Research www.AJOG .

org

OBSTETRICS
Acute pyelonephritis in pregnancy: an 18-year retrospective
analysis
Deborah Ann Wing, MD; Michael John Fassett, MD; Darios Getahun, MD, PhD

OBJECTIVE: We sought to describe the incidence of acute pyelone- likely to be complicated by anemia (26.3% vs 11.4%; OR, 2.6; 95% CI,
phritis in pregnancy, and to assess its association with perinatal 2.4e2.9), septicemia (1.9% vs 0.03%; OR, 56.5; 95% CI,
outcomes in an integrated health care system. 41.3e77.4), acute pulmonary insufficiency (0.5% vs 0.04%; OR,
12.5; 95% CI, 7.2e21.6), acute renal dysfunction (0.4% vs 0.03%;
STUDY DESIGN: A retrospective cohort study was performed using
OR, 16.5; 95% CI, 8.8e30.7), and spontaneous preterm birth (10.3%
medical records on 546,092 singleton pregnancies delivered in all
vs 7.9%; OR, 1.3; 95% CI, 1.2e1.5). Most of the preterm births
Kaiser Permanente Southern California hospitals from 1993 through
occurred between 33-36 weeks (9.1%).
2010. These medical records include the perinatal service system along
with inpatient and outpatient encounter files. Adjusted odd ratios (ORs) CONCLUSION: We characterize the incidence of pyelonephritis in an
and 95% confidence intervals (CIs) were used to estimate associations. integrated health care system where routine prenatal screening for
asymptomatic bacteriuria is employed. Maternal complications are
RESULTS: The incidence of acute antepartum pyelonephritis was
commonly encountered and the risk of preterm birth is higher than the
0.5% (2894/543,430). Women with pyelonephritis in pregnancy were
baseline obstetric population.
more likely to be black or Hispanic, young, less educated, nulliparous,
initiate prenatal care late, and smoke during pregnancy. Pregnancies Key words: perinatal outcomes, pregnancy, preterm labor,
of women with pyelonephritis compared to those without were more pyelonephritis

Cite this article as: Wing DA, Fassett MJ, Getahun D. Acute pyelonephritis in pregnancy: an 18-year retrospective analysis. Am J Obstet Gynecol 2014;210:219.e1-6.

A cute pyelonephritis is one of the

most common nonobstetric in-
dications for antepartum hospitalization
incidence of acute pyelonephritis during
pregnancy has decreased substantially in
developed countries. Pyelonephritis in
The purpose of this study is to
describe the recent trends in acute py-
elonephritis among pregnant women, to
and is estimated to complicate up to 2% pregnancy occurs mostly before delivery, accurately characterize at-risk mothers,
of all pregnancies in the United States.1 with all but 10-20% of cases diagnosed and to examine whether acute pyelone-
While still common in developing coun- in the second and third trimesters.2,3 phritis is associated with increased risk
tries, during the past few decades, the Nearly one quarter of affected women of perinatal outcomes in an integrated
will have
1 recurrences during the same health care system. Previous reports in
pregnancy.4 Women with asymptomatic the literature describing pyelonephritis
From the Department of Obstetrics and
bacteriuria, dened as a urine culture in pregnancy have been set in urban
Gynecology, University of California, Irvine,
School of Medicine, Orange (Dr Wing); from midstream collection with a single university environments.6-8
Department of Obstetrics and Gynecology, isolate of >100,000 colony-forming units
Kaiser Permanente Southern California Medical of a uropathogen2 are at increased risk of M ATERIALS AND M ETHODS
Group, West Los Angeles (Dr Fassett); and developing pyelonephritis in pregnancy This was a retrospective cohort study of
Department of Research and Evaluation, Kaiser
compared to women without bacteriuria. pregnant women delivering singletons in
Permanente Southern California Medical Group,
Pasadena (Dr Getahun), CA, and Department of Screening for and treatment of asymp- Kaiser Permanente Southern California
Obstetrics and Gynecology, University of tomatic bacteriuria in pregnancy reduces (KPSC) hospitals from 1993 through
Medicine and Dentistry of New Jersey, New the risk of subsequent pyelonephritis 2010 (n 546,092). KPSC is one of the
Brunswick, NJ (Dr Getahun). from approximately 20-35% to 1-4%.3 largest integrated health care systems in
Received May 22, 2013; revised Sept. 6, 2013; Other potential risk factors that have the United States with approximately
accepted Oct. 2, 2013. been identied include multiparity, dia- 31,000 annual deliveries, and providing
The authors report no conict of interest. betes mellitus, urinary tract stones or comprehensive care to >3.4 million
Presented as a poster at the 58th annual malformations, and low socioeconomic residents of Southern California.
meeting of the Society for Gynecologic status.4,5 Untreated pyelonephritis can We extracted medical records com-
Investigation, Miami, FL, March 16-19, 2011.
lead to increased risk of maternal and piled electronically from all KPSC
Reprints not available from the authors.
fetal morbidity and mortality including facilities. To ascertain exposures and
0002-9378/$36.00 maternal fever, acute respiratory distress, outcomes, we used a unique maternal
2014 Mosby, Inc. All rights reserved.
http://dx.doi.org/10.1016/j.ajog.2013.10.006 acute renal failure, stillbirth, and preterm medical record number to link 4 different
birth.5 electronic medical records: the perinatal

