A comparison of obstetrical outcomes with labor induction agents used at term.

Objective: To compare the obstetrical outcomes of term pregnancies induced with one of four commonly used labor induction agents.

Methods: This is a retrospective cohort study of induced deliveries between 1 August 1995 and 31 December 2007 occurring at the Los Angeles County + University of Southern California Medical Center. Viable, singleton, term pregnancies undergoing induction were identified. Exclusion criteria included gestational age less than 37 weeks, previous cesarean delivery, breech presentation, stillbirth, premature rupture of membranes, and fetal anomaly. Induction methods studied were oxytocin, misoprostol, dinoprostone and Foley catheter. Our primary outcome was cesarean delivery rate among the four induction agents. Secondary outcomes included rate of failed induction, obstetrical complications, and immediate neonatal complications.

Results: A total of 3707 women were included in the study (1486 nulliparous; 2221 multiparous). Outcomes were compared across induction methods using Chi-square Tests (Pearson or Fisher's, as appropriate). Among the nulliparous patients, there was no statistical difference among the four induction agents with regards to cesarean delivery rate (p = 0.51), frequency of failed inductions (p = 0.49), the cesarean delivery frequency for "fetal distress" (p = 0.82) and five minute Apgar score <7 (p = 0.24). Among parous patients, the cesarean delivery rate varied significantly by induction method (p < 0.001), being lowest among those receiving misoprostol (10%). Those receiving oxytocin and transcervical Foley catheter had cesarean rates of 22%, followed by dinoprostone at 18%. The rate of failed inductions was 2% among those receiving misoprostol, compared to 7-8% among those in the other groups (p < 0.01). Although cases of "fetal distress" between the four induction agents was not significantly different amongst multipara women, the cesarean delivery indication for "fetal distress" was higher among multipara receiving misoprostol (p = 0.004). There was no difference among the different induction agents with regards to five minute Apgar <7 (p = 0.34).

Conclusion: Among nulliparous women, all induction methods have similar rate of cesarean delivery. The use of misoprostol appears to be associated with a lower risk of cesarean birth among parous women induced at our institution.