Objective
The aim of this randomized study was to compare 2 protocols for inducing labor in women with premature rupture of membranes (PROM) at term.
Study Design
Women with PROM and a Bishop score ≤5 were randomly assigned to receive either an intravenous oxytocin infusion (n = 223) or a dinoprostone pessary followed 6 hours later by an intravenous oxytocin infusion (n = 227).
Results
Vaginal delivery within 24 hours of labor induction increased significantly with sustained-released dinoprostone followed by oxytocin infusion (78.5% vs 63.3%; relative risk, 1.23; 95% confidence interval, 1.09–1.39; P = .001). Maternal and neonatal outcomes were similar between the groups.
Conclusion
Sustained-released dinoprostone followed 6 hours later by an oxytocin infusion in term women with PROM was associated with a higher rate of vaginal delivery within 24 hours, and no difference in maternal-neonatal complications was observed compared with oxytocin infusion alone.
Premature rupture of membranes (PROM) is most commonly defined as rupture of membranes before the onset of labor and occurs in 8% of term pregnancies. A prolonged interval from rupture of membranes to delivery is associated with an increase in the incidence of chorioamnionitis and neonatal sepsis. The management of term patients with PROM, especially those with an unfavorable cervix, remains controversial. Management options include immediate induction of labor vs delayed induction or expectant management. Several reports have detailed an increase in maternal and neonatal morbidity with expectant management, whereas active management leads to a shorter interval from PROM to delivery, reducing the risk of postnatal infections. Additionally, active management is preferred by patients.
In 1985, Ekman-Ordeberg et al randomly assigned 20 nulliparous women with PROM at term and unfavorable cervices to immediate induction of labor with either intravenous oxytocin or prostaglandin (PG) E2 intravaginal gel. They noted fewer instrumental deliveries (cesarean sections plus operative vaginal deliveries) in the PG-treated group and concluded that PGE2 use might mitigate any increased risk for cesarean section with labor induction. In contrast, in a randomized trial of 94 nulliparous women with PROM at term and cervices unfavorable for induction of labor, Chua et al found that PGE2 pessaries conferred no advantage over intravenous oxytocin for the outcomes of interest, including length of labor or risk for cesarean delivery.
The Term Prelabor Rupture of Membranes Study demonstrated that the time from randomization to delivery for women in the oxytocin group was significantly less than that for women in the PGE2 group. Additionally, the women in the oxytocin induction group had a lower rate of infection than those in the PG induction group with no statistically significant difference in the cesarean section rate. In contrast, some studies have documented the safety and efficacy of dinoprostone vaginal pessary for induction of labor in term patients with PROM. However, the American College of Obstetricians and Gynecologists recommends that in term PROM, labor should be induced at the time of presentation, generally with oxytocin infusion, to reduce the risk of chorioamnionitis.
We know that neonatal intensive care unit (NICU) admission, variable decelerations, and primary cesarean delivery rates are positively correlated with a longer admission to labor onset interval in women with PROM. Thus, to reduce the time from induction to delivery, concurrent oxytocin infusion with dinoprostone vaginal pessary use for labor induction in women with PROM may be beneficial. Several studies have shown shorter induction to delivery intervals with oxytocin infusion concurrent with a PG at initiation of labor. However, in these studies, there was little information about PROM labor induction. We compared 2 protocols (oxytocin vs sustained-release dinoprostone followed 6 hours later by oxytocin) for inducing labor in women with PROM at term and an unfavorable cervix. Our hypothesis was that no significant difference in vaginal delivery would be observed within 24 hours after treatment with oxytocin or sustained-release dinoprostone followed 6 hours later by oxytocin.
Materials and Methods
The present study was carried out at the Department of Obstetrics and Gynecology, Bakirkoy Women’s and Children’s Teaching Hospital, Istanbul, Turkey, from May 2009 through September 2010. This tertiary teaching hospital has 14,000 deliveries annually. Ethical oversight was provided by the Bakirkoy Women’s and Children’s Teaching Hospital Ethics Committee (reference no. 298). This trial was performed in compliance with the Declaration of Helsinki.