MARCH 2014 American Journal of Obstetrics & Gynecology 219.e1

Research Obstetrics www.AJOG.org

services system, maternal and infant failure (an abrupt or rapid decline in After exclusions, a total of 543,463
hospitalization records, maternal outpa- renal ltration function; ICD-9 codes singleton pregnancies at
20 weeks of
tient health care encounters records, and 584.x and 669.x), respiratory distress gestation remained for analysis.
laboratory records. The perinatal services (ICD-9 codes 518.8x), spontaneous pre-
system records contain maternal socio- term birth (a premature labor and de- Statistical analyses
demographic (age, race/ethnicity, edu- livery occurring at 20-36 completed We estimated the incidence of acute py-
cation) and behavioral (smoking during weeks of gestation [grouped into blocks elonephritis diagnosis among singleton
pregnancy, timing of prenatal care initi- of 20-28, 29-33, and 34-36 weeks of pregnant women delivered in all KPSC
ation) characteristics as well as birth- gestation]), stillbirth (the intrauterine hospitals. Second, we compared the
weight and gestational age at delivery death of an infant >20 completed weeks maternal demographic and behavioral
from the infants birth certicate. The of gestation), chorioamnionitis (an in- characteristics between women with and
maternal and infant hospitalization and ammation at the maternal-fetal inter- without acute pyelonephritis using the
outpatient physician encounter records face; ICD-9 codes 762.7x and 658.4x), c2 test. Differences with P < .05 were
include International Classication of preeclampsia (hypertensive disorder >20 considered statistically signicant. Third,
Diseases, Ninth Revision (ICD-9), Clinical weeks of pregnancy, combined with logistic regression models were applied
Modication (ICD-9-CM) codes from proteinuria and/or edema; ICD-9 codes to examine the associations between
which we derived maternal medical his- 642.4 and 642.5), and neonatal death maternal characteristics and acute pyelo-
tory, obstetrical history, and procedures (the death of a live born infant within nephritis before and after adjusting for
for services throughout KPSC. Microbi- 28 days of life). The exposure variable of several potential confounding factors.
ology laboratory records of each acute interest was acute antepartum pyelone- Fourth, we further examined the associ-
pyelonephritis patient were reviewed for phritis (ICD-9-CM codes 590.1x). ation between acute pyelonephritis and
the presence of uropathogens in the We validated the accuracy of the ICD- perinatal outcomes after accounting for
culture results. The study was approved 9-CM coding by abstracting a random the effects of potential confounding fac-
by the KPSC Institutional Review Board. sample of 400 medical records. For this tors listed in Table 1. The analyses were
Variables that were evaluated as po- validation study, pregnancies resulting also stratied by spontaneous preterm
tential confounders or mediators in- in low birthweight or premature births birth into 3 groups dened above and
cluded maternal age (<20, 20-29, 30-34, were oversampled to ensure adequate by low birthweight (<1500 g and 1500-

