Five-year mortality and related prognostic factors after inpatient stroke rehabilitation: a European multi-centre study.

OBJECTIVE To determine 5-year mortality and its association with baseline characteristics and functional status 6 months post-stroke for patients who received inpatient rehabilitation. DESIGN A prospective rehabilitation-based cohort study. SUBJECTS A total of 532 consecutive stroke patients from 4 European rehabilitation centres. METHODS Predictors were recorded on admission. Barthel Index was assessed at 6 months (BI6mths) and patients were followed for 5 years post-stroke. Survival probability was computed using Kaplan-Meier analysis and compared across 3 BI6mths-classes (0-60, 65-90, 95-100) (log-rank test). Significant independent predictors were determined using multivariate Cox regression analysis (hazard ratio (HR)). RESULTS Five-year cumulative risk of death was 29.12% (95% confidence interval (CI): 22.86-35.38). Age (HR=1.06, 95% CI: 1.04-1.09), cognitive impairment (HR=1.77, 95% CI: 1.21-2.57), diabetes mellitus (HR=1.68, 95% CI: 1.16-2.41) and atrial fibrillation (HR=1.52, 95% CI: 1.08-2.14) were independent predictors of increased mortality. Hyperlipidaemia (HR=0.66, 95% CI: 0.46-0.94), and higher BI6mths (HR=0.98, 95% CI: 0.97-0.99) were independent predictors of decreased mortality. Five-year survival probability was 0.85 (95% CI: 0.80-0.89) for patients in BI6mths-class: 95-100, 0.72 (95% CI: 0.63-0.79) in BI6mths-class: 65-90 and 0.50 (95% CI: 0.40-0.60) in BI6mths-class: 0-60 (p<0.0001). CONCLUSION Nearly one-third of rehabilitation patients died during the first 5 years following stroke. Functional status at 6 months was a powerful predictor of long-term mortality. Maximum functional independence at 6 months post-stroke should be promoted through medical interventions and rehabilitation. Future studies are recommended to evaluate the direct effect of rehabilitation on long-term survival.