EXPRESS: Brain Age Gap from Non-Contrast CT Is Independently Associated with Functional Outcome after Acute Ischemic Stroke.
Authors
Affiliations (5)
Affiliations (5)
- Department of Biomedical Engineering, University of Calgary, Calgary, Alberta, Canada.
- Hotchkiss Brain Institute, Calgary, Alberta, Canada.
- Department of Clinical Neuroscience, Cumming School of Medicine, University of Calgary, Calgary, Alberta, Canada.
- Division of Neurology, Department of Medicine, University of Alberta, Edmonton, Alberta, Canada.
- Department of Electrical and Software Engineering, University of Calgary, Calgary, Alberta, Canada.
Abstract
BackgroundOutcomes after acute ischemic stroke (AIS) vary beyond lesion burden and clinical severity, suggesting a role for baseline structural brain vulnerability. We evaluated whether brain age gap (BrainAGE) derived from routine non-contrast computed tomography (NCCT) is associated with functional outcome after AIS.MethodsWe analyzed 1,173 AIS patients from a prospective multicenter cohort. A deep learning model estimated brain age from contralesional baseline NCCT to minimize lesion-related confounding. BrainAGE was defined as predicted minus chronological age. The primary outcome was 90-day functional status, assessed using proportional odds regression adjusted for age, sex, baseline stroke severity, early ischemic changes, and reperfusion therapy.ResultsBrainAGE was higher in AIS patients than in controls (+3.1 ± 9.1 years; p<0.001). Higher BrainAGE was associated with worse functional outcome after adjustment for age, sex, and other clinical predictors (POR 1.091 per 5 years; 95% CI 1.023-1.162; p=0.008). BrainAGE improved model fit but modestly improved discrimination.ConclusionsBrainAGE from routine NCCT provided modest incremental prognostic information beyond age and conventional clinical predictors. These findings support CT-based BrainAGE as a marker of brain health complementary to established prognostic factors. Whether BrainAGE improves patient-level risk stratification or clinical decision-making requires independent external validation and prospective evaluation.