Poststroke dementia predicts poor survival in long-term follow-up: influence of prestroke cognitive decline and previous stroke.

BACKGROUND: The aim of this study was to investigate the influence of poststroke dementia on long-term survival after acute stroke and also to assess the possible influence of prestroke cognitive decline and previous stroke on this relationship. METHODS: A total of 451 consecutive patients with acute ischaemic stroke admitted to hospital were included in the study and followed up for 12 years. Dementia was diagnosed 3 months after stroke in 115 patients (25.5%). RESULTS: In Kaplan-Meier analysis, poststroke dementia predicted poor long-term survival (5.1 years vs 8.8 years in patients who did not have poststroke dementia; p<0.001). Prestroke cognitive decline had a negative influence on survival in patients with poststroke dementia (3.8 years vs 5.8 years; p<0.001); however, previous stroke did not affect survival in these patients (p = 0.676). In stepwise Cox regression proportional hazards analysis adjusted for significant covariates, poststroke dementia (hazard ratio (HR) 1.53; p = 0.003), advanced age (HR 1.07; p<0.001), severity of stroke (HR 1.91; p<0.001), smoking (HR 1.35; p = 0.035), cardiac failure (HR 1.61; p = 0.003) and atrial fibrillation (HR 1.89; p = 0.035) were all independent predictors of poor long-term survival. Poststroke dementia (HR 2.33; p<0.001), advanced age (HR 1.07; p<0.001) and poor Rankin score (HR 2.15; p = 0.001) were associated with death from brain-related causes, including infarction, haemorrhage and dementia. CONCLUSIONS: Long-term follow-up of our large well-defined poststroke cohort indicated that in patients with acute stroke, dementia is a significant predictor of poor long-term survival and death from brain-associated causes. Prestroke cognitive decline seems to have an additional negative influence on survival, but previous stroke does not seem to affect survival.