Development of a nomogram for prediction of vaginal birth after cesarean delivery

Abstract

Objective: To develop a model based on factors available at the first prenatal visit that predicts chance of successful vaginal birth after cesarean delivery (VBAC) for individual patients who undergo a trial of labor.

Methods: All women with one prior low transverse cesarean who underwent a trial of labor at term with a vertex singleton gestation were identified from a concurrently collected database of deliveries at 19 academic centers during a 4-year period. Using factors identifiable at the first prenatal visit, we analyzed different classification techniques in an effort to develop a meaningful prediction model for VBAC success. After development and cross-validation, this model was represented by a graphic nomogram.

Results: Seven-thousand six hundred sixty women were available for analysis. The prediction model is based on a multivariable logistic regression, including the variables of maternal age, body mass index, ethnicity, prior vaginal delivery, the occurrence of a VBAC, and a potentially recurrent indication for the cesarean delivery. After analyzing the model with cross-validation techniques, it was found to be both accurate and discriminating.

Conclusion: A predictive nomogram, which incorporates six variables easily ascertainable at the first prenatal visit, has been developed that allows the determination of a patient-specific chance for successful VBAC for those women who undertake trial of labor.

Level of evidence: II.