Misoprostol (Cytotec; G.D. Searle LLC, Chicago, IL) is a synthetic prostaglandin E1 that currently is available as a 100-μg tablet for prevention of gastric ulcers. It is also used off-label for preinduction cervical ripening as well as induction of labor. Misoprostol is inexpensive, with a cost of less than $1 per 100-μg tablet; is stable at room temperature; and is easily administered orally. In October 2000, Searle notified physicians that misoprostol was not Food and Drug Administration approved for labor induction. This position was challenged by The American College of Obstetricians and Gynecologists Committee on Obstetrics Practice on the grounds that misoprostol was a safe and effective drug for cervical ripening and labor induction. 1 Indeed, Goldenberg and coauthors reviewed the literature on obstetric use of misoprostol and concluded that “over 200 studies involving a total of more than 16,000 women have evaluated its effectiveness in pregnant women, and the results support its continued use.” 2 The Food and Drug Administration has since modified the misoprostol package insert to no longer prohibit its off-label use in pregnancy.
We sought to assess the effectiveness of oral misoprostol as a means of preempting the requirement for oxytocin infusion in women with spontaneously ruptured membranes but judged to have inadequate uterine activity. The prevailing practice at our institution has been to commence oxytocin stimulation of labor in such women as soon as possible after admission. We were of the view that oral administration of misoprostol, if effective, would significantly simplify stimulation of labor compared with continuous oxytocin, which requires uninterrupted bedside nursing to achieve and maintain an effective and safe infusion.
MATERIALS AND METHODS
Between August 1, 2000 and July 12, 2001, healthy nulliparous parturients with singleton, cephalic-presenting fetuses (at least 36 0/7 to 41 6/7 weeks' gestation) with ruptured membranes and without evidence of labor were evaluated for participation in this study at Parkland Hospital. The study protocol was developed by investigators in the Department of Obstetrics and Gynecology and approved by the Institutional Review Board of the University of Texas Southwestern Medical Center at Dallas.
Ruptured membranes were diagnosed when amnionic fluid was visualized coming through the cervical os on sterile speculum examination. Cervical dilatation was to be 2 cm or less. In addition, these women experienced less than four contractions in a 10-minute period. Before randomization, pregnancies complicated by hypertension, diabetes, fever (at least 38.0C), meconium-stained amnionic fluid, or nonreassuring fetal heart rate tracing were excluded from participation in this study.
After written consent was obtained, each woman randomly received either misoprostol (100 μg) or an identical placebo given orally. Randomization was executed by the Investigational Drug Pharmacy of Parkland Hospital using a computerized randomization sequence, and the caregivers were blinded to the study drug given to each patient.
Continuous electronic fetal monitoring was used, with reassessment every 20 minutes by the labor and delivery unit nursing staff. If there was no cervical change and the patient did not perceive painful uterine contractions, a second study drug dose was administered 4 hours after the initial dose. After an additional 4 hours (for a total of 8 hours from administration of the initial dose), an oxytocin infusion was initiated if there was no cervical change and an absence of painful uterine contractions. Oxytocin was infused at a starting dose of 6 mU per minute with 6 mU per minute increases at 40-minute intervals to a maximum dose of 42 mU per minute. 3 Dystocia necessitating cesarean delivery was diagnosed when labor did not progress in the presence of adequate uterine activity, defined as more than 200 Montevideo units for 2–4 hours. Forceps-assisted delivery was performed for nonreassuring fetal heart rate patterns in the second stage of labor or for prolonged second stage of labor.
Labor epidural analgesia was provided upon the patient's request. The diagnosis of chorioamnionitis was made when the patient developed a fever (at least 38.0C) during her labor course.
Uterine tachysystole was defined when six or more uterine contractions occurred per 10 minutes for a span of 20 minutes in the absence of a nonreassuring fetal heart rate. Uterine hypertonus was defined as a single contraction lasting for longer than 2 minutes without fetal heart rate decelerations. Uterine hyperstimulation syndrome was diagnosed when there was tachysystole or hypertonus with associated fetal heart rate decelerations. 4 When hyperstimulation syndrome was identified, the patient was placed in a lateral position and oxygen (10 L per minute) was applied by face mask.
