Second-by-second monitoring of SpO2 by wearable device during showering identifies patients at risk for worsening heart failure

Abstract Introduction Assessment of exercise tolerance is crucial for predicting heart failure (HF) prognosis. While tests like the 6-minute walk test (6MWT) assess submaximal exercise load and provide prognostic information (1-3), their standardized conditions are difficult to replicate at home. Given the need for a reproducible load test, we explored whether routine activities could serve as an alternative. Showering, a common activity with a moderate load (2–3 METs) (4), often causes dyspnoea in HF patients. We investigated whether continuous SpO₂ monitoring during showering with a waterproof pulse oximeter could provide a reliable home-based assessment. Purpose This study aimed to examine the association between continuous SpO₂ changes during showering (ΔSpO₂-shw) and HF prognosis, and whether ΔSpO₂-shw could serve as a marker for predicting HF exacerbation. Methods In this prospective observational study, 65 patients hospitalized with acute HF were enrolled. Additionally, 12 non-HF patients admitted for pacemaker implantation or battery replacement served as controls. Prior to discharge, all participants underwent 24-hour continuous SpO₂ monitoring using a waterproof pulse oximeter (ATP-W03, Fukuda, Japan), which recorded SpO₂ every second during rest, showering, and the 6MWT. ΔSpO₂-shw was calculated by subtracting the average SpO₂ during showering from resting SpO₂. HF patients were followed for one year to assess a composite outcome of HF hospitalization or cardiovascular death. Results Among HF patients, the mean age was 79.2 years; 49% were female; the mean ejection fraction was 47.2%; and the median NT-proBNP level was 2298 pg/ml. The average SpO₂ values were 95.7% at rest, 92.9% during the 6MWT, and 91.4% during showering. Both showering and 6MWT SpO₂ values were significantly lower than those at rest (p0.001), with showering SpO₂ significantly lower than 6MWT SpO₂ (p0.001) (Figure 1). The median ΔSpO₂-shw was 3.4%, and HF patients were stratified into small (3.4%) and large (≥3.4%) ΔSpO₂-shw groups. No significant differences were found in pulmonary function or echocardiographic data between these groups. Kaplan-Meier analysis revealed a higher cardiac event rate in the large ΔSpO₂-shw group (hazard ratio = 2.02, 95% CI = 1.01–4.09, p=0.047) (Figure 2). The control group had a significantly lower ΔSpO₂-shw (1.0% vs. 3.4%, p0.001). A strong correlation was observed between ΔSpO₂-shw and log NT-proBNP levels (n=77, r=0.548, p0.001). Conclusion Our findings suggest that ΔSpO₂-shw is a promising predictive marker for HF exacerbation. Continuous SpO₂ monitoring during routine activities like showering offers a practical and reproducible way to assess exercise tolerance at home, potentially enhancing HF management by enabling early detection of disease progression.Changes in SpO₂ levels Kaplan-Meier Analysis of Cardiac Events