Research

www. AJOG.org

OBSTETRICS

Administration of cefazolin prior to skin incision is superior to

cefazolin at cord clamping in preventing postcesarean
infectious morbidity: a randomized, controlled trial
Scott A Sullivan, MD, MSCR; Triz Smith, MD; Eugene Chang, MD; Thomas Hulsey, ScD; J. Peter Vandorsten, MD;
David Soper, MD
OBJECTIVE: The objective of the study was to determine whether the
administration of cefazolin prior to skin incision was superior to administration at the time of umbilical cord clamping for the prevention
of postcesarean infectious morbidity.

fectious morbidity in the study group (relative risk [RR] 0.4, 95%
confidence interval [CI] 0.18 to 0.87), decreased endometritis (RR
0.2, 95% CI 0.15 to 0.94). No increase in neonatal sepsis (P .99),
sepsis workups (P .96), or length of stay (P .17) was observed.

STUDY DESIGN: This was a prospective, randomized, double-blind,

CONCLUSION: Administration of prophylactic cefazolin prior to skin

incision resulted in a decrease in both endomyometritis and total postcesarean infectious morbidity, compared with administration at the
time of cord clamping. This dosing did not result in increased neonatal
septic workups or complications.

placebo-controlled trial. Study subjects received cefazolin 15-60 minutes prior to incision and controls received cefazolin at the time of cord
clamping. The occurrence of endomyometritis, wound infection, total
infectious morbidity, and neonatal complications were compared.
RESULTS: There were 357 subjects enrolled. No demographic differ-

ences were observed between groups. There were decreased total in-

Key words: antibiotic prophylaxis, cesarean delivery,

endomyometritis

Cite this article as: Sullivan SA, Smith T, Chang E, et al. Administration of cefazolin prior to skin incision is superior to cefazolin at cord clamping in
preventing postcesarean infectious morbidity: a randomized, controlled trial. Am J Obstet Gynecol 2007;196;455.e1-455.e5.

esarean delivery is the most commonly performed surgery in the

United States, with nearly 1.2 million
procedures performed in 2004.1 Infectious morbidity, consisting primarily of
From the Departments of
Obstetrics/Gynecology (Drs Sullivan, Smith,
Chang, Vandorsten, and Soper) and
Biostatistics, Bioinformatics, and
Epidemiology (Dr Hulsey), Medical
University of South Carolina, Charleston,
SC 29425.
Presented at the 27th Annual Clinical Meeting
of the Society for Maternal-Fetal Medicine,
San Francisco, CA, Feb. 5-10, 2007.
Received Dec. 8, 2006; revised Jan. 25,
2007; accepted Mar. 5, 2007.
Reprints: Scott A Sullivan, MD, Medical
University of South Carolina, 96 Jonathan
Lucas Street, Charleston, SC 29425;
sullivas@musc.edu.
Supported by the Department of Obstetrics
and Gynecology Research Foundation,
Medical University of South Carolina.
0002-9378/$32.00
2007 Mosby, Inc. All rights reserved.
doi: 10.1016/j.ajog.2007.03.022

endomyometritis and wound infection,

remains a leading cause of postoperative
complications.2 Estimates of postcesarean infection rates range from 7% to
20%, depending on demographic and
obstetric variables.3,4 Infection following
cesarean delivery results in not only increased hospital stay but also increases
the cost of care. For example, diagnosing
and treating a single case of endometritis
is estimated to cost $815.5
Prophylactic antibiotics can reduce
the incidence of postcesarean infectious
morbidity by as much as 75%.6 This risk
reduction is observed in both planned
and emergent cesareans.7,8 First-generation cephalosporin antibiotics are the
most commonly used agents and are
usually administered following delivery
of the infant after the cord is clamped.9
Neither the use of broad-spectrum antimicrobials nor the administration of additional doses postoperatively has been
shown to be superior to a single-dose
cephalosporin regimen.10,11

