Background and purpose Vascular structural abnormality-related intracerebral haemorrhage (VSARICH) accounts for 10–20% of cases of intracerebral haemorrhage (ICH), but none of the grading scales for primary ICH are reliable for VSARICH. This study aimed to propose a grading scale based on clinical and anatomical parameters to predict short-term clinical outcome.

Methods Data were prospectively collected from patients with ICH recruited consecutively from 50 secondary and tertiary hospitals in China. Demographic and clinicopathological factors associated with mortality and good clinical outcome were identified and used to develop a grading scale for VSARICH.

Results The VSARICH was 10.8% and 13% in the derivation (n=335) and validation (n=109) cohorts, respectively. Data from 307 patients with VSARICH in the derivation cohort were used to generate a VSARICH score (VSARICHS) system ranging from 0 to 9. Points were assigned based on the Glasgow Coma Scale (GCS) score on admission (GCS 3–4=4 points; 5–12=2 points; 13–15=0 points), age (≥80 years=2 points; 79–60=1 point; ≤59=0 points), presence of subarachnoid haemorrhage (yes=1 point; no=0 points) and presence of herniation (yes=2 points; no=0 points). VSARICHS showed good discrimination in the derivation cohort (area under the receiver operating characteristic curves, AUCs)AUCs 0.837 for good clinical outcome; 0.942 for mortality) and validation cohort (AUCs 0.813 for good clinical outcome; 0.930 for mortality).

Conclusions VSARICHS appears to be a reliable clinical scoring system that may prove useful for guiding risk stratification, clinical treatment and research.