A MASKED RANDOMIZED COMPARISON OF ORAL AND VAGINAL ADMINISTRATION OF MISOPROSTOL FOR LABOR INDUCTION

Objective: To test the null hypothesis that administering misoprostol orally or vaginally will result in no difference in time to vaginal bi rth, and to determine whether different frequencies of tachysystole an d hyperstimulation are associated with route of administration. Method s: Two hundred six women after 37 completed weeks' gestation who prese nted with an indication for induction were randomly assigned to receiv e misoprostol (50 mu g) either orally or vaginally every 4 hours as ne eded to induce labor. placebo use and allocation concealment accomplis hed blinding until data analysis was completed. Sample size was calcul ated to allow a two-tailed alpha of .05 and power (1 - beta) of 80%. A ll fetal heart rate and uterine activity graphs were classified accord ing to Curtis' criteria before induction groups were unmasked. Results : Analysis involved 104 women in the oral group and 102 in the vaginal group. The mean time (+/- standard deviation) to vaginal birth with o ral misoprostol was 1072 (+/-593) minutes compared with 846 (+/-385) m inutes with the vaginal protocol (P = .004). There were no significant differences in cesarean rate, epidural use, or neonatal outcomes. Mor e frequent tachysystole for 20 minutes (P < .01) and hyperstimulation (P < .04) were observed with vaginal misoprostol. No neonatal asphyxia occurred in either group. Conclusion: Misoprostol effectively induces labor, given orally or vaginally. There is a shorter interval to vagi nal birth with vaginal application; however, the more frequent occurre nce of fetal heart rate graph abnormalities in this group suggests tha t, until the optimal dosing interval for vaginal use is determined, th e preferred route of misoprostol administration might be oral. (Obstet Gynecol 1998;92: 481-6. (C) 1998 by The American College of Obstetric ians and Gynecologists.).