2013

http://informahealthcare.com/jmfISSN: 1476-7058 (print), 1476-4954 (electronic)

J Matern Fetal Neonatal Med, Early Online: 1–6! 2013 Informa UK Ltd. DOI: 10.3109/14767058.2013.799653

ORIGINAL ARTICLE

Leukocyte blood count during early puerperium and its relation topuerperal infectionyUri P. Dior1,2*, Liron Kogan1*, Uriel Elchalal1, Neta Goldschmidt3, Ayala Burger2, Ran Nir-Paz4, and Yossef Ezra1

1Department of Obstetrics and Gynecology, 2Braun School of Public Health, 3Department of Hematology, and 4Department of Clinical Microbiology

and Infectious Diseases, Hadassah-Hebrew University Medical Center, Jerusalem, Israel

Abstract

Objective: To describe the white blood cell (WBC) and neutrophil counts in early puerperiumand to investigate their contribution to the diagnosis of puerperal bacterial infection.Methods: A retrospective cohort analysis through which clinical and laboratory data werecollected from 67 695 term live births. Total leukocyte and neutrophil blood count percentileswere established for febrile parturients (FP) with puerperal fever (�38 �C) and for non-FP (NFP),and stratified by mode of delivery. Rates of positive bacterial cultures were compared accordingto the total leukocyte and neutrophil blood counts.Results: Mean WBC counts of parturients delivering vaginally and by cesarean section were12.62� 103 and 12.71� 103/mL for NFP, and 14.38� 103 and 12.74� 103/mL for FP, respectively.The proportions of parturients with a WBC count of �15� 103/mL were 36.4% for FP and 21.8%for NFP (p50.001). Neutrophils comprised 80% or more of the leukocyte count in 57.6% of FPand in 30.6% of NFP (p50.001). However, no statistically significant differences in the rates ofpositive bacterial cultures were observed between those with high and low levels of leukocytesand neutrophils.Conclusions: Leukocytosis and non-extreme neutrophilia were not found to reliably associatewith bacterial infection, and their value in determining antibiotic therapy is questioned.

Keywords

Infection, neutrophilia, puerperal fever,puerperium, white blood cell count

History

Received 15 February 2013Revised 4 April 2013Accepted 19 April 2013Published online 31 May 2013

Introduction

The normal total white blood cell (WBC) count in adults

varies from 4.4 to 11.0� 103/mL. The majority of normal

leukocytes (approximately 60%) are mature neutrophils. In

adults, leukocytosis is defined as a total WBC count of more

than two standard deviations above the mean, or as a value

greater than 11.0� 103/mL, and is most commonly due to an

increase in the absolute number of mature neutrophils

(neutrophilia). Neutrophilia is defined as an absolute neutro-

phil count (ANC) greater than 7.7� 103/mL [1]. Neutrophilic

leukocytosis commonly presents in conjunction with infec-

tion, stress, smoking, pregnancy and following exercise, and

is attributed to the movement of neutrophils from the

marginated pool into the circulating pool [2].

In adults, acute infection is generally suspected with the

occurrence of leukocytosis, though the evidence for such

association is sparse [3]. Pregnancy is associated with

leukocytosis that is primarily related to increased circulation

of neutrophils. The leukocyte count begins to rise during the

first trimester of pregnancy, and reaches a mean of 8.5� 103/

mL, with a range of 5.6–12.2� 103/mL by the second and third

trimester [4]. However, data are limited regarding WBC count

during labor and puerperium. Two series reported the mean

WBC counts of 10.0–16.0� 103/mL in laboring women, with

an upper level as high as 29� 103/mL [5,6]. The mean count

was shown to increase linearly with the duration of labor and

to decrease to the normal non-pregnant range by the sixth day

postpartum [6].

Although postpartum fever does not necessarily reflect a

bacterial infection, empiric antibiotic treatment is often

initiated to avoid adverse consequences. However, antibiotic

overuse has become a global public health issue, especially in

light of the emergence of resistant bacteria [7,8].

