Cardiac Rehabilitation Programs for Chronic Heart Disease: A Bayesian Network Meta-analysis

Background: Cardiac rehabilitation is a medically supervised program after coronary events that involves exercise and dietary modification. We evaluated the comparative benefits and harms of cardiac rehabilitation strategies via a network meta-analysis. Methods: We followed a pre-specified protocol (PROSPERO: CRD42018094998). We searched Embase, MEDLINE, and Cochrane Central Register of Randomized Trials databases for randomized controlled trials that evaluated cardiac rehabilitation vs a second form of rehabilitation or standard/usual care in adults after myocardial infarction, coronary artery bypass grafting, percutaneous coronary intervention, or angiography. Risk of bias and evidence quality was evaluated using the Cochrane tool and Grading of Recommendations Assessment, Development and Evaluation (GRADE), respectively. Pairwise and Bayesian network meta-analyses were performed for 11 clinical outcomes. Results: We included 134 randomized controlled trials involving 62,322 participants. Compared with standard care, exercise-only cardiac rehabilitation reduced the odds of cardiovascular mortality (odds ratio [OR], 0.70; 95% credibility interval [CrI], 0.51-0.96; moderate-quality evidence), major adverse cardiovascular events (OR, 0.57; 95% CrI, 0.40-0.78; low-quality evidence), nonfatal myocardial infarction (OR, 0.71; 95% CrI, 0.54-0.93; moderate-quality evidence), all-cause hospitalization (OR, 0.74; 95% CrI, 0.54-0.98; moderate-quality evidence), and cardiovascular hospitalization (OR, 0.69; 95% CrI, 0.51-0.88; moderate-quality evidence). Exercise-only cardiac rehabilitation was associated with lower cardiovascular hospitalization risk relative to cardiac rehabilitation without exercise (OR, 0.68; 95% CrI, 0.48-0.97; moderate-quality evidence). Conclusions: Cardiac rehabilitation programs containing exercise might provide broader cardiovascular benefits compared with those without exercise.