002;45:1178–85.
44. Raina R et al. Int J Impot Res 2008;20:121–6.
REFERENCES 45. Interstim Device Trademarked by Medtronic Corp.
46. Holley et al. Presentation at SESAUA Annual
1. Rivera R et al. J Urol 2007;177:586–8. Meeting, Orlando, Florida, USA, 2007.
2. Fengler SA, Abcarian H. Am J Surg 1997;173:213–7. 47. Elkelini MS, Hassouna MM. Eur Urol 2006;50:311–6.
3. Dreznik Z et al. Colorectal Dis 2003;5:53–5. 48. Delacroix SE, Winters JC. Clin Colon Rectal Surg
4. Culkin DJ, Ramsey CE. J Urol 2003;169:2181–3. 2010;23(2):104–12.
5. Mason AY. Proc R Soc Med 1970;63(suppl.):91–4. 49. Grass F et al. J Surg Res 2017;207:70–6.
Index
A ACS NSQIP, see American College of ALTA, see Aluminum potassium sulfate
Surgeons National Surgical and tannic acid
AAST, see American Association for the
Quality Improvement Program Altmeier procedure, 225–227;
Surgery of Trauma
Acute diverticulitis, 109, 110 see also Perineal
ABCRS, see American Board of Colon and
Acute respiratory distress syndrome proctosigmoidectomy; Rectal
Rectal Surgeons
(ARDS), 36 prolapse operative repairs
Abdomen imaging, 104–105; see also
Adaptive statistical iterative reconstruction Aluminum potassium sulfate and tannic
Colorectal imaging studies
(ASIR), 108 acid (ALTA), 173
Abdominoperineal resection (APR), 130,
Adenoma detection rate (ADR), 158 Alvimopan, 84, 371
151, 255, 310
Adhesive disease, 54; see also Ambulatory anorectal procedures, 1
abscess formation, 276
Intraoperative challenges American Association for the Surgery of
extralevator operative technique, 275
abdominal and pelvic adhesions, 54 Trauma (AAST), 384
history, 271
bowel adhesion grading, 54 American Board of Colon and Rectal
intraoperative hemorrhage, 274
bowel obstructions, 55, 56 Surgeons (ABCRS), 156
laparoscopic abdominal operative
lack of adhesions, 55, 56 American College of Cardiology
technique, 272–273
Adjunctive radiation treatment (ACC), 2
leg positioning for, 272 acute adverse effects, 316–317 American College of Chest Physicians
methods of closure, 275–276 benefit of, 312–316 (ACCP), 29
nerve supply to rectum, 274 chronic late adverse effects of, American College of Gastroenterology
nonhealing wound and perineal sinus, 318–320 (ACG), 124, 348
276–277 complications after preoperative American College of Surgeons
open abdominal operative irradiation, 317–318 National Surgical Quality
technique, 272 postoperative adjuvant radiation Improvement Program (ACS
patient preparation and positioning, studies, 313 NSQIP), 2, 16
271–272 rectal injury and treatment, 320–321 surgical risk calculator, 10–11
perineal hernia and evisceration, 277 Adjuvant; see also Chemotherapy; American Heart Association (AHA), 2
preservation of sexual and urinary Rectal cancer transanal American Joint Committee on Cancer,
function, 273–274 approaches see AJCC
robotic abdominal operative chemotherapy, 306–307 American Society for Gastrointestinal
technique, 273 radiation therapy, 269–270 Endoscopy (ASGE), 124
standard perineal operative technique, ADR, see Adenoma detection rate American Society of Anesthesiologists
274–275 Advancement flap procedures, 184; see also (ASA), 4, 18, 80, 124
thumbtack occlusion of bleeding Anal fistula American Society of Colon and Rectal
basivertebral vein, 274 AFP, see Anal Fistula Plug Surgeons (ASCRS), 16, 42, 158,
ABS, see Artificial bowel sphincter AHA, see American Heart Association 189, 266
ACC, see American College of Cardiology AIDS, see Acquired immune deficiency Amitriptyline, 213
ACC/AHA cardiac risk assessment syndrome Anal; see also Colorectal trauma; Perianal
algorithm, 3 AJCC (American Joint Committee on Crohn disease; Rectal prolapse
ACCP, see American College of Chest Cancer), 306 operative repairs
Physicians AL, see Anastomotic leak block, 20
ACE, see Antegrade colonic enema ALBIOS, see Albumin Italian Outcomes dilation, 190
Acetaminophen, 24 Sepsis encirclement, 228
Acne inversa, see Hidradenitis suppurativa Albumin in sepsis resuscitation, 34 manometry, 367
Acquired immune deficiency syndrome Albumin Italian Outcomes Sepsis sphincter injury, 392–393
(AIDS), 152 (ALBIOS), 34 stenosis, 343–344
407
408 Index
Anal condyloma, 149 ileal pouch-anal, 133 Anocutaneous flaps, 192; see also Anal
giant condyloma acuminatum, 150 ileorectal, 326 fissure
treatment, 150–151 intestinal, 39 Anoikis, 296
Anal encirclement, 228 Kono side-to-side, 339 Anorectal; see also Colorectal trauma
Anal fissure, 188, 194 lower, 256 foreign bodies, 391–392
advancement flaps, 192 mucosectomy with hand-sewn, 327 operations, 18
anal dilation, 190 primary resection with, 239 spaces, 178
anocutaneous flaps, 192 straight coloanal, 263 STD, 153
botulinum toxin, 190 types, 338–339 Anorectal dysfunction, 93
calcium-channel blockers, 190 Anastomotic hemorrhage, 46–47; see also treatment, 152
chronic, 189 Anastomotic leak; Intraoperative Anorectal melanoma, 151
classification, 188–189 anastomosis clinical presentation, 151
Crohn disease, 193 Anastomotic leak (AL), 66; see also etiology and diagnosis, 152
diamond flap, 193 Cutaneous fistula with AL; Anorectal physiology (ARP), 93; see also
diet and lifestyle modification, 189 Postoperative anastomotic Anorectal physiology testing
house advancement flap, 193 complications limitations
without hypertonicity, 194 antibiotics, 68–69 lab, 94
lateral internal sphincterotomy, colorectal surgery, 68 Anorectal physiology testing limitations,
190–192 contrast extravasation from left-sided, 93, 102–103
location and etiologies of, 189 68 anorectal physiology lab, 94
nitrates, 189–190 damage control philosophy, 69 biofeedback for fecal incontinence, 99,
nonoperative management, 189 diagnosis, 67–68 100
RCT of nitrates vs. surgical drainage, 69 corpus cavernosum, 94
sphincterotomy, 191 endoscopic techniques, 69–70 defecography, 100–101
sexually transmitted diseases, 194 etiologies, 67 electromyography, 97, 98
special situations, 193 imaging, 68 manometry, 94, 96–97
surgical management, 190 management, 68, 69 MRI in constipation, 102
V-Y anoplasty, 193 preventative measures, 67 physiology of fecal continence, 93
Y-V anoplasty, 192 surgery, 69 pudendal nerve terminal motor latency,
Anal fistula, 177; see also Perirectal abscess treatment algorithm for surgical 97, 98
advancement flap procedures, 184 management of, 70 rectal capacity and sensation, 96
anatomical considerations, 177–179 Anastomotic stricture, 71–72; see also rectoanal inhibitory reflex, 96
ano-cutaneous advancement flap, Postoperative anastomotic sphincter pressure measurement, 94,
184–185 complications 96
anorectal spaces, 178 Anastomotic technique, 41; see also transanal ultrasonography, 98–99
classification, 179 Intraoperative anastomosis ultrasound limitations in incontinence,
drainage tubes in fistulas, 182 Anesthesia, 18; see also Intraoperative 99
endorectal advancement flap, 184, 185 positioning; Surgical Care Anoscopic-assisted stapler insertion, 45; see
fibrin glue, 182 Improvement Project also Intraoperative anastomosis
filac, 186 awareness under, 23–24 Anoscopy, 122; see also Transanal
fistula plug, 182–183 central neuraxial blockade, 21–23 endoscopy
fistulotomy with sphincter general, 18–19 ANP, see Autonomic nerve preservation
reconstruction, 185 ilioinguinal and iliohypogastric nerve Antegrade colonic enema (ACE), 373; see
intersphincteric fistulas, 178 block, 23 also Fecal incontinence operative
ligation of internal fistula tract, 183 intraoperative goal-directed therapy, 24 techniques
management of, 181–182 intraoperative opioid sparing, 24 procedure, 220
stem cells, 184 local, 19–20 Antibiotics, 14
video-assisted anal fistula management, monitored anesthetic care, 20–21 Anticholinergic agents, 212
185–186 postoperative pain management, 24–25 Anti-TNF-α agents, see Antitumor necrosis
Anal Fistula Plug (AFP), 182 regional, 21 factor-alpha agents
Anal transition zone (ATZ), 327 technique for anal block, 20 Antitumor necrosis factor-alpha agents
Anastomosis, 338–339; see also Crohn transversus abdominis plane block, 23 (Anti-TNF-α agents), 7
disease management Anesthesiology, 18 Apixaban, 9
colon resection with, 385–387 Anesthetic drugs, 19 APR, see Abdominoperineal resection
with diversion, 353–354 Angiodysplasia, 349 ARDS, see Acute respiratory distress
end-to-end, 42–44 Ano-cutaneous advancement flap, 184–185; syndrome
esophageal, 66 see also Anal fistula ARP, see Anorectal physiology
Index 409
Artificial bowel sphincter (ABS), 219–220; BUN, see Blood urea nitrogen for colon and rectal cancer, 306
see also Fecal incontinence Buschke-Löwenstein tumor, see Giant future direction in colorectal cancer,
operative techniques condyloma acuminatum 308
Artificial urinary sphincters (AUSs), hyperthermic intraperitoneal, 308
396, 404, 405; see also Urologic in metastatic colorectal cancer, 307–308
C
complications in resectable rectal cancer, 307
ASA, see American Society of CABG, see Coronary artery bypass graft CHF, see Congestive heart failure
Anesthesiologists CAHPS, see Consumer Assessment of Child-Turcotte-Pugh (CTP), 5
ASCRS, see American Society of Colon and Healthcare Providers and Chimney effect, 132
Rectal Surgeons Systems Chlamydia trachomatis infections, 153
ASGE, see American Society for Calcium-channel blockers, 190 Chronic anticoagulation and perioperative
Gastrointestinal Endoscopy Capsule endoscopy, 334 management, 8–9; see also
ASIR, see Adaptive statistical iterative Carcinoembryonic antigen (CEA), 289 Preexisting conditions
reconstruction Cardiac evaluation, 1; see also Preexisting Chronic kidney disease, 5
Aspirin, 9 conditions Chronic obstructive pulmonary disease
ATZ, see Anal transition zone Cardiac risk indices, 2 (COPD), 4, 80, 130, 224
AUSs, see Artificial urinary sphincters Cardiopulmonary resuscitation (CPR), 19 Chronic renal failure, 4
Autonomic nerve preservation (ANP), 402 Cardiovascular (CV), 19 Chronic sinuses, 72; see also Postoperative
disease, 1 anastomotic complications
Care paths, 87, 92 CI, see Confidence interval
B
analgesia, 90–91 CIC, see Clean intermittent catheterization
Balloon expulsion test, 102, 367 benefits of, 87–88 Circumferential radial margin (CRM),
Barron-McGivney rubber band ligator, 173; bowel preparation, 89–90 255, 256; see also Rectal cancer
see also Rubber band ligation colorectal surgery care path, 88 surgery, restorative
Bascom cleft lift, 201; see also Pilonidal early ambulation and use of urinary Cirrhosis, 5
disease catheters, 91 CJP, see Colonic J-pouch
Bascom procedure, 199–200; see also elements of, 87, 88 Clean-contaminated wounds, 33
Pilonidal disease during emergency surgery, 91–92 Clean intermittent catheterization (CIC),
β-Blockers, 3 enhanced recovery protocol at Mayo 403
BHPM, see Bis-(p-hydroxyphenyl)-pyridyl- Clinic, 88 Clean wounds, 33
2-methane feeding, 89 Clinical decision support (CDS), 61
Bicap, see Bipolar cautery fluid management, 90 Clopidogrel, 9
Biofeedback, 99; see also Anorectal patient education, 88–89 Closed/Ferguson procedure, 164,
physiology testing limitations protocols, 87 165; see also Excisional
for constipation, 102 Cathartic colon, 370 hemorrhoidectomy
for fecal incontinence, 99–100 Catheter-associated urinary tract infection Clostridium difficile infection (C-Diff), 80
therapy, 371–372 (CAUTI), 80 CMS, see Centers for Medicare and
training for FI, 213–214 CAUTI, see Catheter-associated urinary Medicaid Services
BIOLIFT plug, 183; see also Ligation of the tract infection CNS, see Central nervous system
intersphincteric tract CBC, see Complete blood count Cocaine, 19; see also Anesthesia
Bipolar cautery (Bicap), 76 CC, see Complete cytoreduction Coliseum technique, 302
BIS, see Bispectral index cCR, see Complete clinical response Colitis, 106
Bis-(p-hydroxyphenyl)-pyridyl-2-methane C-Diff, see Clostridium difficile infection Colon algorithm, 386
(BHPM), 370 CDS, see Clinical decision support Colon and rectal surgery, 48; see also
Bispectral index (BIS), 23 CEA, see Carcinoembryonic antigen Intraoperative anastomosis
Bladder dysfunction, 402–404; see also Cecal volvulus, 105 EHR use in, 64
Urologic complications Centers for Medicare and Medicaid Colon cancer, 132; see also Laparoscopic
Bladder injuries, 397–398; see also Urologic Services (CMS), 61 colonic surgery
complications Central nervous system (CNS), 19 adjuvant chemotherapy in resectable,
Bleeding risk, indicators for, 49 Central neuraxial blockade, 21–23; see also 306–307
Blood urea nitrogen (BUN), 370 Anesthesia chemotherapy, 306
BMI, see Body mass index Cerebrospinal fluid (CSF), 21 Colonic J-pouch (CJP), 263
Body mass index (BMI), 6, 130, 326 cGMP, see Cyclic guanosine Colon injury management, 384, 388; see
Botulinum toxin, 190 monophosphate also Colorectal trauma
Bowen disease, 149 Chemoradiotherapy (CRT), 157, 256 colon algorithm, 386
Brooke ileostomy, 324–325; see also Chemotherapy colostomy closure, 387–388
Proctocolectomy adjuvant, 306–307 controversy of primary repair, 384–385
410 Index
Colon injury management (Continued) sagittal T2-weighted image of pelvis, enterography, 106–108, 109, 110
damage control, 387 114 imaging of peritoneal carcinomatosis,
role of resection with anastomosis, small bowel obstruction, 105 297
385–387 T-staging, 114–118 Computed tomography colonography
Colonoscopy, 124; see also Transanal Colorectal liver metastasis, 292; see also (CTC), 104
endoscopy Rectal cancer, recurrent Computed tomography enterography
optimal effectiveness and quality of, 158 Colorectal metastases, 119; see also Hepatic (CTE), 335
proposed quality indicators for, 159–160 metastatic colorectal cancer Computerized provider order entry
Colorectal anastomosis, 66; see also Colorectal-POSSUM (CR-POSSUM), 12 (CPOE), 61
Postoperative anastomotic Colorectal surgery, 66, 68 Confidence interval (CI), 6, 89, 130
complications care path, 88 Congestive heart failure (CHF), 224
Colorectal Cancer (CRC), 121, 296, quality and outcome measures of, Constipation, 93, 365; see also Anorectal
244, 306 157–158 physiology testing limitations
anastomotic bleeding, 247–248 rectal cancer vs. colon cancer surgery, anal manometry, 367
anastomotic leak, 246–247 157 anatomical evaluation and functional
anastomotic stricture, 247 Colorectal trauma, 380; see also Colon studies, 368
autonomic nerve injury, 249–250 injury management; Rectal injury balloon expulsion test, 367
chemotherapy future direction, 308 management biofeedback for, 102
chemotherapy in metastatic, 307–308 AAST colon and rectal injury scales, biofeedback therapy, 371–372
functional outcomes, 250–251 384 chronic, 365
laparoscopic mobilization of splenic abdominal sonography for trauma, 383 colonic transit studies, 101
flexure, 249 anal sphincter injury, 392–393 complications, 375–376
oncologic outcomes in, 245 anorectal foreign bodies, 391–392 defecography, 368
patient-centered outcomes, 251 bullet tract through retroperitoneum, drugs associated with, 366
pelvic hemorrhage, 248 383 evaluation of, 366–367
preoperative evaluation, 245 colon algorithm, 386 functional evaluation and diagnostic
rectal mobilization demonstrating colon injury diagnosis, 382 studies, 367
autonomic nerves, 250 comparison of diagnostic techniques investigations for, 100
splenic injury, 248–249 for, 382 limitations of defecography in, 101
surgical outcomes in, 245 computerized tomography, 383 medical conditions causing, 366
surgical technique, 245 diagnostic laparoscopy, 383–384 medical treatment of, 368–371
ureteral injuries, 249 diagnostic peritoneal lavage, 382 medications commonly used for, 369
Colorectal imaging studies, 104 etiology, 381 MRI in, 102
