Dual-center development and validation of a LDL-C and TC based nomogram for preoperative hemorrhagic risk in adult moyamoya disease.

INTRODUCTION

Moyamoya disease (MMD) is a chronic cerebrovascular disease characterized by progressive stenosis of intracranial arteries and compensatory formation of abnormal collateral vessels. Preoperative intracranial hemorrhage represents a severe and catastrophic complication in adult patients with MMD, while reliable strategies for precise risk stratification are still lacking to date. Furthermore, the potential association of lipid metabolic indices with preoperative hemorrhagic risk in MMD remains poorly clarified and insufficiently explored.

METHODS

A retrospective multicenter study was conducted enrolling 260 adult MMD patients recruited from two medical centers during 2015-2025. All participants were categorized into the hemorrhage group (n = 57) and non-hemorrhage group (n = 203). Univariate analysis, LASSO regression with 10-fold cross-validation, and multivariate logistic regression were sequentially performed to screen and determine independent risk factors. A predictive nomogram was further constructed, and its predictive performance was comprehensively validated via receiver operating characteristic (ROC) curve, calibration curve, decision curve analysis (DCA), and clinical impact curve (CIC).

RESULTS

Elevated LDL-C was confirmed as an independent risk factor for preoperative hemorrhage (OR = 2.46, 95%CI: 1.41-4.28, p = 0.002). Restricted cubic spline regression demonstrated a linearly positive correlation between LDL-C, TC and preoperative hemorrhagic risk, with optimal cut-off values of 3.5 mmol/L and 5.0 mmol/L, respectively. The ROC showed moderate discriminative value with an AUC of 0.78 (95%CI: 0.72-0.85), accompanied by satisfactory sensitivity, specificity and favorable clinical net benefit.

CONCLUSION

Elevated LDL-C is independently correlated with the occurrence of preoperative hemorrhage in adult MMD patients. The nomogram based on LDL-C and TC serves as a concise and clinically practical tool for moderate-risk stratification of preoperative hemorrhage, which is conducive to facilitating individualized clinical management and therapeutic decision-making for MMD.