Using Ngai et al's results, 5 sample size calculations were initially based on an expected 51% requirement for oxytocin infusion with placebo. Using 80% power for a two-sided test of .05 significance, we estimated that 56 women would need to be randomized to demonstrate the reported four-fold reduction in the need for oxytocin. Because of slow recruitment, the inclusion criteria were modified to permit enrollment of patients regardless of the elapsed time from membrane rupture to admission. This significantly increased the pool of patients eligible for this trial. Using these inclusion criteria, the sample size was calculated to be 102 randomized women to demonstrate a reduction in the requirement for oxytocin infusion from 35% to 10%. This revised prediction for the use of oxytocin in the placebo arm was based on the assumption that women with longer ruptured membranes intervals would less often require oxytocin.
Statistical analysis included Pearson χ2, Cochran–Mantel–Haenszel χ2 for the overall fixed estimate of odds ratio, analysis of variance, and Kruskal–Wallis test. P values of less than .05 were considered statistically significant. Analysis was performed using SAS 8.2 (SAS Institute Inc., Cary, NC).
During the 12-month study period, a total of 102 women were enrolled. The trial was originally limited to women with ruptured membranes–admission intervals of less than 3 hours. Subsequently, women with longer ruptured membranes–admission intervals were enrolled to facilitate recruitment. Maternal demographic characteristics for women with ruptured membranes less than 3 hours versus more than 3 hours were compared, as well as the combined misoprostol versus placebo study groups, which are shown in Table 1. There were no significant differences between the subgroups or the combined study groups.
Labor outcomes in women randomized to misoprostol versus placebo are shown in Table 2. There was no significant difference between these groups when comparing admission–study drug administration intervals. The study drug–delivery intervals, however, were significantly shorter in women given misoprostol (median 790 minutes, versus 1228 minutes for placebo; P < .001). The requirement of oxytocin administration was significantly decreased in the women who received misoprostol (37%, versus 90% for placebo; P < .001). Six or more uterine contractions occurring in 10 minutes without fetal heart rate decelerations (uterine tachysystole) occurred in 25% of women receiving misoprostol, compared with 2% of women receiving placebos. Hyper-stimulation syndrome occurred infrequently: There were three cases (6%) within the misoprostol group, compared with none in the placebo group (P = .079).
Significantly more women randomized to the placebo (59%) underwent labor epidural analgesia than women randomized to misoprostol (39%) (P = .048). However, an analysis stratified by oxytocin requirement yielded no significant association between misoprostol versus placebo and epidural analgesia (P = .324). The incidence of chorioamnionitis was not significantly related to misoprostol administration. In addition, there were no significant differences in route of delivery between the two study groups (Table 3).
Birth outcomes did not differ when comparing the two groups. There were no infants who had an Apgar score of 3 or less at 5 minutes, had an umbilical artery blood pH of 7.00 or less, or required admission to the neonatal intensive care unit.
Misoprostol reduced the use of oxytocin infusion from 90% in the placebo group to 37% in the treatment group. This reduction was associated with an admission–delivery interval approximately 7 hours shorter when misoprostol had been used. Serious hyperstimulation (ie, uterine hyperactivity associated with fetal heart rate decelerations) was infrequent. There were three such cases, and none required emergency cesarean delivery. Uterine hyperactivity without accompanying fetal heart rate decelerations was frequent and occurred in 25% of women given misoprostol. Route of delivery and infant outcome were unaffected by misoprostol.
