International Journal of Gynecology and Obstetrics 124 (2014) 5962

Contents lists available at ScienceDirect

International Journal of Gynecology and Obstetrics

journal homepage: www.elsevier.com/locate/ijgo

CLINICAL ARTICLE

Antibiotic prophylaxis in prelabor spontaneous rupture of fetal membranes

at or beyond 36 weeks of pregnancy
Ashraf F. Nabhan , Amr Elhelaly, Mohamed Elkadi
Department of Obstetrics and Gynecology, Ain Shams University, Cairo, Egypt

a r t i c l e

i n f o

Article history:
Received 23 April 2013
Received in revised form 11 July 2013
Accepted 27 September 2013
Keywords:
Antibiotic prophylaxis
Chorioamnionitis
Fetal membranes
Neonatal sepsis
Rupture of membranes

a b s t r a c t
Objective: To assess the effectiveness of prophylactic antibiotics compared with placebo in preventing neonatal
and maternal infection-related morbidity associated with prelabor spontaneous rupture of membranes at or beyond 36 weeks of pregnancy. Methods: In the present randomized controlled trial conducted during 20092011,
1640 women with prelabor spontaneous rupture of fetal membranes at or beyond 36 weeks of pregnancy were
randomly assigned to receive a single dose of prophylactic intravenous antibiotics or placebo on admission to the
labor ward of Ain Shams University, Cairo, Egypt. The participants, caregivers, and investigators were blinded to
the group assignment. The primary outcome measure was early-onset neonatal sepsis. An intention-to-treat
analysis was performed. Results: Early-onset neonatal sepsis occurred in 34 (4.1%) and 24 (2.9%) neonates in
the antibiotics and placebo groups, respectively (risk ratio 1.42; 95% condence interval 0.852.37). Maternal
infection outcomes were not signicantly different between the 2 trial arms. Conclusion: The routine use of prophylactic antibiotics in women with prelabor spontaneous rupture of fetal membranes at or beyond 36 weeks of
pregnancy does not reduce the risk of neonatal and maternal infection-related morbidity.
Trial registration number: ACTRN12608000501347
2013 International Federation of Gynecology and Obstetrics. Published by Elsevier Ireland Ltd. All rights reserved.

1. Introduction
Spontaneous prelabor rupture of fetal membranes (SROM) is encountered in approximately 8% of term pregnancies [1]. Almost twothirds of women with SROM at term will spontaneously experience
regular uterine contractions of labor within 24hours. The events leading
to SROM at term are not clear and are probably different from those
leading to preterm rupture of membranes [1,2]. At term, a weak zone
develops in the fetal membranes overlying the cervix because of
changes indicative of apoptosis and remodeling of collagen [3].
Maternal and fetal morbidity may ensue in cases of prelabor SROM
at or beyond 36 weeks of pregnancy [4,5]. Neonatal infections increase
10-fold and might lead to mortality or serious morbidity [2,6]. Factors
that might increase the risk of infections in affected newborns include
clinical chorioamnionitis, meconium-stained liquor, frequent vaginal
examinations, and an interval of more than 24 hours between SROM
and labor [4]. Maternal risks include chorioamnionitis and endometritis; both are signicantly increased at 12hours and 16hours, respectively, following prelabor SROM at or beyond 36 weeks of pregnancy [5,7].
The pressing clinical dilemma is how healthcare providers should
act to prevent neonatal and maternal morbidity. The 2 questions that

are raised with this common daily practice issue are whether to induce
labor and whether to administer prophylactic antibiotics.
A Cochrane review [8] showed that the induction of labor reduces
the maternal risks, but it is not superior to expectant management in
reducing neonatal infection [9].
It is unclear whether antibiotic prophylaxis in women with prelabor
SROM reduces the risk of maternal and neonatal infections [2]. Some experts [10] recommend prophylactic antibiotics only if the duration of
prelabor SROM exceeds 18 hours, to guard against the possibility of
group B streptococcal (GBS) infection. However, it is not clear whether
this approach reduces infections caused by other organisms [1].
The factor of safety takes precedence in the absence of a clear
benet of the routine use of prophylactic antibiotics in women
with prelabor SROM at term. Intrapartum antibiotics have been
shown to increase the risk of neonatal sepsis [11], including serious
late-onset neonatal sepsis [12].
Given the gap in current knowledge, the objective of the present randomized controlled trial was to assess the effectiveness of prophylactic
antibiotics compared with placebo in preventing early-onset neonatal
sepsis (EONS) in prelabor SROM at or beyond 36 weeks of pregnancy.