of abdomen and pelvis, 104–105 injury scales, 384 pelvic magnetic resonance imaging, 368
acute diverticulitis, 109, 110 perineal injury, 381, 393 radiopaque markers, 367
cecal volvulus, 105 physical examination, 382 recurrence of constipation/outcomes,
coronal orthogonal T2-weighted preoperative assessment, 381–382 376–377
image, 114 primary repair vs. diversion for colon results of total abdominal colectomy
Crohn disease, 107, 108, 109 injury, 385 with ileorectal anastomosis, 372
cross-sectional imaging, 105 rectal algorithm, 389 Rome III diagnostic criteria for, 366
CT angiogram, 111, 112, 113 rectal injury diagnosis, 384 scintigraphy, 367–368
CT enterography, 106, 107 shrapnel injury to small bowel, 381 small bowel transit studies, 101
diverticulitis, 108–110 strategies to minimize infection in, subtypes of, 365
double-contrast barium enema, 105 390–391 surgery for combined disorders,
gastrointestinal bleeding, 110–112 Colostomy, 47, 220–221; see also Fecal 374–375
hepatic metastatic colorectal cancer, incontinence operative surgical options, 372–374
118–120 techniques; Intraoperative ultrasound, 368
imaging interpretation, 113–114 anastomosis Consumer Assessment of Healthcare
imaging technique, 113 Complete blood count (CBC), 1, 121 Providers and Systems (CAHPS),
inflammatory and infectious bowel Complete clinical response (cCR), 280, 282 251
diseases, 106 Complete cytoreduction (CC), 298 Contaminated wounds, 33
IR embolization, 112, 113 Complete pathological response (pCR), 280 Continent ileostomy, 325–326; see also
MRI enterography, 108 Complete spontaneous bowel movements Proctocolectomy
normal small bowel, 108 (CSBMs), 371 Conventional defecography, 368; see also
radiographic examination, 104 Computed tomography (CT), 1, 66, 79, 104, Constipation
rectal bleeding, 111 245, 267, 289 COPD, see Chronic obstructive pulmonary
rectal cancer staging, 112–113, 115 angiogram, 111, 112 disease
Index 411
Coronal orthogonal T2-weighted image, 114 Cutaneous adenocarcinoma in situ; see also Dermatologic neoplasms, 149
Coronary artery bypass graft (CABG), 2 Perianal paget disease Detachable staplers, 45; see also
Corpus cavernosum, 94 Cutaneous fistula with AL, 70; see also Intraoperative anastomosis
CORTICUS Trial (Corticosteroid Therapy Postoperative anastomotic Diabetes, 6; see also Preexisting conditions
of Septic Shock), 35 complications Diagnostic imaging examinations, 104
COX, see Cyclooxygenase antimotility agents and octreotide, 71 Diagnostic peritoneal lavage (DPL), 382; see
CPOE, see Computerized provider order definitive management, 71 also Colorectal trauma
entry sepsis control, 70–71 Diamond flap, 193; see also Anal fissure
CPR, see Cardiopulmonary resuscitation wound care, 71 Diffusion-weighted imaging (DWI, 117
CRC, see Colorectal Cancer CV, see Cardiovascular Digital rectal examination (DRE), 122, 280,
C-reactive protein (CRP), 68, 79, 121, 134 Cyclic guanosine monophosphate (cGMP), 382; see also Colorectal trauma
CRM, see Circumferential radial margin 370 Diosmiplex, 173
Crohn colitis, 339; see also Crohn disease Cyclooxygenase (COX), 24 Direct thrombin inhibitor, 49
management; Perianal Crohn COX-1, 9 Dirty wounds, 33
disease; Small bowel disease COX-2 inhibitors, 75 Disinfectant colitis, 127
management Cytoreduction, 301; see also Peritoneal Dithering, 125
colectomy, 339–340 carcinomatosis secondary to CRC Diverticular bleeding, 347–348; see also
colon cancer, 340 Cytoreductive surgery (CRS), 297, 298–299; Lower gastrointestinal bleeding
colonic strictures, 339 see also Peritoneal carcinomatosis Diverticulitis, 108–110, 135, 235; see also
toxic colitis, 340 secondary to CRC Colorectal imaging studies;
Crohn disease (CD), 107, 108, 109, 134–135, complications, 299 Inflammatory bowel disease
169, 333; see also Inflammatory indications, 299 acute complicated disease, 238
bowel disease palliation, 300–301 acute uncomplicated disease, 237–238
anorectal surgery for anal fissures in PCI staging and prognostic impact, acute/urgent surgery, 238–239
patients with, 193 299–300 Ambrosetti classification of, 237
of pouch, 330 second-look surgery, 300 anastomotic leak, 241
symptoms of, 333 classification, 237
Crohn disease management, 333; see also clinical presentation and diagnosis,
D
Crohn colitis; Perianal Crohn 236–237
disease; Small bowel disease Dabigatran etexilate, 9 complications, 241
management Damage control laparotomy (DCL), 52, elective resection for, 135
anastomosis type, 338–339 382; see also Colorectal trauma elective surgery, 240–241
biologics and surgery, 333–334 DCBE, see Double-contrast barium enema epidemiology, 235–236
endoscopic balloon dilation, 335 DCL, see Damage control laparotomy evaluation after recovery from acute,
intraabdominal abscess, 335 Deep incisional SSI, 33 238
Kono side-to-side anastomosis, 339 Deep vein thrombosis (DVT), 22, 75, failure to reverse, 241–242
multidisciplinary management, 334 82–83; see also Postoperative impact on health care, 235
nonsurgical management, 333 complications intraoperative classification, 237
postoperative medical therapy, 344 Defecation; see also Anorectal physiology mortality, 241
preoperative evaluation, 334 testing limitations; Constipation pathophysiology, 236
strictures, 335 balloon expulsion test, 102 recurrence after previous surgical
thromboembolic disease, 335 colonic transit studies, 101 resection, 242
CR-POSSUM, see Colorectal-POSSUM defecography, 100–101 risks associated with Hartmann
CRS, see Cytoreductive surgery investigations for constipation and reversal, 239–240
CRT, see Chemoradiotherapy disordered, 100 timing of closure, 242
Cryptoglandular theory, 178; see also limitations of defecography in treatment, 237
Perirectal abscess constipation, 101 urgent procedures for, 135–136
CSBMs, see Complete spontaneous bowel normal, 365 Diverting loop ileostomy (DLI), 263
movements small bowel transit studies, 101 DLI, see Diverting loop ileostomy
CSF, see Cerebrospinal fluid Definitive surgery, 71; see also Donabedian model of quality of care, 156
CT, see Computed tomography Postoperative anastomotic outcomes, 157
CTC, see Computed tomography complications process, 156–157
colonography Delayed intraperitoneal chemotherapy, 303; structure, 156
CTE, see Computed tomography see also Peritoneal carcinomatosis Dopamine, 34
enterography secondary to CRC Doppler-guided hemorrhoidal arterial
CTP, see Child-Turcotte-Pugh Delorme procedure, 227–228; see also ligation, 166–167; see also
Cuffitis, 330; see also Fistula Rectal prolapse operative repairs Excisional hemorrhoidectomy
412 Index
Double-contrast barium enema (DCBE), Electronic Medical Record Adoption open/Milligan-Morgan procedure,
104, 105 Model (EMRAM), 61; see also 163–164
Double staple, 44–45; see also Electronic health records stapled hemorrhoidectomy, 165–166
Intraoperative anastomosis HIMSS analytics, 62 Whitehead procedure, 164–165
DPL, see Diagnostic peritoneal lavage ELS, see Electrolyte lavage solution Expanded adipose tissue derived stem cells
DVT, see Deep vein thrombosis EMG, see Electromyography (eADSC), 184
EMRAM, see Electronic Medical Record Extended thromboprophylaxis, 9–11; see
Adoption Model also Preexisting conditions
E
EMT, see Epithelial to mesenchyme External anal sphincter defects, 214; see
eADSC, see Expanded adipose tissue transition also Fecal incontinence operative
derived stem cells Endorectal advancement flap (ERAF), 184, techniques
EAES, see European Association for 185; see also Anal fistula
Endoscopic Surgery Endorectal ultrasound (ERUS), 255, 267
F
Early goal-directed therapy (EGDT), 32, Endoscopic balloon dilation, 335
33 Endoscopic mucosal resection, 270; see Factor Xa inhibitors, 49
EAST, see Eastern Association for the also Rectal cancer transanal FAST, see Focused assessment with
Surgery of Trauma approaches sonography for trauma
Eastern Association for the Surgery of Endoscopic submucosal dissection (ESD), Fast-track surgery (FTS), 88
Trauma (EAST), 388 266 FDA, see U.S. Food and Drug
Eastern Cooperative Oncology Group Endoscopist-directed propofol (EDP), 124 Administration
(ECOG), 299 End-to-end anastomoses (EEA), 42–44; see FDG, see Fluorodeoxyglucose
ECOG, see Eastern Cooperative Oncology also Intraoperative anastomosis FDG-PET, see Fluorodeoxyglucose-
Group Enhanced recovery after surgery (ERAS), positron emission tomography
EDP, see Endoscopist-directed propofol 64, 74, 87, 359 Fecal continence, 93
EGD, see Esophagoduodenoscopy Enhanced recovery pathways (ERPs), 79, Fecal diversion, 343; see also Perianal
EGDT, see Early goal-directed therapy 359 Crohn disease
EGFR, see Epidermal growth factor Enterovesical fistula, 397 Fecal incontinence (FI), 207, 221
receptor Epidermal growth factor receptor (EGFR), anal sphincter, 208
EHR, see Electronic health records 307 anorectal manometry, 211
Elective colectomy, 241 Epithelial to mesenchyme transition biofeedback, 213–214
Electrocautery and energy devices, 50 (EMT), 296 central nervous system, 210
Electrolyte lavage solution (ELS), 121 ERAF, see Endorectal advancement flap colon/stool consistency, 210
Electromyography (EMG), 97, 213, 372; see ERAS, see Enhanced recovery after surgery congenital malformations, 209
also Anorectal physiology testing ERPs, see Enhanced recovery pathways defecography, 211
limitations ERUS, see Endorectal ultrasound diagnosis, 210
anal plug, 97 Erythrocyte sedimentation rate (ESR), 340 epidemiology, 207–208
concentric needle, 97 ESD, see Endoscopic submucosal dissection etiology and scoring, 208
electrode to place transanally, 98 ESICM, see European Society of Intensive history and physical examination,
limitations of, 98 Care Medicine 210–211
pudendal nerve terminal motor latency, Esophageal anastomoses, 66; see also magnetic resonance imaging, 212
97 Postoperative anastomotic medical therapy, 212–213
single-fiber, 97 complications obstetric injury, 208–209
St. Mark’s electrode, 98 Esophagoduodenoscopy (EGD), 110 proposed algorithm for treatment of,
surface EMG/biofeedback, 97 ESR, see Erythrocyte sedimentation rate 212
Electronic health records (EHR), 61, 64–65, European Association for Endoscopic pudendal nerve terminal motor latency,
144–145; see also Medical legal Surgery (EAES), 266 211–212
issues European Society of Intensive Care rectal prolapse, 210
adoption of, 64–65 Medicine (ESICM), 29 surgical injury, 209
basic, 61 EUS, see Endorectal ultrasound testing, 211
challenges with, 63–64 Evacuation proctography, see Conventional treatment, 212
contemporary, 61 defecography ultrasonography, 211
early adopters of, 62 Excisional biopsy, see Local excision Wexner/Cleveland clinic Florida fecal
electronic functions required for Excisional hemorrhoidectomy, 163; see also incontinence scale, 208
hospital adoption of, 62 Hemorrhoidal surgery Fecal incontinence operative techniques,
evidence for, 62–63 closed/Ferguson procedure, 164, 165 214
synoptic reports, 64 Doppler-guided hemorrhoidal arterial antegrade continence enema procedure,
use in colon and rectal surgery, 64 ligation, 166–167 220
Index 413
artificial bowel sphincter, 219–220 Fog reduction/elimination device (FRED), Hanley procedure, 180; see also Perirectal
colostomy, 220–221 137 abscess
gracilis muscle transposition, 217 Four D’s, 388; see also Rectal injury Hartmann colostomy reversal, 239–240; see
injectables, 219 management also Diverticulitis
magnetic anal sphincter, 220 direct repair, 390 Hazard ratio (HR), 315
ongoing investigations, 221 distal washout, 390 Health Evolution through Logical
overlapping sphincteroplasty, 214–216 diversion, 388, 390 Programming (HELP), 62
postanal repair, 216 drainage, 390 Health information technology (HIT), 62
posterior tibial nerve stimulation, rectal washout, 391 Health Information Technology for
218–219 FRED, see Fog reduction/elimination Economic and Clinical Health
radiofrequency therapy, 220 device (HITECH), 61
sacral neuromodulation, 217–218 Frykman-Goldberg procedure, 231–232; see Heated intraperitoneal chemotherapy
sphincteroplasties, 216 also Resection rectopexy; Rectal (HIPEC), 297, 301; see also
FENIX, 220 prolapse operative repairs Peritoneal carcinomatosis
FI, see Fecal incontinence FTS, see Fast-track surgery secondary to CRC
Fibers, 369 choice of chemotherapy agents,
Fibrin glue, 182; see also Anal fistula 302–303
G
Filac, 186; see also Anal fistula closed vs. open technique, 301–302
Finney stricturoplasties, 337; see also Small Gadoxetic acid (Gd-EOB-DTPA), 118 coliseum technique, 302
bowel disease management Gastrointestinal bleeding, 110–112, 347; see hyperthermia, 301
FIQLS, see FI Quality of Life Scale also Colorectal imaging studies; surgical principles, 301
FI Quality of Life Scale (FIQLS), 208; see Lower gastrointestinal bleeding tissue penetration, 301
also Fecal incontinence Gastrointestine (GI), 106 tumor cell entrapment theory, 301
FI severity index (FISI), 208; see also Fecal GC-C, see Guanylate cyclase-C Heineke-Mikulicz strictureplasty, 336;
incontinence Gd-EOB-DTPA, see Gadoxetic acid see also Small bowel disease
FISI, see FI severity index GDMT, see Guideline-directed medical management
Fistula, 178, 330; see also Anal fistula; therapy HELP, see Health Evolution through
Restorative proctocolectomy GI, see Gastrointestine Logical Programming
Crohn disease, 330 Giant condyloma acuminatum, 150 Hemorrhoidal cushions, 161; see also
layer closure, 186 Glutaraldehyde-induced colitis, 127 Hemorrhoidal surgery
plug, 182–183 Glyceryl trinitrate (GTN), 189 Hemorrhoidal surgery, 161
pouch failure and salvage, 330–331 Goldman Cardiac Risk Index, 2 clinical presentation, 162–163
pouchitis and cuffitis, 330 Goodsall’s rule, 179; see also Perirectal comparison and results, 167–168
pouch vaginal fistula, 330 abscess Crohn disease, 169
surgery, 209 GORE BIO-A, 183 etiology, 162
Fistulotomy with sphincter reconstruction, Gracilis muscle transposition, 217; see also excisional hemorrhoidectomy, 163–167
185; see also Anal fistula Fecal incontinence operative immunocompromised state, 170
Flavonoids, 173 techniques Milligan-Morgan open
Flexible lower endoscopy, see Flexible Graciloplasty, see Gracilis muscle hemorrhoidectomy, 163–164
sigmoidoscopy/colonoscopy transposition mixed, 162
Flexible sigmoidoscopy/colonoscopy, 123; GRAS (Generally Recognized As Safe), 173 nonsurgical management, 163
see also Transanal endoscopy GTN, see Glyceryl trinitrate portal hypertension, 169
history and indications, 123–124 Guanylate cyclase-C (GC-C), 371 postoperative pain and bowel
maneuvers for difficult colonoscopy, Guideline-directed medical therapy management, 168
124–125 (GDMT), 2 postpartum hemorrhoids, 169
need for prophylactic antibiotics, 125 special situations, 168
patients on antithrombotic medications, stapled hemorrhoidectomy, 165–166
H
125 use of alternative energy sources, 167
sedation, 124 HAART, see Highly active antiretroviral Hemorrhoids, 161, 171, 348; see also
technique, 124 therapy Excisional hemorrhoidectomy;
Fluid therapy in sepsis resuscitation, 34 HAI, see Hepatic intra-arterial Hemorrhoidal surgery; Rubber
Fluorodeoxyglucose (FDG), 297 embolization band ligation
Fluorodeoxyglucose-positron emission HALS, see Hand-assisted laparoscopic anatomy and pathophysiology, 161–162,
tomography (FDG-PET), 118 surgery 171–172
Focused assessment with sonography for Hand-assisted laparoscopic surgery clinical evaluation, 172
trauma (FAST), 383; see also (HALS), 54, 134, 136, 138; see also comparison of techniques, 175–176
Colorectal trauma Minimally invasive colectomy conservative management, 172–173
414 Index
Hemorrhoids (Continued) HR, see Hazard ratio Crohn disease, 134–135