A total of four randomized studies have been published on the use of oral misoprostol in women with ruptured membranes without labor at term. 5–8 The misoprostol dosages studied ranged from 50 μg to 200 μg. In all cases, misoprostol administration was reported to be an effective and safe method for stimulation of labor. The effects of oral misoprostol were compared to those of a placebo in two of these previously published studies. 5,8 Ngai and colleagues 5 randomized 80 women to a single 200-μg misoprostol dose and found the interval from treatment to delivery to be approximately 11 hours shorter. Similarly, Hoffmann and coauthors 8 randomized 96 women to 100 μg of oral misoprostol, for a maximum of two doses, 6 hours apart, and this resulted in shortening the median treatment to delivery interval from 25 hours to 7.5 hours. Ngai and colleagues 5 also reported approximately a four-fold reduction in use of oxytocin. Our results are consistent with these prior reports on the efficacy of oral misoprostol to shorten labor and often reduce the need for oxytocin infusion in women with ruptured membranes at term. Specifically, the magnitude of reduction in oxytocin use was approximately three-fold in our study, although the absolute proportions of patients given oxytocin in both study areas were much higher than those reported by Ngai. In planning our study, we attribute this variation to unforeseeable differences in management of the patients enrolled in Ngai's and our studies.
Besides efficacy, safety is an important consideration in the evaluation of oral misoprostol given for stimulation of labor. The sample sizes previously reported, individually as well as for our study, are insufficient for an assessment of the safety of oral misoprostol. Shown in Figure 1, in meta-analysis format, are outcomes that could be used to evaluate the safety of oral misoprostol, using the four previously reported trials 5–8 as well as our own results. The aggregate number of pregnancies studied totals 466. There were no perinatal deaths in any of these trials. Based upon this meta-analysis, as well as our own now reported experience, we find that oral misoprostol not only is very effective for stimulation of labor but also appears to be safe. Uterine rupture, a reported complication, 1 did not occur in our study, nor was it reported in other comparable trials analyzed. However, the sample size necessary to address uterine rupture as an outcome clearly exceeds the bounds of our analysis.
Because our study protocol did not permit oxytocin use until 8 hours had elapsed, the results in the placebo arm may provide some insight into management of women with ruptured membranes without labor at term. There are at least three alternative management approaches for these women. One is to promptly commence labor stimulation with oxytocin. Another approach that seems to have fallen out of favor is to wait for spontaneous labor for 24 hours or longer. 9 The third alternative sometimes used is to wait only 8–12 hours before beginning labor stimulation in the hope that spontaneous labor will ensue. Only 10% of the placebo group spontaneously labored within 8 hours in our trial. Moreover, the median elapsed time in the labor unit was increased from 13.1 hours in the misoprostol group to 20.5 hours in the placebo group, suggesting that deliberate expectant management for 8 hours offered no benefit to the great majority of women.
1. American College of Obstetricians and Gynecologists. Response to Searle's drug warning on misoprostol. ACOG committee opinion no. 248. Washington: American College of Obstetricians and Gynecologists, 2002.
2. Goldenberg AB, Greenberg MB, Darney PD. Misoprostol and pregnancy. N Engl J Med 2001;344:38–45.
3. Satin AJ, Leveno KJ, Sherman L, Brewster DS, Cunningham FG. High-versus low-dose oxytocin for labor stimulation. Obstet Gynecol 1992;80:111–6.
4. American College of Obstetricians and Gynecologists. Fetal heart patterns: Monitoring, interpretation, and management. ACOG technical bulletin no. 207. Washington: American College of Obstetricians and Gynecologists, 2002.
5. Ngai SW, To WK, Lao T, Ho PC. Cervical priming with oral misoprostol in pre-labor rupture of membranes at term. Obstet Gynecol 1996;87:923–6.
6. Butt KD, Bennett KA, Crane JMG, Hutchens D, Young DC. Randomized comparison of oral misoprostol and oxytocin for labor induction in term prelabor membrane rupture. Obstet Gynecol 1999;94:994–9.
7. Ngai SW, Chan YM, Lam SW, Lao TT. Labor characteristics and uterine activity: Misoprostol compared with oxytocin in women at term with prelabor rupture of membranes. Br J Obstet Gynaecol 2000;107:222–7.
8. Hoffmann RAM, Anthony J, Fawcus S. Oral misoprostol vs. placebo in the management of prelabor rupture of membranes at term. Int J Gynaecol Obstet 2001;72:215–21.
9. 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. N Engl J Med 1996;334:1005–10.