Due to the unique physiologic state of the postpartum

parturient, diagnosing a puerperal infection poses a clinical

challenge. A complete blood count is one of the common

tools for determining treatment, though there is no solid

evidence for the clinical role and the diagnostic reliability of

the WBC in diagnosing puerperal infection. We hypothesized

that maternal blood count in early peurperium is of limited

significance in diagnosing a puerperal bacterial infection.

Therefore, by using a large cohort, we aimed to determine

the total and differential WBC count in early puerperium,

for febrile parturients (FP) and non-febrile parturients

yThis study was presented orally at the Annual Meeting of the IsraeliSociety of Maternal-Fetal Medicine, 10 November 2011.*These authors contributed equally to this study.

Address for correspondence: Uri P. Dior, MD MPH, Department ofObstetrics and Gynecology, Hadassah-Hebrew University MedicalCenter, P.O. Box 12000, Jerusalem 91120, Israel. Tel: +972-50-5172642. Fax: +972-77-3355207. E-mail: uri.dior@gmail.com

J M

ater

n Fe

tal N

eona

tal M

ed D

ownl

oade

d fr

om in

form

ahea

lthca

re.c

om b

y Y

ork

Uni

vers

ity L

ibra

ries

on

10/1

3/13

For

pers

onal

use

onl

y.

(NFP), according to mode of delivery, and to investigate the

value of this measure for diagnosing puerperal bacterial

infections.

Methods

We conducted a retrospective cohort study based on singleton

live births occurring in a tertiary medical center in Jerusalem,

Israel, between 1 January 2003 and 30 June 2011. This center

is comprised of two medical centers that serve as both

community and tertiary (referral) centers. Inclusion criteria

were parturient age above 18 years, live birth at term or post-

term (37–42 weeks gestation) and availability of postpartum

blood count. We excluded parturients suffering from intra-

partum fever (temperature �38.0) and parturients for whom

antibiotic treatment was initiated before delivery.

Data were collected and analyzed for maternal age, parity,

gestational age at delivery and mode of delivery (vaginal or

cesarean). In our institution, a blood count is drawn routinely

from each parturient 6–24 h after labor. Another blood count

is usually taken if there is a hematologic indication (e.g.

anemia, thrombocytopenia) or if a puerperal infection is

suspected. All blood counts are collected into tubes contain-

ing EDTA and are analyzed in the laboratory by a standard

procedure using an automated cell counter.

Our first variable of interest was the total and differential

WBC count in early puerperium. Early puerperium was

defined as the period of hospitalization starting immediately

after birth, usually up to 48 h after a normal vaginal delivery

and up to 120 h after a non-complicated cesarean delivery.

First, we depicted the range of total and neutrophil blood

counts of term parturients, i.e. mean� SD, minimal and

maximal values, and the 3rd, 5th, 10th, 50th, 90th, 95th and

99th percentiles. Next, we divided the study population into

FP, for whom at least one early puerperal body temperature

measured �38 �C, and non-FP (NFP), with a maximal early

puerperal temperature of 37.9 �C. For the FP, we used the

WBC count obtained up to 24 h after the maximal tempera-

ture measurement.

Total and differential leukocyte blood count percentiles

were established for each group. All analyses were performed

for the total cohort and also stratified by the mode of delivery.

To examine whether an elevated total WBC count or

neutrophilia in early puerperium predicts a bacterial infection,

data regarding blood, urine, wound, uterine and other cultures

of the FP group were collected. Positive cultures were defined

as cultures presenting with bacterial growth that was not

suspected as a contaminant. Vaginal yeasts, Staphylococcus

species, Micrococcus species, Propionilbacterium species and

Bacillus species were considered as contaminants. Culture

results of FP, reflecting bacterial infections, were compared

according to the total leukocyte count and the neutrophil

count. Length of stay at the hospital (delivery to discharge)

was used as a surrogate marker for bacterial infection.

A power analysis was performed in order to determine the

extent to which the sample size would be adequate to detect

the hypothesized effects. According to the analysis, if alpha is

5%, the power is 80%, and the predicted difference between

the groups is at least 2.1-fold [9], we would need a much

smaller study population group and then we actually have to

prove statistically significant association between variables, if

there is actual one. Therefore, if no association was found, it

leads us to the conclusion that those two factors are not

correlated.

