Predictors of Vaginal Delivery after Cervical Ripening using a Synthetic Osmotic Dilator

Abstract Objective To evaluate the determinants of vaginal delivery and safety in women undergoing cervical ripening with a synthetic osmotic dilator (Dilapan-S) prior to induction of labor. Methods We conducted a secondary analysis of an international multicenter prospective observational study of Dilapan-S for cervical ripening in pregnancies greater than 32 weeks. Data were obtained in a standardized fashion and entered into a centralized electronic data capture system. The association between Bishop score and vaginal delivery was further evaluated with a multivariate receiver-operating characteristic (ROC) curve analysis. A Wilcoxon rank test and 1-way ANOVA were used for statistical analysis (significance: P Results Between May 2015 and July 2016, 444 pregnant women were included. Three hundred ten (70%) delivered vaginally. Compared to patients who underwent cesarean delivery, those who delivered vaginally were more likely to have a history of prior vaginal delivery. Vaginal delivery rates were significantly correlated with Bishop scores of pre and post Dilapan-S and difference. After adjusting for age, BMI, number of dilators, cervical ripening time, and gestational age, both prior vaginal delivery and post-Dilapan-S Bishop scores were strong predictors of vaginal delivery (estimate coefficient: 0.1275 ± 0.03 P =  .0002; 0.049 ± 0.01 P =  .0001; respectively). Aggregate ROC accounting for these variables further supported these findings (AUC = 0.734). The lower confidence interval limit of vaginal delivery rates was above 50% when post-Dilapan-S Bishop scores were ≥ 5. Cox regression analyses demonstrated that the duration of labor was significant shorter in women that had vaginal delivery. Conclusion Bishop scores after cervical ripening with Dilapan-S are good predictors of vaginal delivery. Bishop scores