2. Materials and methods

Corresponding author at: Department of Obstetrics and Gynecology, Faculty of
Medicine, Ain Shams University, 50 El-Khalifa El-Maamoun Street, Heliopolis, Cairo
11341, Egypt. Tel.: +20 1148466668; fax: +20 222595775.
E-mail addresses: [email protected], [email protected] (A.F. Nabhan).

The present randomized, controlled, parallel-group trial was conducted at the Labor and Delivery Ward of the Department of Obstetrics,
Ain Shams University, Cairo, Egypt, between June 1, 2009, and January

0020-7292/$ see front matter 2013 International Federation of Gynecology and Obstetrics. Published by Elsevier Ireland Ltd. All rights reserved.
http://dx.doi.org/10.1016/j.ijgo.2013.07.018

60

A.F. Nabhan et al. / International Journal of Gynecology and Obstetrics 124 (2014) 5962

31, 2011. The study protocol was approved by the Research and
Ethics Committee of the department. All participants gave written
informed consent.
The participants included women with a singleton pregnancy with
prelabor SROM at or beyond 36 weeks of pregnancy. These practical
and pragmatic eligibility criteria reect daily clinical practice. The exclusion criteria included rupture of membranes for more than 12 hours,
clinical chorioamnionitis, meconium-stained liquor on admission, fetal
anomalies, and presence of rheumatic valve disease in the mother.
The participants were randomly assigned to receive prophylactic intravenous antibiotics or placebo using a simple randomization procedure with a 1:1 allocation ratio. A computer-generated randomization
list was drawn up by a faculty member who was not involved in the recruitment. The obstetrician on the labor ward enrolled the participants,
and allocation concealment was achieved by obtaining the next available randomization number by telephone. The intervention group
received parenteral ampicillin-based prophylaxis (a single dose of
1500 mg ampicillin/sulbactam) and the control group received a placebo (solvent without the active ingredient). In both arms, the hospital
protocol for SROM at term was followed, meaning that no GBS cultures
were obtained and no routine intrapartum GBS prophylaxis was given.
The 2 groups were managed on admission by induction of labor or by
cesarean delivery as indicated.
Caregivers, participants, and staff assessing outcomes (whether in
the laboratory or in the neonatology department) were blinded to the
group assignment. Blinding (masking) was successfully achieved by a
labor ward nurse who provided the antibiotics or placebo.
The data were stored and cross-checked by 2 research fellows. Furthermore, a random sample (20%) of data entry sheets was checked
by 2 independent faculty members to verify the accuracy of the data.
The prespecied primary outcome was probable or denite EONS.
Other outcomes were clinical chorioamnionitis; clinical endometritis;
induction of labor/cesarean delivery; postpartum hemorrhage; postpartum pyrexia; adverse drug reactions; duration of maternal hospital stay;
pneumonia of the newborn; fetal or early neonatal death (not attributable to congenital abnormalities); Apgar score of less than 7 at 5 minutes; admission to the neonatal intensive care unit; need for and
duration of mechanical ventilation; and duration of stay in the neonatal
intensive care unit.
The criteria for EONS included neonatal infection within the rst
7 days of life. Denite EONS was dened by clinical signs of sepsis plus
at least 1 of the following ndings: (1) a positive blood, cerebrospinal
uid (CSF), urine, or tracheal aspirate culture; (2) a positive CSF Gram
stain; (3) a positive blood, CSF, or urine antigen detection test; or
(4) a positive chest radiograph conrming neonatal pneumonia.
Probable EONS was dened by clinical signs of infection plus either a
complete blood count suggestive of infection or a CSF with an elevated
leukocyte count, a high protein concentration, or a low glucose concentration. Clinical chorioamnionitis was dened by clinical signs of
infection before delivery, including maternal pyrexia and uterine tenderness, plus laboratory ndings of infection. Clinical endometritis
was dened by clinical signs of uterine infection following delivery
plus laboratory ndings of infection. Maternal pyrexia was dened as
a temperature of 38 C or higher.
Prior data [1,2,4,8] indicated a 6% baseline risk of EONS (probable
or denite) in prelabor SROM at or beyond 36 weeks of pregnancy.
Based on the assumption that the true EONS rate with prophylactic
antibiotics is 3%, 820 participants per study arm were required to
be able to reject the null hypothesis that the EONS rates in the intervention and control arms would be equal with a power of 0.8 and a
type I error of 0.05.
The effect size of the intervention was estimated by calculating risk
ratios (RRs) with 95% condence intervals (CIs) for binary outcomes
and mean differences with 95% CIs for continuous outcomes. To test
the hypothesis, the unpaired 2-sided t test was used for continuous
data and the Fisher exact test was used for categoric data. P b 0.05 was