The statistical software package SPSS 20.0 (SPSS Inc.,

Chicago, IL) was used for data analyses. The study was

approved by the Institutional Review Board of Hadassah

Medical Center (0063-11-HMO).

Results

Cohort description

Of 82 526 singleton births that occurred in our medical center

during the study period, 67 695 met inclusion criteria. The

mean maternal age was 29.8� 5.6 and mean parity was

2.88� 1.99. Mean gestational age at delivery was

39.38� 1.23. The mode of delivery was vaginal for 57 374

(84.8%) and cesarean for 10 321 (15.2%). A total of 1088

complete blood counts were withdrawn within 24 h of

maximal early puerperal fever.

Total WBC and neutrophil counts according to themode of delivery

The mean WBC count for the total cohort was 12.67� 103/

mL. Mean WBC counts for vaginal and cesarean deliveries for

the total cohort were 12.65� 103 and 12.76� 103/mL,

respectively. The mean neutrophil percentage (NP) and

ANC of the total cohort were 76.34% and 9.80� 103/mL,

respectively. The mean NP values for vaginal and cesarean

deliveries for the total cohort were 75.49% and 81.10%,

respectively. The mean ANC values for vaginal and cesarean

deliveries for the total cohort were 9.68� 103 and

10.43� 103/mL, respectively.

Total and differential WBC count according to pres-ence of fever

Tables 1 and 2 presents the mean values� SD and the 3rd,

5th, 10th, 50th, 90th, 95th and 99th percentiles for WBC, NP

and ANC of NFP at early puerperium by mode of delivery.

Tables 3 and 4 presents the mean values� SD and the 3rd,

5th, 10th, 50th, 90th, 95th and 99th percentiles for leukocyte

and neutrophil counts of parturients who experienced early

puerperal fever (�38 �C) (FP), by mode of delivery. Mean

WBC counts were 13.86� 5.06� 103/mL for FP and

12.63� 3.56� 103/mL for NFP. Figure 1 presents the distri-

bution of WBC and neutrophil counts for the FP and NFP

Table 1. Leukocyte count (103/mL) of non-febrile parturients in earlypuerperium by mode of delivery.

Total(N¼ 66 332)

Vaginal delivery(N¼ 56 474)

Cesarean section(N¼ 9858)

Mean (SD) 12.63 (3.56) 12.62 (3.54) 12.71 (3.68)3rd Percentile 7.40 7.40 7.405th Percentile 7.90 7.90 7.9010th Percentile 8.70 8.70 8.7050th Percentile 12.10 12.10 12.2090th Percentile 17.20 17.20 17.3095th Percentile 19.10 19.10 19.3899th Percentile 23.40 23.40 23.64

2 U. P. Dior et al. J Matern Fetal Neonatal Med, Early Online: 1–6

J M

ater

n Fe

tal N

eona

tal M

ed D

ownl

oade

d fr

om in

form

ahea

lthca

re.c

om b

y Y

ork

Uni

vers

ity L

ibra

ries

on

10/1

3/13

For

pers

onal

use

onl

y.

groups. FP had higher WBC and neutrophils counts than NFP.

A WBC count above 15� 103/mL was observed in 36.4% of

FP and in 21.8% of NFP (p50.001). Neutrophils comprised

80% or more of the WBC count in 57.6% of FP and in 30.6%

of NFP (p50.001).

Correlation of early puerperal fever and bacterialinfection

Cultures were taken from 72.4% FP (N¼ 788). Of them, 184

patients (23.4%) were cultured once, 187 (23.7%) were

cultured twice from the same site or from two different sites,

276 (35.0%) were cultured three times from the same site or

from three different sites and 141 patients (17.9%) were

cultured four times or more from the same site or from

different sites. Figure 2 presents the frequency of total and

positive cultures examined from different sites. At least one

positive bacterial culture, indicating a bacterial infection,

was detected in 22.7% of the cultures withdrawn.