Optimal timing for prophylactic antibiotic administration is based on animal

studies that demonstrate a maximum
protective effect when adequate tissue
antibiotic levels are present prior to bacterial contamination.12 However, concerns about neonatal exposure to antibiotics and the potential effect on a
neonatal sepsis workup prompted obstetricians to delay the administration until
the time of umbilical cord clamping.13 In
addition, there is concern regarding the
potential selection of resistant pathogens, especially Escherichia coli.14 There
is no prospective evidence to support
these concerns. A recent randomized
trial found no difference in neonatal sepsis, sepsis workups, or neonatal intensive care unit (NICU) admissions between groups administered prophylactic
antibiotics preoperatively or at cord
clamping.15
In the majority of surgical procedures
that require prophylaxis, antibiotics are
usually administered prior to the skin incision.16 Delay in administration has
been associated with increased surgical

MAY 2007 American Journal of Obstetrics & Gynecology

455.e1

Research

Obstetrics

www.AJOG.org

FIGURE

Subject flow through the cefazolin trial

site infection risk.17 Such data suggest a

potential benefit of preoperative antibiotic dosing with cesarean procedures.
There are limited data to address the
timing of cesarean antibiotic prophylaxis. In a small randomized trial [n
90], Wax et al18 concluded that there was
no difference in infectious morbidity between preoperative antibiotics and those
given at cord-clamp. Fejgin et al19 reported decreased wound infections (P
.008) in a prospective trial (n 435)
with use of preoperative antibiotic prophylaxis. More recently Thigpen et al15
demonstrated a decrease in total infectious morbidity with preoperative antibiotics but not specifically with the outcomes of endomyometritis or wound
infection.
A Cochrane database review from
2002 concluded that an adequately

powered randomized trial was needed

to address this issue.9 To this end, we
performed a prospective, doubleblinded, randomized clinical trial to
determine whether antibiotic prophylaxis administered preoperatively was
more effective in preventing infectious
morbidity following cesarean delivery
than administration following cord
clamp.

M ATERIALS AND M ETHODS

The Institutional Review Board at the
Medical University of South Carolina
(MUSC) and the research division of the
Department of Obstetrics and Gynecology approved this project (approval
#11120, January 2003). In addition, the
trial was registered with the federal government and posted on the Clinical

TABLE 1

Demographic variable comparison

Variable
Maternal age

Study group
(n 175)

Control group
(n 182)

28.3 6.1

28.3 6.0

225.3 144.5

228.1 152.9

P value
1.0

..............................................................................................................................................................................................................................................

Maternal weight

.85

..............................................................................................................................................................................................................................................

Parity

1.4

1.2

.07

..............................................................................................................................................................................................................................................

Race (C/AA/H/other)*

72/77/21/5

82/65/31/4

.45

Medicaid

87 (49%)

82 (45%)

.61

Premature delivery (less than 37 wks)

30 (17%)

46 (25%)

.08

..............................................................................................................................................................................................................................................
..............................................................................................................................................................................................................................................
..............................................................................................................................................................................................................................................

* Caucasian, African American, Hispanic, other.

455.e2

American Journal of Obstetrics & Gynecology MAY 2007

Trials.gov (Clinical Trials.gov) clearinghouse (NCT00330278).

This was a randomized, doubleblinded, placebo-controlled trial. Total
postcesarean infectious morbidity was
the primary outcome. Power was calculated a priori using MUSC infection control data, which indicated a historical
postcesarean infection incidence of 17%.
Using a power of 0.80 and an alpha of
0.05, 174 subjects per arm were calculated to be necessary to detect a 50% decrease in overall infectious morbidity for
subjects given preoperative antibiotic
prophylaxis. Primary and secondary
outcomes were defined prior to the beginning of enrollment.
Subjects were eligible for the trial if
they were older than 24 weeks estimated
gestational age and required cesarean delivery. Exclusion criteria included cephalosporin allergy, age less than 18 weeks,
exposure to any antibiotic agent within 1
week of delivery, or the need for emergent cesarean delivery. Following informed consent, simple randomization
using a random number table was performed by the investigational pharmacy
staff. Consolidated Standard of Reporting Trials (CONSORT) standards for
randomization were followed.
Infusion bags were prepared in accordance with randomization and the bags
labeled as A or B. One bag contained 1 g
of cefazolin mixed with 50 cc of normal
saline. The other bag contained only 50
cc of normal saline. Bag A was administered at least 15 minutes prior to skin incision but no more than 60 minutes before. This interval was chosen based on
previously published maternal pharmacokinetics of cefazolin.20 Bag B was administered at the time of cord clamping
by the anesthesia staff present for the
procedure. Providers and patients were
blinded to the contents of the bags until
the conclusion of the study.
Subjects were followed through their
hospital course and up to the 6-week
postpartum visit. Infectious morbidity
was documented using established clinical criteria.21
Endomyometritis was diagnosed if
maternal fever greater than 100.4F on 2
separate occasions along with uterine
fundal tenderness, tachycardia, or leuko-

