Introduction: The role of cardiac rehabilitation (CR) in the premature coronary artery disease (premCAD) population has not been well evaluated. We sought to evaluate predictors of CR referral and completion and the association between CR and survival in subjects with premCAD.
Methods: The Total Cardiology-Rehabilitation and Alberta Provincial Project for Outcomes Assessment in Coronary Heart Disease (APPROACH) databases were used to obtain data on all subjects in Calgary, Canada with premCAD (>/=1 vessel CAD on angiogram in women <65 and men <55 years) between 1996-2016. Logistic regression models were constructed to assess predictors of CR referral and completion. A Cox proportional hazard model evaluated the associations between CR status (not referred; referred, did not complete; completed CR) and survival. All models were adjusted for available demographics, disease severity, and co-morbidities.
Results: A total of 11,119 subjects (3,608 women; median age men, 49.7; women, 56.9 years) with premCAD were identified. Nearly two-thirds of subjects (7,071, 63.1%) were referred and 3,982 (56.3% of those referred; 35.5% of the total cohort) completed CR. Men were more likely than women to be referred (67.0% vs 55.0%, p<0.0001) and to complete CR if referred (58.6% vs 50.5%, p<0.0001). In multivariable models, advancing age in years predicted non-referral (OR 0.97, 95%CI 0.96, 0.97); having diabetes (OR 0.65, 95%CI 0.57, 0.73) and current smoking (OR 0.48, 95%CI 0.43, 0.53) predicted CR non-completion. Median follow-up was 9.5 years (IQR 5.2, 14.4) and there were 1,329 deaths. In a fully adjusted model, CR completion predicted better survival (HR 0.48, 95%CI 0.41, 0.56) but referral alone did not (HR 0.87, 95%CI 0.74, 1.01).
Conclusions: Only about one third of eligible premCAD subjects complete CR. Given the significant survival benefit associated with CR completion, efforts to ensure referral to and completion of CR must be undertaken to optimize long-term outcomes.