Most bacterial infections were urinary or blood. Mean

lengths of hospital stay for parturients with negative and

positive bacterial cultures were 105 and 143 h, respectively

(p50.001).

Table 5 compares results of bacterial cultures between high

and low levels of leukocytes, ANC and NP retrieved within

24 h of early puerperal fever. No statistically significant

differences in the rates of positive bacterial cultures were

observed between those with high and low levels of leuko-

cytes, ANC or NP. Elevated levels of leukocytes and of

neutrophils were not associated with positive bacterial

cultures. Even at very elevated levels of total leukocyte

count, no statistically significant difference was observed

between the percentage of positive and negative cultures

(Figure 3a). Yet, for very extreme values of neutrophil

percentiles, but not for extreme values of neutrophil absolute

count, higher percentages of positive bacterial cultures were

Table 4. Neutrophil counts of febrile parturients in early puerperium by mode of delivery.

Total Vaginal delivery Cesarean section

Absolute count*(N¼ 1037) % (N¼ 1082)

Absolute count*y(N¼ 706) %z (N¼ 737)

Absolute count*z(N¼ 331) %z (N¼ 345)

Mean (SD) 11.38 (4.68) 80.66 (6.0) 11.83 (4.84) 80.75 (7.50) 10.42 (4.17) 80.48 (6.24)3rd Percentile 4.48 66.34 4.80 66.40 3.90 66.035th Percentile 5.10 68.91 5.29 68.79 4.46 69.5310th Percentile 5.90 72.03 6.17 71.90 5.60 72.3650th Percentile 10.80 81.30 11.20 81.45 10.17 81.0090th Percentile 17.60 88.80 18.30 89.40 16.00 87.8495th Percentile 19.90 90.70 20.56 91.00 18.50 89.7799th Percentile 25.82 93.53 26.39 93.62 22.49 93.06

*103/mL. yp50.0001. zNot significant.

Table 2. Neutrophil counts of non-febrile parturients in early puerperium by mode of delivery.

Total Vaginal delivery Cesarean section

Absolute count*(N¼ 64 839) % (N¼ 66 100)

Absolute count*(N¼ 55 225) % (N¼ 56 267)

Absolute count*(N¼ 9614) % (N¼ 9833)

Mean (SD) 9.76 (3.30) 76.25 (7.04) 9.65 (3.28) 75.41 (6.88) 10.38 (3.32) 81.04 (5.94)3rd Percentile 5.0 62.20 4.90 61.70 5.60 69.205th Percentile 5.40 64.30 5.40 63.80 6.00 71.1010th Percentile 6.10 67.40 6.00 66.80 6.70 73.9050th Percentile 9.20 76.70 9.10 75.80 9.90 81.6090th Percentile 14.00 84.70 13.90 83.60 14.60 87.8095th Percentile 15.90 86.90 15.70 85.90 16.50 89.4099th Percentile 20.20 90.80 20.10 90.20 20.70 92.20

*103/mL.

Table 3. Leukocyte count (103/mL) of febrile parturients in early puerperium by mode of delivery.

Total (N¼ 1088) Vaginal delivery (N¼ 741) Cesarean section (N¼ 347)

Mean (SD) 13.86 (5.06) 14.38 (5.21)* 12.74 (4.55)*3rd Percentile 6.10 6.45 5.205th Percentile 6.80 7.18 6.4010th Percentile 7.90 8.10 7.4650th Percentile 13.40 13.80 12.4090th Percentile 20.60 21.00 18.6295th Percentile 23.10 23.50 21.5099th Percentile 28.62 29.05 25.47

*p50.0001.

DOI: 10.3109/14767058.2013.799653 Leukocyte blood count during early puerperium 3

J M

ater

n Fe

tal N

eona

tal M

ed D

ownl

oade

d fr

om in

form

ahea

lthca

re.c

om b

y Y

ork

Uni

vers

ity L

ibra

ries

on

10/1

3/13

For

pers

onal

use

onl

y.

detected than negative (Figure 3b). A sub-analysis, through

which only FP with at least two measurements of elevated

puerperal temperature were included, had a trivial effect on

the results.

