Predictive model and scoring system for delayed cerebral ischemia following aneurysmal subarachnoid hemorrhage: A ten-year prospective analysis of observational data.

Introduction: Cerebral vasospasm after aneurysmal subarachnoid hemorrhage (aSAH) may cause delayed cerebral ischemia (DCI), a major determinant of poor outcomes. Existing predictive models lack dynamic clinical assessment, limiting timely intervention. Research question: This study aimed to develop and validate a model integrating baseline and dynamic predictors of DCI. Materials and methods: We analyzed data from 524 aSAH patients admitted between 2015 and 2024. Multivariate logistic regression identified predictors of DCI, which were incorporated into the Delayed Cerebral Ischemia Scoring System (DCISS). Model performance was internally validated. Results: DCI occurred in 157 patients (30 %). Seven independent predictors were identified: age <55 years (aOR 2.24), modified Fisher grades 3-4 (aOR 2.92), pathological blood pressure (aOR 1.76), pathological body temperature (aOR 2.38), restlessness (aOR 2.10), new neurological deficit (aOR 4.85), and Glasgow Coma Scale deterioration ≥2 points (aOR 6.33). The DCISS demonstrated excellent discrimination (AUC .89, 95 % CI .86-.92) with 88.5 % sensitivity and 82.0 % specificity at a cut-off ≥8. Risk groups were defined: low (0-3 points, 3.2 % risk), moderate (4-7, 18.5 %), high (8-13, 46.7 %), and very high (≥14, 82.5 %). Higher scores correlated with unfavorable outcomes (mRS 3-6) at discharge and 3 months (p < .001). Discussion and conclusion: The DCISS, incorporating static and dynamic parameters, offers robust prediction of DCI after aSAH. By enabling continuous risk re-stratification and individualized management, it may improve early identification of high-risk patients and reduce morbidity.