Exercise-based cardiac rehabilitation after stroke and the risk of incident dementia: an observational study from a global federated network

Background Dementia is a common and disabling complication after stroke. Exercise-based cardiac rehabilitation (EXCR) improves vascular risk factors, physical fitness, and systemic inflammation factors that may be relevant to post-stroke cognitive decline. Purpose To investigate the association between initiation of EXCR after stroke and the subsequent risk of developing incident dementia. Methods We used TriNetX, a global federated electronic health record (EHR) network, to conduct a retrospective observational cohort study of adults with ischemic and/or hemorrhagic stroke and baseline NIHSS ≤16. Patients were grouped according to initiation of EXCR between 1 day and 6 months after the index stroke and were propensity score–matched. The primary outcome was incident composite dementia (Alzheimer’s disease, vascular dementia, or unspecified dementia). Secondary outcomes included dementia subtypes, all-cause mortality, stroke recurrence, rehospitalization, and acute myocardial infarction (AMI), evaluated using time-to-event analyses over 5 years (primary analysis) and 3 years (sensitivity analysis). To mitigate potential reverse causation, we also conducted a 1-year lagged analysis excluding events occurring within the first year after the index date; the association with incident dementia remained statistically significant. Results The study included 20,200 stroke patients who underwent EXCR (mean age 67.0 ± 13.5 years; 54.9% male) and 1,706,668 stroke patients who did not undergo EXCR (mean age 66.5 ± 15.3 years; 48.4% male). After propensity score matching (PSM), EXCR was associated with a lower 5-year risk of incident dementia (HR 0.68; 95% CI 0.64–0.72) and lower all-cause mortality (HR 0.52; 95% CI 0.50–0.54). EXCR was also associated with reduced stroke recurrence (HR 0.51; 95% CI 0.46–0.56) and fewer rehospitalizations (HR 0.75; 95% CI 0.73–0.77). In a 1-year lagged analysis excluding events occurring within the first year after the index date, the association with incident dementia remained significant (HR 0.85; 95% CI 0.79–0.92). Results from the 3-year follow-up analysis were consistent with the primary findings. Conclusions EXCR after stroke was associated with a lower 5-year risk of incident dementia and with lower mortality and recurrent stroke. Results were consistent at 3 years. Prospective studies are warranted to confirm these findings and clarify mechanisms.