In summary, we did not find a correlation between total

WBC and neutrophil counts and between bacterial infections

at early puerperium.

Discussion

In this large cohort, we characterized WBC and neutrophil

counts in early puerperium, according to febrile status and

mode of delivery. We did not find a correlation between high

leukocyte and neutrophil counts and between bacterial

infections.

We present reference values for total WBC and neutrophil

counts at early puerperium in a large cohort. Although there

are very limited data regarding physiological and patho-

physiological blood counts in early puerperium, our findings

correspond to the data available for WBC count in the third

trimester of pregnancy [10].

Cesarean section may be considered a significant insult to

the body [11] and thus expected to result in an elevation of the

total leukocyte count. Nevertheless, we found no clinically

Figure 1. Leukocyte and neutrophil counts by febrile status. Distributions of leukocyte (a) and neutrophil (b) counts are depicted for febrile parturients(body temperature� 38 �C) and non-febrile parturients (body temperature538 �C).

4 U. P. Dior et al. J Matern Fetal Neonatal Med, Early Online: 1–6

J M

ater

n Fe

tal N

eona

tal M

ed D

ownl

oade

d fr

om in

form

ahea

lthca

re.c

om b

y Y

ork

Uni

vers

ity L

ibra

ries

on

10/1

3/13

For

pers

onal

use

onl

y.

significant difference in leukocyte and neutrophil counts

between NFP undergoing vaginal and cesarean deliveries.

This finding may be explained by a dominance of the

cytokine induced (e.g. granulocyte colony stimulating factor

(G-CSF)) and molecular pathways that determine the leuko-

cyte count in pregnancy and labor [12], rather than those that

determine leukocyte count during surgery-related stress.

However, for the FP group, the mean leukocyte count was

higher in women who underwent cesarean rather than vaginal

delivery. The reason for this finding is not clear.

As expected, mean levels of total leukocyte count and

absolute and relative neutrophil count were higher among

parturients who experienced puerperal fever than among

those who did not. We set out to investigate whether the

regular assumption of correlation of temperature above 38 �Celevated total leukocyte and neutrophil blood count, and

bacterial infection is relevant in the condition of postpartum

fever.

Puerperal fever �38 �C presented in 2.2% of the women

included in this study. Puerperal fever is a relatively common

postpartum phenomenon that can reflect a postpartum

complication [13]. Although puerperal fever can result from

bacterial infection, viral infection and from non-infectious

causes, such as ovarian vein thrombosis [14,15], it can also be

a normal finding in postpartum women. The initial differen-

tial diagnosis of bacterial infection related to puerperal fever

includes endometritis, urinary tract infection, mastitis, pneu-

monia, pyelonephritis and surgical site infection [16,17].

Diagnosing the etiology of puerperal fever poses a

challenge for the clinician in the maternity department. As

anamnesis and physical examination do not always reveal the

underlying cause of fever, leukocyte count is used routinely as

an adjunctive measure. An elevated total leukocyte count and

a ‘‘left shift’’ of the differential count, i.e. an elevated total

and relative number of neutrophils, usually indicates a

bacterial infection rather than a viral infection or a non-

infectious fever etiology [18]. In some cases, when no definite

diagnosis is established, the presence of elevated body

temperature and elevated WBC count prompt clinicians to

initiate broad-spectrum antibiotic empiric treatment [19].

However, immoderate antibiotic usage can result in antibiotic

resistance development and in a wide range of adverse effects

[7,8].

Surprisingly, positive bacterial cultures were no more

prevalent among those with elevated leukocyte and neutrophil

counts than among those with low counts. This lack of

association was observed also for very extreme values of total

leukocyte count but not for extreme values of neutrophil

count. For this group, we found higher rates of positive

bacterial cultures among parturients with extremely elevated

neutrophil percentiles (90th and 95th percentiles). Although

we are unaware of comparative investigations among parturi-

ents, one small study of 172 non-pregnant patients presenting

to an emergency department demonstrated an association

between extreme leukocytosis (425 000) and infectious

disease [3]. The association observed in this study suggests

that during early puerperium, a very high neutrophil count,

but not a high leukocyte count, may be attributed to bacterial

infection. Notably, though not statistically significant, at

extreme values of WBC, there was a larger difference in the

Figure 3. Positive culture rates according to blood count percentile: rateof positive cultures above and below percentile groups of WBC (a) andneutrophil (b) counts. NS = Not significant.