Doppler-guided hemorrhoidal arterial HS, see Hidradenitis suppurativa diverticulitis, 135–136
ligation, 166–167 HSV, see Herpes simplex virus rectal prolapse, 136
external and internal, 162 Human immunodeficiency virus (HIV), stoma creation, 136
grading of symptomatic internal, 172 152–153 ulcerative colitis, 133–134
hemorrhoidal cushions, 161 Human papillomavirus (HPV), 149, 150 Inflatable penile prosthesis (IPP), 396, 404;
infrared photocoagulation, 175 Hypercoagulable conditions, 9; see also see also Urologic complications
prolapsing, 163, 172 Preexisting conditions Informed consent, 142–143; see also
rubber band ligation, 173–175 Hyperthermic intraperitoneal Medical legal issues
sclerotherapy, 173 chemotherapy (HIPEC), 308; see Infrared coagulator, 173; see also Rubber
stapled hemorrhoidectomy, 165–166 also Chemotherapy band ligation
strangulated, 168–169 Infrared photocoagulation (IRC), 175; see
thrombosed external, 168, 169 also Hemorrhoids
I
treatment, 172 Initial resuscitation of sepsis, 34; see also
Heparin, 9 IAP, see Intraabdominal pressure Sepsis
Heparin-induced thrombocytopenia IBD, see Inflammatory bowel disease Injection sclerotherapy, 173; see also
(HIT), 9 IBS, see Irritable bowel syndrome Hemorrhoids
Hepatic chemoembolization techniques, 294 ICG, see Indocyanine green Injury to normal organs, 82; see also
Hepatic intra-arterial embolization (HAI), Idiopathic hypertrophic subaortic stenosis Postoperative complications
293 (IHSS), 22 Innervation of lower urinary tract, 403; see
Hepatic metastases, 118 IDSA, see Infectious Diseases Society of also Urologic complications
Hepatic metastatic colorectal cancer, 118; America Inotropic therapy in sepsis resuscitation,
see also Colorectal imaging IHSS, see Idiopathic hypertrophic subaortic 34
studies stenosis International Multicentre Prevalence Study
axial noncontrast T2 image, 119 Ileal mesenteric window, 41; see also on Sepsis, The (IMPreSS), 31
axial postcontrast T1 fat-saturated Intraoperative anastomosis Intersphincteric fistulas, 178; see also Anal
image, 119, 120 Ileal pouch-anal anastomosis (IPAA), 133 fistula
coronal noncontrast T2 image, 119 Ileorectal anastomosis (IRA), 326 Intersphincteric proctectomy (ISP), 259,
FDG-PET/CT, 118 Ileostomy, 47; see also Intraoperative 260–261; see also Rectal cancer
Herpes simplex virus (HSV), 153 anastomosis surgery, restorative
Hidradenitis suppurativa (HS), 202; see also Ileus, 83–84; see also Postoperative Intestinal anastomosis, 39; see also
Pilonidal disease complications Intraoperative anastomosis
chronically recurrent Hurley stage III IMA, see Inferior mesenteric artery Intestinal stomas, 57; see also
HS, 203 Imaging interpretation, 113–114; see also Intraoperative challenges
classification, 203 Colorectal imaging studies end-loop stoma, 59
diagnosis and natural history, 203 Imaging technique, 113; see also Colorectal ideal stoma creation, 58
excisional therapy, 204–205 imaging studies ostomy triangle, 58
nonexcisional therapy, 204 Immunonutrition (IMN), 7 pseudo-loop colostomy, 58
pathophysiology, 202 Immunosuppression, 7–8; see also Intraabdominal pressure (IAP), 53
severe Hurley III perianal HS, 205 Preexisting conditions Intraoperative anastomosis, 38
Highly active antiretroviral therapy IMN, see Immunonutrition adjuvants and drains, 47
(HAART), 152, 177 IMPreSS, see International Multicentre anastomotic hemorrhage, 46–47
HIMSS analytics EMRAM, 62; see also Prevalence Study on Sepsis, The anastomotic technique, 41
Electronic health records Improving Surgical Care and Recovery anastomotic testing, 45–46
HIPEC, see Heated intraperitoneal (ISCR), 64 anoscopic-assisted stapler
chemotherapy; Hyperthermic Indocyanine green (ICG), 57 insertion, 45
intraperitoneal chemotherapy Infection, 33; see also Sepsis; Surgical site bowel preparation, 39
HIT, see Health information infection bowel status, 39–40
technology; Heparin-induced Infectious Diseases Society of America challenges, 46
thrombocytopenia (IDSA), 33 detachable staplers, 45
HITECH, see Health Information Inferior mesenteric artery (IMA), 253 double staple, 44–45
Technology for Economic and Inflammatory and infectious bowel end-to-end, 42–44
Clinical Health diseases, 106; see also Colorectal exposure, 40
HIV, see Human immunodeficiency virus imaging studies ileal mesenteric window, 41
House advancement flap, 193; see also Anal Inflammatory bowel disease (IBD), 6, ileostomy vs. colostomy, 47
fissure 133, 210, 203, 335; see also inadequate anastomotic lumen, 46
HPV, see Human papillomavirus Laparoscopic colonic surgery intestinal anastomosis, 39
Index 415
isoparastaltic side-to-side functional Intravenous fluid (IVF), 84 Laparoscopic lavage (LL), 135
end-to-end anastomotic Intravenous pyelogram (IVP), 402 Laparoscopic mesh rectopexy, 230; see also
technique, 46 IORT, see Intraoperative radiation Rectal prolapse operative repairs
leakage, 46 treatment Laparoscopy, 232; see also Rectal prolapse
mobilizing right colon, 42 IPAA, see Ileal pouch-anal anastomosis operative repairs
obtaining adequate length, 40–41 IPP, see Inflatable penile prosthesis LAR, see Low anterior resection
operative principles, 38–39 IRA, see Ileorectal anastomosis LARS, see Low anterior resection syndrome
preanastomotic considerations, 38 IRC, see Infrared photocoagulation LAST, see Local anesthetic systemic
preoperative discussion and IR embolization, 112, 113 toxicity
planning, 38 Irritable bowel syndrome (IBS), 210 Lateral internal sphincterotomy (LIS), 190–
proximal protection, 47 ISCR, see Improving Surgical Care and 192, 209; see also Anal fissure
pulley sutures, 44 Recovery LAVA, see Liver acquisition volume
purse-string suture, 44 Isoparastaltic side-to-side functional end- acceleration
side to side, 45 to-end anastomotic technique, LE, see Local excision
staples vs. sutures, 42 46; see also Intraoperative Leiomyoma, 277
stirrups for modified lithotomy anastomosis Length of stay (LOS), 131
position, 40 ISP, see Intersphincteric proctectomy Lesions size (LS), 299
triple stapling, 44–45 Ivalalon sponge rectopexy, 231; see also LGIB, see Lower gastrointestinal bleeding
types, 43 Rectal prolapse operative repairs Lidocaine toxicity, 19
Intraoperative challenges, 49, 60 IVF, see Intravenous fluid LIFT, see Ligation of the intersphincteric
adhesive disease, 54–56 IV ibuprofen, 24 tract
damage control, 52–54 IVP, see Intravenous pyelogram Ligation of hemorrhoid mass, see Rubber
intraoperative hemorrhage, 49–52 IV patient-controlled analgesia (IVPCA), band ligation
lesion identification, 56–57 22 Ligation of the intersphincteric tract
operative technique of rectus abdominis IVPCA, see IV patient-controlled analgesia (LIFT), 181, 183, 209, 342; see also
muscle welding, 51 Anal fistula
preoperative colonoscopy with mucosal BIOLIFT plug, 183
J
endoclip placement, 57 Limited sinus excision, 199; see also
preoperative evaluation, 49 Jackson-Pratt drain, 81; see also Pilonidal disease
sutures tied over free muscle flap, 52 Postoperative complications Linaclotide, 370
trouble with stoma creation, 57–59 Jeep disease, see Pilonidal disease Liposomal bupivacaine, 20
untied sutures applied to adjacent tissue LIS, see Lateral internal sphincterotomy
near bleeding site, 51 Liver acquisition volume acceleration
K
Intraoperative goal-directed therapy, 24; (LAVA), 108
see also Anesthesia Karydakis flap, 200–201; see also Pilonidal Liver disease, 5–6; see also Preexisting
Intraoperative positioning, 25; see also disease conditions
Anesthesia; Surgical Care Ketamine, 75 Liver metastases (LM), 299
Improvement Project Keyhole technique, 357; see also Stoma Liver resection, 292; see also Rectal cancer,
lateral decubitus, 26 Kono side-to-side anastomosis, 339; see also recurrent
lithotomy, 26–27 Crohn disease management liver ablation, 293–294
modified left lateral/Sims position, 26 liver first approach, 293
nerves at risk for injury during, 25 primary first approach, 292–293
L
prone, 25–26 simultaneous approach, 293
supine, 25 Lactulose, 370 LL, see Laparoscopic lavage
Intraoperative radiation treatment (IORT), Laparoscopic colonic surgery, 129 LM, see Liver metastases
292 advantages, 130 LMWH, see Low molecular weight heparin
Intraperitoneal chemotherapy (IPC), 297, colon cancer, 132 LNM, see Lymph node metastases
298–299; see also Peritoneal disadvantages, 130–132 Local anesthetic systemic toxicity (LAST),
carcinomatosis secondary to hand-assisted laparoscopic surgery, 138 19
CRC inflammatory bowel disease, 133–136 Local excision (LE), 266, 315; see also
complications, 299 rectal cancer, 132–133 Rectal cancer management
delayed recurrent, 303 single-site laparoscopic surgery, 138 for CCR, 285–286
indications, 299 straight laparoscopic colectomy, after nCRT as diagnostic approach, 285
palliation, 300–301 137–138 Local recurrence (LR), 268
PCI staging and prognostic impact, technical considerations in minimally Loop-end colostomy, 362; see also Ostomies
299–300 invasive colectomy, 136 LOS, see Length of stay
second-look surgery, 300 treatable conditions, 132 Low anterior resection (LAR), 259, 271, 310
416 Index
Low anterior resection syndrome (LARS), MAP, see Mean arterial pressure single-site laparoscopic surgery, 138
264 Marsupialization, 198, 199; see also straight laparoscopic colectomy,
Lower gastrointestinal bleeding (LGIB), Pilonidal disease 137–138
347 Matrix metalloproteinases (MMPs), 67 Minimally invasive techniques in Rectal
colitis, 348 MBP, see Mechanical bowel preparation cancer surgery, 262–263
diagnosis and treatment algorithm for McGown suction rubber band ligator, 173; MiraLAX prep, 15
acute, 350 see also Rubber band ligation MMPs, see Matrix metalloproteinases
diverticular bleeding, 347–348 MDT, see Multidisciplinary team Model for End-Stage Liver Disease
endoscopic strategies, 349 Mean arterial pressure (MAP), 34 (MELD), 5
etiology, 347, 348 Mechanical bowel preparation (MBP), Modified left lateral, 26; see also
hemorrhoids, 348 14, 15–16, 67, 89–90; see also Intraoperative positioning
inflammatory bowel disease, 348 Preoperative bowel preparation MODS, see Multiple organ dysfunction
initial evaluation, 349 Medical legal issues, 141, 146 syndrome
management, 349 anatomy of malpractice suit, 145–146 Monitored anesthetic care (MAC), 20–21
neoplasia, 348 charting, 143–144 Motor unit potential (MUP), 97