Obstetrics

www.AJOG.org

TABLE 2

Obstetric variable comparisons

Variable

Study group
(n 175)

Control group
(n 182)

P value

Indication for C/S (Arr, NRFS, NL, other)

50/36/51/38

54/39/44/45

.68

Diabetes

17 (10%)

29 (16%)

.10

Multiple gestation

12 (7%)

19 (10%)

.31

Preeclampsia

18 (10.3%)

25 (13.7%)

.4

Estimated blood loss (cc)

878 112

895 127

.18

..............................................................................................................................................................................................................................................
..............................................................................................................................................................................................................................................
..............................................................................................................................................................................................................................................
..............................................................................................................................................................................................................................................
..............................................................................................................................................................................................................................................

6.1 5.0

ROM time (h)

6.8 5.4

.20

25 (13.7%)

.87

13 (7%)

.12

..............................................................................................................................................................................................................................................

Internal monitors

23 (13%)

..............................................................................................................................................................................................................................................

Subcutaneous drain

6 (3.4%)

Operative time (min)

45.3 13.6

..............................................................................................................................................................................................................................................

48 14.9

.07

..............................................................................................................................................................................................................................................

Arr, arrest disorders; C/S, cesarean section; NRFS, nonreassuring fetal status; NL, not laboring; ROM, rupture of membranes.

cytosis. Wound infection was diagnosed

if there was purulent discharge, erythema, and induration of the incision
site. Hematomas, seromas, or wound
breakdowns in the absence of previously
discussed signs were not considered
wound infections. Pyelonephritis was diagnosed by maternal temperature, flank
pain, and urine culture showing more
than 100,000 colonies of a gram negative
uropathogen.
Neonatal sepsis was diagnosed by a
positive blood culture. Organism, antibiotic resistances, and clinical course
data were recorded. Length of stay, admission status, and decision to undertake a sepsis workup were determined by
the staff neonatologists who were also
blinded to group assignment.
Study data were analyzed using SPSS
version 12.0 (Chicago, IL). Normality of
the data was tested with KolmogorovSmirnov and 2 goodness-of-fit tests
when appropriate. 2 comparisons were
performed for categorical variables and

relative risks (RR) with 95% confidence

interval were calculated. Analysis of variance testing was performed when multiple groups of categorical variables were
encountered. Student t tests were used
for continuous variable analysis. A logistic regression was also performed to obtain adjusted relative risk values. A P
value of less than .05 was considered statistically significant.

R ESULTS
There were 367 subjects consented for
this trial. Ten subjects did not complete
randomization because of either medication delay, previously undiscovered
exclusion criteria, or development of a
need for emergent cesarean delivery.
Three hundred fifty-seven subjects completed randomization and received the
study medications. Eight subjects were
lost to 6-week postpartum follow-up but
were included in the final analysis because their hospital course data were

Research

available. The Figure illustrates the randomization scheme. Completion of the

trial was accomplished in 26 months.
Table 1 summarizes the demographic
variables measured in the study and control groups. There were no significant
differences observed between the study
and control groups. Table 2 illustrates
the clinical obstetric variables measured
between the two groups. There were
again no significant differences observed
between the preoperative or cordclamping groups. There were no cases of
maternal anaphylaxis or other adverse
events related to cefazolin use reported
during the trial.
Infectious morbidity outcomes are
summarized in Table 3. There were more
cases of endomyometritis in the control
group (n 10) than in the study group
(n 2), and this difference was statistically significant (RR 0.2, 95% confidence interval [CI] 0.15 to 0.94). There
were also more cases of wound infection
in the control group (n 10) than in the
study group (n 5); however, this difference did not reach statistical significance (RR 0.52, 95% CI 0.18 to 1.5).
Seven cases of hematomas or seromas
that did not meet criteria for wound infections were noted. There was 1 case of
pyelonephritis in the control group and
1 case of pneumonia in the study group.
Overall, total infectious morbidity in the
control (n 21) group was higher than
in the study group (n 8), and this difference was statistically significant (RR
0.4, 95% CI 0.18 to 0.87). A logistic
regression model was performed to adjust for 6 demographic and clinical variables that are associated with infectious
risk and were noted to trend toward the
control group. The adjusted odds ratios

