| |

Received: 17 May 2020    Revised: 23 June 2020    Accepted: 30 June 2020

DOI: 10.1002/rth2.12414

ORIGINAL ARTICLE

Anticoagulation practice patterns in COVID-19: A global survey

Rachel P. Rosovsky MD, MPH1  | Kristen M. Sanfilippo MD, MPHS2  |

Tzu Fei Wang MD, MPH3  | Sandeep K. Rajan MBBS, MD4 | Surbhi Shah MD5 |
Karlyn A. Martin MD6 | Fionnuala Ní Áinle PhD, MB, MRCPI, FRCPath7  |
Menno Huisman MD8 | Beverley J. Hunt MD, OBE9  | Susan R. Kahn MD, MSc10,11  |
Barry Kevane MB, PhD12  | Agnes Y. Y. Lee MD, MSc, FRCPC13  |
Claire McLintock MD14  | Lisa Baumann Kreuziger MD, MS15
1
Department of Medicine, Massachusetts
General Hospital, Harvard Medical School, Abstract
Boston, MA, USA
2
Background: Best practice for prevention, diagnosis, and management of venous
Department of Medicine, Washington
University St. Louis School of Medicine thromboembolism (VTE) in patients with coronavirus disease 2019 (COVID-19) is un-
and John Cochran Veterans Administration known due to limited published data in this population.
Medical Center, St. Louis, MO, USA
3 Objectives: We aimed to assess current global practice and experience in manage-
Department of Internal Medicine, Ohio
State University Wexner Medical Center, ment of COVID-19–associated coagulopathy to identify information to guide pro-
Columbus, OH, USA
spective and randomized studies.
4
Department of Medicine, University
of Nebraska Medical Center College of Methods: Physicians were queried about their current approach to prophylaxis, di-
Medicine, Memphis, TN, USA agnosis, and treatment of VTE in patients with COVID-19 using an online survey tool
5
Department of Medicine, University of
distributed through multiple international organizations between April 10 and 14,
Minnesota, Minneapolis, MN, USA
6
Department of Medicine, Northwestern
2020.
University Feinberg School of Medicine, Results: Five hundred fifteen physicians from 41 countries responded. The major-
Chicago, IL, USA
7
ity of respondents (78%) recommended prophylactic anticoagulation for all hospital-
Department of Hematology, Mater
Misericordiae University Hospital and ized patients with COVID-19, with most recommending use of low-molecular-weight
Rotunda Hospital, University College Dublin heparin or unfractionated heparin. Significant practice variation was found regarding
(UCD) School of Medicine, Dublin, Ireland
8 the need for dose escalation of anticoagulation outside the setting of confirmed or
Department of Thrombosis and
Hemostasis, Leiden University Medical suspected VTE. Respondents reported the use of bedside testing when unable to
Center, Leiden, The Netherlands
perform standard diagnostic imaging for diagnosis of VTE. Two hundred ninety-one
9
Department of Haematology, King's
College, Pathology & Rheumatology, Guy's & respondents reported observing thrombotic complications in their patients, with 64%
St Thomas' NHS Foundation Trust, London, noting that the complication was pulmonary embolism. Of the 44% of respondents
England
10
who estimated incidence of thrombosis in patients with COVID-19 in their hospital,
Department of Medicine, McGill
University, Montreal, QC, Canada estimates ranged widely from 1% to 50%. One hundred seventy-four respondents
11
Division of Internal Medicine & Center noted bleeding complications (34% minor bleeding, 14% clinically relevant nonmajor
for Clinical Epidemiology, Jewish General
Hospital, Montreal, QC, Canada
bleeding, and 12% major bleeding).
12
Department of Hematology, Mater Conclusion: Well-designed epidemiologic studies are urgently needed to under-
University Hospital, University College of stand the incidence and risk factors of VTE and bleeding complications in patients
Dublin (UCD), School of Medicine, Dublin,
Ireland

This is an open access article under the terms of the Creative Commons Attribution-NonCommercial-NoDerivs License, which permits use and distribution in
any medium, provided the original work is properly cited, the use is non-commercial and no modifications or adaptations are made.
© 2020 The Authors. Research and Practice in Thrombosis and Haemostasis published by Wiley Periodicals LLC on behalf of International Society on Thrombosis
and Haemostasis..

Res Pract Thromb Haemost. 2020;4:969–983.  |

wileyonlinelibrary.com/journal/rth2     969
 |
970       ROSOVSKY et al.

13
Department of Medicine, University of
British Columbia, Vancouver, BC, Canada with COVID-19. Randomized clinical trials addressing use of anticoagulation are also
14
National Women's Health, Auckland City needed.
Hospital, Auckland, New Zealand
15
Department of Medicine, Blood Research
Institute, Versiti, Medical College of
Wisconsin, Milwaukee, WI, USA

Correspondence
Rachel P. Rosovsky, Massachusetts General
Hospital, Zero Emerson, Suite 118, Office
114, Boston, MA 02114, USA.
Email: [email protected]

Handling Editor: Cihan Ay

KEYWORDS

anticoagulants, bleeding, blood coagulation, COVID-19, venous thromboembolism

Essentials

• Physicians were surveyed about current venous thromboembolism (VTE) practice patterns.
• Anticoagulant recommendations for coronavirus disease 2019 (COVID-19) vary.
• Estimates of VTE incidence and dose and duration of anticoagulation varied among respondents.
• Randomized trials of anticoagulation in patients with COVID-19 are urgently needed.

1 |  INTRODUCTION about thromboprophylaxis, diagnosis, and treatment of VTE in pa-

tients with COVID-19.
The novel coronavirus disease 2019 (COVID-19) was first identi-
fied in Wuhan, China, in December 2019, and now has infected
over 8.6 million people and caused in excess of 460  000 deaths 2 | METHODS
worldwide as of June 20, 2020 (www.stati​s ta.com). 1,2 Studies
have demonstrated that patients who died of COVID-19 had higher Physicians were surveyed using an online survey tool
levels of plasma D-dimers on admission compared with those who (SurveyMonkey). The survey was sent by email to members of the
survived.3-7 Furthermore, autopsy studies of patients with COVID- Hemostasis and Thrombosis Research Society, Venous thrombo-
19 have found fibrin thrombi within the pulmonary vasculature Embolism Network US, the Latin American Cooperative Group for
8,9
supporting the presence of a hypercoagulable state. The over- Hemostasis and Thrombosis, Unit for Thrombosis and Hemostasis
all incidence of venous thromboembolism (VTE) in patients with at the Hospital de Clínicas in Uruguay, the Mexican Society of
COVID-19 is unknown. Studies evaluating incidence have been lim- Thrombosis and Hemostasis, the Asia Pacific Society of Thrombosis
ited to case reports and case series, with estimates ranging from and Haemostasis, the Thrombosis and Haemostasis Society of
as low as 1% in the general wards to as high as 69% in intensive Australia and New Zealand, the Irish Network for VTE Research,
care units using screening ultrasound. 3-7,10-27 The limited data pre- and the ISTH between April 10 and 14, 2020. Social media links
sent a challenge for health care providers in prevention, diagnosis, were also provided. The survey was written in English, but the in-
and management of VTE in patients with COVID-19. We therefore troduction was translated into Spanish for the Latin American
sought to assess current global practice patterns in the management Thrombosis and Hemostasis societies. The survey included direct
of COVID-19–associated coagulopathy and to identify unanswered questions, with the option of writing in a response if a specific one
questions that may guide prospective and randomized studies. We was not provided in the selections listed (see Appendix S1). Topics
asked clinicians to share their experience and recommendations of the 28-question survey included estimates of thrombotic and
ROSOVSKY et al. |
      971

TA B L E 1   Demographics and characteristics of survey TA B L E 1   (Continued)

respondents
Number of
Number of Characteristic respondents
Characteristic respondents
<10 180 (49.9)
Country n = 357 (%) 11-25 85 (23.6)
United States 96 (26.9) 26-50 43 (11.9)
Spain 55 (15.4) 51-100 35 (9.7)
United Kingdom 51 (14.3) 101-250 14 (3.9)
Canada 48 (13.5) 251-500 3 (0.8)
Ireland 13 (3.6) >500 1 (0.3)
Belgium 10 (2.8)
COVID, coronavirus disease 2019.
Italy 9 (2.5)
Australia 7 (2.0) hemorrhagic complications, management of various clinical scenar-
Peru 6 (1.7) ios in patients with COVID-19, type and intensity of anticoagulation,
Netherlands 5 (1.4%) and laboratory testing and diagnostic approaches. Multiple selec-
New Zealand 5 (1.4) tions were permitted (such that the combined percentage in each
Argentina 5 (1.4) question could exceed 100%) given the complexity of potential sce-