radiologic strategies, 349–350 documentation, 143 MPFF, see Micronized purified flavonoid
surgical strategies, 350–351 electronic health record, 144–145 fraction
vascular malformation, 348–349 high–risk areas in colorectal treatment, MR, see Magnetic resonance
Low molecular weight heparin (LMWH), 142 MRE, see Magnetic resonance
22 informed consent, 142–143 enterography
LR, see Local recurrence liability and insurance, 141–142 MRF, see Mesorectal fascia
LS, see Lesions size medical liability crisis, 142 MRI, see Magnetic resonance imaging
Lubiprostone, 370 medical malpractice insurance, 141 MSC, see Mesenchymal stromal cells
Lymph node metastases (LNM), 267 physician–patient relationship, 142 Mucosectomy with hand-sewn
Medical outcomes and quality, 155, 158, anastomosis, 327; see also
160 Restorative proctocolectomy
M
colonoscopy, 158 Multidisciplinary team (MDT), 157
MAC, see Monitored anesthetic care Donabedian model of quality of care, Multiple organ dysfunction syndrome
MACE, see Myocardial infarction or major 156–157 (MODS), 29, 31
adverse cardiac event history of quality assessment and MUP, see Motor unit potential
Magnetic resonance (MR), 282 improvement, 155–156 Myocardial Infarction and Cardiac Arrest
Magnetic resonance enterography (MRE), OSTRiCH Consortium, 158 (MICA), 2
106–108, 109, 335 proposed quality indicators for Myocardial infarction or major adverse
Magnetic resonance imaging (MRI), 98, colonoscopy, 159–160 cardiac event (MACE), 2
101, 104, 177, 267, 289; see also quality and outcome measures specific
Anorectal physiology testing to colorectal surgery, 157–158
N
limitations MELD, see Model for End-Stage Liver
limitations of, 102 Disease NaP, see Sodium phosphate
Malnutrition, 6–7; see also Preexisting Mercedes closure, 363; see also Ostomies Nasogastric tube, 81; see also Postoperative
conditions Mesenchymal stromal cells (MSC), 184 complications
Malpractice claims in colorectal disease, Mesh sling repair, 229; see also Rectal National Cancer Data Base (NCDB), 267
142 prolapse operative repairs National Comprehensive Cancer Network
Malpractice suit, 145; see also Medical legal Mesorectal fascia (MRF), 113 (NCCN), 266
issues Metabolic equivalents (METs), 2 National Healthcare Safety Network
deposition pitfalls, 145–146 METs, see Metabolic equivalents wound classification system, 33
initial phase, 145 MICA, see Myocardial Infarction and National surgical adjuvant breast and
ploys, 146 Cardiac Arrest bowel program (NSABP), 312
pretrial discovery, 145 Michelassi strictureplasty, 337; see also National Surgical Quality Improvement
trial, 146 Small bowel disease management Program (NSQIP), 90, 157, 337
Manometry, 94; see also Anorectal Micronized purified flavonoid fraction Nationwide Inpatient Sample (NIS), 391
physiology testing limitations (MPFF), 173 NBCA, see N-butyl-2-cyanoacrylate
catheter, 95 Milligan-Morgan open hemorrhoidectomy, N-butyl-2-cyanoacrylate (NBCA), 111
investigations for incontinence, 94 163–164; see also Excisional NCCN, see National Comprehensive
rectal capacity and sensation, 96 hemorrhoidectomy Cancer Network
rectoanal inhibitory reflex, 96 Minimally invasive colectomy, 136; see also NCCTG, see North Central Cancer
tracing, 95 Laparoscopic colonic surgery Treatment Group
value and limitations of, 96–97 hand-assisted laparoscopic surgery, 138 NCDB, see National Cancer Data Base
Index 417
Neisseria gonorrhoea, 153 bowel preparation, 16–17 PE, see Pulmonary embolism
Neoadjuvant chemoradiation therapy Organ/space SSI, 33 PEG, see Polyethylene glycol
(nCRT), 256–258, 279; see also OS, see Overall survival Pelvic sepsis, 329; see also Restorative
Rectal cancer management; OSA, see Obstructive sleep apnea proctocolectomy
Rectal cancer surgery, restorative; Osmotic agents, 15 Penetrating Abdominal Trauma Index
Tumor response assessment Osmotic laxatives, 369 (PATI), 384
Neoadjuvant radiotherapy, 257 Ostomies, 352; see also Stoma Penrose drain, 81; see also Postoperative
acute adverse effects, 316–317 considerations, 353 complications
benefit of, 312–316 end stoma, 354 Percutaneous coronary intervention
chronic late adverse effects of, 318–320 historical perspective, 352 (PCI), 2
preoperative irradiation complications, intraoperative technical tips, 360–361 Perianal abscess, 341; see also Perianal
317–318 loop-end colostomy, 362 Crohn disease
preoperative studies, 313 marking obese patient, 360 Perianal Crohn disease, 340; see also
rectal injury and treatment, 320–321 Mercedes or triangular closure, 363 Crohn colitis; Crohn disease
Neuraxial techniques, 22–23; see also ostomy reversal, 362–363 management; Small bowel
Anesthesia postoperative support, 361–362 disease management
NIS, see Nationwide Inpatient Sample pre-op preparation, 359–360 anal stenosis, 343–344
Nitrates, 189–190 prevention, 359 carcinoma in fistula, 343
NM, see Nuclear medicine primary anastomosis with diversion, fecal diversion, 343
Nonsteroidal anti-inflammatory drugs 353–354 fissures, 343
(NSAIDs), 24, 91 stoma complications, 354–359 perianal abscess and fistula, 341
Norepinephrine, 34 stomal siting, 360 rectovaginal fistula, 342–343
Normal small bowel, 108 surgical decision-making, 353 skin tags, 343
North Central Cancer Treatment Group Ostomy triangle, 58 surgical management, 341–342
(NCCTG), 312 OSTRiCH Consortium, 158 PeriAnal fistula, 341; see also Perianal
NSABP, see National surgical adjuvant Overall survival (OS), 268 Crohn disease
breast and bowel program Overlapping sphincteroplasty, 214–216; see Perianal Paget disease, 151
NSAIDs, see Nonsteroidal anti- also Fecal incontinence operative Perineal hernias, 277; see also
inflammatory drugs techniques Abdominoperineal resection
NSQIP, see National Surgical Quality Overwhelming postsplenectomy infection Perineal injury, 393; see also Colorectal
Improvement Program (OPSI), 50 trauma
Nuclear medicine (NM), 104, 110 Perineal proctosigmoidectomy, 225–227;
see also Rectal prolapse operative
P
repairs
O
Pain management, 24; see also Anesthesia Perineal sinus, 276; see also
OAP, see Oral antibiotics multimodality pain management, 25 Abdominoperineal resection
Obesity, 6; see also Preexisting conditions Parastomal hernia, 356–358; see also Stoma Perineal wound, 275; see also
Obstructive sleep apnea (OSA), 4 Parastomal varices, 359; see also Stoma Abdominoperineal resection
OC, see Optical colonoscopy Participant Use Data File (PUF), 157 Perioperative β-blockade, 3–4; see also
Odds ratio (OR), 8, 130 PATI, see Penetrating Abdominal Trauma Preexisting conditions
OHP, see Open Hartmann procedure Index Peripheral nerve evaluation (PNE), 375
Open abdominal repairs, 228; see also Patient-controlled analgesia (PCA), 75 Perirectal abscess, 177; see also Anal fistula
Rectal prolapse operative repairs Patient-controlled device, 90 anatomical considerations, 177–179
Open Hartmann procedure (OHP), 135 Patient-controlled epidural analgesia catheter drainage of abscess, 180
Open/Milligan-Morgan procedure, 163, (PCEA), 21 cryptoglandular theory, 178
see Milligan-Morgan open PCA, see Patient-controlled analgesia diagnosis of abscesses, 179
hemorrhoidectomy PCCRC, see Peritoneal carcinomatosis drainage of abscesses, 180–181
Opioids, 24 secondary to CRC Goodsall’s rule, 179
Opioid-sparing anesthetics, 24; see also PCEA, see Patient-controlled epidural Hanley procedure, 180
Anesthesia analgesia management of abscesses, 179–180
Opium derivatives, 212 PCI, see Percutaneous coronary Peritoneal cancer index (PCI), 297; see
OPSI, see Overwhelming postsplenectomy intervention; Peritoneal cancer also Peritoneal carcinomatosis
infection index secondary to CRC
Optical colonoscopy (OC), 104 pCR, see Complete pathological response staging and prognostic impact, 299–300
OR, see Odds ratio PCR, see Polymerase chain reaction Peritoneal carcinomatosis secondary to
Oral antibiotics (OAP), 14; see also PDAI, see Pouchitis disease activity index CRC (PCCRC), 296, 303–304
Preoperative bowel preparation PDPH, see Postdural puncture headache chemotherapy, 298
418 Index
Peritoneal carcinomatosis secondary to Polyethylene glycol (PEG), 15, 121, 370 Postpolypectomy syndrome, 126; see also
CRC (PCCRC) (Continued) Polymerase chain reaction (PCR), 153 Transanal endoscopy
CT imaging of peritoneal Polytetrafluoroethylene (PTFE), 229 Postpoylpectomy coagulation syndrome,
carcinomatosis, 297 Portal venous hypertension, 169; see also see Postpolypectomy syndrome
cytoreduction, 301 Hemorrhoidal surgery Pouch failure, 330; see also Fistula
cytoreductive surgery, 298–301 Portomesenteric venous thrombosis, 83 Pouchitis, 330; see also Fistula
delayed intraperitoneal chemotherapy, Positron emission tomography (PET), 104, Pouchitis disease activity index (PDAI),
303 289, 297 330
diagnostic laparoscopy, 297–298 POSSUM (Physiological and Operative Pouch stricture, 329; see also Restorative
etiology, 296–297 Severity Score for enumeration of proctocolectomy
heated intraperitoneal chemotherapy, Mortality and Morbidity), 12 Pouch vaginal fistula (PVF), 330; see also
301–303 Postdural puncture headache (PDPH), 23 Fistula
imaging, 297 Posterior mesh fixation, 230–231; see also PPCs, see Postoperative pulmonary
intraperitoneal chemotherapy, 298–301 Rectal prolapse operative repairs complications
peritoneal metastatic cascade, 296 Posterior tibial nerve stimulation PPH, see Prolapsed hemorrhoids
Peritoneal metastatic disease, see Peritoneal (PTNS), 218–219; see also PRA, see Primary resection with
carcinomatosis secondary to CRC Fecal incontinence operative anastomosis
Peritonitis, 238 techniques Preexisting conditions, 1
PET, see Positron emission tomography Postoperative anastomotic complications, ACC/AHA cardiac risk assessment
Phenylephrine, 34 66 algorithm, 3
Pilonidal disease, 195, 201–202; see also anastomotic leak, 67 ACS NSQIP surgical risk calculator,
Hidradenitis suppurativa anastomotic stricture, 71–72 10–11
acute pilonidal abscess, 196 chronic anastomotic sinus, 72 ambulatory anorectal procedures, 1
advanced disease, 200 cutaneous fistula with AL, 70 cardiac evaluation, 1
advanced flaps, 201 Postoperative complications, 74, 84–85 cardiac risk indices, 2
algorithm for management of, 202 ASEPSIS score, 78 chronic anticoagulation and
Bascom cleft lift, 201 deep vein thrombosis, 82–83 perioperative management, 8–9