Figure 2. Total and positive cultures: frequency of total and positivecultures examined from different sites.

Table 5. Results of bacterial cultures by total leukocyte and neutrophilcount*y.

Leukocytecountz

Absoluteneutrophil

countzNeutrophilpercentagez

415 000 �15 000 410 000 �10 000 480% �80%

Positive culture 22.9 22.6 20.5 23.9 22.5 23.0Negative culture 77.1 77.4 79.5 77.1 77.5 77.0

*Results presented in percentages; yBlood count within 24 h after fevermeasurement;zNot significant

DOI: 10.3109/14767058.2013.799653 Leukocyte blood count during early puerperium 5

J M

ater

n Fe

tal N

eona

tal M

ed D

ownl

oade

d fr

om in

form

ahea

lthca

re.c

om b

y Y

ork

Uni

vers

ity L

ibra

ries

on

10/1

3/13

For

pers

onal

use

onl

y.

rates of positive bacterial cultures. Given the fact that extreme

leukocytosis (419 000/mL) appears only in 5% of NFP, it may

be that over this threshold, a closer clinical surveillance is

warranted.

Our findings support the existence of a particular patho-

physiology of leukocytosis and neutrophilia during preg-

nancy, labor and puerperium. We speculate a role in

puerperium of mechanisms that have been described to be

involved in pregnancy or labor. The elevation of plasma

G-CSF [12] may have an effect. Another possibility is the

steroid effect during pregnancy and labor, as demonstrated by

elevated cortisol levels, which have been shown to reach

levels similar to those observed in women with Cushing’s

syndrome [20,21], and which are related to reduced neutro-

phil adhesion and increased release from marrow stores [22].

Leukocytosis may also represent an inflammatory response

that is possibly increased postpartum, as well as other

parameters such as C-reactive protein, fibrinogen and ESR.

Due to our common clinical practice, it was not possible to

provide these additional data; this increased background

response may prevent to isolate the effect on infection rather

than inflammatory response on increased WBC count.

Missing data, due to retrospective design, are the main

limitation of this study. Blood counts and cultures were not

performed for all parturients experiencing fever. However,

since we do not expect differences in characteristics between

parturients for whom blood counts and cultures were and

were not available, we assume that only a non-differential

misclassification may have resulted. The very large cohort is

an important strength of this study.

To estimate possible information bias due to the number of

elevated temperature measurements, a sub-analysis was

carried out through which only parturients with two or

more measurements of temperature �38 �C were included.

This analysis yielded similar results, i.e. no significant

differences were encountered between positive culture rates

among high and low leukocytes. In addition, lending support

to our findings, a significant difference in positive culture

rates was encountered above and below very high neutrophil

percentile groups.

In summary, this study depicted the values of total

leukocyte and neutrophil counts in early puerperium in a

large cohort, according to the mode of delivery and

postpartum fever. The determination of such reference

values is potentially important as a diagnostic tool and for

research purposes. Nevertheless, the total leukocyte and

neutrophil count in early puerperium appears not to be a

reliable tool for diagnosing a puerperal bacterial infection.

These findings should encourage physicians to rely more on

their clinical judgment and not only on laboratory tests.

Verification of our findings and elucidation of underlying

pathophysiological mechanisms and clinical implications of

leukocytosis and neutrophilia during early puerperium are

warranted.

Declarations of interest

The authors report no declarations of interest.

References

1. Hollowell JG, van Assendelft OW, Gunter EW, et al. Hematologicaland iron-related analytes-reference data for persons aged 1 yearand over: United States, 1988–94. Vital Health Stat 2005;11:1–156.

2. Boxer LA, Allen JM, Baehner RL. Diminished polymorphonuclearleukocyte adherence. Function dependent on release of cyclic AMPby endothelial cells after stimulation of beta-receptors by epineph-rine. J Clin Invest 1980;66:268–74.

