Optimal Surgical Timing and Outcome Prediction in Hemorrhagic Moyamoya Disease: A Retrospective Cohort Study

Objective To develop and validate a prognostic nomogram incorporating optimal surgical timing thresholds for predicting long‐term outcomes in adult hemorrhagic moyamoya disease (MMD) patients undergoing encephaloduroarteriosynangiosis (EDAS). Methods We conducted a retrospective cohort study of 256 consecutive adults with hemorrhagic MMD treated with EDAS (2013–2017) at a tertiary neurosurgical center. Using least absolute shrinkage and selection operator (LASSO) regression with 10‐fold cross‐validation, we identified key predictors for nomogram development. The model's performance was rigorously evaluated through discrimination (C‐index, AUC‐ROC), calibration, and decision curve analysis (DCA) in both training (70%) and validation (30%) cohorts. Results Our hemorrhage‐specific nomogram incorporated five independent predictors: (1) age at initial hemorrhage, (2) surgical timing (optimal 3–6 month window post‐ictus), (3) preoperative functional status (modified Rankin Scale), (4) perioperative complications, and (5) recurrent hemorrhage. The model demonstrated excellent discrimination (C‐index: 0.889 training, 0.819 validation) and calibration. DCA confirmed superior clinical utility across 15%–80% probability thresholds compared to conventional approaches. Restricted cubic spline analysis validated the non‐linear relationship between surgical timing and outcomes. Conclusions This study provides the first validated clinical decision tool for adult hemorrhagic MMD, establishing 3–6 months post‐hemorrhage as the optimal surgical window for EDAS. The nomogram's robust performance metrics and RCS‐validated temporal risk stratification offer neurosurgeons an evidence‐based approach to optimize revascularization timing and improve functional outcomes.