Bascom procedure, 199–200 future directions, 84 considerations for extended
chronic pilonidal disease, 196 ileus, 83–84 thromboprophylaxis, 9–11
depilation, 197 infection, 77–80 diabetes, 6
etiology, 195–196 injury to normal organs, 82 hypercoagulable conditions, 9
fibrin, 197 Jackson-Pratt drain, 81 immunosuppression, 7–8
karydakis flap, 200–201 minimizing bleeding complications, liver disease, 5–6
limited excision, 199 75–77 malnutrition, 6–7
management of, 196–197 nasogastric tube, 81 obesity, 6
marsupialization, 198, 199 nausea and vomiting, 83 perioperative β-blockade, 3–4
minimal/moderate disease, 198 pain management, 75 preoperative testing, 1
nonoperative and minimally invasive Penrose drain, 81 pulmonary assessment, 4
management, 197 postoperative pulmonary renal disease, 4–5
phenol, 198 complications, 80 risk assessment tools, 11–12
pilonidal cyst, 196 radiopaque markers, 81 testing and interventions, 2–3
presentation and diagnosis, 196 retained Ray-Tec sponge, 82 Preoperative bowel preparation, 14, 121; see
rhomboid flap, 201, 202 retained surgical items, 81–82 also Transanal endoscopy
surgical treatment, 198 splenic injury, 82 irritant laxatives, 122
U-flap technique, 201 thumbtack occlusion of bleeding mechanical bowel preparation, 15–16
unroofing and wide local excision, basivertebral vein, 76 oral antibiotic bowel preparation,
198–199 time out and wrong site surgery, 84 16–17
Pit picking, 199 ureteral injury, 82 polyethylene glycol, 121
Plecanatide, 371 urinary tract infections and urinary single vs. split dosing, 122
PNE, see Peripheral nerve evaluation retention, 80–81 sodium phosphate, 121–122
PNTML, see Pudendal nerve terminal Postoperative day 3 (POD3), 79 special circumstances, 17
motor latency Postoperative ileus (POI), 130 systemic antibiotic prophylaxis, 14–15
POD3, see Postoperative day 3 Postoperative pain control, 90 Preoperative colonoscopy with mucosal
Podophyllin, 150 Postoperative pulmonary complications endoclip placement, 57; see also
POI, see Postoperative ileus (PPCs), 4 Intraoperative challenges
POISE (Perioperative Ischemic Evaluation Postpartum hemorrhoids, 169; see also Primary resection with anastomosis (PRA),
Study), 3–4 Hemorrhoidal surgery 239
Index 419
Proctocolectomy; see also Ulcerative colitis benefits of adjuvant and neoadjuvant, rationale for organ preservation,
surgical management 312–316 279–280
with Brooke ileostomy, 324–325 pre-vs. postoperative radiation studies, response prediction, 280
with continent ileostomy, 325–326 314 tools in tumor response assessment,
with IPAA, see Restorative radiation dermatitis, 317 282–283
proctocolectomy radiation proctitis, 320 tumor response assessment, 281
Proctoscopy, rigid, 122; see also Transanal rectal injury and treatment, 320–321 watch and wait strategy, 279, 283–285
endoscopy surgical complications after Rectal cancer, recurrent, 288–289; see also
complications, 123 preoperative, 317–318 Rectal cancer
indications, 122–123 Radiofrequency ablation (RFA), 293 colorectal liver metastasis, 292
technique, 123 Radiofrequency therapy (RF therapy), initial evaluation, 289–290
Prokinetic agents, 370 220; see also Fecal incontinence liver resection, 292–294
Prolapsed hemorrhoids (PPH), 163, see operative techniques management, 290–291
Stapled hemorrhoidectomy Radiopaque markers, 367 palliative treatment, 291
Prosecretory agents, 370 RAIR, see Rectoanal inhibitory reflex pulmonary metastasis, 294
Prothrombin time/international Randomized control trial (RCT), 24, 130, treatment outcomes, 291–292
normalized ratio (PT/INR), 22 198 Rectal cancer staging, 112–113; see also
Pruritus ani, 148 Rapid plasma reagin (RPR), 153 Colorectal imaging studies;
clinical presentation, 148 RBCs, see Red blood cells Rectal cancer
etiology and diagnosis, 148–149 RBL, see Rubber band ligation carcinoma staging, 311
with excoriations, 149 RCCC, see Rectal Cancer Coordinating structured report template for, 115
systemic conditions associated with, 149 Committee Rectal cancer surgery, restorative, 253; see
treatment, 149 RCRI, see Revised Cardiac Risk Index also Rectal cancer management
Pseudo-loop colostomy, see Loop-end ReCCoEs, see Rectal Cancer Centers of autonomic nerve preservation, 264
colostomy Excellence bowel function, 264
PTFE, see Polytetrafluoroethylene Rectal algorithm, 389 circumferential radial margin, 256
PT/INR, see Prothrombin time/ Rectal bleeding, 111 distal margin, 255–256
international normalized ratio Rectal cancer, 132–133; see also functional outcomes, 264
PTNS, see Posterior tibial nerve Laparoscopic colonic surgery; laparoscopic and robotic radical rectal
stimulation Rectal cancer surgery, restorative resections, 262
Pudendal nerve terminal motor latency chemotherapy for, 306 lower anastomoses, 256
(PNTML), 96, 97, 98, 211; see also chemotherapy in resectable, 307 margins, 255
Anorectal physiology testing management, 158, 266 milestones in, 254
PUF, see Participant Use Data File surgery, 157 minimally invasive techniques for,
Pulley sutures, 44; see also Intraoperative technical improvements, 253–255 262–263
anastomosis Rectal cancer adjunctive treatment, 310; neoadjuvant chemoradiotherapy,
Pulley technique, 127 see also Radiation therapy; Rectal 256–258
Pulmonary assessment, 4; see also cancer management preoperative evaluation, 255
Preexisting conditions postoperative adjuvant radiation reservoir and pouch reconstruction, 263
Pulmonary embolism (PE), 83 studies, 313 sexual function, 264
Pulmonary metastasis, 294; see also Rectal preoperative neoadjuvant radiation sphincter conservation, 256–258
cancer, recurrent studies, 313 surgery for sphincter conservation,
Purse-string suture, 44; see also rectal carcinoma staging, 311 258–262
Intraoperative anastomosis Rectal Cancer Centers of Excellence technical improvements, 253–255
PVF, see Pouch vaginal fistula (ReCCoEs), 158 temporary diversion, 263
Pyoderma gangrenosum, 355; see also Rectal Cancer Coordinating Committee Rectal cancer transanal approaches,
Stoma (RCCC), 158 266; see also Rectal cancer
Rectal cancer management, 279, 286; management
see also Rectal cancer adjunctive adjuvant treatment, 269–270
R
treatment; Rectal cancer comparison of techniques, 269
Radiation, see Radiation therapy transanal approaches complications, 269
Radiation therapy (RT), 280, 310, 321; adjuvant treatment, 285 endoscopic treatment, 270
see also Rectal cancer adjunctive baseline staging and indications for initial evaluation, 267
treatment nCRT, 280 local excision after chemoradiation, 270
acute adverse effects, 316–317 local excision, 285–286 local recurrence and survival, 267–268
adverse effects of adjuvant and neoadjuvant treatment options, preoperative preparation, 268
neoadjuvant, 318–320 280–281 technical, 268
420 Index
Rectal cancer transanal approaches Renal injuries, 401–402; see also Urologic SAFE, see Saline versus Albumin Fluid
(Continued) complications Evaluation
transanal endoscopic microsurgery, Resection rectopexy, 231–232; see also Sagittal T2-weighted image of pelvis, 114
268–269 Rectal prolapse operative repairs Saline versus Albumin Fluid Evaluation
transanal excision, 268 Restorative proctocolectomy (RP), 254, (SAFE), 34
transanal minimally invasive surgery, 324, 326; see also Ulcerative Salvage procedures, 330–331; see also
268–269 colitis surgical management Fistula
Rectal injury management, 388, 390; see complications, 328 SAS, see Synthetic absorbable suture
also Colorectal trauma differences in stages, 327–328 SCA, see Straight coloanal anastomosis
compared to colon injury, 388 fistula, 330–331 SCCA, see Squamous cell carcinoma
data reviewed in 2016 EAST guidelines, functional results, 328–329 SCCM, see Society of Critical Care Medicine
389 hemorrhage, 329 SCIP, see Surgical Care Improvement
diagnosis, 384 ileal pouch configurations, 327 Project
four D’s, 388–390 indications, 326 SCOAP, see Surgical Care Outcomes
rectal algorithm, 389 mucosectomy and hand-sewn vs. Assessment Program
rectal injury identification, 383 double-stapled anastomosis, 327 SEER, see Surveillance, Epidemiology, and
Rectal neoplasms, 258 obtaining adequate length, 328 End Results
Rectal procidentia, see Rectal prolapse septic complications, 329 Senna, 370
Rectal prolapse, 136, 224; see also small bowel obstruction, 329 Sepsis, 29, 30, 31; see also Infection; Sepsis
Inflammatory bowel disease stricture, 329 treatment; Surviving Sepsis
additional studies, 225 technique, 326–327 Campaign
anal physiology, 225 Retained Ray-Tec sponge, 82 and associated conditions, 29
classification, 224 Retained surgical items, 81–82; see also bundles, 31–33
patient evaluation and investigations, Postoperative complications early goal-directed therapy, 32, 33
225 Retractors, 40 evaluation and diagnosis, 31
physical exam, 225 Retrograde urethrogram (RUG), 396 multiple organ dysfunction syndrome,
recurrent, 232–233 Return of bowel function (ROBF), 134 29, 31
Rectal prolapse operative repairs, 225 Revised Cardiac Risk Index (RCRI), 2 sepsis-3, 29, 30, 31
anal encirclement, 228 RFA, see Radiofrequency ablation septic shock, 31
Delorme procedure, 227–228 RF therapy, see Radiofrequency therapy severe, 30
ivalalon sponge rectopexy, 231 Rhomboid flap, 201, 202; see also Pilonidal Surviving Sepsis Campaign, 31
laparoscopic mesh rectopexy, 230 disease systemic inflammatory response
laparoscopy, 232 Ripstein procedure, 229; see also Mesh sling syndrome, 29
mesh sling repair, 229 repair; Rectal prolapse operative Sepsis-3, 29, 30, 31
open abdominal repairs, 228 repairs Sepsis supportive therapy, 35; see also
perineal proctosigmoidectomy, Rivaroxaban, 9 Sepsis treatment
225–227 ROBF, see Return of bowel function blood product administration, 35
perineal repairs, 225 RPR, see Rapid plasma reagin deep vein thrombosis prophylaxis, 35
posterior mesh fixation, 230–231 RR, see Relative risk glucose control, 35
resection rectopexy, 231–232 RT, see Radiation therapy goal setting for care, 35
results of mesh rectopexy, 230 Rubber band ligation (RBL), 171, 173; see nutrition, 35
suture rectopexy, 231 also Hemorrhoids other recommendations, 35–36
ventral mesh rectopexy, 229–230 banding internal haemorrhoid, 174 Sepsis treatment, 34; see also Initial
Rectal tumor, 114 Barron-McGivney rubber band ligator, resuscitation of sepsis; Sepsis