35 years); race/ethnicity categorized number of these risk factors to be 2499 g) categories to determine if asso-
as non-Hispanic white (white), non- reviewed. Because we applied a stratied ciations were modied by these factors.
Hispanic black (black), Hispanic, Asian/ sampling approach, the accuracy mea- The strength of the associations was
Pacic Islander, and other race/ethnicity; sures were estimated using weighted an- explored based on the odds ratios (ORs)
maternal education (<12, 12, and
13 alyses. Abstracted records were compared and their 95% condence intervals (CIs).
years of completed schooling); maternal with diagnosis codes collected electroni- All analyses were performed using soft-
smoking during pregnancy (yes/no); cally. After adjusting for sampling frac- ware (SAS, version 9.2; SAS Institute,
timing of initial prenatal care (rst tions, the estimated sensitivity, specicity, Cary, NC).
trimester, later or none); parity (0, 1, and positive and negative predictive

2); medical conditions (chronic hy- values were 97%, 99%, 97%, and 99% for R ESULTS
pertension and pregestational diabetes); anemia; 100%, 99%, 92%, and 100% for During the 18-year study period, 2894
gestational diabetes; and birth year (1993 chorioamnionitis; 82%, 95%, 74%, and cases of acute antepartum pyelonephritis
through 1995, 1996 through 1998, 1999 97% for group B streptoccus infection; were identied, for an incidence level of
through 2001, 2002 through 2004, 2005 92%, 98%, 80%, and 99% for gestational 5.3 per 1000 births. Rates gradually in-
through 2007, and 2008 through 2010). fever; and 97%, 97%, 68%, and 100% for crease from 4.6 per 1000 births in 1993
Since the annual incidence rates for acute preeclampsia, respectively. These ndings to 5.9 per 1000 births in 2010, reecting
pyelonephritis were low, we chose to support the validity of the diagnosis codes a relative increase over the time period of
combine 3 years of records for statistical in our study. 29%; P value for linear trends < .001.
stability. Gestational age data are based on a Most cases of acute pyelonephritis were
We used (ICD-9) codes 590.x to clinical estimate of gestational age and diagnosed in the second and third
identify acute pyelonephritis. This and all were categorized into 3 groups: 20-28, trimester of pregnancy, accounting for
subsequent diagnoses were made clini- 29-33, and 34-36 weeks as well as term 90.8% of cases in this analysis. Women
cally and conrmed by laboratory tests. birth (37-42 weeks). were hospitalized for a mean of 2.8 days
The outcome of interest examined in this From 1993 through 2010, there were (SD 1.7). Women who were diagnosed
study were: anemia (anemia compli- 546,092 singleton live births and fetal with acute antepartum pyelonephritis
cating pregnancy childbirth or the puer- deaths recorded in all KPSC hospitals. were more likely to be younger, have
perium; ICD-9 codes 648.2x), septicemia We sequentially excluded births at <20 fewer years of education, be of black
(systemic inammatory response syn- weeks gestation (n 1707), and early or Hispanic ethnicity, smoke during
drome; ICD-9 codes 995.9x), acute renal termination of pregnancy (n 922). pregnancy, initiate prenatal care late in