France 4 (1.1) narios. Demographics of respondents collected included practice

patient population (adult or pediatric), practice location, country of
Other 43 (12.0)
practice, and years of experience.
Practice n = 353 (%)
Survey results were summarized using descriptive statistics.
Adult 305 (86.4)
Student t tests and chi-square tests were used to evaluate for asso-
Adults/Pediatrics 37 (10.5)
ciations between responses to questions and demographic factors
Pediatrics 11 (3.1)
or between survey responses.
Specialty n = 347 (%)
Hematology 224 (64.6%)
General internal medicine/Hospitalist 69 (19.9) 3 | RESULTS
Pulmonary/Critical care 38 (11.0)
Vascular medicine 24 (6.9) 3.1 | Demographics of respondents
Cardiology 18 (5.2)
Medical ncology 15 (4.3) Five hundred fifteen participants answered at least one ques-

General pediatrics 4 (1.2)

tion on the survey and 71% of participants completed the survey.
Baseline characteristics of survey respondents are listed in Table 1
Years in practice n = 352 (%)
and Figure 1. Based on the responses of 278 respondents that listed
<5 44 (12.5)
their affiliated hospital, 266 hospitals were represented across 41
5-10 62 (17.6)
countries.
11-15 62 (17.6)
16-20 52 (14.8)
>20 132 (37.5) 3.2 | Anticoagulation
Number of patients who are COVID n = 365 (%)
positive at practicing hospital
All of the participants provided responses to the question to whom
<100 134 (36.7) they would recommend thromboprophylaxis. The majority of re-
100-250 84 (23.0) spondents (78%) recommended prophylactic anticoagulation for all
251-500 68 (18.6) hospitalized patients with COVID-19, while 43% also selected that
501-1000 41 (11.2) they would follow institutional guidelines for criteria for prophy-
1001-3000 18 (4.9) laxis. Eight percent of respondents recommended thromboprophy-
>3000 5 (1.4) laxis for all patients with COVID-19, irrespective whether inpatient
Unknown 15 (4.1) or outpatient. Multiple selections for choice of anticoagulant were

Number of patients cared for who are n = 361 (%)

allowed. Of the 453 respondents who indicated use of low-molec-
COVID positive ular-weight heparin (LMWH) for thromboprophylaxis, 61% (n = 278)
recommended fixed-dose LMWH, and 62% (n = 281) recommended
(Continues) weight-adjusted LMWH. One-third recommended unfractionated
 |
972       ROSOVSKY et al.

heparin (UFH) (22% prophylactic fixed dose and 12% weight- membrane oxygenation (ECMO) catheters leading them to recom-
adjusted dose). Direct oral anticoagulants (DOACs) were recom- mend escalation to therapeutic anticoagulation for those affected
mended by 6% (n = 27), with another 6% recommended a variety of patients. Ninety-six percent of the 391 respondents recommended
other regimens including escalated doses of LMWH for all patients LMWH, 48% UFH, 27% DOACs, 13% vitamin K antagonists, 10%
or based on D-dimer or disease severity. fondaparinux, and 7% intravenous (IV) direct thrombin inhibitors for
In response to when dose escalation of prophylactic antico- therapeutic anticoagulation.
agulation to intermediate dose was considered, 28% (n  =  122) of Extended VTE prophylaxis (after discharge), depending on the
respondents did not recommend escalated doses of prophylactic an- presence of risk factors, was recommended by 276 of the 449 (62%)
ticoagulation for any indication (Figure 2). If recommended, a variety respondents who answered this question. The most common risk
of factors were used to select patients for dose escalation (Figure 2), factor for recommending this approach was a history of VTE before
and LMWH was the most commonly mentioned agent (98%; n = 279) COVID-19 (31%, n = 141), but other indications included a history of
followed by UFH (26%; n  =  73). Recommendations to escalate to cancer (24%; n = 109), patients who required admission to the inten-
intermediate dosing did not differ between physicians practicing in sive care unit (ICU; 22%; n = 98), and patients meeting inclusion cri-
the United States and other countries, between providers practicing teria of prior trials for extended prophylaxis for medically ill patients
in hospitals with <250 patients who were COVID positive or >250 (21%; n = 93). Other risk factors identified include hospitalization for
patients who were COVID positive, or between hematologists and COVID-19 (20%; n = 87), D-dimer greater than two times the upper
other medical specialties. Eighty-two percent of physicians recom- limit of normal (ULN; 18%; n = 79), obesity (15%; n = 69), and preg-
mending dose escalation also noted that their patients experienced nancy (12%; n = 54). Most respondents recommend LMWH (78%,
thromboembolism compared to 69% of physicians who did not rec- n  =  207) for extended VTE prophylaxis, followed by rivaroxaban
ommend intermediate prophylaxis (P < .01). (32%; n = 86), apixaban (24%; n = 65), and betrixaban (2%, n = 6).
Recommendations for escalation to a therapeutic dose of anti- Recommendations for extended prophylaxis did not differ between
coagulation was reported by 398 (77%) of respondents. Indications physicians practicing in the United States and other countries but
for dose escalation included a new diagnosis of atrial fibrillation or was more often recommended by those who practiced in hospitals
VTE (86%; n = 341) and high clinical suspicion of VTE but unable to with more than 250 COVID-19 admissions (67% vs 54%; P = .02), he-
obtain diagnostic testing (78%; n = 310). No respondents reported matologists/oncologists compared to other medical specialties (67%
that they would escalate to therapeutic anticoagulation in all pa- vs 47%; P < .01), or from physicians who noted their patients experi-
tients hospitalized with COVID-19, while 2% reported that there enced thrombotic complications (66% vs 38%; P < .01).
was no indication that would lead them to escalate to therapeutic
anticoagulation (Figure 2). Other indications for which respondents
would escalate to therapeutic doses of anticoagulant therapy varied, 3.3 | Diagnosis
including some respondents using certain clinical scoring systems to
guide escalation to full-dose anticoagulation (Figure 2). In addition, Three hundred ninety-one (75%) participants responded to ques-
seven respondents reported clotting of circuits such as continuous tions regarding using compression ultrasound (CUS) to diagnose
renal replacement therapies (CRRT), dialysis filters, or extracorporeal deep vein thrombosis (DVT) in patients who are COVID-19 positive.

F I G U R E 1   Percentage of survey respondents by county. Respondents from the United States were identified by region. This figure
represents the nationalities reported by each respondent. The expanded area is the breakdown of United States by region
ROSOVSKY et al. |
      973

35

Intermediate Dose
30

Therapeutic Dose

25
Percent of Respondents

20

15

10

0
U

LN

en
e

tin
LN
e

se
ed

n
e

LN

or
IC
on

e
or

ni
or

og

rri

be

ct
or
liz

U
U

po
Sc

Sc
N

fa
Fe
Sc
ta

in
x

O
3x

5x

tro
>6
FA
pi

br
IC

k
2-

ed
4-
C

ris
os

Fi

g
D

er
SI
SO

at
er

in
er
lH

TE
ed
ed

im

is
ev
im
ed

im
ed
Al

R
-d

at
at

lV
El
-d
at

-d

ev
ev

D
at

na
D
ev

D
ev

El
El

tio
El
El

di
Ad
Clinical Factor

F I G U R E 2   Percentage of survey respondents recommending escalated doses of anticoagulation to intermediate or therapeutic dosing
based on clinical scenarios. This figure highlights the indications for which respondents would elect to escalate prophylactic anticoagulation
to intermediate or therapeutic doses. DIC, disseminated intravascular coagulation; ICU, intensive care unit; SIC, sepsis-induced
coagulopathy; SOFA, sequential organ failure assessment; ULN, upper limit of normal; VTE, venous thromboembolism

Eighty percent reported obtaining CUS only in patients with clinical answered questions regarding ongoing laboratory monitoring.
symptoms of DVT, 17% of participants reported testing based on the The most frequent baseline laboratory test ordered was a com-
D-dimer results, and 8% reported testing in all ICU patients (some plete blood cell count (CBC; 93.9%). Laboratory tests ordered
even reported monitoring periodically or based on D-dimer trends). at baseline by more than 75% of respondents included D-dimer,
An additional 2% reported obtaining CUS in all patients upon hospital prothrombin time (PT), activated partial thromboplastin time
admission. If unable to obtain standard imaging, 59% reported diag- (aPTT), fibrinogen, and C-reactive protein (CRP). Additional lab-
nosing patients with pulmonary embolism (PE) based on worsening oratory tests frequently ordered included baseline basic meta-
respiratory status or right-heart strain on bedside echocardiogram. bolic panel (BMP) or comprehensive metabolic panel, ferritin,
Fifty-five percent reported using hemodynamic instability, 48% re- and lactate dehydrogenase. Similarly, the most frequent labora-
ported unilateral limb swelling, and 42% reported using clinical scor- tory tests ordered to monitor COVID-19 patients at least three
ing tools. An additional 35% reported using increasing D-dimer, 11% times per week was a CBC. Additional laboratory tests ordered
reported IV-line malfunction or increase in ventilated to perfused lung at least three times per week by >50% of respondents included
areas (ie, dead space), and 5% reported the need for proning as sur- BMP, D-dimer, PT, aPTT, fibrinogen, and CRP. The infrequently
rogates for thrombosis. ordered (<5%) coagulation tests at baseline or for routine moni-
toring included antithrombin activity, ADAMTS-13 activity,
antiphospholipid antibodies (APLAs), thromboelastography, tro-
3.4 | COVID-19 laboratory monitoring ponin, and von Willebrand factor activity. Finally, when asked
about changes in practice, several respondents indicated a shift
Three hundred ninety-two (76%) respondents answered ques- from monitoring UFH using aPTT to monitoring with anti-Xa lev-
tions regarding baseline laboratory ordering practices, while 380 els, while some respondents indicated incorporating the practice
974       | ROSOVSKY et al.

