Imaging Scores in Subarachnoid Hemorrhage: Performance on Prediction of Functional Outcome, Mortality, and...

Background/Objectives: Several imaging scores have been developed for subarachnoid hemorrhage (SAH), but their prognostic performance for long-term functional outcome and post-hospital complications remains insufficiently characterized. We evaluated whether five admission imaging scores (modified Fisher, Claassen, Hijdra, Graeb, IVH) independently predict 12-month functional outcome and major secondary endpoints. Methods: We performed a retrospective cohort study of 479 consecutive patients with atraumatic SAH recorded in a prospectively maintained institutional registry. Admission CT/MRI was scored by two board-certified neuroradiologists blinded to clinical outcomes. The primary endpoint was unfavorable functional outcome at 12 months (modified Rankin scale [mRS] 4-6). Secondary endpoints included 12-month mortality, delayed cerebral ischemia (DCI), post-hemorrhagic epilepsy, shunt-dependent hydrocephalus, return to work, and patient-reported health. Receiver operating characteristic (ROC) analyses and multivariable logistic regression adjusted for established predictors were conducted. Results: All imaging scores were significantly associated with the primary endpoint and demonstrated adequate discrimination (area under the curve [AUC] ~0.70-0.74), with the Graeb and IVH scores performing highest for long-term functional outcome, mortality, and shunt dependence. Associations with DCI and epilepsy were modest. In multivariable analyses, all imaging scores remained independently associated with mRS 4-6. Subgroup analyses showed stronger prognostic performance in good-grade SAH, aneurysmal SAH, and cases with concomitant intraventricular hemorrhage. Conclusions: Admission imaging burden independently predicts 12-month functional outcome, mortality, and shunt dependence after SAH. Incorporating IVH-oriented measures alongside established clinical grading may improve individualized risk stratification, particularly in good-grade and aneurysmal SAH.