Nomogram Predicting Survival in Gastroesophageal Variceal Bleeding Treated With Sclerotherapy and Cyanoacrylate

This study aimed to develop and validate a survival prediction model for patients with gastroesophageal variceal bleeding treated with combined sclerotherapy and cyanoacrylate injection.

Patient data were retrospectively collected and randomly divided into training and validation cohorts (7:3). Key prognostic variables in the training cohort were identified using random survival forest (RSF), least absolute shrinkage and selection operator (LASSO) regression, and extreme gradient boosting (XGBoost). A nomogram was developed with the Cox proportional hazards model. Kaplan-Meier (KM) curves assessed stratified survival, SHapley Additive exPlanations (SHAP) analysis interpreted variable contributions, and bootstrap resampling was used for internal validation. Model performance was evaluated by receiver operating characteristic (ROC) curves, calibration plots, and decision curve analysis (DCA).

The concordance index for 1- and 3-year survival was 0.736 and 0.735 in the training cohort and 0.847 and 0.865 in the validation cohort. Time-dependent ROC analysis showed AUCs of 0.736 (95% CI: 0.648-0.824) and 0.735 (95% CI: 0.657-0.813) in the training cohort and 0.869 (95% CI: 0.794-0.944) and 0.879 (95% CI: 0.812-0.945) in the validation cohort. Calibration curves demonstrated good agreement, and DCA confirmed substantial net clinical benefit across threshold probabilities ranging from 0.2 to 0.6.

The prognostic nomogram integrating age, MELD score, and the risk of rebleeding after combined endoscopic therapy demonstrated good clinical applicability and may provide a valuable reference for individualized treatment and follow-up management.