considered statistically signicant. An intention-to-treat analysis was

performed using SPSS version 19 (IBM, Armonk, NY, USA).

3. Results
The study included 1640 participants; 820 received antibiotics and
820 received placebo (Fig. 1). There was no protocol deviation. After
discharge, all women and newborns were seen at 7 and 14 days. Attrition rates of 2.6% and 5.0% occurred in the intervention and control
arms, respectively, and were associated with incomplete collection
of data.
The baseline clinical characteristics were similar between the 2
groups (Table 1). All women with diabetes received an insulin regimen.
Of the women with hypertension, 10 (7 [0.9%] in the antibiotics arm and
3 [0.4%] in the placebo arm) had chronic hypertension, which was
controlled using -methyldopa. Pre-eclampsia was classied as severe
in 7/48 (14.6%) women, who received magnesium sulfate as per hospital protocol. A single care giver examined each woman digitally during
labor. A maximum of 4 vaginal examinations were performed during
the 1st stage of labor. None of the women had internal fetal monitoring.
External electronic fetal monitoring was performed if deemed to be
indicated by the attending consultant.
Probable EONS occurred in 25 (3.0%) and 17 (2.1%) neonates in the
antibiotics and placebo groups, respectively (RR 1.47; 95% CI, 0.82.7)
(Table 2). Denite EONS occurred in 9 (1.1%) and 7 (0.9%) neonates
in the antibiotics and placebo groups, respectively (RR 1.29; 95% CI,
0.483.44).
Maternal infection outcomes including postpartum pyrexia and endometritis were not signicantly different between the 2 trial arms
(Table 3). Maternal clinical chorioamnionitis occurred in 40 (2.4%)
women (21 [2.6%] in the antibiotics arm and 19 [2.3%] in the placebo
arm), who received therapeutic doses of a parenteral antibiotic regimen
as per hospital standard of care.
There were no cases of neonatal pneumonia, early neonatal death, or
maternal mortality. All perinatal deaths occurred during the intrapartum period and were attributed to acute fetal hypoxia owing to
placental abruption (2 [0.1%]), cord prolapse (1 [0.1%]), or oxytocin
use (5 [0.3%]).
The frequency of postpartum hemorrhage was signicantly different
between the 2 trial arms (35 [4.3%] and 16 [2.0%] in the antibiotics and
placebo groups, respectively; RR 2.19 [95% CI, 1.223.92]; P = 0.009).
However, a subgroup analysis revealed a nonsignicant difference in
the frequency of postpartum hemorrhage among women who had a cesarean delivery (antibiotics group 12/165 [7.3%], placebo group 4/122
[3.3%]; RR 2.22 [95% CI, 0.736.71]; P = 0.158). Among women who delivered vaginally, the difference was signicant (antibiotics group 23/
655 [3.5%], placebo group 12/698 [1.7%]; RR 2.04 [95% CI 1.03 to 4.07];
P = 0.042), and 8 (1.2%) women in the antibiotics arm and 2 (0.3%)
women in the placebo arm had cervical tears and/or vaginal lacerations
during delivery.
Although the maternal duration of hospital stay after a cesarean delivery was signicantly different between the 2 groups (Table 3), the mean
difference (1.96 hours; 95% CI, 0.523.40) is not of clinical signicance.
There were no adverse effects attributable to the use of antibiotic
prophylaxis.

4. Discussion
The current trial provides evidence that the routine prophylactic use
of antibiotics in women with term prelabor SROM does not provide any
maternal and neonatal benets. Antibiotic prophylaxis was not superior
to placebo in reducing EONS (whether probable or denite) or maternal
infection-related outcomes including chorioamnionitis, postpartum
pyrexia, and endometritis.

A.F. Nabhan et al. / International Journal of Gynecology and Obstetrics 124 (2014) 5962

61

Assessed for eligibility

(n=2162)

Excluded (n=522)
Did not meet inclusion criteria
(n=306)
Declined to participate (n=216)

Randomized
(n=1640)

Received prophylactic antibiotics

(n=820)

Received placebo
(n=820)

Lost to follow -up (incomplete data)

(n=21)

Lost to follow -up (incomplete data)

(n=41)

Included in the intention -to-treat analysis

(n=820)

Included in the intention -to-treat analysis

(n=820)

Fig. 1. Flow diagram of the participants.

The present trial followed current clinical practice patterns for

women with prelabor SROM at term. The study had a robust design, the attrition rate was low, and all prespecied outcomes
were reported.
The ndings from the present study add to the body of evidence
available for the update of the Cochrane review on antibiotics for
prelabor rupture of membranes at or near term [2]. The conclusions in
the current version of the Cochrane review [2] are based on 2 studies
[13,14] with a small sample size. Furthermore, these 2 trials used management protocols where antibiotics were administered for approximately 48 hours and induction of labor was delayed for 24 hours.
However, these ndings cannot be generalized to current practice,

Table 1
Clinical characteristics of women with prelabor spontaneous rupture of fetal membranes
at or beyond 36 weeks of pregnancy.

Variable

Antibiotics
(n = 820)

Placebo
(n = 820)

P value

Maternal age, y
Parity
Gestational age on delivery, wk
Pulse on admission, bpm
Temperature on admission, C
Admission to delivery interval, h
Prior cesarean delivery
Hypertension with pregnancy
Diabetes with pregnancy
Induction of labor
Cesarean delivery
Intra partum fetal distress
Failure of progress
Repeat cesarean delivery
Pre-eclampsia
Breech presentation
Other indications

26.05 5.06
1.17 1.27
39.26 1.24
82.49 3.78
36.85 0.32
5.30 3.45
35 (4.3)
31 (3.8)
16 (2.0)
513 (62.6)
165 (20.1)
51 (6.2)
39 (4.8)
29 (3.5)
14 (1.7)
11 (1.3)
21 (2.6)

c25.59 4.63
1.12 1.15
39.20 1.27
82.70 3.56
36.84 0.307
5.38 3.60
31 (3.8)
17 (2.1)
19 (2.3)
511 (62.3)
122 (14.9)
41 (5.0)
31 (3.8)
15 (1.8)
6 (0.7)
10 (1.2)
19 (2.3)

0.060
0.336
0.302
0.255
0.505
0.646
0.707
0.056
0.733
0.959
0.006
0.334
0.393
0.046
0.113
N0.99
0.873

Values are given as mean SD or number (percentage).

where the latency period before labor induction in women with

prelabor SROM at term is limited to less than 12 hours. We have
addressed these drawbacks in the present trial, which therefore reects
the current practice in most obstetric facilities.
The Cochrane review [2] showed that the use of antibiotics in
women with prelabor SROM at or beyond 36 weeks of pregnancy
may reduce the frequency of maternal endometritis. By contrast,
the present data do not indicate any benet. The small benet for
maternal infection-related morbidity shown in the Cochrane review
[2] might be explained by the prolonged duration of SROM in the 2
randomized controlled trials [13,14] included in the review. In
current clinical practice, most hospitals adopt a strategy of planned
delivery by immediate labor induction using various methods
[79,1520]. Therefore, the nding from the previous trials [13,14]
is not applicable to current practice.
Many factors increase the risk of neonatal infection associated
with prelabor SROM at or beyond 36 weeks of pregnancy. These
include chorioamnionitis and increased time from prelabor SROM
to onset of labor and delivery [7,21]. In the absence of such factors
on admission, women and their neonates will not benet from antibiotic prophylaxis. The limited availability of resources in lowincome settings and the possible long-term effects of antibiotic therapy on the infant add further weight to the recommendation not to
use prophylactic antibiotics on a routine basis in women with
prelabor SROM at or beyond 36 weeks of pregnancy.
The limitations of the present study include the lack of long-term
follow-up and the absence of a subgroup analysis of women with a
higher risk for infection, including those with diabetes or meconiumstained liquor. Neonates were not followed for late-onset neonatal
sepsis, and drug-resistant infections among newborns were not
reported. The rate of EONS in the control arm was only 2.9%.
Based on a post-hoc power analysis, future studies would need
to include over 1,700 women in each group to detect a 50% reduction in the rate of newborn infections with an alpha of 0.05 and a
power of 0.80.

62

A.F. Nabhan et al. / International Journal of Gynecology and Obstetrics 124 (2014) 5962

Table 2
Prespecied neonatal outcomes among women with prelabor spontaneous rupture of fetal membranes at or beyond 36 weeks of pregnancy.
Outcome

Antibiotics
(n = 820)

Placebo
(n = 820)

Effect size
RR (95% CI)

Birth weight, g
Early-onset neonatal sepsis
Probable early-onset neonatal sepsis
Denite early-onset neonatal sepsis
Apgar score b7 at 5 minutes
Admission to the neonatal intensive care unit
Intrapartum fetal death
Mechanical ventilation
Duration of neonatal intensive care unit stay, d
Duration of mechanical ventilation, d

3243.04 142.27
34 (4.1)
25 (3.0)
9 (1.1)
15 (1.8)
43 (5.2)
6 (0.7)
8 (0.98)
0.30 1.51
0.04 0.43

3255.25 157.73
24 (2.9)
17 (2.1)
7 (0.9)
7 (0.9)
37 (4.5)
2 (0.2)
5 (0.6)
0.25 1.31
0.03 0.41

12.21
1.42
1.47
1.29
2.14
1.16
3.00
1.60
0.05
0.01

(26.78 to 2.36)
(0.852.37)
(0.82.7)
(0.483.44)
(0.885.23)
(0.761.78)
(0.6114.82)
(0.534.87)
(0.09 to 0.19)
(0.03 to 0.05)

Abbreviations: CI, condence interval; RR, risk ratio.

a
Values are given as mean SD or number (percentage).

Table 3
Prespecied maternal outcomes among women with prelabor spontaneous rupture of fetal membranes at or beyond 36 weeks of pregnancy.
Outcome

Antibiotics
(n = 820)

Placebo
(n = 820)

Effect size
RR (95% CI)

Chorioamnionitis
Postpartum pyrexia
Endometritis
Manual removal of the placenta
Duration of maternal hospital stay after
Vaginal delivery, h
Cesarean delivery, h

21
28
5
11

19
29
2
9

1.11
0.97
2.5
1.22

(2.6)
(3.4)
(0.6)
(1.3)

8.29 8.50
28.91 16.84

(2.3)
(3.5)
(0.2)
(1.1)

8.24 7.47
26.95 12.56

(0.602.04)
(0.581.61)
(0.4912.85)
(0.512.93)

0.05 (0.73 to 0.83)

1.96 (0.523.40)

Abbreviations: CI, condence interval; RR, risk ratio.

a
Values are given as mean SD or number (percentage).

In conclusion, the use of prophylactic antibiotics in women with

prelabor SROM at or beyond 36 weeks of pregnancy does not reduce
the risk of neonatal and maternal infection-related morbidity.
Acknowledgments
Intramural support was received from the Department of Obstetrics
and Gynecology, Ain Shams University.
Conict of interest
The authors have no conicts of interest.
References
[1] Marowitz A, Jordan R. Midwifery management of prelabor rupture of membranes at
term. J Midwifery Womens Health 2007;52(3):199206.
[2] Flenady V, King J. Antibiotics for prelabour rupture of membranes at or near term.
Cochrane Database Syst Rev 2002;3:CD001807.
[3] Moore RM, Mansour JM, Redline RW, Mercer BM, Moore JJ. The physiology of fetal
membrane rupture: insight gained from the determination of physical properties.
Placenta 2006;27(1112):103751.
[4] Seaward PG, Hannah ME, Myhr TL, Farine D, Ohlsson A, Wang EE, et al. International
multicenter term PROM study: evaluation of predictors of neonatal infection in infants born to patients with premature rupture of membranes at term. Premature
Rupture of the Membranes. Am J Obstet Gynecol 1998;179(3 Pt 1):6359.
[5] Seaward PG, Hannah ME, Myhr TL, Farine D, Ohlsson A, Wang EE, et al. International
Multicentre Term Prelabor Rupture of Membranes Study: evaluation of predictors of
clinical chorioamnionitis and postpartum fever in patients with prelabor rupture of
membranes at term. Am J Obstet Gynecol 1997;177(5):10249.
[6] Belady PH, Farkouh LJ, Gibbs RS. Intra-amniotic infection and premature rupture of
the membranes. Clin Perinatol 1997;24(1):4357.

[7] Tran SH, Cheng YW, Kaimal AJ, Caughey AB. Length of rupture of membranes in the
setting of premature rupture of membranes at term and infectious maternal
morbidity. Am J Obstet Gynecol 2008;198(6):700.e15.
[8] Dare MR, Middleton P, Crowther CA, Flenady VJ, Varatharaju B. Planned early birth
versus expectant management (waiting) for prelabour rupture of membranes at
term (37 weeks or more). Cochrane Database Syst Rev 2006;1:CD005302.
[9] Mozurkewich E. Prelabor rupture of membranes at term: induction techniques. Clin
Obstet Gynecol 2006;49(3):67283.
[10] Money DM, Dobson S, Canadian Paediatric Society, Infectious Diseases Commitee.
The prevention of early-onset neonatal group B streptococcal disease. J Obstet
Gynaecol Can 2004;26(9):82640.
[11] Glasgow TS, Speakman M, Firth S, James B, Byington CL, Young PC. Clinical and
economic outcomes for term infants associated with increasing administration of
antibiotics to their mothers. Paediatr Perinat Epidemiol 2007;21(4):33846.
[12] Glasgow TS, Young PC, Wallin J, Kwok C, Stoddard G, Firth S, et al. Association of
intrapartum antibiotic exposure and late-onset serious bacterial infections in infants.
Pediatrics 2005;116(3):696702.
[13] Cararach V, Botet F, Sentis J, Almirall R, Prez-Picaol E. Administration of antibiotics to patients with rupture of membranes at term: a prospective, randomized, multicentric study. Collaborative Group on PROM. Acta Obstet Gynecol
Scand 1998;77(3):298302.
[14] Ovalle A, Gomez R, Martinez MA, Giglio MS, Bianchi R, Diaz L, et al. Antibiotic
treatment of patients with term Premature Rupture of Membranes: A randomized
clinical trial. Prenat Neonat Med 1998;3(6):599606.
[15] American College of Obstetricians and Gynecologists. ACOG Practice Bulletin No. 120: Use
of prophylactic antibiotics in labor and delivery. Obstet Gynecol 2011;117(6):147283.
[16] Ezra Y, Michaelson-Cohen R, Abramov Y, Rojansky N. Prelabor rupture of the membranes
at term: when to induce labor? Eur J Obstet Gynecol Reprod Biol 2004;115(1):237.
[17] Mozurkewich E. Management of premature rupture of membranes at term: an
evidence-based approach. Clin Obstet Gynecol 1999;42(4):74956.
[18] Mozurkewich EL, Wolf FM. Premature rupture of membranes at term: a metaanalysis of three management schemes. Obstet Gynecol 1997;89(6):103543.
[19] Duff P. Premature rupture of membranes at term: a medical and economic rationale
for active management. CMAJ 1997;157(11):15412.
[20] Ingemarsson I. Controversies: premature rupture of membranes at termno advantage of delaying induction N 24 hours. J Perinat Med 1996;24(6):5739.
[21] Kovavisarach E, Sermsak P. Risk factors related to premature rupture of membranes in
term pregnant women: a case-control study. Aust N Z J Obstet Gynaecol 2000;40(1):302.