3. Lawrence YR, Raveh D, Rudensky B, Munter G. Extremeleukocytosis in the emergency department. QJM 2007;100:217–23.

4. Gordon MC. Maternal physiology. In: Gabbe SG, Niebyl JR,Landon MB, et al. eds, Obstetrics: normal and problem pregnan-cies. 6th ed. Philadelphia (PA): Elsevier Inc.; 2012:53–54.

5. Molberg P, Johnson C, Brown TS. Leukocytosis in labor: what areits implications? Fam Pract Res J 1994;14:229–36.

6. Acker DB, Johnson MP, Sachs BP, Friedman EA. The leukocytecount in labor. Am J Obstet Gynecol 1985;153:737–9.

7. Alumran A, Hou XY, Hurst C. Validity and reliability ofinstruments designed to measure factors influencing the overuseof antibiotics. J Infect Public Health 2012;5:221–32.

8. Dellit TH, Owens RC, McGowan Jr JE, et al. Infectious DiseasesSociety of America and the Society for Healthcare Epidemiology ofAmerica guidelines for developing an institutional program toenhance antimicrobial stewardship. Clin Infect Dis 2007;44:159–77.

9. Coburn B, Morris AM, Tomlinson G, Detsky AS. Does this adultpatient with suspected bacteremia require blood cultures? JAMA2012;308:502–11.

10. Lurie S, Rahamim E, Piper I, et al. Total and differentialleukocyte counts percentiles in normal pregnancy. Eur JObstetGynecolReprod Biol 2008;136:16–9.

11. Desborough JP. The stress response to trauma and surgery. Br JAnaesth 2000;85:109–17.

12. Umesaki N, Fukumasu H, Miyama M, et al. Plasma granulocytecolony stimulating factor concentrations in pregnant women.Gynecol Obstet Invest 1995;40:5–7.

13. Burrows LJ, Meyn LA, Weber AM. Maternal morbidity associatedwith vaginal versus cesarean delivery. Obstet Gynecol 2004;103:907–12.

14. Dunnihoo DR, Gallaspy JW, Wise RB, Otterson WN. Postpartumovarian vein thrombophlebitis: a review. Obstet Gynecol Surv1991;46:415–27.

15. Klima DA, Snyder TE. Postpartum ovarian vein thrombosis. ObstetGynecol 2008;111:431–5.

16. Duff P. Pathophysiology and management of post cesareanendomyometritis. Obstet Gynecol 1986;67:269–76.

17. Charles J, Charles D. Postpartum infection. In: Charles D, ed.Obstetric and perinatal infections. St Louis (MO): Mosby YearBook; 1993:60.

18. Holland SM, Gallin JI. Disorders of granulocytes and monocytes. In:Longo D, Fauci A, Kasper D, Hauser S, et al. eds, Harrison’sprinciples of internal medicine. 18th ed. New York: McGraw-Hill;2011:477.

19. French LM, Smaill FM. Antibiotic regimens for endometritis afterdelivery. Cochrane Database Syst Rev 2004;4:CD001067.

20. Rainey WE, Rehman KS, Carr BR. Fetal and maternal adrenals inhuman pregnancy. Obstet Gynecol Clin North Am 2004;31:817–35.

21. Tuimala RJ, Kauppila AJ, Haalahti J. Response of pituitary-adrenalaxis on partal stress. Obstet Gynecol 1975;46:275–8.

22. Dale DC, Fauci AS, Guerry DI, Wolff SM. Comparison of agentsproducing a neutrophilic leukocytosis in man: hydrocortisone,prednisone, endotoxin, and etiocholanolone. J Clin Invest 1975;56:808–13.

6 U. P. Dior et al. J Matern Fetal Neonatal Med, Early Online: 1–6

J M

ater

n Fe

tal N

eona

tal M

ed D

ownl

oade

d fr

om in

form

ahea

lthca

re.c

om b

y Y

ork

Uni

vers

ity L

ibra

ries

on

10/1

3/13

For

pers

onal

use

onl

y.