External validation of a prediction model for successful external cephalic version

Area of Science:

• Obstetrics and Gynecology
• Maternal-Fetal Medicine
• Clinical Prediction Modeling

Background:

• External cephalic version (ECV) is a procedure to turn breech presentation to cephalic.
• Accurate prediction of ECV success is crucial for patient counseling and clinical decision-making.
• Existing prediction models require external validation to assess their generalizability.

Purpose of the Study:

• To externally validate a prediction model for the probability of successful ECV.
• To evaluate the model's calibration and discrimination performance.
• To develop a practical score chart for estimating ECV success probability in clinical practice.

Main Methods:

• External validation of a previously developed prediction model using a cohort of 320 women undergoing ECV.
• Assessment of model performance using calibration (underestimation/overestimation) and discrimination (area under the receiver operating characteristic curve - ROC).
• Development of a score chart for clinical application.

Main Results:

• The prediction model showed moderate discrimination (AUC 0.66; 95% CI: 0.60-0.72).
• The model generally underestimated the success rate by 4-14% but effectively distinguished between high (>50%) and low (<20%) predicted success probabilities.
• ECV was successful in 117 out of 320 women (37%).

Conclusions:

• The prediction model for ECV success is validated externally and holds predictive value.
• The developed score chart can assist clinicians in counseling and decision-making for ECV in term singleton breech presentations.
• The model supports informed choices regarding ECV procedures for breech babies.