Labor is induced in up to 20 percent of pregnancies, and cervical ripening is required in about one half of inductions. Several systems have been developed to apply the prostaglandin dinoprostone to the cervix for ripening, but little is known about the comparative effectiveness of the different methods. Perry and Leaphart compared intracervical placement of sustained-release dinoprostone with posterior fornix placement.

They studied 63 mothers admitted to a university medical center for induction of labor. Participants were required to have a singleton, cephalic fetus of at least 36 weeks of gestation and a Bishop score of less than 6. Patients with more than 2-cm cervical dilation, recent use of cervical ripening agents, a history of cesarean delivery or hysterotomy, or any contraindication to vaginal delivery were excluded from the study. The mothers were assigned randomly to intracervical or posterior fornix placement of dinoprostone. Patients and staff were blinded to the location of the ripening agent, apart from the physician who performed the procedure. The insert was retained for 12 hours, and continuous fetal monitoring and tocodynamometry were used. Cervical change was assessed after 12 hours and if the dinoprostone was removed for any reason.

The two groups of patients were comparable in all important variables. Overall, placement in the cervix was associated with a lower median time to delivery (22.4 hours compared with 25.1 hours for placement in the posterior fornix), but the difference was not statistically significant. Nine mothers in the intracervical group and eight mothers in the posterior fornix group required multiple inserts of dinoprostone. When data on the 46 mothers who required only one dose of dinoprostone were analyzed, intracervical placement was associated with significantly shorter times to delivery and to initial request for pain medication (see accompanying table). The time to onset of labor or regular contractions was not significantly different, but 42 percent of the intracervical group did not require oxytocin compared with only 27 percent of the posterior fornix group. The groups were comparable in all other measures of obstetric and neonatal outcome.

The authors conclude that intracervical (rather than posterior fornix) placement of dinoprostone decreases time to delivery without increasing complications of labor in mothers who respond to a single dose of the sustained-release preparation.