Premature rupture of membranes (PROM) at term occurs in about 8% of term pregnancies.1 Expedited labor induction in PROM reduces chorioamnionitis, endometritis, and neonatal infection.2 Oxytocin infusion compared with prostaglandin for labor induction is associated with less chorioamnionitis and neonatal infection, but epidural anesthesia and the requirement for internal fetal monitoring is increased.3 Recent guidance supports the use of oxytocin infusion as the first-line labor induction method in PROM.1
Premature rupture of membranes, nulliparity, and induction of labor are recognized major risk factors for failure to progress during the first stage of labor.4 Induction of labor in PROM compared with when membranes are intact is associated with a longer second stage and a higher rate of cesarean delivery due to failure to progress.5 Neonatal intensive care admission, variable decelerations, and primary cesarean delivery rates are positively correlated with a longer admission-to-labor-onset interval in women with PROM.6 Nulliparas with PROM and unfavorable cervices needing labor induction remain a clinical challenge.
Concurrent oxytocin infusion with prostaglandin in the labor induction of nulliparas with intact membranes at term has been shown to improve maternal satisfaction7 and is associated with a shorter induction-to-delivery interval and reduced drug costs.8 Concurrent oxytocin and prostaglandin for labor induction with intact membranes does not increase the risk of adverse outcomes due to uterine hyperstimulation.7–11 There is a paucity of information on concurrent therapy in term PROM labor induction.
We hypothesized that concurrent treatment will result in a faster and more satisfactory labor in PROM labor induction. We sought to estimate the effect of front-loading a single dose of vaginal dinoprostone in nulliparas with PROM and unfavorable cervices who are receiving titrated oxytocin infusion as the standard method for induction of labor.
MATERIALS AND METHODS
We performed a double-blind randomized trial of a single dose of 3-mg dinoprostone pessary and titrated intravenous oxytocin infusion compared with placebo pessary and titrated intravenous oxytocin infusion. The study was approved by the University of Malaya Medical Centre Medical Ethics committee (Institutional Review Board Reference No. 607.9). All participants provided written informed consent. Enrolment was from November 12, 2007, to December 1, 2008. Participants were followed up until hospital discharge. Potential candidates for labor induction were identified by care providers in our delivery suite after the diagnosis of uncomplicated PROM at term and a decision to induce labor. Women were given verbal and written information before recruitment and randomization. All participants gave written consent.
Inclusion criteria were ruptured membranes confirmed clinically by the demonstration of pooling of liquor at the upper vagina on speculum assessment, nulliparity (no previous delivery with more than 20 weeks of gestation), gestational age more than 36 weeks, unfavorable Bishop Score 6 or less,7–10 less than one contraction in 15 minutes if any was present, singleton fetus in cephalic presentation, a reassuring cardiotocogram, and decision already reached to induce labor. Exclusion criteria were previous uterine incision, meconium-stained liquor, known severe fetal anomaly, maternal asthma, and allergy to prostaglandin. In our center, women with confirmed uncomplicated PROM were given the options of immediate labor induction or expectant inpatient management for up to 24 hours.
The randomization sequence was generated by a computerized random number generator in blocks of 8 and prepared by an investigator (P.C.T.). Allocation to treatment arms was effected by the sequential opening of sealed numbered envelopes, each containing either 3-mg dinoprostone pessary or an identical-looking placebo pessary. The sealed numbered envelopes were made up periodically as required and kept in a refrigerator until use.
The allocated pessary was inserted into the posterior fornix. At the same time, oxytocin infusion at 2 milliunits/min was started for all participants. Oxytocin infusion was doubled every 30 minutes to a maximum of 32 milliunits/min or until four contractions in 10 minutes was achieved.
Continuous cardiotocogram monitoring was maintained throughout the induction and labor. Standard management of labor induction and labor was applied as previously described.12 Antibiotic prophylaxis against early onset neonatal Group B streptococcal sepsis was routinely administered during labor if PROM of 18 hours or more was reached or maternal fever 38°C or more developed. The usual prophylactic regimen was an initial dose of 2 g of ampicillin intravenously followed by 1 g ampicillin intravenously every 4 hours until delivery.