219.e2 American Journal of Obstetrics & Gynecology MARCH 2014

www.AJOG.org Obstetrics Research
their pregnancy, and be nulliparous
compared to women who were not TABLE 1
diagnosed with acute pyelonephritis in Distribution of maternal and infant characteristics based on
pregnancy (Table 1). pyelonephritis status
The association between maternal No pyelonephritis Pyelonephritis
characteristics and acute pyelonephritis Characteristics n [ 540,536 (%) n [ 2894 (%)
during pregnancy is shown in Table 2. Maternal age, ya
Compared with women 20-29 years old, <20 7.7 21.1
women who were <20 years old were
20-29 47.9 52.9
more likely to have a pregnancy com-
plicated by acute pyelonephritis (OR, 2.0; 30-34 27.0 16.5
95% CI, 1.8e2.3). Compared with white
35 17.4 9.5
women, Hispanic and black women were a
Maternal race
at signicantly increased risk of acute
pyelonephritis during their pregnancy, Non-Hispanic white 27.0 21.9
however, the risk for black women dis- Non-Hispanic black 10.5 11.1
appeared after adjusting for multiple Hispanics 49.2 58.4
confounders. Compared with women
Asian/Pacific Islanders 10.9 6.2
who had
13 years of formal education,
women who completed 12 years of edu- Others/unknown 2.4 2.4
cation had signicantly increased risk Maternal education, y a

of acute pyelonephritis (OR, 1.3; 95%

<12 13.7 22.2
CI, 1.2e1.5) as were women who had
<12 years of formal education (OR, 1.5; 12 30.8 37.3
95% CI, 1.4e1.7). Late initiation of
13 50.8 36.4
prenatal care and smoking during preg-
Late initiation of prenatal carea 15.0 20.1
nancy were each associated with 1.1-fold
a
increased risk of acute pyelonephritis. Smoking during pregnancy 9.9 11.9
a
Pregnancy complicated by pregestational Gravida
diabetes was associated with 1.7-fold Primigravida 29.2 37.5
(95% CI, 1.3e2.1) increased risk of
Multigravida 70.8 62.5
acute pyelonephritis.
We did not nd signicant differences Parity a

in the frequencies of medical complica- Nullipara 39.9 48.6

tions such as chronic hypertension and
Multipara 60.1 51.4
gestational diabetes between those women
with and without the diagnosis of py- Chronic hypertension 3.3 3.1
elonephritis. Table 3 shows ndings on Pregestational diabetes 2.2 2.9
the association between acute pyelone-
Gestational diabetes 9.5 8.2
phritis during pregnancy and adverse a
perinatal outcomes. Women with acute Year of delivery
pyelonephritis during pregnancy were at 1993 through 1995 15.0 15.4
signicantly increased risk of anemia 1996 through 1998 17.2 16.0
(OR, 2.6; 95% CI, 2.4e2.9), septicemia
1999 through 2001 17.6 14.7
(OR, 56.5; 95% CI, 41.3e77.4), acute
renal failure (OR, 16.5; 95% CI, 2002 through 2004 16.4 14.4
8.8e30.7), respiratory distress (OR, 12.5; 2005 through 2007 16.6 16.6
95% CI, 7.2e21.6), spontaneous pre- 2008 through 2010 17.2 22.9
term birth (OR, 1.3; 95% CI, 1.2e1.5), a
Differences between pyelonephritis and no pyelonephritis by maternal and infant characteristics were statistically significant
and low birthweight birth (OR, 1.3; 95% (P < .001).
CI, 1.1e1.5) compared to women Wing. Acute pyelonephritis in pregnancy. Am J Obstet Gynecol 2014.
without. Nearly 10% of women diag-
nosed with acute pyelonephritis de-
livered prematurely, compared to 7.9% between 33-36 weeks gestational age. (OR, 1.3; 95% CI, 1.1e1.5) and primary
without the condition. Most of these Acute pyelonephritis was also associated cesarean delivery (OR, 1.2; 95% CI,
spontaneous preterm births occurred with increased risk of chorioamnionitis 1.1e1.3). However, acute pyelonephritis

