A META-ANALYSIS OF CONTROLLED-RELEASE PROSTAGLANDIN FOR CERVICAL RIPENING AND LABOUR INDUCTION.

OBJECTIVES: to examine the effectiveness and safety of controlled-release prostaglandin for cervical ripening. DATA SOURCES: we conducted a computerized search of Medline using the keywords "Cervidil"; or "Propess"; or "prostaglandin/dinoprostone" and "pessary"; or "prostaglandin/dinoprostone" and "controlled" and "release"; or "prostaglandin/dinoprostone" and "vaginal" and "insert"; identifying studies in the English literature published between January 1980 and December 1999. STUDY SELECTION: we included all randomized trials that compared controlled-release vaginal prostaglandin with a retrieval string with any other method of cervical ripening or with a placebo. Nine studies were identified.TABULATION, INTERACTION, AND RESULTS: the two coauthors independently abstracted the data and evaluated the studies for methodologic quality, blinded to each other's results. Homogeneity of results was evaluated, and if present, the summary odds ratio and 95 percent confidence intervals were determined. Compared with placebo, controlled-release prostaglandin resulted in cervical change (summary odds ratio [OR] = 3.99, 95% confidence interval [CI] = 2.71-5.86), a higher rate of vaginal delivery in 12 hours (OR = 29.01, 95% CI = 7.08-118.87), less need for oxytocin (OR = 0.14, 95% CI = 0.06-0.32), but a higher incidence of excessive uterine activity (P<0.0001) and hyperstimulation (P = 0.004). When compared with Prepidil, there was a higher rate of excessive uterine activity with controlled-release prostaglandin (P = 0.03), but less need for oxytocin (OR = 0.09, 95% CI = 0.01-0.53). When compared with Prepidil/immediate oxytocin, there was a lower rate of active labour in 12 hours with controlled-release prostaglandin (OR = 0.27, 95% CI = 0.10-0.72). There was a lower rate of vaginal delivery in 12 hours and higher incidence of oxytocin use with controlled-release prostaglandin as compared with misoprostol (OR = 0.53, 95% CI = 0.34-0.83 and OR = 1.58, 95% CI = 1.08-2.32 respectively). The induction to delivery interval was shorter with controlled-release prostaglandin than with placebo or Prepidil, but longer than with Prepidil/immediate oxytocin, or misoprostol. Although no differences were seen with maternal morbidity (such as Caesarean delivery) or neonatal outcomes, the sample size was not adequate to evaluate these outcomes. CONCLUSION: controlled-release prostaglandin appears to be an effective cervical ripening agent as compared with Prepidil, but may result in an increased incidence of excessive uterine activity. Controlled-release prostaglandin may not be as effective as misoprostol or Prepidil/immediate oxytocin. Further randomized trials with larger sample sizes are needed to evaluate maternal morbidity and neonatal outcomes.