In the event of a nonreassuring cardiotocogram associated with uterine hyperactivity, the following actions were suggested: reduce or stop oxytocin infusion, remove pessary (if still present), administer subcutaneous terbutaline for tocolysis or expedite operative delivery depending on individual circumstances and the severity of the cardiotocogram abnormality. The cardiotocogram was assessed by a blinded investigator after delivery to identify hyperstimulation and cardiotocogram abnormalities.
Participants’ characteristics and outcomes were extracted onto a standardized case report form. Case notes and hospital records were scrutinized after delivery to retrieve relevant clinical outcome data.
Primary outcomes were 1) vaginal delivery within 12 hours of induction and 2) maternal satisfaction score for the birth process obtained within 24 hours of delivery. A visual analog scale (VAS) with a range of 0 to 10, with higher score denoting greater satisfaction, was used to gauge maternal satisfaction. Secondary outcomes were induction-to-delivery interval, mode of delivery, analgesia use in labor, uterine hyperstimulation (eg, uterine tachysystole—defined as six or more uterine contractions per 10 minutes over two consecutive 10-minute periods without fetal heart rate abnormalities; uterine hypertonus—uterine contraction lasting more than 2 minutes; or hyperstimulation syndrome—defined as six or more uterine contractions per 10 minutes with fetal heart rate abnormalities), meconium stained liquor, peridelivery blood loss and blood transfusion, maternal fever from recruitment to discharge (temperature 38°C or more), various neonatal outcomes (special care nursery admission, umbilical cord blood pH and base excess, Apgar score, and phototherapy).
Because a study of concurrent oxytocin and prostaglandin for PROM labor induction was not available at trial inception to our knowledge, power calculation was based on a study of labor induction in women with intact membranes using concurrent oxytocin with prostaglandin compared with prostaglandin, which showed a 39% compared with 15% vaginal delivery rate at 12 hours for nulliparas.8 Setting significance at 5%, power at 80%, and one-to-one recruitment ratio, 53 women were required in each arm. Allowing for a 10% dropout rate, we planned to recruit a total of 116 women.
Analysis was performed on intention-to-treat basis. Data were entered into SPSS 15 (SPSS Inc., Chicago IL). GraphPad Instat software (GraphPad Software Inc., San Diego CA) was also used. The one-sample Kolmogorov-Smirnov test was used to check for normality of data distribution. The Student t test was used on continuous data, and the Mann-Whitney U test was used for ordinal and nonnormally distributed data. Fisher exact test was applied for 2×2 categorical data sets and χ2 test for larger than 2×2 categorical data sets. Relative risk (RR) and its 95% confidence interval (CI) were calculated using GraphPad Instat (using Fisher exact test). All tests used were two-tailed. A P<.05 in any test was considered significant.
Figure 1 displays the recruitment flow. One hundred sixteen women were randomly assigned: 58 to dinoprostone and oxytocin and 58 to placebo and oxytocin. All assigned women received allocated treatment. One woman allocated placebo was found to be in breech presentation cesarean delivery was offered, accepted, and quickly performed. Another woman allocated to dinoprostone was discovered to have a history of asthma soon after pessary insertion, and the pessary was immediately removed. Labor induction was continued with oxytocin only. These two women were excluded. There were another five women (four with Bishop score 7 and one with Bishop score 8), four of whom were assigned to concurrent therapy and one to oxytocin only who despite not fulfilling inclusion criterion of Bishop Score 6 or less were included. The errors incurred in the summation of their Bishop score from its various components were not discovered until at data analysis (they were thought to have a Bishop score of 6 when recruited). These five participants were managed as per trial protocol throughout their induction and labor.
Patient characteristics are listed in Table 1. The groups were comparable in all characteristics except for maternal age. The women assigned to concurrent dinoprostone and oxytocin was younger (mean age±standard deviation 26.9±2.9 years compared with 28.5±4.2 years; P=.017).
