Purpose
To construct and validate a nomogram to predict the risk of adverse events (intraoperative massive hemorrhage or retained products of conception) during the termination of Cesarean scar pregnancy (CSP).
Method
Data from patients diagnosed with CSP who underwent Dilation and Curettage (D&C) at two hospitals were retrospectively collected. This data formed both internal and external cohorts for analysis. The internal cohort was split randomly, with 70% of the data allocated to a training set and 30% to an internal validation set. The external cohort was used exclusively as the external validation set. LASSO and logistic regression were utilized to select variables and construct a nomogram. The nomogram's performance was assessed using various methods including C-index, calibration curve, decision curve analysis (DCA), and clinical impact curve analysis (CICA) to evaluate its ability for identification, calibration, and clinical effectiveness.
Results
The prediction nomogram incorporated several predictors, including scar thickness, type of CSP, gestational sac diameter, and blood flow. This nomogram exhibited strong discrimination, as evidenced by a C-index of 0.829 (95% confidence interval: 0.770–0.887). Moreover, even in the interval validation set, a high C-index value of 0.784 was achieved, and in the external validation set, it reached 0.833. Further assessment through calibration curve analysis, DCA, and CICA revealed a robust agreement between the nomogram's predictions and actual observations, underscoring its utility and reliability.
Conclusion
The validated nomogram effectively predicts adverse events in CSP, showing good discrimination and calibration, making it useful in clinical settings.