51 A Multivariable Predictive Model for Success of External Cephalic Versions

Abstract

To create a user-friendly prediction model for external cephalic version (ECV) success using patient characteristics in a US-based population. A retrospective case control was performed of women who underwent an ECV attempt between 2006 and 2016 who were at least 18 years of age with a non-anomalous, singleton pregnancy. Variables assessed included maternal age, height, weight, BMI, parity, fetal sex, and obstetric information known at the ECV procedure (gestational age, estimated fetal weight, fetal position, placental location, and amniotic fluid volume). Univariable and multivariable logistic regressions were used to determine the association of patient characteristics with ECV success. Estimated odds ratios and corresponding 95% confidence intervals were calculated for each variable, and a parsimonious model for ECV success was determined using bootstrapping (N=2000) to find the model with the highest discriminatory capacity [as determined by the area under the receiver operating characteristic curve (AUC)]. 1138 patients were included in this analysis. The overall ECV success rate was 40.6%. Factors significantly associated with ECV success were maternal age, BMI, parity, placental location, estimated fetal weight, and fetal position. A model with BMI, parity, placental location, and fetal position (Table) had the highest AUC [0.667 (CI 0.634-0.701)], resulted in good calibration (Figure), and is represented by the following equation: 1/(1+eˆ(-x)) where x=1.1726 - 0.0314 (BMI) - 0.9299 (nulliparity) + 1.0218 (transverse/oblique position at ECV) – 0.5113 (anterior placenta). An interactive version of this equation was created and can be accessed at http://www.ecvcalculator.com/. A user-friendly “ECV Success Calculator” can be created that uses easily obtainable patient factors and can be used as an aid in patient counseling.View Large Image Figure ViewerDownload Hi-res image Download (PPT)