Ultrasound assessment of the cervix in predicting successful membrane sweeping: a prospective observational study

Objective Membrane sweeping has been shown to potentially reduce the need for formal induction of labor. The primary aim of this study was to elucidate the role of ultrasound assessment of the cervix in predicting successful membrane sweeping in singleton pregnancies at term; the secondary aim was to build a multiparametric prediction model integrating, maternal, pregnancy and ultrasound characteristics, able to anticipate spontaneous delivery at term. Methods Prospective observational study including singleton pregnancies at term undergoing membrane sweeping. Cervical length (CL) and posterior cervical angle (PCA) were assessed on ultrasound immediately before the procedure. Primary outcome was successful membrane sweeping, defined as spontaneous vaginal birth without formal induction within the 24hours. A subgroup analysis was computed considering women experiencing spontaneous vaginal birth within 48 hours from the procedure. The secondary outcome was to explore the diagnostic performance of a multiparametric model including maternal, pregnancy, and ultrasound assessment of the cervix in predicting spontaneous vaginal birth following membrane sweeping. Multivariate logistic regression and area under the curve (ROC) analyses were used to compute the data. Results One hundred fifty-nine singleton pregnancies undergoing membrane sweeping were included in the analysis. Successful membrane sweeping within 24 hours occurred in 68/159 women (36.5%). Parity (aOR = 1.87, 95% confidence interval [CI] 1.2-2.44), gestational age (aOR = 1.32, 95% CI 1.14-1.76), CL (aOR = 0.47, 95%CI 0.31-0.69) and PCA (aOR = 1.22, 95%CI 1.07-1.41) were independently associated with spontaneous vaginal birth within 24 hours from sweeping. The AUC of the constructed model was 0.796 (95% CI 0.727-0.865). Likewise, CL (aOR = 0.80, 95%CI 0.72-0.89), PCA (aOR = 1.19, 95%CI 1.10-1.28) and gestational age at the procedure (aOR = 1.65, 95%CI 1.09-1.86; p = .04) were independently associated with delivery within 48 hours with an AUC of 0.737 (95%CI 0.659-0.815). Conclusions Cervical ultrasound assessment of the cervix prior to membrane sweeping is associated with spontaneous vaginal birth within 24 and 48 hours from the procedure. The combination of cervical ultrasonographic parameters with parity and gestational age can predict the chances of delivery within 24 or 48 hours from membrane sweeping. The findings from this study support the use of ultrasound assessment of the cervix prior to membrane sweeping in order to more accurately predict the likelihood of spontaneous vaginal delivery.