Exercise training workloads in cardiac rehabilitation are associated with clinical outcomes in patients with heart failure

Background In patients with coronary heart disease, the exercise workload (i.e., metabolic equivalents of task, METs) at which patients exercise train upon entry and completion of cardiac rehabilitation (CR) are independently related to prognosis. Unknown is the association between exercise training workloads in CR and clinical outcomes in patients with heart failure (HF). Methods Patients with HF who participated in an early outpatient CR program were used in this retrospective analysis. Exercise workloads upon entry and completion of CR were converted to METs. The primary outcome was all-cause mortality and the secondary outcome was HF hospitalization. Cox regression analysis was used to assess the adjusted risk between MET levels in CR and clinical outcomes. Results Among 707 patients, the median exercise training workload at the start and end of CR was 2.5 METs (IQR 2.1 to 3.1 METs) and 3.2 METS (IQR 2.7 to 4.1 METs), respectively, for men and 2.2 METs (IQR 1.9 to 2.6 METs) and 2.9 METS (IQR 2.3 to 3.4 METs), respectively, for women. There were 242 deaths and 266 HF hospitalizations. METs achieved at the end of CR had the strongest independent association with all-cause mortality (adjusted HR, 95% CI: 0.58, 0.48-0.70) and HF hospitalization (adjusted HR, 95% CI: 0.62, 0.52-0.74). Each 1 MET higher work load at the end of CR was associated with a 42% and 38% lower adjusted risk for all-cause mortality and HF hospitalization, respectively. Conclusions In a diverse cohort of patients with chronic HF our data suggests that an easily accessible measure of exercise capacity (i.e., METs) that is collected during CR is independently associated with the adjusted risk for both all-cause mortality and HF-specific hospitalization. Training at MET levels <3.5 METs identifies patients that might benefit from closer clinical surveillance and reinforced adherence to medical and lifestyle preventive strategies.