Women with a live singleton term fetus (37-42 weeks of gestation, as determined by the last menstrual period or by a first- or second-trimester ultrasound scan) in cephalic presentation and a reactive non-stress test, who presented with PROM and a Bishop score of ≤5 before the onset of labor, and with no spontaneous contractions (<4 contractions within 20 minutes) were included. Ruptured membranes were diagnosed when amniotic fluid drained from the cervical os during sterile speculum examination; there was a pool of fluid in the posterior fornix; or with a positive result to the AmniSure ROM test (PAMG-1 immunoassay; N-Dia, New York, NY). Women in active labor or with previous cesarean section or other type of uterine surgery, antepartum hemorrhage, chorioamnionitis, contraindication to PG use (eg, bronchial asthma or glaucoma), contraindication to vaginal delivery, multiple pregnancy, nonvertex presentation, or major fetal anomalies were excluded.
Women who fulfilled the appropriate criteria were invited to participate in the study, and those who agreed provided informed consent. The randomization sequence was generated by a computerized random number generator in blocks of 10 and prepared by 2 of the investigators (O.C.G. and O.A.). Allocation to treatment arms was by the sequential opening of sealed, numbered envelopes, each of which contained intravenous oxytocin infusion (group A or traditional treatment group) or a dinoprostone pessary followed 6 hours later by intravenous oxytocin infusion (group B or experimental treatment group). Blinding of the participants and the obstetrician or midwife was not possible because of the obvious differences in techniques used and we did not find identical drugs for placebo matching. The study was not placebo controlled; thus, women and clinicians were aware of the treatment allocation scheme.
Demographic background data and obstetric history were obtained by a single operator (O.A.), who interviewed the women immediately before labor induction. The duration of leakage and the color and content of the fluid were noted. Sterile vaginal examination was done by a single operator to ascertain the Bishop score (O.A.). Ultrasound was done to assess fetal growth parameters, amniotic fluid index, and placental location. External cardiotocography was performed to assess fetal well-being and to confirm the absence of contractions.
Patients in group A received an oxytocin infusion at 2 mU/min, and it was doubled every 30 minutes to a maximum of 32 mU/min or until 4 contractions in 10 minutes were achieved. Patients in group B received a single dose of sustained-released dinoprostone (Propess; Vitalis, Ankara, Turkey) into the posterior vaginal fornix. This sustained-released product releases dinoprostone at a low but steady rate (0.3 mg/h). It remained in the vagina for up to 12 hours, as recommended by the manufacturer. Standard intravenous oxytocin was administered 6 hours after insertion of the vaginal pessary. An initial dose of 2 mU/min was increased by 2 mU/min at 30-minute intervals to a maximum dose of 32 mU/min or until 4 contractions in 10 minutes were achieved. Once started, oxytocin infusion was continued to delivery unless otherwise indicated. In group B, the pessary was removed if it was still present 12 hours after placement or if a worrisome fetal heart rate (FHR) pattern persisted.
All women were monitored for maternal pulse, temperature, respiratory rate, and blood pressure. FHR was monitored by continuous electronic means in all patients. The same examiner (O.A.) did a vaginal examination at 6-hour intervals to assess the progress of labor. Cervical dilatation, station, and position were noted at each examination. Antibiotic prophylaxis against early onset neonatal group B streptococcal sepsis was routinely administered during labor if PROM of ≥18 hours occurred or maternal fever ≥38°C developed. The usual prophylactic regimen was an initial dose of 2 g ampicillin intravenously, followed by 1 g ampicillin intravenously every 4 hours until delivery. Paracetamol and/or pethidine were used for labor analgesia. Epidural analgesia was not available for pain relief in labor.