of using anti-Xa levels to monitor dosing of prophylactic or ther- did not know the incidence in their hospitalized and ICU patients,
apeutic LMWH. respectively (Figure  3). Of the respondents who estimated the in-
cidence of thrombosis, approximates ranged from 1% to >50%
(Figure 3). Incidence of thrombosis was estimated to be higher in ICU
4 |  OUTCOMES patients compared to all hospitalized patients (P = .02). Of the 261
respondents who provided what dose of anticoagulation patients
4.1 | Bleeding complications were on when thrombotic complications occurred, 39% (n  =  101)
reported none, 84% (n = 218) reported prophylactic, 18% (n = 47)
Questions about bleeding complications were answered by 377 (73%) reported intermediate, and 11% (n = 30) reported therapeutic dose.
physicians. Over half (n = 203) reported they had witnessed no bleed- Of the 367 respondents who reported on thrombotic complica-
ing complications, minor bleeding was reported by 34% (n = 129), clini- tions, 21% (n  =  76) reported no thrombotic complications in their
cally relevant nonmajor bleeding by 14% (n = 54), and major bleeding COVID-19 patients. The majority of thrombotic complications re-
by 12% (n = 46). The most common bleeding sites reported included ported were PE (64%; n  =  234) followed by lower-extremity DVT
cutaneous/line related (41%; n  =  65), mucous membranes (41%; (49%; n = 181), upper-extremity DVT (19%; n = 71), and superficial
n  =  65), gastrointestinal (27%; n  =  43), hemoptysis/alveolar hemor- vein thrombosis (9%; n = 34). Few participants identified thrombosis
rhage (22%; n = 35), genitourinary (16%; n = 27), retroperitoneal (13%; in unusual locations with 5% (n = 19) reporting intracardiac throm-
n = 21), neurologic (10%; n = 16), and muscular (3%; n = 2). Bleeding bosis, 3% (n = 11) splenic vein thrombosis, and 4% (n = 13) cerebral
complications were most often reported in patients on therapeutic an- vein thrombosis. For arterial thrombosis, 20% (n  =  75) of respon-
ticoagulation (65%; n = 97), followed by intermediate (27%; n = 41) or dents reported ischemic stroke, 12% (n = 52) myocardial infarction,
prophylactic dose (41%; n = 62). Nine percent of respondents reported and 9% (n  =  34) peripheral artery embolism had occurred in their
that bleeding occurred without anticoagulation. patients. Twenty nine percent (n = 105) of respondents reported a
high clinical suspicion for VTE in patients for whom they were un-
able to obtain diagnostic testing. In addition, 16% (n = 59) reported
4.2 | Thrombotic complications sudden death with concern for thrombosis. For adjunctive therapies,
12% (n  =  45) reported thrombotic complications associated with
When queried about the approximate incidence of VTE in patients mechanical circulatory support (ie, ECMO, ventricular assist device),
with COVID-19, 56% of the 293 and 55% of the 290 participants and 25% (n = 90) reported these complications with dialysis or CRRT.

60

5 | DISCUSSION
All hospitalized patients
50
The COVID-19 pandemic has affected all countries and has required
ICU patients
rapid adaptation to clinical practice with limited published evidence.
Our survey identified several areas with consensus on management
40
Percent of Respondents

of COVID-19, including the use of therapeutic anticoagulation for all

confirmed or clinically suspected VTE and universal use of thrombo-

30 prophylaxis for all hospitalized patients. Common VTE risk factors

identified that might prompt consideration of higher-dose prophy-
laxis or extended prophylaxis included ICU care and cancer. Most
20 clinicians recommended LMWH or UFH as anticoagulants of choice.
Many organizations including American Society of Hematology
(ASH), ISTH, World Health Organization, American College of
10 Cardiology (ACC), Anticoagulation Forum, British Thoracic Society,
and CHEST have published guidance for the prevention of VTE in pa-
tients with COVID-19 (Table 2). 28-38 These guidance statements rec-
0 ommend standard thromboprophylaxis in all hospitalized patients
Unknown 1-5 6-10 11-15 16-30 >30 with COVID-19 infection unless a strong contraindication is present.
Estimated incidence of thrombosis Our survey responses reflect adoption of this guidance into practice.
The use of LMWH as the most commonly recommended regimen is
F I G U R E 3   Reported incidence of thrombosis in hospitalized and
also consistent with published guidance. LMWH in addition to hepa-
ICU patients with COVID-19. This figure represents the estimated
rin has been shown to have anti-inflammatory properties, which may
incidence of thrombosis, reported by each respondent, for all
hospitalized patients and ICU patients. COVID-19, coronavirus be an added benefit in COVID-19 infection where proinflammatory
disease 2019; ICU, intensive care unit. cytokines are markedly elevated.39-41
ROSOVSKY et al. |
      975

TA B L E 2   Summary of recommendations in VTE prophylaxis from various guidance and survey responses for patients with COVID-19

Acute or Escalation from prophylactic

Acutely ill; no Critically ill; no bleeding Acute or critically ill critically ill with dose to therapeutic or Extended prophylaxis post
Study bleeding risk risk with bleeding CrCl < 30 mL/min intermediate dose discharge (criteria, agent)

ASH-FAQ33 LMWH or Same as acutely ill and Mechanical prophylaxis UFH (twice daily to Unknown in critically ill Consider based on
fondaparinux recommend participation three times daily) Reasonable to consider in inclusion criteria from
favored over UFH in well-designed patients who experience previous trials (eg,
to reduce contact clinical trials and/or recurrent clotting of access combination of age,
unless the risk of epidemiologic studies devices comorbidities, and
bleeding is judged when available Recommend participation in D-dimer >2 times ULN).
to exceed the risk well-designed clinical trials Any decision needs to
of thrombosis and/or epidemiologic studies consider VTE risk factors
when available. (including reduced
mobility) and bleeding
risk as well as feasibility.
Consider betrixaban
orrRivaraban or
ASA45,46,72,73

Anticoagulation Standard-dose VTE Suggest increased doses Mechanical prophylaxis Dose adjust for Suggest against intensification Consider on a case by case
forum36 prophylaxis as per of VTE prophylaxis with regular renal function of anticoagulant dosing based basis in patients with
existing societal (intermediate dose reassessment for only on biomarkers. However, ongoing VTE risk factors
guidelines LMWH/UFH or low- conversion to acutely worsening clinical and low bleeding risk.
intensity UFH infusion), pharmacologic status in conjunction with
based largely on expert prophylaxis. laboratory value changes,
opinion. may necessitate further
Reasonable to employ thromboembolic workup or
both pharmacologic empiric treatment.
and mechanical VTE
prophylaxis if no
contraindication

ACC28 LMWH may be LMWH may be Mechanical prophylaxis UFH (twice daily to Insufficient data Consider up to 45 d
advantageous over advantageous over UFH three times daily) Majority reccommend against if elevated VTE risk
UFH to reduce to reduce exposure escalation without high bleeding
exposure 32% favored intermediate and risk.
5% therapeutic Panelist breakdown if
considering:
51% DOAC
24% LMWH

ISTH29 LMWH in the LMWH in the absence of Not Specified UFH (twice daily to Escalation to therapeutic if Not specified
absence of any any contraindications three times daily) presumed VTE
contraindications (active bleeding
(active bleeding and platelet count
and platelet count <25 × 109/L
<25 × 109/L

SSC of the Standard-dose UFH Prophylactic-dose UFH or Mechanical VTE prophylaxis Patients with obesity as Consider for all
ISTH35 or LMWH should LMWH. Intermediate- thromboprophylaxis recommendations defined by actual body weight hospitalized patients
be used after dose LMWH (50% of should be or BMI should be considered that meet high VTE risk
careful assessment respondents) can also be modified based for a 50% increase in dose of criteria
of bleed risk, with considered in high-risk on deteriorating thromboprophylaxis Either LMWH (30%) or a
LMWH as the patients renal function Treatment-dose heparin should DOAC (ie, rivaroxaban
preferred agent a  Multimodal not be considered for primary or betrixaban 30% of
Intermediate-dose thromboprophylaxis with prevention until results of respondents can be used.
LMWH may also be mechanical methods randomized controlled trials Duration can be
considered (30% of should be considered are available approximately 14 d
respondents) (60% of respondents) at least (50% of
respondents), and
up to 30 d (20% of
respondents)

BTS31 Standard risk: Standard risk: prophylactic Not specified Not specified Not possible to advocate Consider in high-risk
prophylactic LMWH (daily) any particular escalation patients (h/o VTE, cancer,
LMWH (daily) High risk: LMWH (twice approach and suggest reduced mobility, or ICU
High risk: LMWH daily) developing local protocols for admission) and if risk of
(twice daily) risk stratification in patients VTE is greater than risk
with COVID-19 of bleeding
Consider intermediate dose If considering: prophylactic
in high risk patients and LMWH or DOAC
therapeutic in proven or
suspected acute VTE.

