Location-Specific Hematoma Volume Predicts Early Neurological Deterioration in Supratentorial ICH

Early neurological deterioration (END) adversely affects outcomes in patients with intracerebral hemorrhage (ICH). This study aimed to determine the location-specific hematoma volumes for END in supratentorial ICH patients.

We retrospectively analyzed supratentorial ICH patients presenting from two prospective cohorts. END was defined as a ≥ 4-point increase in the National Institutes of Health Stroke Scale (NIHSS) score or a ≥ 2-point decrease in the Glasgow Coma Scale score within 24 h of admission. The training cohort was used to determine location-specific hematoma volume cutoffs for END and to develop the location-specific hematoma volume for early neurological deterioration (LIVED) score. Internal validation used 30% of the cohort, with external validation in a separate cohort.

A total of 1199 patients with supratentorial ICH were included, divided into training (n = 633), internal validation (n = 272), and external validation (n = 294) cohorts. Hematoma volume thresholds for END were 21 mL for basal ganglia, 12 mL for thalamus, and 32 mL for lobar hemorrhages. Multivariable logistic regression identified location-specific hematoma volume, right-sided ICH, prior ischemic stroke, and NIHSS score as independent predictors, forming the LIVED score (range 0-5 points). The LIVED score showed superior predictive performance for END compared with established ICH scores, with the highest area under the curve (AUC) across cohorts. Additionally, it exhibited strong discrimination for 3-month outcomes, including functional independence, poor outcomes, and mortality, with AUCs > 0.70 in all cohorts.

Location-specific hematoma volume thresholds independently predicted END, and the LIVED score demonstrated reliable performance for risk stratification in supratentorial ICH.