External validation of a prediction model for successful external cephalic version

Abstract

We sought external validation of a prediction model for the probability of a successful external cephalic version (ECV). We evaluated the performance of the prediction model with calibration and discrimination. For clinical practice, we developed a score chart to calculate the probability of a successful ECV. We studied 320 women undergoing ECV, of which 117 (37%) were successful. The model underestimated the success rate by 4 to 14%. The area under the receiver operating characteristic curve was moderate (0.66; 95% confidence interval: 0.60 to 0.72), but the model was able to make good distinction between women with a higher predicted probability of a successful ECV (more than 50%) compared with women with a lower predicted probability of a successful ECV (less than 20%). Our model to predict the outcome of ECV holds in external validation. It can be used to support patient counseling and decision making for ECV in singleton term breech presentations.