TABLE 3

Infectious morbidity
Summary of postcesarean infectious morbidity observed
Outcome

Study group
(n 175)

Control group
(n 182)

Relative risk
95% CI

Adjusted OR
95% CI

Endomyometritis

2 (1%)

10 (5%)

0.2 (0.2 to 0.94)

0.22 (0.05 to 0.9)

Wound infections

5 (3%)

10 (5%)

0.52 (0.18 to 1.5)

0.4 (0.1 to 1.3)

Total infectious morbidity

8 (4.5%)

21 (11.5%)

0.4 (0.18 to 0.87)

0.35 (0.14 to 0.82)

................................................................................................................................................................................................................................................................................................................................................................................
................................................................................................................................................................................................................................................................................................................................................................................
................................................................................................................................................................................................................................................................................................................................................................................

OR, odds ratio.

MAY 2007 American Journal of Obstetrics & Gynecology

455.e3

Research

Obstetrics

www.AJOG.org

TABLE 4

Neonatal outcomes
Variable

Study group
(n 185)

Control group
(n 194)

P value

Birthweight

3034 732

2947 989

.35

Gestational age

37.5 2.8

37 3.1

.11

Sepsis

6 (3%)

7 (3.6%)

.99

Septic workup

35 (19%)

36 (18.5%)

.96

NICU admission

25 (13.5%)

33 (17%)

.40

Intermediate admission

35 (19%)

32 (16.4%)

.65

NICU days

14.2 15.8

19.7 24.9

.01

6.6 9.9

8.5 15.8

.17

..............................................................................................................................................................................................................................................
..............................................................................................................................................................................................................................................
..............................................................................................................................................................................................................................................
..............................................................................................................................................................................................................................................
..............................................................................................................................................................................................................................................
..............................................................................................................................................................................................................................................
..............................................................................................................................................................................................................................................

Length of stay

..............................................................................................................................................................................................................................................

pH less than 7.1

10 (5%)

are reported in Table 3 and were not significantly different from the crude relative risks.
The neonatal outcome variables are illustrated in Table 4. There were no significant differences observed between
the 2 groups in neonatal sepsis (P .99),
NICU admission (P .4), total length of
stay (P .17), metabolic acidosis (P
.88), or sepsis workups (P .96). There
were significantly fewer NICU admission days in the preoperative antibiotic
group (P .01).
In addition, the neonatal cases of sepsis did not demonstrate different causative organisms between groups or an in-

12 (6%)

.88

creased incidence of antibiotic resistant

organisms. Table 5 lists the causative organisms and their resistance profiles.

C OMMENT
Postcesarean infectious morbidity is
common and costly and rarely can lead
to life-threatening conditions such as
necrotizing fasciitis.22 Cesarean infectious rates have been shown to be
higher than in comparable surgical
procedures.23 Whereas it is impossible
to compare outcomes across different
procedures, 1 possible explanation for
this difference is the common practice

of delaying prophylactic antibiotics

until cord clamping for neonatal concerns. The neonatal concerns often
cited to justify this practice have not
been validated by prospective trials.
On the contrary, previous studies have
demonstrated no increase in neonatal
sepsis or sepsis workups when prophylactic antibiotic are given preoperatively.15 Our study confirms these findings and, in fact, revealed a trend
toward decreased NICU admission
and length of stay in the preoperative
antibiotic group. We do not believe
this is causative but merely more evidence of no neonatal harm.
Our study confirms the findings in the
animal models of antibiotic prophylaxis
of Burke.12 He demonstrated that the
size of Staphylococcus-induced lesions
were reduced when antibiotic tissue levels were present before bacterial contamination. It is probable that it is this mechanism that explains the benefit of
preoperative antibiotics. Waiting until
cord clamping to administer antibiotics
allows bacterial contamination of the
uterus and subcutaneous tissues well before adequate tissue levels can be
achieved.
Strengths of this study include its design as a randomized, double-blind, placebo trial. The a priori power was