MARCH 2014 American Journal of Obstetrics & Gynecology 219.e3

Research Obstetrics www.AJOG.org

The frequencies of uropathogenic bac-

TABLE 2 teria isolated included: Escherichia coli
Association between maternal characteristics and acute pyelonephritis (82.5%), Streptococcus species (21.4%),
OR (95% CI) Klebsiella pneumoniae (7.6%), Staphylo-
Characteristics Crude Adjusteda coccus species (6.5%), Proteus mirabilis
(4.9%), and Enterococcus species (5.7%).
Maternal age, ya
Of those women who developed adult
<20 2.5 (2.3e2.7) 2.0 (1.8e2.3) respiratory distress syndrome (ARDS)
20-29 1.0 (Ref) 1.0 (Ref) in association with pyelonephritis, the
30-34 0.6 (0.5e0.7) 0.6 (0.5e0.7) causative microorganisms were: Escher-
ichia coli (50%), Staphylococcus species

35 0.5 (0.4e0.6) 0.5 (0.4e0.6)
(14.3%), P mirabilis (7.1%), Enterococcus
a
Maternal race species (7.1%), and Streptococcus species
Non-Hispanic white 1.0 (Ref) 1.0 (Ref) (7.1%).
Non-Hispanic black 1.3 (1.1e1.5) 1.1 (0.9e1.2)
C OMMENT
Hispanics 1.5 (1.3e1.6) 1.2 (1.1e1.3) Overall, 0.5% of the pregnant women
Asian/Pacific Islanders 0.7 (0.6e0.8) 0.8 (0.7e0.9) included in this study were hospitalized
Others/unknown 1.2 (1.0e1.6) 1.1 (0.8e1.4) for acute pyelonephritis, which is less
a
than historical reports.9 We speculate
Maternal education, y that the implementation of the guide-
<12 2.3 (2.0e2.5) 1.5 (1.4e1.7) lines for screening for asymptomatic
12 1.7 (1.6e1.8) 1.3 (1.2e1.5) bacteriuria in pregnancy by the US Pre-
ventative Health Task Force10 and the

13 1.0 (Ref) 1.0 (Ref)
American College of Obstetricians and
a
Late initiation of prenatal care 1.4 (1.3e1.5) 1.1 (1.0e1.2) Gynecologists11 may be partly respon-
Smoking during pregnancy a
1.2 (1.1e1.4) 1.1 (1.0e1.3) sible for this, leading to greater detection
and treatment of this predecessor to
Gravidaa
many cases of pyelonephritis. It is also
Primigravida 1.0 (Ref) 1.0 (Ref) possible the high rates of early entry into
Multigravida 0.7 (0.6e0.8) 1.1 (0.9e1.2) prenatal care and the population char-
Parity a acteristics might have contributed to the
observed low rate.
Nullipara 1.0 (Ref) 1.0 (Ref) Compared with women without a
Multipara 0.7 (0.6e0.8) 0.9 (0.8e1.0) diagnosis of acute pyelonephritis, women
Chronic hypertension 0.9 (0.8e1.4) 1.0 (0.8e1.3) who were diagnosed with acute pyelo-
nephritis in pregnancy in this managed
Pregestational diabetes 1.3 (1.1e1.6) 1.7 (1.3e2.1)
care organization were more likely to be
Gestational diabetes 0.9 (0.7e1.0) 1.0 (0.9e1.1) <29 years of age, less educated, Hispanic
Year of delivery a or black, and nulliparous. As with previ-
1993 through 1995 1.0 (Ref) 1.0 (Ref)
ous investigations we found associations
with lower educational achievement
1996 through 1998 0.9 (0.8e1.0) 0.9 (0.8e1.1) levels and late entry to prenatal care.
1999 through 2001 0.8 (0.7e0.9) 0.8 (0.7e1.0) These may be markers of lower socio-
2002 through 2004 0.9 (0.7e1.0) 0.9 (0.8e1.1) economic status, a known association
with the development of acute pyelone-
2005 through 2007 1.0 (0.9e1.1) 1.1 (0.9e1.2)
phritis in pregnancy.4,7,8,12,13 We did not
2008 through 2010 1.3 (1.2e1.5) 1.5 (1.3e1.7) nd associations with chronic hyperten-
CI, confidence interval; OR, odds ratio; Ref, reference. sion, nor did we nd associations with
a
Adjustments were made for other variables in table. gestational diabetes.
Wing. Acute pyelonephritis in pregnancy. Am J Obstet Gynecol 2014. We also demonstrated increasing fre-
quencies of pyelonephritis when ana-
lyzing our data in 3-year epochs. We
was not associated with increased risk In those pregnant women with acute speculate that despite increased adher-
of very low birthweight, preeclampsia, pyelonephritis, urine cultures identied ence to guidelines recommending routine
stillbirth, or neonatal mortality. pathogens in 1887 (65.2%) subjects. screening for asymptomatic bacteruria in

