Background: Older age, larger hematoma volume, and intraventricular hemorrhage (IVH) are established predictors of poor prognosis after intracerebral hemorrhage (ICH). We hypothesized that higher cerebral small vessel disease (CSVD) burden substantially affects functional outcome and its detailed characterization in addition to already established predictors will enhance our prognostic accuracy.
Methods: We studied consecutive patients with spontaneous supratentorial ICH, functionally independent at baseline (modified Rankin Scale (mRS) ≤2), with brain MRI within 1 month of ICH and known 3-month functional status. We collected clinical and imaging data and data on: Cerebral microbleeds (CMBs), enlarged perivascular spaces (EPVS) severity, white matter disease (WMD) (Fazekas scale), chronic lacunar and non-lacunar infarcts, cerebral atrophy. Functional endpoints were (a) 3-month functional independence (mRS ≤2) (b) independent ambulation at 3 months. Hematoma expansion was defined as relative increase of ≥ 33% or absolute increase of >6ml.
Results: 111 patients (70.7±13.9 years, 40% female) met our inclusion criteria. 43(38%) achieved functional independence,71(64%) independent ambulation. In univariate analyses, older age, hematoma volume, IVH, cerebral atrophy, severe WMD but not EPVS severity or chronic infarcts were associated with worse functional and ambulatory outcome. CMBs presence was associated with poor functional outcome but not ambulation status. In multivariable logistic regression, volume, IVH, CMBs (OR 3.95, CI: 1.4-12.1, p=0.009), brain atrophy (OR 3.99, CI: 1.1-16.2 p=0.03) but not age or WMD severity remained associated with functional independence (model p<0.0001, ROC=0.92). Hematoma volume and brain atrophy but not age, IVH or WMD severity were associated with independent ambulation. We found no association between CSVD markers and hematoma expansion.
Conclusion: CMBs presence and cerebral atrophy conferred a 4-fold higher risk of functional dependence after ICH. The effect of hematoma size remained very robust but addition of CSVD markers and brain atrophy neutralized the previously observed effect of age. This observation merits further exploration and validation in a larger, prospective cohort.