The primary outcome measure was the number of women who succeeded in delivering vaginally within 24 hours of the initiation of the protocol. Secondary outcomes were incidence of excess uterine activity (uterine tachysystole with or without FHR deceleration), labor induction to delivery interval, labor induction to active phase interval (at least 70% effaced cervix, and cervical dilatation ≥4), meconium-stained liquor, mode of delivery, instrumental delivery rate, maternal satisfaction score for the birth process obtained within 24 hours of delivery (a visual analog scale [VAS] of 0-10, with higher score denoting greater satisfaction), VAS pain score (0 representing no pain to 10 representing unbearable pain), and maternal and neonatal complications.
Maternal complications included the incidence of maternal side effects (nausea, vomiting, diarrhea, and pyrexia), postpartum hemorrhage (blood loss >500 mL), third- or fourth-degree lacerations, intrapartum chorioamnionitis (defined as temperature ≥38°C accompanied by maternal or fetal tachycardia [>160 beats/min], uterine tenderness, malodorous amniotic fluid discharge, and/or maternal leukocytosis [white blood cell count >15,000 cell/min 3 ]), and postpartum endometritis (defined as temperature ≥38°C accompanied by uterine tenderness and/or purulent or foul-smelling lochiae beyond the first 24 hours after delivery). Neonatal complications noted were Apgar scores of <7 at 5 minutes, neonatal jaundice, neonatal sepsis, and admission to the NICU.
All cardiotocograms were reviewed by one of the authors to identify and classify abnormal patterns. Tachysystole with or without FHR abnormalities was defined as ≥6 uterine contractions per 10 minutes over 2 consecutive 10-minute periods. Tachysystole with FHR abnormalities was managed conservatively, with the women placed in the left lateral position, reducing or stopping oxytocin infusion, infusion of a bolus of 500 mL crystalloid solution, and administration of oxygen via a face mask. If the worrisome FHR pattern persisted, the dinoprostone pessary was removed in group B, and 250 μg of intravenous ritodrine was administered to both groups. If the pattern resolved, the oxytocin infusion was restarted after 15 minutes at half of the maximum previous dose.
There is no standard definition for what constitutes failed labor induction. Here, we defined failed induction of labor as when the patient was not in active phase after 12 hours. The decision to perform a cesarean delivery was made based on our usual obstetric practice, and the indication for the cesarean section was recorded (failed induction of labor, failure to progress in established labor, or nonreassuring fetal status [based on FHR patterns]).
At the start of this randomized controlled trial, no other published trial had compared these 2 induction regimens. Thus, we conducted a nonblinded pilot trial of 50 patients in each group before the full trial. The number of women who succeeded in delivering vaginally within 24 hours was 70% for intravenous oxytocin infusion and 82% for dinoprostone pessary, followed 6 hours later by intravenous oxytocin infusion. On the basis of these sample data, to keep the power of the study at 80% with an α level of 0.05, a sample size of 213 patients in each group was required. We recruited 225 patients in each group.
Med Calc, version 9.3 (Med Calc, Mariakerke, Belgium), was used for statistical analyses. Analysis was done by intention to treat. Normal distribution of continuous variables was assessed by the Kolmogorov-Smirnov test. The χ 2 test and Fisher exact test were used for analysis of categorical variables, Student t test was used for normally distributed variables in the analysis of continuous variables, and the Mann-Whitney U test was used for variables that were not normally distributed. Relative risk (RR) with 95% confidence interval (CI) was calculated. P values < .05 were considered to indicate statistical significance. Survival curves were created using the Kaplan-Meier method. The trial was registered at clinicaltrials.gov ( NCT01099280 ).
Results
In total, 450 women were randomized to treatment with oxytocin (n = 223) or dinoprostone followed by oxytocin (n = 227). Of these, 6 women withdrew from the study (2 in the oxytocin or traditional treatment group and 4 in the dinoprostone–oxytocin or experimental treatment group) ( Figure 1 ). Maternal demographic and obstetric characteristics were similar between the groups, with no difference in gestational age, preinduction Bishop score, PROM to induction interval, or antenatal risk factors. Both groups also had similar numbers of multiparous and nulliparous patients ( Table 1 ).