Rectoanal inhibitory reflex (RAIR), 96, 211, 173 supportive therapy
367; see also Anorectal physiology complications, 174–175 antibiotic therapy, 36
testing limitations infrared coagulator, 173 source control, 36
Rectovaginal fistula (RVF), 342; see also McGown suction rubber band ligator, steroid use, 34–35
Perianal Crohn disease 173 surgery in septic patient, 36
Rectum, 210; see also Fecal incontinence RUG, see Retrograde urethrogram Septic shock, 31
Recurrent prolapse, 232–233; see also RVF, see Rectovaginal fistula Sequential Organ Failure Assessment
Rectal prolapse (SOFA), 31
Red blood cells (RBCs), 111 score, 32
S
Relative risk (RR), 6, 16, 130 Setons, 181
Renal disease, 4–5; see also Preexisting Sacral nerve stimulation (SNS), 217, 375; see Severe sepsis, 30
conditions also Fecal incontinence operative Sexual dysfunction, 404; see also Urologic
Renal failure, 4–5 techniques complications
Index 421
Sexually transmitted diseases (STDs), 153 planes for intersphincteric resection, translocation of, 358
and anal fissures, 194 261 Z-plasty repair for stenosis, 356
SFM, see Splenic flexure mobilization sphincter preservation and neoadjuvant STOP-IT, see Study to Optimize Peritoneal
Side-to-side isoperistaltic strictureplasty, treatment, 258 Infection Therapy
337; see also Small bowel disease splenic flexure and left colon Straight coloanal anastomosis
management mobilization, 262 (SCA), 263
SILS, see Single incision laparoscopic surgical techniques for, 258 Straight laparoscopic colectomy (SLC),
surgery; Single-site laparoscopic total mesorectal excision, 259 136, 137–138; see also Minimally
surgery transanal total mesorectal excision, invasive colectomy
Simplified peritoneal cancer index (SPCI), 261–262 Strangulated hemorrhoids, 168–169; see
297 transanal transabdominal proctectomy, also Hemorrhoidal surgery
Sims position, 26; see also Intraoperative 260–261 Study to Optimize Peritoneal Infection
positioning Sphincteroplasties, 216; see also Fecal Therapy (STOP-IT), 36, 69
Single incision laparoscopic surgery (SILS), incontinence operative Sugarbaker technique, 357; see also Stoma
134 techniques Superficial incisional SSI, 33
Single-site laparoscopic surgery (SILS), 136, Sphincter pressure measurement, 94, 96; Surgical Care and Outcomes Assessment
138; see also Minimally invasive see also Anorectal physiology Program, 129
colectomy testing limitations Surgical Care Improvement Project (SCIP),
SIRS, see systemic inflammatory response Splenic flexure mobilization (SFM), 262; see 18, 156; see also Anesthesia;
syndrome also Sphincter conservation Intraoperative positioning
SIS, see Surgical Infection Society Splenic injury, 82; see also Postoperative process and outcome measures, 27
Skin tags, 343; see also Perianal Crohn complications Surgical Care Outcomes Assessment
disease following colonoscopy, 127 Program (SCOAP), 7
Small bowel disease management, 336; see Squamous cell carcinoma (SCC), 203, Surgical Infection Society (SIS), 29
also Crohn colitis; Crohn disease 395 Surgical prophylaxis, 33
management; Perianal Crohn SSC, see Surviving Sepsis Campaign Surgical site infection (SSI), 6, 14, 67, 74,
disease Stapled hemorrhoidectomy, 165– 130; see also Infection; Sepsis;
Finney stricturoplasties, 337 166; see also Excisional Sepsis supportive therapy
Heineke-Mikulicz strictureplasty, 336 hemorrhoidectomy incisional, 33
ileocolic resection, 338 STDs, see Sexually transmitted diseases management of, 34
laparoscopic vs. open intervention, 338 Stem cells, 184; see also Anal fistula organ/space, 33
side-to-side isoperistaltic Steroids, 7 Surveillance, Epidemiology, and End
strictureplasty, 337 Stimulant laxatives, 370 Results (SEER), 110
strictureplasty, 336–338 Stirrups for modified lithotomy position, Surviving Sepsis Campaign (SSC), 31; see
Small bowel obstruction (SBO), 105, 40; see also Intraoperative also Initial resuscitation of sepsis;
130, 329; see also Restorative anastomosis Sepsis
proctocolectomy Stoma, 352; see also Inflammatory bowel bundles, 32
SNS, see Sacral nerve stimulation disease; Ostomies Suture rectopexy, 231; see also Rectal
Society of Critical Care Medicine (SCCM), complications, 354 prolapse operative repairs
29 creation, 136 Suturing, 42
Sodium phosphate (NaP), 15 dehydration, 354–355 Sympathetic nerve fibers, 273
formulations, 121–122 formation, 352 Synoptic reports, 64; see also Electronic
Sodium picosulfate, 122 intraperitoneal repairs, 357 health records
SOFA, see Sequential Organ Failure ischemia/necrosis, 355–356 Synthetic absorbable suture (SAS), 396
Assessment keyhole technique, 357 Syphilis, 153
Sorbitol, 370 mesh positions, 357 Systemic antibiotic prophylaxis, 14–15;
SPCI, see Simplified peritoneal cancer mucocutaneous separation, 355 see also Preoperative bowel
index parastomal hernia, 356–358 preparation
Sphincter conservation, 256–258; see also parastomal varices, 359 Systemic inflammatory response syndrome
Rectal cancer surgery, restorative prolapse, 358 (SIRS), 29, 30
anterior resection in upper rectal pyoderma gangrenosum, 355
cancers, 260 risk factors for, 354
T
concepts and preparation, 258–259 skin complications, 355
dissection planes for low anterior stenosis, 356 TAE, see Transanal excision
resection, 260 stomal varices, 358–359 TAMIS, see Transanal minimally invasive
intraoperative positioning and stoma retraction, 355 surgery
retraction, 261 Sugarbaker technique, 357 TATA, see Transanal transabdominal
422 Index
TaTME, see Transanal total mesorectal B&K 3D self-contained, 99 Ulcerative colitis surgical management, 323
excision Transjugular intrahepatic portosystemic biologics, 331
TAUS, see Transanal ultrasound shunt (TIPS), 5, 169 continent ileostomy, 325
TCP, see Transverse coloplasty pouch Transmural burn syndrome, see indications for surgery, 323–324
TEM, see Transanal endoscopic Postpolypectomy syndrome proctocolectomy with Brooke ileostomy,
microsurgery Transrectal ultrasound (TRUS), 112; see 324–325
Thiersch repair, 228; see also Anal also Transanal ultrasound; proctocolectomy with continent
encirclement; Rectal prolapse Transvaginal and transrectal ileostomy, 325–326
operative repairs ultrasonography restorative proctocolectomy, 326–331
Thromboembolic disease, 335; see also Transvaginal and transrectal surgical options, 324
Crohn disease management ultrasonography (TTUS), 368; total abdominal colectomy with
Thrombosed external hemorrhoids, 168, see also Transanal ultrasound; ileorectal anastomosis, 326
169; see also Hemorrhoidal Transrectal ultrasound Truelove and Witts classification
surgery Transverse coloplasty pouch (TCP), 263 system, 324
TIPS, see Transjugular intrahepatic Transversus abdominis plane (TAP), 23, 75; Ultrasound, 99; see also Anorectal
portosystemic shunt see also Anesthesia physiology testing limitations
TME, see Total mesorectal excision TRG, see Tumor regression grade Unroofing, 198; see also Pilonidal disease
TNF-α, see Tumor necrosis factor-alpha Triangular closure, 363; see also Ostomies UNYSQI, see Upstate New York Surgical
Total abdominal colectomy (TAC), 134, Triple stapling, 44–45; see also Quality Initiative
324, 326; see also Ulcerative Intraoperative anastomosis Upper gastrointestinal bleeding
colitis surgical management TRUS, see Transrectal ultrasound (UGIB), 347; see also Lower
Total mesorectal excision (TME), 133, 157, T-staging, 114–118; see also Colorectal gastrointestinal bleeding
253, 259, 266, 279, 402; see also imaging studies Upstate New York Surgical Quality
Sphincter conservation axial diffusion-weighted image of Initiative (UNYSQI), 68
Total parenteral nutrition (TPN), 35 pelvis, 116 Ureter, 398
TPN, see Total parenteral nutrition morphologic nodal characterization Ureteral injuries, 82, 398; see also
Transanal endoscopic microsurgery (TEM), with MRI, 117 Postoperative complications;
266, 268, 285; see also Rectal oblique axial T2 image of pelvis, 116, Urethral injuries; Urologic
cancer transanal approaches 117 complications
Transanal endoscopy, 121, 127 oblique coronal T2 image of pelvis, 116 anatomy, 398
anoscopy, 122 true axial T2 image of pelvis, 117 bladder flap procedure, 401
bowel preparation, 121–122 tumor staging in rectal cancer, 116 delayed recognition, 401
flexible colonoscopy, 123–125 TTUS, see Transvaginal and transrectal devascularization, 399
hemorrhage, 126 ultrasonography distal one-third ureter, 400–401
incomplete colonoscopy, 127 Tumor cell entrapment theory, 301 iatrogenic, 398
infectious complications, 126–127 Tumor necrosis factor-alpha (TNF-α), 204, laceration, 399
perforation, 126 331 ligation, 399
postcolonoscopy colitis, 127 Tumor regression grade (TRG), 282 location-dependent repair of iatrogenic,
postpolypectomy syndrome, 126 Tumor response assessment, 281, 284; see 400
rigid proctoscopy, 122–123 also Rectal cancer management middle one-third ureter, 400
splenic injury, 127 clinical and endoscopic assessment, 282 prevention, 398–399
technical complications, 125 complete clinical response, 282 proximal one-third ureter, 400
Transanal excision (TAE), 266, 268; see also diffusion-weighted magnetic resonance, psoas bladder hitch, 401
Rectal cancer transanal approaches 283 thermal, 399
Transanal minimally invasive surgery radiological assessment, 282–283 transureteroureterostomy, 402
(TAMIS), 266, 268; see also Rectal timing for, 281 types of, 399
cancer transanal approaches tools in, 282 ureter, 398
Transanal total mesorectal excision variation in total lesion glycolysis, 283 ureteral replacement by ileum, 400
(TaTME), 256, 261–262; see also ureteroureterostomy, 399
Rectal cancer surgery, restorative Ureteroureterostomy, 399
U
Transanal transabdominal (TATA), Urethral injuries, 395–396; see also Ureteral
260–261; see also Rectal cancer UC, see Ulcerative colitis injuries; Urologic complications
surgery, restorative U-flap technique, 201; see also Pilonidal Urinary fistulas, 396
Transanal ultrasound (TAUS), 96, 98; disease Urinary tract infections (UTI), 80, 91
see also Transrectal ultrasound; UGIB, see Upper gastrointestinal bleeding Urologic complications, 395; see also
Transvaginal and transrectal Ulcerative colitis (UC), 133–134; see also Ureteral injuries
ultrasonography Inflammatory bowel disease artificial devices, 404–405
Index 423
bladder dysfunction, 402–404 Venous thromboembolism (VTE), 6, 83, W