(Continues)
976       | ROSOVSKY et al.

TA B L E 2   (Continued)

Acute or Escalation from prophylactic

Acutely ill; no Critically ill; no bleeding Acute or critically ill critically ill with dose to therapeutic or Extended prophylaxis post
Study bleeding risk risk with bleeding CrCl < 30 mL/min intermediate dose discharge (criteria, agent)

Thrombosis-UK/ LMWH or LMWH + mechanical Mechanical prophylaxis UFH (twice daily to Therapeutic for presumed PE Not specified
BSH32,65 fondaparinux compression stockings (in bleeding and if three times daily)
according to license platelets <30 × 10 9/L) or reduced dose
LMWH

Dutch50 Prophylactic LMWH Prophylactic LMWH Not specified Not specified Therapeutic if VTE is Not specified
irrespective of risk irrespective of risk score confirmed
score If imaging not possible
and D-dimer increases
progressively, consider
therapeutic AC

Chinese30 Use Padua or LMWH over UFH Mechanical Prophylaxis UFH (BID) If VTE suspected and unable Consider if persistent
IMPROVE RAM to to be confirmed due to risk of VTE at time of
calculate risk and restricted conditions, discharge
if high or moderate curative anticoagulation LMWH favored over
risk, LMWH recommended in absence of DOAC due to potential
recommended. contraindications drug-drug interactions
and/or frequent
comorbidities

CHEST34 LMWH or LMWH over UFHb ; and Mechanical prophylaxis Not specified Insufficient data to Extended
fondaparinux over LMWH or UFH over justify increased- thromboprophylaxis
UFHb ; and LMWH, fondaparinux or DOACc  intensity anticoagulant in patients at low risk
fondaparinux or thromboprophylaxis in the of bleeding should
UFH over DOACc  absence of randomized be considered, if
controlled trials emerging data on the
postdischarge risk of VTE
and bleeding indicate
a net benefit of such
prophylaxis

VENUS Survey 78% VTE prophylaxis 33% use intermediate Not specifically queried Not specifically 28% no escalation 39% no postdischarge,
61% LMWH dose in ICU patients queried 72% escalation for select 61% postdischarge (risk
33% UFH 9% use therapeutic dose in patients factors including ICU
6% DOAC ICU patients stay, D-dimer, obesity,
cancer, h/o VTE)
Respondent breakdown:
78% LMWH,
24% apixaban,
2% betrixaban,
32% rivaroxaban

ACC, American College of Cardiology; ASA, aspirin; ASH, American Society of Hematology; BMI, body mass index; BTS, British Thoracic Society;
COVID-19, coronavirus 2019; CrCl, creatinine clearance; DOAC, direct oral anticoagulant; FAQ, frequently asked questions; H/o, history of; ICU,
intensive care unit; LMWH, low-molecular-weight heparin; N/A, not applicable; RAM, risk assessment models; SIC, sepsis-induced coagulopathy;
SSC, Scientific and Standardization Committee; SSH, Swedish Society of Hematology; UFH, unfractionated heparin; ULN, upper limit of normal;
VENUS, Venous thromboEmbolism Network United States; VTE, venous thromboembolism.
a
VTE prophylaxis recommendations should be modified based on extremes of body weight, severe thrombocytopenia (ie, platelet counts of
50 000 × 109/L or 25 000 × 109/L). DOACs should be considered with caution as coadministration of immunosuppressant, antiviral and other
experimental therapies may potentiate or interfere with DOAC therapy.
b
Favors this approach to limit staff exposure.
c
Cautions against the use of DOACs in these patients secondary to the high risk of rapid clinical deterioration in these patients.

Our study also demonstrated significant differences in practice than non–critical care patients (Table 3). Almost one-third of our re-
patterns such as the decision to escalate the dose of anticoagulation spondents recommended dose escalation in obese patients, likely
and/or to consider extended prophylaxis and in the use of laboratory based on prior studies: One retrospective analysis demonstrated de-
monitoring. The indication to consider dose escalation from prophy- creased VTE incidence in obese patients with escalated prophylactic
lactic to intermediate doses of anticoagulation is a subject of debate dosing,42 and another showed that weight-based dosing of enoxa-
that is not currently informed by high-quality data. International parin, 0.5 mg/kg twice daily, achieved higher anti-Xa levels in obese
guidance reflects this discrepancy in practice, with some organiza- patients compared to prophylactic or fixed dosing.43 However, major
tions suggesting intermediate-dose anticoagulation for prophylaxis, VTE guidelines have not addressed weight-based dosing due to lack
whereas others have not (Table 2). Escalation to intermediate dosing of randomized data, and additional studies are needed.
for thromboprophylaxis in ICU patients may be based on the limited We also found that while the majority of respondents recom-
existing data that suggests ICU patients have a higher risk of VTE mended escalation to therapeutic anticoagulation only in those
ROSOVSKY et al. |
      977

TA B L E 3   Published studies evaluating incidence of VTE in patients with COVID-19

Anticoagulant and
Study Country Study design N Population dose Rate of VTE
12,53
Klok et al Netherlands Retrospective 184 ICU Prophylactic or Original cumulative
cohort intermediate incidence of VTE and ATE
(median of 7 d): 31%. (VTE:
27%)
Updated cumulative
incidence of VTE and ATE
(median 14 d)a : 49%
Cui et al10 China Retrospective 81 ICU None 25% (20/81) (DVT)
cohort
Helms et al19 France Prospective 150 ICU Prophylactic: 70% 18.7% (28/150)
cohort Therapeutic: 30%
Poissy et al13 France Retrospective 107 ICU Prophylactic: 91% 15-d cumulative incidence:
cohort Therapeutic: 9% 20.4% (PE)
Desborough et al17 United Kingdom Retrospective 66 ICU Prophylactic: 83% 15% (10/66) PE in 30 d
cohort Therapeutic: 17%
Fraissé et al18 France Retrospective 92 ICU Prophylactic: 47% 31/92 (33.7%)
cohort Therapeutic: 53%
Thomas et al24 United Kingdom Retrospective 63 ICU Prophylactic Cumulative incidence: 27%
cohort
Llitjos et al20 France Retrospective 26 ICU Prophylactic: 31% 69% (18/26)
cohort Screening Therapeutic: 69%
100%
Longchamp27 France Retrospective 25 ICU Prophylactic: 92% 32% (8/25)
cohort Screening (weight-based)
100% Therapeutic: 8%
Nahum et al23 France Retrospective 34 ICU Prophylactic 27/34 (79%)
cohort Screening
100%
Voicu et al25 France Prospective 56 ICU Prophylactic: 78% 26/56 (46%) DVT
cohort Screening 100% Therapeutic: 13%
Middeldorp et al22 Netherland Retrospective 75 ICU Prophylactic or 21-d cumulative incidencea :
cohort Screening 27% intermediate 59% (symptomatic: 34%)
Goyal et al11 United States Retrospective 393 Invasive Not specified 7.7% (10/130)
(New York) cohort mechanical 3/263 (1.1%)
ventilation
Noninvasive
mechanical
ventilation
Lodigiani et al21 Italy Retrospective 388 ICU: 16% Prophylactic, Cumulative rate (VTE + ATE)
cohort Regular ward: intermediate, or 21% (27.6% ICU, 6.6%
84% therapeutic (100% ward)
in ICU, 75% in
ward)
Bompard et al15 France Retrospective 135 ICU (18%) Prophylactic or Total 32/135 (24%) PE
cohort Regular ward intermediate ICU 50%
(35%) (obese and ICU) in Others 18%
ER (47%) inpatients
Al-Samkari et al7 United States Retrospective 400 ICU (36%) Prophylactic Total: 4.8% radiographically
(Boston) cohort Regular ward confirmed VTE (4.13 per
(64%) 100 patient-weeks)
ICU – 7.6% (4.76 per 100
patient-weeks)
Regular ward- 3.1% (3.49 per
100 patient-weeks)

(Continues)
 |
978       ROSOVSKY et al.