Outcomes are shown in Table 2. There were no significant differences in the predefined primary outcomes: vaginal delivery rates within 12 hours were 25 of 57 (43.9%) compared with 27 of 57 (47.4%) (relative risk 0.9, 95% CI 0.6–1.4, P=.85) and median VAS satisfaction score was 8 (interquartile range 2) compared with 8 (interquartile range 2), P=.38 for concurrent treatment compared with oxytocin only, respectively.
There was no significant difference in any of the secondary outcomes recorded. Induction-to-delivery interval and mode of delivery were very similar. Although not significant, 14% (compared with 5.3%; P=.20) of women assigned to concurrent treatment had uterine hyperstimulation, and 8.8% (compared with 1.8%; P=.44) of them also required stoppage of oxytocin infusion due to uterine hyperactivity. One woman allocated to oxytocin only also required terbutaline tocolysis for uterine hyperstimulation syndrome. There was no cesarean delivery indicated by uterine hyperstimulation syndrome within the trial.
Two women assigned to concurrent treatment had significant morbidity. In one woman, labor was induced 9.5 hours after PROM. Intrapartum, she had epidural anesthesia, developed a fever of 40.1°C, and was treated with intravenous ampicillin. She had a normal vaginal delivery 7 hours after labor induction. A combination of a cervical tear and uterine atony caused a primary postpartum hemorrhage with cumulative blood loss of 3 L. She required multiple uterotonics, examination under anesthesia, cervical suturing, and eventually laparotomy and bilateral internal iliac ligation to control persistent bleeding from the cervix. Blood transfusion was given, and she spent several days in the intensive care unit. Her neonate was also admitted to the special nursery for suspected congenital pneumonia. The mother made a full recovery, with an intact uterus. Her neonate also recovered fully. Her provider requested the identity of the allocated treatment at the time of these complications and her treatment was revealed to the provider. In the second woman, labor was induced 2 hours after PROM, and an epidural was placed during labor. Labor was slow, and progress arrested at 9-cm cervical dilatation. A healthy 4.1-kg neonate was delivered by cesarean delivery 19 hours after induction. During the cesarean delivery, hemorrhage due to uterine atony was noted. Uterine atony did not respond to multiple uterotonics. A B-Lynch suture was performed to control hemorrhage. Blood loss was 1.5 L, and blood was transfused. The mother made a full recovery, with an intact uterus.
One woman assigned to placebo had a prolonged episode of fetal bradycardia on cardiotocogram soon after commencement of labor induction that required removal of the pessary and stoppage of oxytocin infusion. However, there was no evidence of uterine hyperactivity at the time of fetal bradycardia. She eventually had a cesarean delivery.
There was no significant difference in any neonatal outcome recorded. There were four neonatal admissions—three newborns were from women assigned to placebo and one from a woman assigned to concurrent therapy. These admissions were indicated by respiratory distress (2) and presumed sepsis (2); there was no admission for birth asphyxia. All four recovered fully. There was a single baby with umbilical arterial cord blood pH <7.1 but did not need admission. No neonate had an Apgar score less than 7 at 5 minutes. Phototherapy for neonatal jaundice was less commonly encountered for the neonates of women assigned to concurrent therapy, 3.5% compared with 8.8% (RR 0.4, 95% CI 0.1–2.0; P=.44), but this difference was not significant.
Our study showed that the front loading of vaginal dinoprostone concurrently with oxytocin infusion to induce labor in term PROM did not hasten delivery or decrease operative delivery compared with oxytocin infusion. Intrapartum analgesic need was similar. Maternal satisfaction with the birth process was also similar.
Concurrent prostaglandin and oxytocin to induce labor in the context of intact membranes has been shown to result in a shorter induction-to-delivery interval but has not significantly decreased the cesarean delivery rate.8,10,11 Maternal satisfaction is also higher with concurrent therapy.7 Membrane sweeping in conjunction with dinoprostone or amniotomy at labor induction (intact membranes) has been shown to shorten induction-to-delivery interval, increase spontaneous vaginal delivery rate, and improve maternal satisfaction.12 Our finding was in contrast to the positive effect on delivery timing that concurrent front-loading regimens have in labor induction when membranes were intact. This difference might be due to the fact that we were comparing concurrent therapy against oxytocin in this study. Previous trials of concurrent prostaglandin and oxytocin for labor induction with intact membranes typically compared against prostaglandin arms. We titrated the rate of oxytocin infusion against uterine contractions. This meant that any slack in uterine activity would be quickly dealt with by an increased delivery of oxytocin, leaving little room for dinoprostone to demonstrate optimization of uterine activity.