219.e4 American Journal of Obstetrics & Gynecology MARCH 2014

www.AJOG.org Obstetrics Research

TABLE 3
Associations between pyelonephritis and perinatal outcome measures
No pyelonephritis Pyelonephritis Unadjusted Adjusteda
Outcomes n [ 540,536 (%) n [ 2894 (%) OR (95% CI) OR (95% CI)
Anemia 11.4 26.3 2.8 (2.6e3.0) 2.6 (2.4e2.9)
Septicemia 0.03 1.9 59.9 (44.3e81.1) 56.5 (41.3e77.4)
Acute renal failure 0.03 0.4 14.8 (8.0e27.4) 16.5 (8.8e30.7)
Respiratory distress/ARDS 0.04 0.5 12.9 (7.5e22.3) 12.5 (7.2e21.6)
Spontaneous preterm birth, wk
<37 7.9 10.3 1.3 (1.2e1.5) 1.3 (1.2e1.5)
20-27 0.6 0.7 1.1 (0.7e1.7) 1.1 (0.7e1.8)
28-32 1.1 1.2 1.2 (0.8e1.6) 1.2 (0.8e1.6)
33-36 6.2 9.1 1.4 (1.2e1.6) 1.4 (1.2e1.6)
Low birthweight, g
<1500 1.2 1.4 1.2 (0.9e1.7) 1.2 (0.9e1.7)
1500-2499 4.2 5.5 1.3 (1.1e1.6) 1.3 (1.1e1.5)
Chorioamnionitis 3.4 4.5 1.3 (1.1e1.6.) 1.3 (1.1e1.5)
Preeclampsia 5.1 5.5 1.1 (.0.9e1.3) 1.0 (0.9e1.2)
Primary cesarean 13.5 14.9 1.1 (1.0e1.3) 1.2 (1.1e1.3)
Stillbirth 0.4 0.3 0.8 (0.4e1.5) 0.9 (0.4e1.8)
Neonatal death 0.3 0.2 0.8 (0.4e1.7) 0.8 (0.4e1.7)
ARDS, adult respiratory distress syndrome; CI, confidence interval; OR, odds ratio.
a
Analyses were adjusted for maternal age, race/ethnicity, education, prenatal care, gravida, chronic hypertension, pregestational and gestational diabetes, smoking during pregnancy, and year of
delivery.
Wing. Acute pyelonephritis in pregnancy. Am J Obstet Gynecol 2014.