bladder injuries, 397–398 335
Warfarin, 9, 49
renal injuries, 401–402 Ventral mesh rectopexy (VMR), 136,
Watch and wait strategy (WW strategy),
sexual dysfunction, 404 229–230; see also Rectal prolapse
279, 284; see also Rectal cancer
urethral injuries, 395–396 operative repairs
management
U.S. Food and Drug Administration Vertical rectus abdominis myocutaneous
follow-up, 283–284
(FDA), 15, 122, 217 flap (VRAM flap), 288, 291; see
outcomes, 284–285
USMSTF, see U.S. Multi-Society Task Force also Rectal cancer, recurrent
WBC, see White blood cell
U.S. Multi-Society Task Force (USMSTF), 121 Veterans Health Information Systems
Wells operation, 230–231; see also Posterior
UTI, see Urinary tract infections and Technology Architecture
mesh fixation; Rectal prolapse
(VISTA), 62
operative repairs
Video-assisted anal fistula treatment
V White blood cell (WBC), 121
(VAAFT), 185–186; see also Anal
Whitehead procedure, 164–165; see also
VAAFT, see Video-assisted anal fistula fistula
Excisional hemorrhoidectomy
treatment Videoproctography (VP), 368
WOC, see Wound, ostomy, and continence
Vascular cushions, see Hemorrhoidal VISTA, see Veterans Health Information
WOCN, see Wound, Ostomy, and
cushions Systems and Technology
Continence Nurse
Vascular endothelial growth factor Architecture
Wound, ostomy, and continence (WOC),
(VEGF), 8 VMAT, see Volumetric-modulated arc
89
Vasopressin, 34 therapy
Wound, Ostomy, and Continence Nurse
Vasopressors in sepsis resuscitation, 34 Volumetric-modulated arc therapy
(WOCN), 58
VDRL, see Venereal Disease Research (VMAT), 312
WW strategy, see Watch and wait strategy
Laboratory VP, see Videoproctography
VEGF, see Vascular endothelial growth VRAM flap, see Vertical rectus abdominis
factor myocutaneous flap
Y
Venereal Disease Research Laboratory VTE, see Venous thromboembolism
(VDRL), 153 V-Y anoplasty, 193; see also Anal fissure Y-V anoplasty, 192; see also Anal fissure