Also, membrane rupture is associated with the considerable local release of prostaglandin at the site of membrane.13 At term PROM, 72% would go into labor spontaneously within 24 hours14 suggesting PROM alone is an effective pathway for labor onset for many. The landmark TERMPROM trial15 has shown that immediate induction with oxytocin compared with prostaglandin is associated with significantly shorter time to active labor, duration of active labor and interval of membrane rupture to delivery, suggesting that the criterion standard agent to use in term PROM labor induction should be oxytocin even without considering its advantage of less infective morbidity.16 Our finding of no benefit and of possible harm with concurrent use of dinoprostone and oxytocin might be explained by the rationale that PROM has released the optimal amount of prostaglandin for synergism with titrated oxytocin infusion to ripen the cervix and successfully induce labor. Additional exogenous prostaglandins in conjunction with oxytocin stimulation might tip the uterine response toward hyperactivity.
Overall, the cesarean delivery rate in our trial was 38%, and 63% of the cesarean deliveries were indicated by failure to progress. A previous trial from our center of concurrent oxytocin infusion and vaginal dinoprostone compared with vaginal dinoprostone for labor induction of term nulliparas with intact membranes and unfavorable cervices reported cesarean delivery rates of 41.9% compared with 44.7%, respectively.7 These cesarean rates were high and broadly similar to that of this study. A major challenge is still in place to find an effective labor induction regime in high-risk nulliparas with unfavorable cervices with and without PROM.
There were two women from the concurrent therapy group with significant morbidity due to hemorrhage due in part to uterine atony. Prolonged exposure to prostaglandin does not result in oxytocin receptor down-regulation or reduction in myometrial contractility in an animal model.17 Oxytocin receptor down-regulation and desensitization occurs with prolonged exposure to oxytocin,18,19 but the total oxytocin infused is nonsignificantly lower in the concurrent therapy group. Induction-to-delivery interval was also very similar. There was no clear rationale to account for an excess of uterine atony with concurrent therapy. No woman required emergency cesarean delivery for uterine hyperstimulation syndrome, a more immediate adverse effect anticipated of concurrent therapy.
Our study has limitations. The mean age of women in the randomized groups was slightly but significantly different. We believe that this is a chance event. Controlling for maternal age in the analysis did not make any difference to the effect of concurrent therapy on the primary outcome of vaginal delivery within 12 hours of labor induction. Trial participants were heterogeneous, comprising women who opted for immediate labor induction on confirmation of PROM and those who opted for initial expectant management but later required labor induction. Power calculation was based on labor induction in women with intact membranes, where a relatively large treatment effect on delivery within 12 hours has been demonstrated with concurrent therapy.8 Our study would be underpowered if the treatment effect were more modest. However, we did not observe any trend of more effective labor induction with concurrent therapy in our analysis.
Concurrent treatment with single-dose vaginal dinoprostone and titrated oxytocin infusion is no better than placebo pessary and similarly titrated oxytocin infusion in inducing labor for PROM at term in nulliparas with unfavorable cervices. There is concern about increased uterine hyperstimulation with concurrent therapy.
1. ACOG Committee on Practice Bulletins-Obstetrics. ACOG Practice Bulletin No. 80: premature rupture of membranes. Clinical management guidelines for obstetrician-gynecologists. Obstet Gynecol 2007;109:1007–19.
2. 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.
3. Tan BP, Hannah ME. Prostaglandins versus oxytocin for prelabour rupture of membranes at term. The Cochrane Database of Systematic Reviews 2000, Issue 2. Art. No.: CD000159. DOI: 10.1002/14651858.CD000159.