TABLE 5

Neonatal sepsis organisms

Subject

Treatment group

Cultured organism

Day of life

Resistances

Staphylococcus aureus

14

None

Staphylococcus aureus

24

Methicillin

Enterobacter cloacae

14

None

Staphylococcus aureus

Enterobacter cloacae

Escherichia coli

10

Ampicillin

Pseudomonas aeriginosa

35

None

Staphylococcus aureus

26

Methicillin

................................................................................................................................................................................................................................................................................................................................................................................
................................................................................................................................................................................................................................................................................................................................................................................
................................................................................................................................................................................................................................................................................................................................................................................

Methicillin

None

................................................................................................................................................................................................................................................................................................................................................................................
................................................................................................................................................................................................................................................................................................................................................................................
................................................................................................................................................................................................................................................................................................................................................................................
................................................................................................................................................................................................................................................................................................................................................................................
................................................................................................................................................................................................................................................................................................................................................................................

Serratia marcesens

10

Staphylococcus aureus

11

None

Methicillin

11

Staphylococcus aureus

11

None

12

Staphylococcus aureus

None

13

Klebsiella pneumoniae

23

None

................................................................................................................................................................................................................................................................................................................................................................................
................................................................................................................................................................................................................................................................................................................................................................................
................................................................................................................................................................................................................................................................................................................................................................................
................................................................................................................................................................................................................................................................................................................................................................................
................................................................................................................................................................................................................................................................................................................................................................................

Treatment groups: 1, study group; 2, control group.

455.e4

American Journal of Obstetrics & Gynecology MAY 2007

Obstetrics

www.AJOG.org
achieved, and the randomization
scheme achieved its purpose of no major
variable differences between groups. A
weakness of the trial is the high-risk nature of the study population. This population was more obese, more likely to
have diabetes, preterm delivery, multiple
gestation, and be of minority ethnicity
than the general population. This is not
surprising, given that the location of the
trial was in a tertiary care center. In addition, all cesareans in this study were
performed by resident physicians as the
primary surgeons. This resulted in a
longer average surgery time, which is a
risk factor for infectious morbidity.24
The marked reduction of infectious
morbidity with preoperative antibiotics
seen in this trial may be less dramatic in a
less obese, lower-risk population.
Overall, given the significant reductions in both endomyometritis and total
postcesarean infectious morbidity, we
would recommend the routine administration of prophylactic antibiotics preoperatively for cesarean section. Exact
timing would depend on the pharmacokinetics of the individual agent because
we used 15-60 minutes prior to skin incision with cefazolin to achieve tissue
levels.20 This recommendation would
appear not to represent a neonatal risk
for either sepsis or selection of resistant
organisms based on our and previous
studies findings. There is no known maternal risk to preoperative dosing of antibiotics as well. Given that there is no
increase in cost associated with this recommendation and the expected reduction in infectious morbidity, it is likely to
be cost effective as well. Future studies
designed to confirm the potential cost
savings of this recommendation are
needed.
f

ACKNOWLEDGMENTS
We thank Myla Ebeling, MS; Charles Rittenberg, MD; Kathryn Menard, MD, MPH; Kimberly
Porter, PharmD; and the MUSC resident
physicians.