Beck, David E. - Kann, Brian R. - Margolin, David A. - Vargas, H. David - Whitlow, Charles B - Improving Outcomes in Colon and Rectal Surgery (2019, CRC Press - Taylor & Francis Group)

Uploaded by

Copyright:

Available Formats

Beck, David E. - Kann, Brian R. - Margolin, David A. - Vargas, H. David - Whitlow, Charles B - Improving Outcomes in Colon and Rectal Surgery (2019, CRC Press - Taylor & Francis Group)

Uploaded by

Document Information

Original Title

Copyright

Available Formats

Share this document

Share or Embed Document

Sharing Options

Did you find this document useful?

Is this content inappropriate?

Copyright:

Available Formats

Beck, David E. - Kann, Brian R. - Margolin, David A. - Vargas, H. David - Whitlow, Charles B - Improving Outcomes in Colon and Rectal Surgery (2019, CRC Press - Taylor & Francis Group)

Uploaded by

Copyright:

Available Formats

Improving Outcomes in

Colon and Rectal Surgery

© 2018 by Taylor & Francis Group, LLC

No claim to original U.S. Government works

Printed on acid-free paper

International Standard Book Number-13: 978-1-138-62683-6 (Pack- Hardback and eBook)

Library of Congress Cataloging-in-Publication Data

Visit the Taylor & Francis Web site at

2 Preoperative bowel preparation 14

3 Anesthesia and intraoperative positioning 18

5 Intraoperative anastomotic challenges 38

6 Other intraoperative challenges 49

7 Optimizing use of electronic health records 61

8 Postoperative anastomotic complications 66

9 General postoperative complications 74

10 Care paths and optimal postoperative management 87

11 Limitations of anorectal physiology testing 93

12 Limitations of colorectal imaging studies 104

13 Transanal endoscopy 121

14 Laparoscopy for colorectal disease 129

15 Medical legal issues 141

16 Miscellaneous conditions 148

17 Outcomes and quality 155

18 Hemorrhoidal surgery 161

19 Nonoperative therapy for hemorrhoidal disease 171

21 Surgery and nonoperative therapy of anal fissure 188

22 Surgery for pilonidal disease and hidradenitis suppurativa 195

23 Surgical treatment of fecal incontinence 207

24 Surgery for rectal prolapse 224

25 Management of diverticulitis 235

26 Abdominal surgery for colorectal cancer 244

27 Abdominal restorative surgery for rectal cancer 253

28 Transanal approaches to rectal cancer 266

29 Abdominoperineal resection 271

32 Evaluation and management of peritoneal metastatic disease 296

33 Chemotherapy for colon and rectal cancer 306

35 Surgical management of ulcerative colitis 323

36 Surgery for Crohn disease 333

37 Evaluation and management of lower gastrointestinal bleeding 347

39 Operative and nonoperative therapy for chronic constipation 365

40 Colorectal trauma 380

41 Prevention and management of urologic complications after colorectal surgery 395

Brian R. Kann, MD, FACS, David E. Beck, MD, FACS,

Joseph C. Adongay MD Huisar Dao Campi MD

Alison Althans BA Thomas E. Cataldo MD, FACS, FASCRS

Traci L. Hedrick MD, MS, FACS, FASCRS Aaron L. Klinger MD

Terry C. Hicks MD, FACS, FASCRS Jonathan Lu MD

Farouq Manji MD Jeffery Mino MD

Shannon McChesney md Daniel E. Sarmiento MD

Allison B. McCoy PhD Nicole M. Saur MD, FACS, FASCRS

Jacob A. McCoy MD Skandan Shanmugan MD

Genevieve B. Melton-Meaux MD, PhD, FACS, FASCRS Benjamin D. Shogan MD

Robert J. Sinnott DO, FACS, FASCRS Charles C. Vining MD

Arida Siripong MD, FACS, FASCRS Katerina O. Wells MD

Sharon L. Stein MD, FACS, FACRS J. Christian Winters MD

Emily Steinhagen MD Moriah E. Wright MD

W. David Sumrall, III MD Massarat Zutshi MD, FACS, FASCRS

H. David Vargas MD, FACS, FASCRS

ARIDA SIRIPONG AND FAROUQ MANJI

27. Linder JA et al. AMIA Annu Symp Proc. 2006;2006: