Kunze, Mirjam MD; Hart, Jan E. MSPH; Lynch, Anne M. MD, MSPH; Gibbs, Ronald S. MD
Premature rupture of membranes is one of the most common complications of both term and preterm pregnancies, yet there is a gap in knowledge regarding how management affects cesarean delivery rate. Among pregnancies with preterm premature rupture of membranes (PROM), gestational age at delivery is the critical factor influencing perinatal outcome and therefore the management is usually expectant until near term.1–3 In PROM at term, the risk for infectious morbidity of the mother and fetus is increased with longer length of rupture of membranes,4–6 and, therefore, expectant management is very brief, with induction of labor within approximately 12 hours.4 However, induction of labor, either at or before term, may have a high failure rate and increase the risk for infection in some cases.
Thus, from a data set of women with PROM, the aim of this study was to estimate whether it is possible to define clinically a subgroup of women who have so high a cesarean delivery rate as to avoid spontaneous onset of labor or induced labor.
MATERIALS AND METHODS
This study was an analysis of records in our department's perinatal database of women who delivered at the University of Colorado Hospital, a tertiary care referral facility, between October 2005 and January 2010 (corresponding to the time from initiation of the database to conduct of the study). The data coordinating center is located at National Jewish Health, Denver, Colorado. The data are entered daily by a group of trained perinatal research assistants. Initial quality control on the data is provided by the data coordinating center (range checks). In addition, on an annual basis, the accuracy of 10–15% of the records is checked by the database manager. The study was approved by the Colorado multiple institutional review board. From the data set of 12,953 total deliveries, we conducted a retrospective cohort study of women who had PROM greater than 24 weeks, a singleton pregnancy, and a viable fetus without congenital anomalies. We excluded cases with multiple pregnancies (n=133), congenital anomalies (n=84), unconfirmed PROM (n=11), intrauterine fetal demise (n=13) and gestational age less than 24 weeks (n=19) resulting in 1,026 (7.9%) cases.
The primary outcome was cesarean delivery. We divided women into the following groups (Fig. 1): group 1, cesarean delivery with a contraindication to labor or to vaginal delivery; group 2, spontaneous labor onset resulting in a vaginal delivery; group 3, spontaneous labor onset resulting in a cesarean delivery; group 4, induction of labor resulting in a vaginal delivery; and group 5, induction resulting in a cesarean delivery. In the event of any concern regarding misclassification of the exposures and outcomes, the medical chart was reviewed by one of the authors (M.K. or R.S.G.).
Diagnosis of rupture of membranes was based on the history of amniotic fluid leakage combined with presence of pooling of amniotic fluid on sterile speculum examination, a positive Nitrazine, a ferning test, or all. When rupture of membranes occurred at any time before the onset of labor, we diagnosed PROM. When PROM occurred before term, we diagnosed preterm PROM. Gestational age was based on the date of last menstrual period and prenatal ultrasonographic examination at less than 22 weeks. The diagnosis of clinical chorioamnionitis was made by the physicians caring for the patient and was based on clinical criteria of maternal temperature, uterine tenderness, malodorous vaginal discharge, maternal leukocytosis, maternal tachycardia, fetal tachycardia, or all of these.7 In our data set, we did not distinguish chorioamnionitis diagnosed before compared with after onset of labor. The diagnosis of placental abruption was based on clinical criteria of painful uterine contractions accompanied by vaginal bleeding. The ultrasonographic examination was used to detect retroplacental hemorrhage, although we recognize its diagnostic insensitivity.
After diagnosis of preterm PROM, all patients were managed in a common way regarding indications for delivery. Expectant management was carried out until 34 0/7 weeks. Induction of labor was indicated for gestational age more than 34 0/7 weeks or diagnosis of clinical chorioamnionitis, abruption, or other indications for delivery unless labor or vaginal delivery were contraindicated. Cesarean delivery was performed for standardized obstetric indications such as ominous abnormal fetal heart patterns or malpresentation. We did not review the fetal heart rate tracings but simply accepted the assessment of attending and resident physicians. Induction of labor was carried out with intravenously oxytocin when the cervix was ripe or with a vaginal prostaglandin preparation if initially unripe.
In addition, corticosteroids were administered between 24 0/7 and 32 0/7 weeks for fetal lung maturity enhancement. Women in this gestational age bracket also received a 7-day course of antibiotics to prolong pregnancy (most commonly ampicillin and erythromycin). Other common indications for antibiotics were group B streptococci prophylaxis, urinary tract infection, and prophylaxis of postcesarean infection. Tocolytics were used infrequently at discretion of the attending physician before 32 weeks and were not used later in gestations with preterm PROM. We did not use intramuscular or vaginal progesterone in this situation.
Maternal demographic characteristics analyzed as possible risk factors were age, prepregnant body mass index (calculated as weight (kg)/[height (m)]2), race or ethnicity, parity, history of preterm birth, education, and cigarette smoking. Pregnancy characteristics included hours between rupture of membranes and delivery and use of intrapartum antibiotics, tocolytics, or corticosteroids.
Neonatal outcomes are inextricably linked to gestational age at delivery and indication for delivery. However, we do provide brief descriptive data, including neonatal death and admission to the neonatal intensive care unit (NICU).
The data were analyzed in SAS 9.2. Means, medians, and ranges for continuous variables are reported in the tables, but nonparametric methods (Wilcoxon rank-sum and Kruskal-Wallis test) were used to test for differences in medians among groups. Associations among categorical variables were tested using the chi-square test (P<.05).
From the original data set of 12,953 deliveries, we identified 1,286 (9.9%) women with PROM and of these, 1,026 (7.9%) patients met the inclusion criteria (Fig. 1). Of these, 404 (39.4%) were preterm at delivery. Of the total study population, 13.4% (n=137) had a contraindication to labor or vaginal delivery. The most common contraindications were malpresentation (breech, transverse lie) and previous cesarean delivery (either multiple or with maternal desire for repeat). In Table 1 we show selected demographics and clinical characteristics of the cohort stratified by gestational age groups. Overall, the cohort (n=1,026) was 39.1% non-Hispanic white, 40.3% Hispanic, 14.3% African American, and 6.3% Asian or other. There were no differences in race or ethnicity across gestational age groups. Women who delivered prematurely were significantly more likely to smoke and less likely to have attended college, but absolute differences were small.
When we excluded women with a cesarean delivery as a result of a contraindication to labor or vaginal delivery, we found significant differences across the four remaining delivery groups with regard to body mass index, hours between rupture of membranes and delivery, parity, chorioamnionitis, and placental abruption (Tables 2 through 4).
Overall, the incidence of vaginal delivery was 84.9% (n=755). Women with induced labor had a significantly higher rate of cesarean delivery compared with women with spontaneous onset of labor (18% and 13%, respectively, P=.03). As shown in Table 5, within the induced (P=.97) and within the spontaneous (P=.93) labor groups, cesarean delivery rates were remarkably similar across gestational age groups. In Table 6 we show cesarean delivery rates among women with PROM and other risk factors, including low gestational age at delivery, chorioamnionitis, nulliparity, and placental abruption. In no group was the observed cesarean delivery rate greater than 27% and, in fact, it was much lower in most subgroups.
Among women who had either spontaneous or induced labor after PROM, 13.2% (117 of 889) developed clinical chorioamnionitis. This diagnosis was made in 6–7% of pregnancies greater than or equal to 34 weeks, but in 39% of pregnancies less than 34 weeks. Among women who had no contraindication to labor, we found the rate of chorioamnionitis was higher in those receiving compared with not receiving tocolytics (28% [28 of 100] compared with 11% [89 of 789], respectively, P<.001). No significant differences were seen in the rate of chorioamnionitis whether labor was spontaneous or induced (13% in both groups).
When we excluded women with a cesarean delivery resulting from a contraindication to labor, we found no significant differences between the induced and spontaneous labor groups with regard to NICU admission, neonatal death, and clinically suspected sepsis.
Premature rupture of the membranes occurs in 1–8% of all pregnancies8–10 and is the clinical antecedent to approximately 30–50% of preterm births. When PROM occurs at term (37 weeks or more), management options are induction of labor compared with expectant management to await spontaneous onset of labor. In a recent Cochrane review, the authors concluded that induction of labor reduces the risk of some maternal infectious morbidity but does not increase either cesarean or operative vaginal delivery.9 However, it is problematic to translate these findings to contemporary practice because of changes in practice because all of these trials were conducted over 10–20 years ago. Since then, general consensus has developed in the United States for management of PROM at term.11 The components of care include induction of labor, either with oxytocin or prostaglandins, shortly after admission. National guidelines to prevent perinatal group B streptococcal infections have standardized practice.12
For PROM before term (preterm PROM), there are national guidelines on use of antibiotic prophylaxis to prevent group B streptococca infection,12,13 to prolong latency,11 and to use corticosteroids,11 yet there is a lack of consensus on the optimal gestational age for delivery in cases of preterm PROM, although in the United States, 56% of maternal-fetal medicine specialists reported proceeding with delivery at 34 weeks.14 In cases of preterm PROM, tocolytics are used less commonly in preterm PROM than in preterm labor with intact membranes.
Given the variations in practice, systematic reviews have become available recently.8–10 However, all trials included in the reviews were published in 1998 or before. In one review, it was concluded that there is “insufficient evidence to guide clinical practice” in deciding whether induction or expectant management is superior.9 In another review, the authors concluded that “intentional delivery” may be more favorable because it led to decreased chorioamnionitis and decreased length of maternal stay.8
None of the trials included in these analyses used methods completely consistent with current practices in the United States, and none inform health care practitioners about the cesarean delivery rate in clinically important subgroups. Accordingly, we sought to provide data from a recent timeframe, in a diverse American population, using nationally recommended practices regarding corticosteroids and antibiotics (for both group B streptococcal prophylaxis and for prolonging latency) and using the common practice of expectant management until 34 weeks of gestation with delivery thereafter.
In our study, we found that, except for 13% of women with a contraindication to either labor or vaginal delivery, the rate of cesarean delivery was relatively low whether there was induction of labor (18%) or spontaneous onset of labor (13%).
Chorioamnionitis developed in a similarly low rate with either spontaneous or induced labor (13% in each group, P=nonsignificant). The rate of chorioamnionitis was 6–7% in pregnancies 34 weeks or more of gestation but fully 39% in cases at less than 34 weeks. Among women who received tocolysis, the rate of chorioamnionitis was significantly higher.
Of high clinical importance, we were able to explore cesarean delivery rates in various subgroups of women with combinations of risk factors. It was our intention to identify any subgroup with a rate of cesarean delivery so high as to recommend proceeding directly to cesarean delivery either at onset of labor or by avoiding induction of labor. Indeed, except for women with a contraindication to labor or vaginal delivery, we were not able to define a subgroup of women with an observed cesarean delivery rate greater than 27%. We acknowledge that the numbers in several subgroups are small. In the groups with either spontaneous or induced labor, we compared neonatal outcomes by route of delivery and found overall that there were no differences in neonatal death or NICU admission.
Our study has strengths of a relatively large number of patients, managed with common indications for induction, at one hospital, over a recent 5-year timeframe (2005–2010). The practices reflect those of many contemporary tertiary centers and national guidelines. We recognize several limitations. This was a retrospective study of women from a perinatal database. Although data were collected during the delivery hospitalization, we did not collect some data that are of potential interest. We did not have data on maternal or neonatal complications occurring after the delivery admission and we did not include some data elements of interest such as cervical examination on admission, group B streptococcal culture results, timing of diagnosis of chorioamnionitis, or previous cesarean delivery in the induced or spontaneous labor groups.
