Safety of cardiac rehabilitation following cerebrovascular accident: a systematic review
Dan Yacubovich, Nicholas R. Eltman, Rebecca J. Raskin-Wish, Rayna A. Patel, Randel L. Swanson IIAbstract
Context
The brain is highly vulnerable to circulatory disruptions, including cerebrovascular accident (CVA or stroke). Stroke survivors can benefit from cardiac rehabilitation (CR).
Objectives
This systematic review evaluated the safety and efficacy of CR in stroke survivors.
Methods
PubMed and Embase databases were screened from 1993 to May 2025 for articles utilizing the Preferred Reporting Items for Systematic Reviews and Meta-Analyses (PRISMA) criteria. We included randomized controlled trials (RCTs), nonrandomized clinical trials/studies, cohort studies (prospective or retrospective), case-control studies, and large case series (≥10 participants). Two independent researchers screened articles utilizing the Participant, Interventions, Comparisons, Outcomes, and Study design (PICOS) criteria for inclusion in the review, and a third person was utilized to resolve any conflicts. Data were abstracted from selected studies to a structured spreadsheet, which included information on study characteristics, adverse events, withdrawals, and functional outcomes.
Results
A total of 2,239 records were identified through database searching, including PubMed (n=2039) and Embase (n=200). After removal of duplicates (n=180), 2059 records underwent title and abstract screening. Of these, 1986 records were excluded, and 73 studies were sought for retrieval. Of the 73 studies, 9 were not retrieved. The remaining 64 full-text articles were assessed for eligibility, of which 52 were excluded due to incorrect study design (n=28), failure to report adverse events (n=15), incorrect intervention (n=6), or incorrect patient population (n=3). A total of 12 studies were included in the review. Out of the studies included, five studies reported adverse events in ranging severity. All events were considered unrelated to the use of CR, due to lower-graded events being expected occurrences of physical exertion, whereas severe and fatal events were attributed to pre-existing comorbidities, unrelated illnesses, or external factors. Withdrawal rates and events were reported in all 12 studies, which showed that withdrawal was unrelated to CR and based more on logistical and medical events present within the patient population. All studies reported improvement in functional outcomes; however, a mix of statistically significant and nonsignificant results were observed, contingent upon the specific outcome measure evaluated.
Conclusions
Collectively, the data suggest that CR is a safe intervention for poststroke rehabilitation and has the potential to improve functional outcomes in patients. However, variability in functional outcomes highlights the need for individualized CR protocols.