TA B L E 3   (Continued)

Anticoagulant and
Study Country Study design N Population dose Rate of VTE
26
Zhang et al China Retrospective 143 Regular ward Prophylactic: 37.1% 66/143 (46.1%) LE DVT
cohort Screening 100%
Middeldorp et al22 Netherland Retrospective 123 Regular ward Prophylactic 21-d cumulative incidence
cohort Screening 27% of both any VTE and
symptomatic VTEa : 9%
Artifoni et al14 France Retrospective 71 Regular ward Prophylactic 16/71 (22.5%) [including 7
cohort Screening 100% (9.8%) PE]
Demelo-Rodriguez Spain Prospective 156 Regular ward Prophylactic: 98% 23/156 (14.7%) DVT
et al16 cohort Screening 100%

ATE, arterial thrombotic event; COVID-19, coronavirus disease 2019; DVT, deep vein thrombosis; ER, emergency room; ICU, intensive care unit; PE,
pulmonary embolism; VTE, venous thromboembolism.
a
These studies released updates or were finalized after the surveys were submitted.

with confirmed or high suspicion of VTE, a small number of clini- costs of different drugs across different regions and countries. In ad-
cians recommended escalation to therapeutic anticoagulation in dition, the need for thromboprophylaxis in patients who are COVID-
patients with additional risk factors (such as elevated D-dimer). 19 positive treated as outpatients is another commonly encountered
The risks and benefits of this approach remain unknown and sev- question for clinicians, and practices vary. Prospective studies are
eral multicenter international trials are under way to address the needed to better define the management of extended prophylaxis in
utility of escalating doses of anticoagulation in hospitalized pa- patients with COVID-19.
tients with COVID-19. This survey also draws attention to some of the challenges in
There is also uncertainty in the role for extended prophy- diagnosing VTE events. Patients’ clinical instability and the scarcity
laxis. While the majority of respondents (62%) recommended of personal protective equipment may affect the ability to obtain
this practice for selected patients with COVID-19, there were a prompt diagnostic imaging. In addition, renal function can be com-
variety of clinical factors influencing this practice. Furthermore, promised in patients with severe COVID-19 infection and limit imag-
while LMWH was the most commonly recommended agent for ing requiring contrast. The majority of the participants report using
extended prophylaxis, additional anticoagulants, including various bedside Doppler ultrasound and echocardiogram to diagnose a VTE.
DOACs, were suggested. These responses highlight the variabil- Case reports have highlighted the importance of awareness for PE as
ity of current clinical practice and the uncertainty surrounding a potential cause for acute decompensation in patients with COVID-
optimal management. The role for extended prophylaxis follow- 19.51 Consensus recommendations from Obi et al52 provide practi-
ing hospitalization has been previously studied because of a high cal guidance in the diagnosis and treatment of VTE in patients with
percentage of VTE events (as high as 57%) occurring after hospital COVID-19 if imaging is unavailable. As is practiced by almost 80% of
discharge.44 Two DOACs, rivaroxaban and betrixaban, have been respondents, empiric anticoagulation without confirmatory imaging
approved for extended prophylaxis (30-45  days);45,46 however, should be considered in patients with a high clinical suspicion of VTE
they are not approved or reimbursed for this indication in all coun- while balancing the risk of bleeding.
tries, and how often these regimens are used in practice remains The number of studies reporting VTE continues to increase,
unclear. Moreover, our study demonstrates that the acutely ill pop- although rates of VTE vary dramatically (Table  3). Our survey re-
ulation in these studies are not the only factors influencing the sponses reflect the differences and heterogeneity in the literature
decision to recommend extended prophylaxis. Extended VTE pro- currently available. When screening ultrasound was performed on
phylaxis has been shown to be beneficial in clinical settings such as admission, Llitjos et al20 reported a cumulative incidence of VTE of
following orthopedic surgery, and abdominal and pelvic surgery for as high as 69%, 23% of which were PE, despite 30% of patients re-
patients with cancer,47 as well as in high-risk ambulatory patients ceiving prophylactic anticoagulation and 70% receiving therapeutic
with cancer receiving chemotherapy.48,49 anticoagulation. In another study of three ICUs in the Netherlands,
Our study supports that clinicians are concerned that patients Klok et al12,53 reported a cumulative rate of all thrombosis of 49%,
with COVID-19 are at increased risk of VTE, leading to recommen- despite all the patients receiving a prophylactic dose of anticoagula-
dations for a role for extended thromboprophylaxis following dis- tion. However, it is important to note that the prophylactic LMWH
charge; a practice that is considered in many guidelines. 28-33,36,50 dose initially used in two of the three ICUs was lower than what
However, almost 40% of providers do not recommend VTE prophy- is recommended by the manufacturer.54 In another study in ICUs
laxis after hospital discharge. Our survey also highlights a variabil- in France, when compared with a matched historical control of pa-
ity in current practice on the type of anticoagulant used following tients who do not have COVID-19 but have acute respiratory dis-
hospital discharge. This variability may reflect local availability and tress syndrome (ARDS), patients with COVID-19 with ARDS had
ROSOVSKY et al. |
      979

TA B L E 4   Recommended laboratory monitoring of hospitalized patients with COVID-19

Thrombosis
ASH63,70 ISTH64 ACC28 UK/BSH32,65 SSH61

Laboratory parameter
aPTT x x
ALT x
Creatinine x
D-dimer x x x x
Fibrinogen x x x x
LDH x
Platelets x x x x
PT/INR x x x x
DIC transfusion management
Nonbleeding Routine blood Maintain: PLT > 25 × 109/L Maintain: Routine blood products not Not
patient products not PLT > 20 × 109/L in recommended specified
recommended those with a high
bleeding risk or
requiring an invasive
procedure
Bleeding patient Maintain: Maintain: PLT > 50 × 109/L Maintain: Maintain: PLT > 50 × 109/L Not
PLT ≥ 50 × 109/L Fibrinogen > 1.5 g/L PLT > 50 × 109/L Fibrinogen ≥ 1.5 g/L specified
Fibrinogen ≥ 1.5 g/L PT Ratio < 1.5 Fibrinogen ≥ 1.5 g/L aPTT or PT ≤ 1.5 × ULN
INR < 1.8 aPTT or TA 1 g IV for patients
PT ≤ 1.5 × ULN without DIC
rVIIa and PCC are not
recommended given they
are prothrombotic

ACC, American College of Cardiology; ALT, alanine transaminase; aPTT, activated partial thromboplastin time; ASH, American Society of
Hematology; INR, International normal ratio; LDH, lactate dehydrogenase; PCC, prothrombin complex concentrate; PLT, platelet count; PT, protime;
rVIIa, recombinant activated factor VIIa; SSH, Swedish Society of Hematology; TA, tranexamic acid; ULN, upper limit of normal.

a 2.6-fold increased odds of thromboembolic complications and a extracorporeal circulation cannula hematoma.19 In the Boston area
6-fold increased odds of PE.19 In contrast to these high rates of VTE, study, the overall rate of hemorrhage was 4.8%, which was similar
retrospective reports from the United States found much lower to the overall rate of VTE reported.7 In a recent study from the
rates of VTE. In 393 consecutive patients admitted to two New York United Kingdom of patients admitted to a critical care unit, 11%
hospitals, the rate of VTE in patients on mechanical ventilation was suffered from a major bleed.17 Similarly, in a French study of 92
only 7.7%, and in the first 400 patients admitted to five affiliated ICU patients, the overall rate of hemorrhagic events was 21%, and
Boston hospitals, the overall rate of VTE in ICU patients was similar, notably, 84% of those were on therapeutic anticoagulation.18 In
7,11
at 7.6%. Differences in the VTE incidence may be reflective of the current survey, 43% of respondents noted that their patients
differences in screening, disease severity (such as ICU status), pa- had bleeding complications. Importantly, this response does not
tient characteristics, or other concurrent therapies, or detection of reflect a true bleeding incidence; it is the percentage of bleeding
immunothromboses that are counted as PE instead of in situ throm- per respondent and not per patient. Furthermore, it may be influ-
bosis, and other factors that have yet to be identified.55 An online enced by recall bias in our survey, lack of documentation of bleed-
COVID registry of thrombosis is being planned, and although obser- ing complications in previous studies, and difficulty with obtaining
vational, it aims to provide representative data on the magnitude of data on hemorrhage without intensive chart review. The majority
56
this problem. of bleeding events (65%) were reported to be on therapeutic an-
Not to be overlooked, anticoagulation, particularly with inter- ticoagulation, which is similar to what has been described in the
mediate or therapeutic doses, may increase the risk of bleeding. latest studies. In a recent nationwide data set from China, patients
Hemorrhage was not initially perceived as a major complication found to have a high risk of VTE using the Padua Prediction Score
in most studies of patients with COVID-19; however, more re- on admission were also found to have a high risk of bleeding. 57
cent studies have reported higher numbers. In a study of 150 Thus, attention should be paid to the balance of bleeding and
French ICU patients, only 4 (2.7%) were reported to have bleed- thrombosis in the management of patients with COVID-19, and
ing complications, which included intracerebral hemorrhage and more studies evaluating these risks are much needed.
 |
980       ROSOVSKY et al.

