Effectiveness, safety, costs, and delivery considerations of exercise-based cardiac rehabilitation in low- and middle-income countries: A systematic review and meta-analysis of randomised controlled trials.

INTRODUCTION: Although cardiac rehabilitation (CR) is recognised as a cornerstone of heart disease management, its availability in low- and middle-income countries (LMICs) remains limited, highlighting a critical evidence gap. To address this, we conducted an updated systematic review and meta-analysis of randomised controlled trials (RCTs) evaluating exercise-based CR in LMICs. METHODS: We initially identified RCTs of exercise-based CR programmes in LMICs from a previously published systematic review and updated with searches of bibliographic databases from May 2020 through to June 2025. Study quality was assessed using the Cochrane risk of bias tool. Mortality, hospitalisation, exercise capacity, health-related quality of life (HRQoL), and mental health outcomes were pooled using random-effects meta-analyses. Safety, cost, and cost-effectiveness data were narratively synthesised. Meta-regression was used to explore potential effect modification and the certainty of evidence assessed using the Grading of Recommendations Assessment, Development and Evaluation (GRADE) framework. RESULTS: Twenty-nine RCTs (6,682 participants) from South Asia (10 studies), East Asia & Pacific (7 studies), Europe & Central Asia (6 studies), Latin America & Caribbean (4 studies), and Middle East & North Africa (2 studies) were included; overall risk of bias was moderate to high. Trials varied in their patient populations, CR interventions, and delivery modes. Pooled analysis showed that exercise-based CR was associated with a moderate improvement in exercise capacity (standardised mean difference: 0.71, 95% CI: 0.47 to 0.96) compared to control. Despite high levels of statistical heterogeneity (I2: 72%), there was no evidence of effect modification by study-level factors, including CR delivery model, exercise dose, duration of follow-up, or risk of bias. CR participation was associated with improvements in some HRQoL measures (SF-12/36 physical component: mean difference 5.0, 95% CI: 0.5 to 9.4) and mental health outcomes (HADS-Anxiety: mean difference 1.19, 95% CI: 0.97 to 1.42; HADS-Depression: mean difference 1.03, 95% CI: 0.20 to 1.85); however, other measures (PHQ-9, GAD-7) showed no consistent effect. Few trials reported data on mortality, hospitalisation, safety, costs, or provided details of CR programme delivery. CONCLUSIONS: This updated review indicates that exercise-based CR in LMICs improves exercise capacity and HRQoL consistent with evidence from high-income settings. However, due to substantial statistical heterogeneity, risk of bias, and inconsistent data collection, the overall certainty of these findings is low. High-quality RCTs that collect patient-relevant outcomes from CR programmes appropriately designed for the low-resource and cultural contexts of LMICs remain a strategic priority.