Patient-centered wearable-integrated rehabilitation program for patients with heart failure: a multicenter randomized controlled trial

Abstract Introduction Exercise training is a key therapeutic component for patients with heart failure (HF). However, participation in and adherence to home-based exercise regimens are frequently inadequate. To address this gap, we have developed a motivational exercise support program that integrates wearable devices with real-time activity monitoring, interactive feedback, and algorithm-based goal setting (Figure 1). Purpose This study aimed to evaluate the efficacy and safety of the exercise app in patients with HF. Methods In this investigator-initiated, open-label, multicenter, randomized controlled trial conducted at three university hospitals in Japan, patients with HF were randomly assigned (1:1) to either the app-based exercise support program or standard exercise prescriptions between October 2022 and January 2025. Over 24 weeks, the app group performed resistance training using video content within the app three days per week and maintained 2000–3000 steps per day at an intensity greater than their target heart rate. The app incorporated multiple educational components to facilitate sustained self-management. Exercise guidance was primarily provided by physicians and therapists in both groups. The primary endpoint was the change in peak oxygen uptake (VO2) at 12 weeks; secondary endpoints included changes in peak VO2 at 24 weeks and skeletal muscle strength. Analyses were conducted in the full analysis set, with a prespecified per-protocol set defined as ≥30 days of app use within 12 weeks. Results are reported as two-tailed 95% confidence intervals (CIs) and corresponding p-values. Informed consent was obtained from all the participants before their participation. Results A total of 100 patients (median age: 63 years; 13 females; all NYHA I–II) were randomized into two groups: an app group (n = 49) and a control group (n = 51). The between-group differences in changes in peak VO2 at 12 and 24 weeks were 0.23 mL/kg/min (95% CI, −0.56 to 1.02; p=0.561) and 0.72 mL/kg/min (95% CI, −0.15 to 1.59; p=0.103), respectively (Figure 2). In contrast, significant and progressive improvements were observed at 12 and 24 weeks in grip strength in both hands and bilateral knee extension muscle strength. No serious adverse events related to the intervention were observed. In the per-protocol set, which included 42 app users and 50 non-users, the corresponding differences were 0.28 mL/kg/min (95% CI: −0.55 to 1.11; p = 0.509) and 0.93 mL/kg/min (95% CI: 0.02 to 1.84; p = 0.045). Conclusion Among patients with HF, the exercise app users did not significantly enhance exercise tolerance compared with non-users. However, this trial highlights the potential utility of wearable device-assisted rehabilitation strategies in improving physical function, an important determinant of HF prognosis.For image description, please refer to the figure legend and surrounding text. For image description, please refer to the figure legend and surrounding text.