(1210) Nomogram for Predicting Shoulder Dystocia After Term Vaginal Delivery

To develop a predictive model and nomogram for shoulder dystocia following term vaginal delivery.

Study Design:

This retrospective cohort study was conducted at a single tertiary care center. We included all vaginal deliveries at ≥37 weeks of gestation (either normal vaginal delivery [NVD] or vacuum extraction [VE]) between July 2012 and December 2023. Shoulder dystocia was the primary outcome and was defined as a ≥1-minute delay in the delivery of the shoulders following the head, or the requirement of specific maneuvers. The full multivariate logistic regression included variables such as age, BMI, parity, diabetes (pregestational and gestational), gestational age, mode of delivery, clinical fetal weight estimation, oxytocin use, second-stage duration, neuraxial analgesia and newborn gender. The stepwise model, selected based on the Akaike Information Criterion, retained the variables parity, clinical weight estimation, mode of delivery (NVD or VE), gestational diabetes mellitus A2 (GDM-A2), and the pregestational diabetes. A nomogram was constructed based on the final logistic regression model. Internal validation and calibration were performed.

Results:

Out of 74,146 eligible term vaginal deliveries, 278 cases (0.4%) of shoulder dystocia were identified. The final logistic regression model showed that increased parity (p < 0.001), higher clinical fetal weight estimation (p < 0.001), and vacuum-assisted delivery (p < 0.001) were independently associated with higher risk of shoulder dystocia. The presence of GDM-A2 and pregestational diabetes showed positive trends but did not reach statistical significance. Their respective predicted probabilities are represented in the nomogram (Figure 1).

Conclusion:

This nomogram provides an individualized tool for estimating the risk of shoulder dystocia in term vaginal deliveries. Its use may support clinical decision-making and intrapartum planning.