Laboratory monitoring of hospitalized patients with COVID-19 continues to evolve rapidly. Our survey results reflect practice in
provides a means to guide care but also may provide predictive and April 2020 which may change over time.
prognostic information. One of the most commonly reported ab- Identification of current practice patterns about prevention, di-
normal laboratory findings in patients with COVID-19 is highly ele- agnosis, and treatment of VTE in patients with COVID-19 has im-
vated D-dimer levels.3-7 Others include elevated fibrinogen, normal portant implications. Our survey highlights consensus including the
or mildly decreased platelets, and normal or near-normal aPTT and use of VTE prophylaxis with LMWH or UFH in hospitalized patients.
PT.58-60 The laboratory testing recommended by major organiza- However, there are many unanswered questions, as is reflected in
tions aligns with the reported practice by the survey respondents the heterogeneity of current available literature as well as of our
(Table 4). 28,61-65 The association of laboratory findings with disease survey responses, including the true incidence of VTE in different
severity also has been identified in several studies. In a single-center patient populations with COVID-19, the use and effects of escalated
cohort study of 198 patients, an elevated D-dimer was associated doses of anticoagulation, and whether extended prophylaxis should
22
with a 50% increased risk of developing VTE. Moreover, labora- be considered and changes outcomes. Well-conducted epidemio-
tory data may have prognostic value in patients with COVID-19.66 logic studies and clinical trials are urgently needed, and randomized
Higher levels of D-dimer have been associated with increased risk of trials addressing these issues are under way.
mortality in COVID-19.6,7,67,68 In a study of 343 patients, those with
a D-dimer of ≥2.0 μg/mL had a 51.5-fold increased risk of in-hospital AC KNOW L ED G M ENTS
mortality compared to those with a D-dimer of <2.0 μg/mL.69 Our The authors thank the Hemostasis and Thrombosis Research Society,
survey findings reflect the ubiquitous monitoring of D-dimer; 88% Venous thromboEmbolism Network United States, Latin American
of respondents recommend obtaining this test at baseline, and 72% Cooperative Group for Hemostasis and Thrombosis, Unit for
recommend monitoring it three times a week. Thrombosis and Hemostasis at the Hospital de Clínicas in Uruguay,
Although rare, COVID-19 can be associated with disseminated and the Mexican Society of Thrombosis and Hemostasis, the Asia
intravascular coagulation (DIC), usually in the later stages of infec- Pacific Society of Thrombosis and Haemostasis, the Thrombosis
tion.4 While the current guidelines provide recommendations for the and Haemostasis Society of Australia and New Zealand, the Irish
management of COVID-19–associated DIC, these recommendations Network for VTE Research, and the International Society on
are conflicting. 28,29,64,70 The interim ISTH guidance supports keep- Thrombosis and Haemostasis for support of the survey. Research re-
9
ing the platelet count >25 × 10 /L in patients with DIC even in the ported in this publication was supported by the National Center for
absence of bleeding (Table 4). 29,64 However, the recently published Advancing Translational Sciences of the National Institutes of Health
guidance from ACC and the ASH guidance statement recommend Award Number UL1-TR002494. The content is solely the responsi-
use of blood products only in the setting of DIC-related bleeding bility of the authors and does not necessarily represent the official
and/or for those in need of invasive procedures or with high risk of views of the National Institutes of Health. Dr Kahn is a Tier 1 Canada
28,70
bleeding (Table 4) Although none of the guidelines suggest es- Research Chair holder, and an investigator of the CanVECTOR
calation of anticoagulation in DIC, 12% of the survey participants Network, which receives grant funding from the Canadian Institutes
reported escalating to intermediate-dose anticoagulation in patients of Health Research (Funding Reference: CDT-142654).
with an elevated DIC score.
Routine testing for APLAs is not recommended in patients with R EL AT I O NS HI P D I S C LO S U R E
COVID-19,6 and only 5% of survey participants reported checking TZW, SS, KAM, BJH, SRK, BK, and CM declare no conflicts of in-
APLAs at baseline. Disease severity and medications used in COVID-19 terest. RPRhas received grants to her institution from BMS and
can affect lupus anticoagulant testing. Additional coagulation laboratory Janssen and consultant/advisory board for BMS, Dova, Janssen,
testing is either not recommended for routine patient management (eg, and Portola, all outside the submitted work; KS is a consultant
thromboelastography)70 or is indicated only in special circumstances.61 for Pfizer Inc and Bayer HealthCare Pharmaceuticals Inc; has re-
In line with this counsel, <5% of respondents pursue these tests. ceived grant funding from the National Heart, Lung, and Blood
Our study provides valuable information to reflect the cur- Institute and American Cancer Society Institutional Research
rent practice pattern of a diverse background of clinicians from 41 Grant; and has received travel support from AstraZeneca
countries. However, we note a few limitations. Due to the nature Pharmaceuticals Inc, all outside of the work submitted). SKRajan
of the survey, recall biases of the perceived rates of bleeding or has received research support from Octapharma, Uniqure,
thrombotic complications are likely. Although physicians from 41 Kedrion, Appelis and consultancy & speakers' honorarium from
countries responded to the survey, there is limited representation Alexion, Takeda, Biomarin, Sanofi-Genzyme, Bayer and Novo
from the Asian or African countries. Furthermore, we are unable Nordisk, all outside the submitted work; FNÁhas received in-
to provide the percentage of respondents, as we could not col- vestigator-initiated grants (paid to university) from Leo Pharma,
lect information on the total number of people who were invited Actelion, and Bayer, all outside the submitted work; MH has
to participate. Our survey was sent via email to multiple interna- received grants from ZonMW Dutch Healthcare Fund and
tional thrombosis groups as well as available on social media. Lastly, grants and personal fees from Boehringer-Ingelheim, Pfizer-
studies of COVID-19 patients are published daily and literature BMS, Bayer Health Care, Aspen, and Daiichi-Sankyo, outside
ROSOVSKY et al. |
      981

