Home- versus clinic-based exercise intervention in patients with heart failure (EXIT-HF trial): a pragmatic randomized controlled trial

Abstract

Abstract Background The limited accessibility partially explains the low uptake of cardiac rehabilitation (ExCR) for heart failure (HF). Home-based programs might address those obstacles but few data from randomized trials are available. Purpose To compare the effectiveness of a home-based versus clinic-based ExCR intervention in HF patients. Methods Single-center, single-blind, parallel groups, non-inferiority pragmatic randomized control trial (1:1 and after COVID-19 2:1 home- vs clinic-based). Adult HF patients referred to our cardiac rehabilitation (CR) unit were randomized to either a clinic-based (24 supervised sessions at a CR unit of a tertiary hospital over 12 weeks) or home-based (4 supervised ExT sessions to familiarize with the training protocol and the remaining sessions continued at home) over 12 weeks. Exercise training protocol consisted of a combination of endurance (at 60%-80% of peak oxygen uptake [VO2peak] and resistance training (elastic bands). The primary endpoint was the change in VO2peak at the end of the 12-week program. Secondary outcomes included adherence (>80% of sessions attendance) and change of Minnesota Living with HF Questionnaire (MLHFQ) scores. Results 120 patients (mean age 64±11 years; 64% men) were randomized: 44 to clinic- and to 76 home-based intervention. 41% had an ischemic cardiomyopathy, 27% had an ICD/CRT, the mean LVEF was 36±11% (62% had LVEF <40%), 73% were on NYHA class II, and the median NT-proBNP was 361 [IQR 170-1008] ng/mL. Intervention groups were well balanced, including basal VO2peak that was similar between groups (17.8±4.8 vs 18.1±5.2 mL/min/Kg; p=0.75). Adherence was similar between groups (87.5% vs 82.5%; p=0.50). At week 12, on average, patients increased VO2 peak by 0.9±2.5 mL/min/Kg compared to baseline (p<0.001). No significant differences were found between clinic- and home-based interventions (p=0.25). Consistently, the MLHFQ total score decreased significantly overall (-10.3±1.7 points; p<0.001), with no significant differences between groups (p=0.91). The drop-out rate was 14% and 17% in clinic- and home-based groups, respectively. Conclusions The clinical impact of home-based CR program on peakVO2 and health-related quality of life of HF patients was similar to a standard supervised clinic-based one. Our data support the home-based program as an effective alternative way of delivering CR in HF.