The main implication of this study for practice is that the likelihood of vaginal delivery for those women with spontaneous onset of labor or induction of labor is high (87% and 82%, respectively). No clinically defined subgroup of women had a cesarean delivery rate of greater than 27%, even when we built in multiple risk factors. Even in our data set of over 1,000 patients with PROM, we acknowledge that the numbers in several subsets are small, resulting in wide confidence intervals around the observed rate of cesarean delivery. Neonatal death and NICU admission are influenced by gestational age at delivery and by indication for delivery but (after excluding those born of pregnancies with a contraindication to either labor or vaginal delivery) were not significantly different when stratified by route of delivery or by induction or spontaneous onset of labor.
Thus, unless there is a contraindication to labor or to vaginal delivery, we recommend that 1) patients be counseled about the high rate of vaginal delivery without attributable adverse neonatal effect; and 2) induction of labor be undertaken when indicated and spontaneous onset of labor be allowed, both with anticipation of a vaginal delivery.
1. Mercer BM. Preterm premature rupture of the membranes. Obstet Gynecol 2003;101:178–93.
2. Hutzal CE, Boyle EM, Kenyon SL, Nash JV, Winsor S, Taylor DJ, et al.. Use of antibiotics for the treatment of preterm parturition and prevention of neonatal morbidity: a metaanalysis. Am J Obstet Gynecol 2008;199:620.e1–8.
3. Manuck TA, Maclean CC, Silver RM, Varner MW. Preterm premature rupture of membranes: does the duration of latency influence perinatal outcomes? Am J Obstet Gynecol 2009;201:414.e1–6.
4. 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;198:700.e1–5.
5. Belady PH, Farkouh LJ, Gibbs RS. Intra-amniotic infection and premature rupture of the membranes. Clin Perinatol 1997;24:43–57.
6. Gibbs RS, Eschenbach DA. Use of antibiotics to prevent preterm birth. Am J Obstet Gynecol 1997;177:375–80.
7. Gibbs RS. Management of clinical chorioamnionitis at term. Am J Obstet Gynecol 2004;191:1–2.
8. Hartling L, Chari R, Friesen C, Vandermeer B, Lacaze-Masmonteil T. A systematic review of intentional delivery in women with preterm prelabor rupture of membranes. J Matern Fetal Neonatal Med 2006;19:177–87.
9. 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). The Cochrane Database of Systematic Reviews 2006, Issue 1. Art. No.: CD005302. DOI: 10.1002/14651858.CD005302.pub2.
10. Buchanan SL, Crowther CA, Levett KM, Middleton P, Morris J. Planned early birth versus expectant management for women with preterm prelabour rupture of membranes prior to 37 weeks' gestation for improving pregnancy outcome. The Cochrane Database of Systematic Reviews 2010, Issue 3. Art. No.: CD004735. DOI: 10.1002/14651858.CD004735.pub3.
11. Premature rupture of membranes. Clinical management guidelines for obstetrician-gynecologists. ACOG Practice Bulletin No. 80. American College of Obstetricians and Gynecologists. Obstet Gynecol 2007;109:1007–19.
12. Verani JR, McGee L, Schrag SJ; Division of Bacterial Diseases, National Center for Immunization and Respiratory Diseases, Centers for Disease Control and Prevention (CDC). Prevention of perinatal group B streptococcal disease—revised guidelines from CDC. MMWR Recomm Rep 2010;59(RR-10):1–36.
13. Centers for Disease Control and Prevention (CDC). Decreasing incidence of perinatal Group B streptococcal disease—United States, 1993–1995. MMWR Morb Mortal Wkly Rep 1997;46:473–7.
14. Ramsey PS, Nuthalapaty FS, Lu G, Ramin S, Nuthalapaty ES, Ramin KD. Contemporary management of preterm premature rupture of membranes (PPROM): a survey of maternal-fetal medicine providers. Am J Obstet Gynecol 2004;191:1497–502.