pregnancy, other age-related comorbid- rapid initiation of antimicrobial therapy. review of admissions for antepartum
ities may place the population at increased It is also possible that underreporting pyelonephritis in a major metropolitan
risk for pyelonephritis. is responsible, a limitation imposed by center over 2 years, 17% of patients were
The uropathogens identied in this retrospective analysis. Contrary to pre- also septicemic.8 A separate analysis
investigation were consistent with pre- vious reports however,18 the most com- noted that 12% of antepartum admis-
vious reports.3,8,14,15 Because of the na- mon uropathogen identied in women sions to the obstetric intensive care unit
ture of the investigation, we could not with ARDS and pyelonephritis was at the same institution were for sepsis
prole chronologic differences in the Escherichia coli, not K pneumoniae. In caused by pyelonephritis.19,20 Our re-
appearance of some of these uropath- fact, none of the cases of ARDS were due sults indicated a nearly 2% frequency
ogens as has been done by others,8 to K pneumoniae. of septicemia in association with acute
although the appearance in later gesta- Approximately 15-20% of women pyelonephritis, nearly 50 times that of
tions concurs with ndings of others.16 with pyelonephritis will have bacter- women without. This may reect dif-
Anemia, the most common compli- emia.8,14,15 Gram-negative bacteria pos- ferences in socioeconomic status of our
cation associated with pyelonephritis, sess endotoxin that, when released into subjects diagnosed with gestational py-
occurs historically in approximately 25% the maternal circulation, can lead to a elonephritis and septicemia that permit
of patients; our ndings are similar.8 The cascade response of cytokines, hista- earlier access to medical care and/or
low frequency of ARDS in our investi- mine, and bradykinin. The resulting adherence to standardized protocols for
gation (0.5%) was considerably less than capillary endothelial damage, dimin- treatment of this infectious complication
historic reports that reect a range from ished vascular resistance, and alterations of pregnancy that lead to improvements
1-8%.7,17 Our result may reect im- in cardiac output may lead to serious and perhaps prevention of this added
provements in treatment for acute complications such as septic shock, complication.
antepartum pyelonephritis with judi- disseminated intravascular coagulation, Because blood cultures were not uni-
cious intravenous uid management and respiratory insufciency, or ARDS. In a versally obtained, we could not assess the

