To assess the effect of RAS-blocking drugs in mortality and ischemic stroke risk in patients with dementia.

RAS system overactivation has been linked to the development of Alzheimer’s disease and there is growing evidence to support a beneficial effect in cognition of RAS-blocking drugs. Furthermore, they are the most commonly prescribed medications to manage hypertension, and patients with dementia are commonly excluded in clinical trials. We aimed to provide epidemiological evidence in these patients of the effect of these drugs on ischemic stroke and mortality risk.

Our study is a nationwide cohort study of 48.711 patients prospectively registered, 2008-2015 in SveDem (national registry of incident dementia in Sweden). Survival analysis, with  cox regression models, were ran in a propensity score matched database, considering cofounding variables (age, gender, comorbidities and other drugs use), being the outcomes ischemic stroke occurrence and death. RAS-drugs-users (20.439) three years before dementia were matched to RAS-drugs-non-users (28.332). 6.189 patients were matched to 6.189 controls for the analysis in the matched cohort for all-cause mortality, and 5416  for the analysis of ischemic stroke. The exposure to the medication was calculated in cumulative DDDs (Defined Daily Dose) three years before diagnosis, to look for a potential dose-response effect.   

A 9% risk reduction in all-cause mortality risk was found among the RAS-drugs-users (HR 0,91 ( CI 95% 0,87-0,94)) in the matched cohort. A statistically significant dose-response protective effect was not confirmed. Neither was a protective effect in ischemic stroke occurrence (HR 1,02 (CI 95% 0,90-1,17)). A sub-analysis comparing ACEIs vs ARBs showed no difference in mortality or stroke risk.

A protective effect in all-cause mortality when receiving RAS-blocking drugs before dementia diagnosis was found. No effect was found for ischemic stroke risk and there were no differences on their effect in survival when comparing sub-groups (AECIs vs ARBs).