Abstract WP216: MRI Biomarkers Improve Prognostic Accuracy Beyond the ICH Score in Acute Intracerebral Hemorrhage

Background: The ICH Score is the standard early prognostic tool in spontaneous intracerebral hemorrhage (ICH) but does not incorporate MRI markers of small vessel disease or secondary ischemic injury. Cerebral microbleeds (CMBs), visible on GRE/SWI MRI and reflecting chronic vascular injury, and acute diffusion-weighted imaging (DWI) lesions may improve prediction of long-term outcomes after ICH. Whether inclusion of these imaging biomarkers can improve functional outcome prediction beyond the ICH Score in acute ICH has not been fully defined. Objective: To determine whether CMB burden and DWI lesion presence improve 90-day functional outcome prediction beyond the ICH Score. Methods: We retrospectively identified adults with spontaneous ICH who underwent MRI including DWI and GRE/SWI sequences, performed either after minimally invasive surgery (MIS) in surgically evacuated patients or post-onsent in nonsurgical patients. DWI lesion presence was recorded. Cerebral microbleeds were visually counted for total burden by two blinded raters. The primary outcome was 90-day modified Rankin Scale (mRS). Logistic regression compared the ICH Score, MRI biomarkers, and combined models, with discrimination assessed by AUC. Results: A total of 102 patients (mean age 59.7 ± 12.9 years, mean admission GCS 10.1 ± 3.9, 59% male) with spontaneous ICH were included. The ICH Score alone predicted poor 90-day functional outcome (mRS > 2) with an AUC of 0.76. A combined model incorporating the ICH Score and MRI biomarkers achieved an AUC of 0.83, representing a statistically significant improvement in discrimination compared with the ICH Score alone (ΔAUC = 0.07, p = 0.049, DeLong test). These findings indicate that MRI biomarkers provide added prognostic value beyond the standard clinical score. Conclusions: CMB burden and DWI lesion presence add significant prognostic value beyond the ICH Score in acute ICH. Incorporating these MRI markers may improve individualized risk stratification and outcome prediction.