the submitted work; AYYLee has received honoraria for con- manifestations of SARS-CoV2 infection. Blood. 2020. https://doi.
org/10.1182/blood.20200​06520
sultancy for Bayer, BMS, LEO Pharma, Servier, and Pfizer and
8. Fox SE, Akmatbekov A, Harbert JL, Li G, Brown JQ, Vander Heide
research support from BMS, all outside the submitted work; RS. Pulmonary and cardiac pathology in Covid-19: the first autopsy
LBKis a consultant for the DHHS Vaccine Injury Compensation series from New Orleans. medRxiv. 2020:2020.04.06.20050575.
Program and has received honoraria for participation in advi- 9. Luo W, Hong YU, Gou J, Li X, Sun Y, Li J, Liu L. Clinical pathology
of critical patient with novel coronavirus pneumonia (COVID-19).
sory boards for CSL Behring and Quercegen Pharmaceuticals,
Preprints (not peer reviewed). Posted 10 April 2020; 2020.
all outside the submitted work. 10. Cui S, Chen S, Li X, Liu S, Wang F. Prevalence of venous thrombo-
embolism in patients with severe novel coronavirus pneumonia. J
AUT HOR CONTRI B UTI O N S Thromb Haemost. 2020;18(6):1421–24.
RR and LBK created the concept and design, and contributed to the 11. Goyal P, Choi JJ, Pinheiro LC, Schenck EJ, Chen R, Jabri A, et al.
Clinical characteristics of Covid-19 in New York City. N Engl J Med.
analysis and interpretation of data, creation of tables, critical writ-
2020;382(24):2372–74.
ing and revision of the intellectual content; and final approval. KMS, 12. Klok FA, Kruip M, van der Meer NJM, Arbous MS, Gommers D, Kant
TFW, SKR, SS, KM contributed to the analysis and interpretation of KM, et al. Incidence of thrombotic complications in critically ill ICU
data, writing and revision of the intellectual content; and final ap- patients with COVID-19. Thromb Res. 2020;191:145–7.
13. Poissy J, Goutay J, Caplan M, Parmentier E, Duburcq T, Lassalle F,
proval. FNA, MH, BJH, SRK, BK, AYYL, CM contributed to the de-
et al. Pulmonary embolism in COVID-19 patients: awareness of an
sign, revision of the intellectual content, and final approval. increased prevalence. Circulation. 2020;142(2):184–186.
14. Artifoni M, Danic G, Gautier G, Gicquel P, Boutoille D, Raffi F, et al.
O RCI D Systematic assessment of venous thromboembolism in COVID-19 pa-
tients receiving thromboprophylaxis: incidence and role of D-dimer
Rachel P. Rosovsky  https://orcid.org/0000-0002-2392-7365
as predictive factors. J Thromb Thrombolysis. 2020;50:211–6.
Tzu Fei Wang  https://orcid.org/0000-0003-2407-9000 15. Bompard F, Monnier H, Saab I, Tordjman M, Abdoul H, Fournier L,
Beverley J. Hunt  https://orcid.org/0000-0002-4709-0774 et al. Pulmonary embolism in patients with Covid-19 pneumonia.
Susan R. Kahn  https://orcid.org/0000-0002-5667-8916 Eur Respir J. 2020.
16. Demelo-Rodriguez P, Cervilla-Munoz E, Ordieres-Ortega L,
Claire McLintock  https://orcid.org/0000-0002-4771-8760
Parra-Virto A, Toledano-Macias M, Toledo-Samaniego N, et al.
Incidence of asymptomatic deep vein thrombosis in patients with
T WITTE R COVID-19 pneumonia and elevated D-dimer levels. Thromb Res.
Rachel P. Rosovsky @Rosovsky.Rachel 2020;192:23–6.
17. Desborough MJR, Doyle AJ, Griffiths A, Retter A, Breen KA,
Kristen M. Sanfilippo @KSanfilippoMD
Hunt BJ. Image-proven thromboembolism in patients with severe
Fionnuala Ní Áinle @fniainle COVID-19 in a tertiary critical care unit in the United Kingdom.
Beverley J. Hunt @BHWords Thromb Res. 2020;193:1–4.
Susan R. Kahn @SusanRKahn1 18. Fraisse M, Logre E, Pajot O, Mentec H, Plantefeve G, Contou D.
Barry Kevane @BarryKevane Thrombotic and hemorrhagic events in critically ill COVID-19 patients:
a French monocenter retrospective study. Crit Care. 2020;24:275.
Agnes Y. Y. Lee @AggieLeeMD
19. Helms J, Tacquard C, Severac F, Leonard-Lorant I, Ohana M,
Claire McLintock @doctormclintock Delabranche X, et al. High risk of thrombosis in patients with se-
Lisa Baumann Kreuziger @Lbkreuziger vere SARS-CoV-2 infection: a multicenter prospective cohort study.
Intensive Care Med. 2020;46(6):1089–98.
20. Llitjos JF, Leclerc M, Chochois C, Monsallier JM, Ramakers M,
REFERENCES
Auvray M, et al. High incidence of venous thromboembolic events
1. Dong E, Du H, Gardner L. An interactive web-based dashboard to in anticoagulated severe COVID-19 patients. J Thromb Haemost.
track COVID-19 in real time. Lancet Infect Dis. 2020;20:533–4. 2020;18(7):1743–6.
2. Huang C, Wang Y, Li X, Ren L, Zhao J, Hu Y, et al. Clinical features 21. Lodigiani C, Iapichino G, Carenzo L, Cecconi M, Ferrazzi P, Sebastian
of patients infected with 2019 novel coronavirus in Wuhan, China. T, et al. Venous and arterial thromboembolic complications in
Lancet. 2020;395:497–506. COVID-19 patients admitted to an academic hospital in Milan, Italy.
3. Chen N, Zhou M, Dong X, Qu J, Gong F, Han Y, et al. Epidemiological Thromb Res. 2020;191:9–14.
and clinical characteristics of 99 cases of 2019 novel coronavirus pneu- 22. Middeldorp S, Coppens M, van Haaps TF, Foppen M, Vlaar AP,
monia in Wuhan, China: a descriptive study. Lancet. 2020;395:507–13. Muller MCA, et al. Incidence of venous thromboembolism in hospi-
4. Tang N, Li D, Wang X, Sun Z. Abnormal coagulation parameters are talized patients with COVID-19. J Thromb Haemost. 2020. https://
associated with poor prognosis in patients with novel coronavirus doi.org/10.1111/jth.14888
pneumonia. J Thromb Haemost. 2020;18(4):844–7. 23. Nahum J, Morichau-Beauchant T, Daviaud F, Echegut P, Fichet J,
5. Wang D, Hu B, Hu C, Zhu F, Liu X, Zhang J, et al. Clinical character- Maillet JM, et al. Venous thrombosis among critically ill patients
istics of 138 hospitalized patients with 2019 novel coronavirus-in- with coronavirus disease 2019 (COVID-19). JAMA Netw Open.
fected pneumonia in Wuhan, China. JAMA. 2020;323(11):1061. 2020;3:e2010478.
6. Zhou F, Yu T, Du R, Fan G, Liu Y, Liu Z, et al. Clinical course 24. Thomas W, Varley J, Johnston A, Symington E, Robinson M, Sheares
and risk factors for mortality of adult inpatients with COVID- K, et al. Thrombotic complications of patients admitted to intensive
19 in Wuhan, China: a retrospective cohort study. Lancet. care with COVID-19 at a teaching hospital in the United Kingdom.
2020;395:1054–62. Thromb Res. 2020;191:76–7.
7. Al-Samkari H, Karp Leaf RS, Dzik WH, Carlson JC, Fogerty AE, 25. Voicu S, Bonnin P, Stepanian A, Chousterman BG, Le Gall A, Malissin
Waheed A, et al. COVID and coagulation: bleeding and thrombotic I, et al. High prevalence of deep vein thrombosis in mechanically
982      | ROSOVSKY et al.

