Cardiac rehabilitation reduces risk of overall mortality within 12 months after a heart failure hospitalization: data linkage analysis of 9,023 patients in Australia

Abstract

Abstract Background Meta-analyses of randomized trials show that cardiac rehabilitation (CR) prevents cardiovascular (CV) and heart failure (HF)-specific hospitalizations. However, globally only 30-50% of eligible patients participate in CR which might affect CR effectiveness. Purpose To investigate CR utilization and clinical outcomes after an admission for HF. Methods This is a retrospective cohort study of patients >=18 years old discharged from public hospitals in South Australia with a diagnosis of HF from 2016-2021. We linked the statewide clinical CR database to hospital administrative and death databases to assess CR utilization (non-attendance vs attendance without completion vs completion of >= 70% of the sessions), the composite primary outcome (CV re-admission or CV death within 12 months after the index hospitalization) and the secondary outcome (all-cause death within 12 months after the index hospitalization). To investigate factors associated with not attending CR, we adjusted a logistic model by age, sex, Index of Relative Socioeconomic Advantage and Disadvantage (IRSAD), remoteness; admission during the COVID pandemic, acute myocardial infarction (MI), procedures, arrhythmias, Charlson comorbidities index, history of diabetes, hypertension, obesity, prior stroke, and peripheral vascular disease. To investigate the association of the primary and secondary outcomes with CR utilization, we adjusted Cox survival models by age; sex; IRSAD; remoteness; ACS; revascularization, valve and implantable devices procedures; AF; other arrhythmias; Charlson comorbidities index; history of diabetes, hypertension, obesity, prior stroke and peripheral vascular disease; remoteness. Results Of the 9023 eligible separations (52.0% male with mean age 77, SD 13.8 years), 8572 (95%) did not attend CR. Among the 451 patients commencing CR, 352 (78.0%) completed it. CABG (118; 1.3%), PCI (252; 2.8%), diabetes (3031; 33.6%), hypertension (1332; 14.8%) and MI (607; 6.7%) were common diagnoses. Women (OR1.62; 95%CI 1.31-2.02); and older age (OR1.03; 95%CI 1.02-1.04) were associated with a higher risk of not attending CR. Within 12 months of the index hospitalization, there were 4056 (45.0%) CV readmissions/CV deaths and 3486 (38.6%) deaths. After adjustment for demographic, clinical and social factors, CR attendance without completion (p=0.60) or completion (p=0.43) was not associated with CV readmission/CV death. The risk of all-cause death was lower in the groups attending without completing (HR 0.57; 95%CI 0.36-0.91; p=0.02) and completing CR (HR 0.52; 95%CI 0.39-0.71; p=0.02) compared to those not attending CR (Figure 1). Conclusion CR attendance and completion reduces the risk of death within 12 months after a HF hospitalization. Strategies focusing on increasing CR participation should target women and older people who are less likely to attend. A limitation of this study is that we could not control for the utilization of HF-specialist clinics.Survival within 12 months