4. Sheiner E, Levy A, Feinstein U, Hallak M, Mazor M. Risk factors and outcome of failure to progress during the first stage of labor: a population-based study. Acta Obstet Gynecol Scand 2002;81:222–6.
5. Park KH, Hong JS, Ko JK, Cho YK, Lee CM, Choi H, et al. Comparative study of induction of labor in nulliparous women with premature rupture of membranes at term compared to those with intact membranes: duration of labor and mode of delivery. J Obstet Gynaecol Res 2006;32:482–8.
6. Hallak M, Bottoms SF. Induction of labor in patients with term premature rupture of membranes. Effect on perinatal outcome. Fetal Diagn Ther 1999;14:138–42.
7. Tan PC, Valiapan SD, Tay PY, Omar SZ. Concurrent oxytocin with dinoprostone pessary versus dinoprostone pessary in labour induction of nulliparas with an unfavourable cervix: a randomised placebo-controlled trial. BJOG 2007;114:824–32.
8. Stewart JD, Rayburn WF, Farmer KC, Liles EM, Schipul AH Jr, Stanley JR. Effectiveness of prostaglandin E2 intracervical gel (Prepidil), with immediate oxytocin, versus vaginal insert (Cervidil) for induction of labor. Am J Obstet Gynecol 1998;179:1175–80.
9. Bolnick JM, Velazquez MD, Gonzalez JL, Rappaport VJ, McIlwain-Dunivan G, Rayburn WF. Randomized trial between two active labor management protocols in the presence of an unfavorable cervix. Am J Obstet Gynecol 2004;190:124–8.
10. Christensen FC, Tehranifar M, Gonzalez JL, Qualls CR, Rappaport VJ, Rayburn WF. Randomized trial of concurrent oxytocin with a sustained-release dinoprostone vaginal insert for labor induction at term. Am J Obstet Gynecol 2002;186:61–5.
11. Hennessey MH, Rayburn WF, Stewart JD, Liles EC. Pre-eclampsia and induction of labor: a randomized comparison of prostaglandin E2 as an intracervical gel, with oxytocin immediately, or as a sustained-release vaginal insert. Am J Obstet Gynecol 1998;179:1204–9.
12. Tan PC, Jacob R, Omar SZ. Membrane sweeping at initiation of formal labor induction: a randomized controlled trial. Obstet Gynecol 2006;107:569–77.
13. Leaver HA, Richmond DH, MacPherson HD, Hutchon DJ. Prostaglandin E2 in induction of labour: a pharmacokinetic study of dose and treatment protocols. Prostaglandins Leukot Essent Fatty Acids 1988;31:1–7.
14. Cammu H, Verlaenen H, Perde MP. Premature rupture of membranes at term in nulliparous women: a hazard? Obstet Gynecol 1990;76:671–4.
15. Hannah ME, Ohlsson A, Farine D, Hewson SA, Hodnett ED, Myhr TL, et al. Induction of labor compared with expectant management for prelabor rupture of the membranes at term. TERMPROM Study Group. N Engl J Med 1996;334:1005–10.
16. Tan BP, Hannah ME. Prostaglandins for prelabour rupture of membranes at or near term. The Cochrane Database of Systematic Reviews 2000, Issue2. Art. No.:CD000178. DOI: 10.1002/14651858.CD000178.
17. Engstrom T, Atke A, Vilhardt H. Oxytocin receptors and contractile response of the myometrium after long term infusion of prostaglandin F2 alpha, indomethacin, oxytocin and an oxytocin antagonist in rats. Regul Pept 1988;20:65–72.
18. Phaneuf S, Asboth G, Carrasco MP, Europe-Finner GN, Saji F, Kimura T, et al. The desensitization of oxytocin receptors in human myometrial cells is accompanied by down-regulation of oxytocin receptor messenger RNA. J Endocrinol 1997;154:7–18.
19. Phaneuf S, Asboth G, Carrasco MP, Linares BR, Kimura T, Harris A, et al. Desensitization of oxytocin receptors in human myometrium. Hum Reprod Update 1998;4:625–33.