ventilated COVID-19 patients. J Am Coll Cardiol. 2020. https://doi. 43. Freeman A, Horner T, Pendleton RC, Rondina MT. Prospective
org/10.1016/j.jacc.2020.05.053 comparison of three enoxaparin dosing regimens to achieve target
26. Zhang L, Feng X, Zhang D, Jiang C, Mei H, Wang J, et al. Deep vein anti-factor Xa levels in hospitalized, medically ill patients with ex-
thrombosis in hospitalized patients with coronavirus disease 2019 treme obesity. Am J Hematol. 2012;87:740–3.
(COVID-19) in Wuhan, China: prevalence, risk factors, and outcome. 4 4. Amin AN, Varker H, Princic N, Lin J, Thompson S, Johnston S.
Circulation. 2020.142(2):114–128. Duration of venous thromboembolism risk across a continuum in
27. Longchamp A, Longchamp J, Manzocchi-Besson S, Whiting L, Haller medically ill hospitalized patients. J Hosp Med. 2012;7:231–8.
C, Jeanneret S, et al. Venous thromboembolism in critically ill pa- 45. Cohen AT, Harrington RA, Goldhaber SZ, Hull RD, Wiens BL, Gold
tients with Covid-19: results of a screening study for deep vein A, et al. Extended thromboprophylaxis with betrixaban in acutely ill
thrombosis. Res Pract Thromb Haemostas. 2020;4(5):842–847. medical patients. N Engl J Med. 2016;375:534–44.
28. Bikdeli B, Madhavan MV, Jimenez D, Chuich T, Dreyfus I, Driggin E, 46. Spyropoulos AC, Ageno W, Albers GW, Elliott CG, Halperin JL, Hiatt
et al. COVID-19 and thrombotic or thromboembolic disease: impli- WR, et al. Rivaroxaban for thromboprophylaxis after hospitalization
cations for prevention, antithrombotic therapy, and follow-up. J Am for medical illness. N Engl J Med. 2018;379:1118–27.
Coll Cardiol. 2020;75(23):2950–73. 47. Streiff MB, Holmstrom B, Angelini D, Ashrani A, Bockenstedt PL,
29. Thachil J, Tang N, Gando S, Falanga A, Cattaneo M, Levi M, et al. Chesney C, et al. NCCN guidelines insights: cancer-associated ve-
ISTH interim guidance on recognition and management of coagu- nous thromboembolic disease, Version 2.2018. J Natl Compr Canc
lopathy in COVID-19. J Thromb Haemost. 2020;18:1023–6. Netw. 2018;16:1289–303.
3 0. Zhai Z, Li C, Chen Y, Gerotziafas G, Zhang Z, Wan J, et al. Prevention 48. Carrier M, Abou-Nassar K, Mallick R, Tagalakis V, Shivakumar S,
and treatment of venous thromboembolism associated with coro- Schattner A, et al. Apixaban to prevent venous thromboembolism
navirus disease 2019 infection: a consensus statement before in patients with cancer. N Engl J Med. 2019;380:711–9.
guidelines. Thromb Haemost. 2020;120(06):937–48. 49. Khorana AA, Soff GA, Kakkar AK, Vadhan-Raj S, Riess H, Wun T,
31. Society BT. BTS Guidance on Venous Thromboembolic Disease in et al. Rivaroxaban for thromboprophylaxis in high-risk ambulatory
Patients with COVID-19; 2020. https://www.brit-thora​cic.org.uk/ patients with cancer. N Engl J Med. 2019;380:720–8.
about​- us/covid​-19-infor​m atio​n -for-the-respi​r ator​y-commu​n ity/ 50. Oudkerk M, Buller HR, Kuijpers D, van Es N, Oudkerk SF, McLoud
Accessed May 4, 2020. TC, et al. Diagnosis, prevention, and treatment of thromboem-
32. Hunt B, Retter A, McClintock C. Practical Guidance for the bolic complications in COVID-19: report of the National Institute
Prevention of Thrombosis and Management of Coagulopathy and for Public Health of the Netherlands. Radiology. 2020; https://doi.
Disseminated Intravascular Coagulation of Patients Infected with org/10.1148/radiol.20202​01629
COVID-19. British Society of Hematology; 2020. https://b-s-h.org. 51. Casey K, Iteen A, Nicolini R, Auten J. COVID-19 pneumonia with he-
uk/media/​18171/​th-and-covid​-25-march​-2020-final.pdf Accessed moptysis: acute segmental pulmonary emboli associated with novel
March 25, 2020. coronavirus infection. Am J Emerg Med. 2020;38(7):1544.e1–1544.e3.
33. Hematology ASo. COVID-19 and VTE/Anticoagulation: Frequently 52. Obi AT, Barnes GD, Wakefield TW, Brown Rvt S, Eliason JL, Arndt
Asked Questions; Version 2.1; 2020. https://www.hemat​ology.org/ E, et al. Practical diagnosis and treatment of suspected venous
covid​-19/covid​-19-and-vte-antic​oagul​ation Accessed April 17, 2020. thromboembolism during COVID-19 Pandemic. J Vasc Surg Venous
34. Moores LK, Tritschler T, Brosnahan S, Carrier M, Collen JF, Doerschug Lymphat Disord. 2020;8(4):526–34.
K, et al. Prevention, diagnosis and treatment of venous thromboem- 53. Klok FA, Kruip M, van der Meer NJM, Arbous MS, Gommers D, Kant
bolism in patients with COVID-19: CHEST guideline and expert panel KM, et al. Confirmation of the high cumulative incidence of throm-
report. Chest. 2020. https://doi.org/10.1016/j.chest.2020.05.559 botic complications in critically ill ICU patients with COVID-19: an
35. Spyropoulos AC, Levy JH, Ageno W, Connors JM, Hunt BJ, Iba T, updated analysis. Thromb Res. 2020;19:148–50.
et al. Scientific and Standardization Committee Communication: 5 4. Aspen Pharmacare Canada Inc. Product Monograph: Fraxiparine®.
clinical guidance on the diagnosis, prevention and treatment of ve- Fraxiparine-Product-Monograph-ENG-20190924.pdf. https://pdf.
nous thromboembolism in hospitalized patients with COVID-19. J hres.ca/dpd_pm/00049​107.PDF Accessed September 24, 2019.
Thromb Haemost. 2020. https://doi.org/10.1111/jth.14929 55. Maley JH, Petri CR, Brenner LN, Chivukula RR, Calhoun TF,
36. Barnes GD, Burnett A, Allen A, Blumenstein M, Clark NP, Cuker Vinarsky V, et al. Anticoagulation, immortality, and observations
A, et al. Thromboembolism and anticoagulant therapy during the of COVID-19. Res Pract Thrombosis Haemostasis. 2020;4(5):​
COVID-19 pandemic: interim clinical guidance from the anticoagu- 674–676.
lation forum. J Thromb Thrombolysis. 2020;50(1):72–81. 56. Barco S, Konstantinides SV, Investigators TC-T. Thrombosis and
37. Cannegieter SC, Klok FA. COVID-19 associated coagulopathy and thromboembolism related to COVID-19 A clarion call for obtain-
thromboembolic disease: commentary on an interim expert guid- ing solid estimates from large-scale multicenter data. Res Pract
ance. Res Pract Thromb Haemost. 2020;4:439–45. Thrombosis Haemostasis. 2020;4(5):741–743.
38. Cohoon KP, Mahé G, Tafur AJ, Spyropoulos AC. Emergence of insti- 57. Wang T, Chen R, Liu C, Liang W, Guan W, Tang R, et al. Attention
tutional antithrombotic protocols for coronavirus 2019. Res Pract should be paid to venous thromboembolism prophylaxis in the man-
Thrombosis Haemostasis. 2020;4(4):510–517. agement of COVID-19. Lancet Haematol. 2020;7:e362–e363.
39. Alhazzani W, Moller MH, Arabi YM, Loeb M, Gong MN, Fan E, et al. 58. Lippi G, Plebani M. Laboratory abnormalities in patients with
Surviving sepsis campaign: guidelines on the management of criti- COVID-2019 infection. Clin Chem Lab Med. 2020;58(7):1131–4.
cally ill adults with coronavirus disease 2019 (COVID-19). Crit Care 59. Panigada M, Bottino N, Tagliabue P, Grasselli G, Novembrino C,
Med. 2020;48(6):e440–e469. Chantarangkul V, et al. Hypercoagulability of COVID-19 patients in inten-
4 0. Poterucha TJ, Libby P, Goldhaber SZ. More than an anticoagu- sive care unit: a report of thromboelastography findings and other param-
lant: do heparins have direct anti-inflammatory effects? Thromb eters of hemostasis. J Thrombosis Haemostasis. 2020;18(7):1738–42.
Haemost. 2017;117:437–44. 60. Spiezia L, Boscolo A, Poletto F, Cerruti L, Tiberio I, Campello E,
41. Liu J, Li J, Arnold K, Pawlinski R, Key NS. Using heparin molecules to et al. COVID-19-related severe hypercoagulability in patients ad-
manage COVID-2019. Res Pract Thromb Haemost. 2020;4:518–23. mitted to intensive care unit for acute respiratory failure. Thromb
42. Wang TF, Milligan PE, Wong CA, Deal EN, Thoelke MS, Gage BF. Haemost. 2020;120(06):998–1000.
Efficacy and safety of high-dose thromboprophylaxis in morbidly 61. La Alessandroa C, Annec A, Pierrea F, Bernhardd G, Lukase G,
obese inpatients. Thromb Haemost. 2014;111:88–93. Ingaf H, et al. Suggestions for thromboprophylaxis and laboratory
ROSOVSKY et al. |
      983

monitoring for in-hospital patients with COVID-19. Swiss Med 70. Hematology ASo. COVID-19 and Coagulopathy: Frequently Asked
Wkly. 2020; https://doi.org/10.4414/smw.2020.20247 Questions; 2020. https://www.hemat​ology.org/covid​-19/covid​
62. Hematology ASo. COVID-19 and aPL Antibodies: Frequently Asked -19-and-coagu​lopathy Accessed April 14, 2020.
Questions; 2020. https://www.hemat​ology.org/covid​-19/covid​ 71. Anderson DR, Dunbar M, Murnaghan J, Kahn SR, Gross P, Forsythe
-19-and-apl-ab Accessed April 20, 2020. M, et al. Aspirin or rivaroxaban for VTE prophylaxis after hip or
63. Hematology ASo. COVID-19 and D-dimer: Frequently Asked knee arthroplasty. N Engl J Med. 2018;378:699–707.
Questions; 2020. https://www.hemat​ology.org/covid​-19/covid​ 72. Weitz JI, Raskob GE, Spyropoulos AC, Spiro TE, De Sanctis Y, Xu J,
-19-and-d-dimer Accessed April 20, 2020. et al. Thromboprophylaxis with rivaroxaban in acutely ill medical
6 4. Thachil J, Tang N, Gando S, Falanga A, Levi M, Clark C, patients with renal impairment: insights from the MAGELLAN and
et al. Laboratory haemostasis monitoring in COVID-19. J Thromb MARINER trials. Thromb Haemost. 2020;120:515–24.
Haemost. 2020. https://doi.org/10.1111/jth.14866
65. Hunt B, Retter A, McClintock C. Practical Guidance for the Prevention
of Thrombosis and Management of Coagulopathy and Disseminated
S U P P O R T I N G I N FO R M AT I O N
Intravascular Coagulation of Patients Infected with COVID-19; 2020.
https://throm​bosis​uk.org/downl​oads/T&H%20and​%20COV​ID.pdf Additional supporting information may be found online in the
66. Thachil J, Cushman M, Srivastava A. A proposal for staging Supporting Information section.
COVID-19 coagulopathy. Res Pract Thrombosis Haemostasis.
2020;4(5):731–736.
67. Guan WJ, Ni ZY, Hu Y, Liang WH, Ou CQ, He JX, et al. Clinical char-
How to cite this article: Rosovsky RP, Sanfilippo KM, Wang TF,
acteristics of coronavirus disease 2019 in China. N Engl J Med.
2020;382:1708–20.
et al. Anticoagulation practice patterns in COVID-19: A global
68. Wu C, Chen X, Cai Y, Xia J, Zhou X, Xu S, et al. Risk factors associ- survey. Res Pract Thromb Haemost. 2020;4:969–983. https://
ated with acute respiratory distress syndrome and death in patients doi.org/10.1002/rth2.12414
with coronavirus disease 2019 pneumonia in Wuhan, China. JAMA
Internal Med. 2020;180(7):934.
69. Zhang L, Yan X, Fan Q, Liu H, Liu X, Liu Z, et al. D-dimer levels on
admission to predict in-hospital mortality in patients with Covid-19.
J Thromb Haemost. 2020;18(6):1324–9.