MARCH 2014 American Journal of Obstetrics & Gynecology 219.e5

Research Obstetrics www.AJOG.org

frequency of bacteremia in our cohort. examination, we were unable to evaluate 9. Bacak SJ, Callaghan WM, Dietz PM,
In fact, in only 841 of the 2949 cases the temporality of the appearance of Crouse C. Pregnancy-associated hospitaliza-
tions in the United States, 1999-2000. Am J
of acute pyelonephritis cases in preg- these risk factors or complications Obstet Gynecol 2005;192:592-7.
nancy were blood cultures performed. related to the diagnosis of acute pyelo- 10. US Preventive Services Task Force.
The limited utility of blood cultures in nephritis in pregnancy. Therefore, the Screening for asymptomatic bacteriuria in
the management of pyelonephritis in associations found between acute py- adults, topic page. Available at: http://www.
pregnancy has been commented on elonephritis and the various perinatal uspreventiveservicestaskforce.org/uspstf/
uspsbact.htm. Accessed June 11, 2013.
previously.21 outcome measures in this study do not 11. American Academy of Pediatrics, American
The exact risk of preterm labor and imply causality. Lastly, the small number College of Obstetricians and Gynecologists.
delivery directly attributable to pyelo- of patients with blood culture tests pre- Chapter 4: Antepartum care. Guidelines for
nephritis in pregnancy is difcult to es- cluded assessment of bacteremia in our perinatal care, 6th ed. Elk Grove Village, IL: AAP;
timate, particularly because delivery may cohort. Washington, DC: ACOG; 2007.
12. Jolley JA, Wing DA. Pyelonephritis in preg-
not occur during the admission for the The ndings of this study conrm that nancy: an update on treatment options for
acute disease, and the risk factors for the diagnosis of acute antepartum py- optimal outcomes. Drugs 2010;70:1643-55.
both pyelonephritis and preterm de- elonephritis increased during the study 13. Scholes D, Hooton TM, Roberts PL,
livery overlap.22 Nonetheless, our results period. It further underscores that acute Gupta K, Stapleton AE, Stamm WE. Risk factors
indicate that women with a diagnosis of pyelonephritis is an important compli- associated with acute pyelonephritis in healthy
women. Ann Intern Med 2005;142:20-7.
acute pyelonephritis in pregnancy have a cation of pregnancy, and a major cause 14. Wing DA, Hendershott CM, Debuque L,
modestly higher likelihood of preterm of perinatal morbidity. - Millar LK. Outpatient treatment of acute pyelo-
delivery compared to women without nephritis in pregnancy after 24 weeks. Obstet
(10.3% vs 7.9%, P < .001). Our results Gynecol 1999;94:683-8.
are substantially higher than those of 368 15. Wing DA. Pyelonephritis in pregnancy:
REFERENCES treatment options for optimal outcomes. Drugs
women delivered at Parkland Hospital
1. Gilstrap LC III, Ramin SM. Urinary tract in- 2001;61:2087-96.
(Dallas, TX) with a history of pyelone- fections during pregnancy. Obstet Gynecol Clin 16. Archabald KL, Friedman A, Raker CA,
phritis during the pregnancy. Of them, North Am 2001;28:581-91. Anderson BL. Impact of trimester on morbidity of
19 of 368 (5%) delivered at <37 weeks, 2. Nicolle LE, Bradley S, Colgan R, Rice JC, acute pyelonephritis in pregnancy. Am J Obstet
and only 4 of 368 (1%) delivered pre- Schaeffer A, Hooton TM. Infectious Diseases Gynecol 2009;201:406.e1-4.
Society of America guidelines for the diagnosis 17. Cunningham FG, Lucas MJ, Hankins GD.
term during their admission for acute
and treatment of asymptomatic bacteriuria in Pulmonary injury complicating antepartum py-
pyelonephritis.8 adults. Clin Infect Dis 2005;40:643-54. elonephritis. Am J Obstet Gynecol 1987;156:
The strengths of this study include its 3. Duff P. Pyelonephritis in pregnancy. Clin 797-807.
population-based nature that includes a Obstet Gynecol 1984;27:17-31. 18. Towers CV, Kaminskas CM, Garite TJ,
racially and ethnically diverse popula- 4. Shefeld JS, Cunningham FG. Urinary tract Nageotte MP, Dorchester W. Pulmonary injury
infection in women. Obstet Gynecol 2005;106: associated with antepartum pyelonephritis: can
tion of pregnant women and taking into
1085-92. patients at risk be identied? Am J Obstet
account several potential confounding 5. Farkash E, Weintraub AY, Sergienko R, Gynecol 1991;164:974-80.
factors. The large number of cases and Wiznitzer A, Zlotnik A, Sheiner E. Acute ante- 19. Zeeman GG. Obstetric critical care: a blue-
standardized approaches to medical care partum pyelonephritis in pregnancy: a critical print for improved outcomes. Crit Care Med
with basis in a large integrated health analysis of risk factors and outcomes. Eur J 2006;34(Suppl):S208-14.
care system are added benets. Another Obstet Gynecol Reprod Biol 2012;162:24-7. 20. Zeeman GG, Wendel GD Jr, Cunningham FG.
6. Gilstrap LC III, Cunningham FG, Whalley PJ. A blueprint for obstetric critical care. Am J
strength of the study was repeated vali- Acute pyelonephritis in pregnancy: an ante- Obstet Gynecol 2003;188:532-6.
dation of the clinical diagnosis codes rospective study. Obstet Gynecol 1981;57: 21. Wing DA, Park AS, Debuque L, Millar LK.
using medical record gold standards. 409-13. Limited clinical utility of blood and urine cultures
Limitations of this study include its 7. Cunningham FG, Morris GB, Mickal A. Acute in the treatment of acute pyelonephritis during
retrospective nature, and the possibility pyelonephritis of pregnancy: a clinical review. pregnancy. Am J Obstet Gynecol 2000;182:
Obstet Gynecol 1973;42:112-7. 1437-40.
of potential confounders that were not 8. Hill JB, Shefeld JS, McIntire DD, 22. Lucas M, Cunningham FG. Urinary in-
accounted for (residual confounding) Wendel GD Jr. Acute pyelonephritis in preg- fections in pregnancy. Clin Obstet Gynecol
in this study. By the nature of this nancy. Obstet Gynecol 2005;105:18-23. 1993;36:855-68.

219.e6 American Journal of Obstetrics & Gynecology MARCH 2014