A Nomogram Based on Ultrasound Features for Predicting Major Intra-Operative Hemorrhage in Patients With Placenta Accreta Spectrum (PAS).

OBJECTIVE

To develop a nomogram for predicting major intra-operative hemorrhage in patients with placenta accreta spectrum (PAS) based on ultrasound features, aiming to guide clinical decision-making in hemorrhage prevention and management.

METHODS

This retrospective cohort study enrolled patients diagnosed with PAS who underwent pre-operative ultrasound evaluation at Fujian Maternity and Child Health Hospital. Participants were randomly allocated to training and validation cohorts in a 7:3 ratio. The training cohort was stratified into the non-major hemorrhage group (n = 77) and major hemorrhage group (n = 61). univariate and multi-variate logistic regression analyses were performed to identify independent risk factors for major intra-operative hemorrhage and to develop a nomogram, and its predictive performance was evaluated using receiver operating characteristic (ROC) curves, calibration curves and Decision Curve Analysis (DCA).

RESULTS

A total of 196 patients were included. Multi-variable logistic regression analysis identified four independent predictors of major intra-operative hemorrhage: placental lacunae feeder vessels , sub-placental hypervascularity/bridging vessels, intra-cervical lakes, and incomplete cervical morphology. These predictors were incorporated into a nomogram. The model demonstrated discriminative capacity with areas under the ROC curves values of 0.877 (95% confidence interval (CI) 0.819-0.935) and 0.851 (95% CI 0.754-0.948) in training and validation cohorts, respectively. A total of 1000 bootstrap-resampling confirmed model stability. Calibration curves showed good consistency between predicted and observed probabilities, and the DCA demonstrated clinical utility.

CONCLUSION

This study developed and validated a nomogram for predicting intra-operative major hemorrhage in patients with PAS, which has clinically applicable performance and robustness, providing actionable perioperative decision-making.