REFERENCES
1. National Institutes of Health State of the Science Conference Statement cesarean delivery on maternal request. Obstet Gynecol
2006;107:1386-97.
2. Henderson E, Love EJ. Incidence of hospitalacquired infections associated with caesarean
section. J Hosp Infect 1995;29:245-55.
3. Yokoe DS, Christiansen CL, Johnson R, et al.
Epidemiology of and surveillance for postpartum infections. Emerg Infect Dis 2001;7:
837-41.
4. Ramsey PS, White AM, Guinn DA, et al. Subcutaneous tissue reapproximation, alone or in
combination with drain, in obese women undergoing cesarean delivery. Obstet Gynecol
2005;105(5 Pt 1):967-73.
5. Chelmow D, Hennesy M, Evantash E. Prophylactic antibiotics for non-laboring patients
with intact membranes undergoing cesarean
delivery: an economic analysis. Am J Obstet
Gynecol 2004;191:1661-5.
6. Chelmow D, Ruehli MS, Huang E. Prophylactic use of antibiotics for nonlaboring patients
undergoing cesarean delivery with intact membranes: a meta-analysis. Am J Obstet Gynecol
2001;184:656-61.
7. Mohamed K. A double-blind randomized
controlled trial on the use of prophylactic antibiotics in patients undergoing elective cesarean
section. Br J Obstet Gynecol 1988;95:689-92.
8. Noyes N, Berkeley AS, Freedman K, Ledger
W. Incidence of postpartum endomyometritis
following single-dose antibiotic prophylaxis with
either ampicillin/sulbactam, cefazolin, or cefotetan in high-risk cesarean section patients.
Infect Dis Obstet Gynecol 1998;6:220-3.
9. Smaill F, Hofmeyr GJ. Antibiotic prophylaxis
for cesarean section. Cochrane Database Syst
Rev 2002;CD00093.
10. Gall S, Hill G. Single-dose versus multipledose piperacillin prophylaxis in primary cesarean operation. Am J Obstet Gynecol
1987;157:502-6.
11. Hager WD, Rapp RP, Billeter M, Bradley
BB. Choice of antibiotic in nonelective cesarean
section. Antimicrob Agents Chemother
1991;35:1782-4.

Research

12. Burke JF. The effective period of preventive

antibiotic action in experimental incisions and
dermal lesions. Surgery 1961;50:161-8.
13. Hopkins L, Smaill F. Antibiotic prophylaxis
regimens and drugs for cesarean section Cochrane Database Syst Rev 2000;CD001136.
14. Edwards RK, Clark P, Sistrom CL, Duff P.
Intrapartum antibiotic prophylaxis I: relative effects of recommended antibiotics on gramnegative pathogens. Obstet Gynecol 2002;
100:534-9.
15. Thigpen B, Hood WA, Chauhan S, et al.
Timing of prophylactic antibiotic administration
in the uninfected gravida: a randomized clinical
trial. Am J Obstet Gynecol 2005;192:1864-71.
16. Page CP, Bohnen JM, Fletcher JR, McManus AT, Solomkin JS, Wittman DH. Antimicrobial prophylaxis for surgical wounds. Guidelines
for clinical care Arch Surg 1993;128:79-88.
17. Classen DC, Evans RS, Pestotnik SL, Horn
SD, Menlove RL, Burke JP. The timing of prophylactic administration of antibiotics and the
risk of surgical-wound infection. N Engl J Med
1992;326:281-6.
18. Wax JR, Hersey K, Philput C, et al. Single
dose cefazolin prophylaxis for postcesarean infections: before vs. after cord clamping. J Matern Fetal Med 1997;6:61-5.
19. Fejgin MD, Markov S, Goshen S, Segal J,
Arbel Y, Lang R. Antibiotic for cesarean section:
the case for true prophylaxis. Int J Gynaecol
Obstet 1993;43:257-61.
20. Fiore Mitchell T, Pearlman MD, Chapman
RL, Bhatt-Mehta V, Faix RG. Maternal and
transplacental pharmacokinetics of cefazolin.
Obstet Gynecol 2001;98:1075-9.
21. Faro S, Soper DE. Infectious disease in
women. Philadelphia (PA): Saunders, 2001. p.
181-91.
22. Goepfort AR, Guinn DA, Andrews WW,
Hauth JC. Necrotizing fasciitis after cesarean
delivery. Obstet Gynecol 1977 Mar;89(3):
409-12.
23. Gaynes RP, Culver DH, Horan TC, Edwards
JR, Richards C, Tolson JS. Surgical site infection (SSI) rates in the United States, 1992-98:
the National Nosocomial Infections Surveillance
System basic SSI risk index. Clin Infect Dis
2001;33(Suppl 2):S69-77.
24. Mangran AJ, Horan TC, Pearlson ML, Silver
LC, Jarvis WR. Guideline for prevention of surgical site infection. Infect Control Hosp Epidemiol 1999;20:247-78.

MAY 2007 American Journal of Obstetrics & Gynecology

455.e5