Non-exercise estimation of peak oxygen uptake in patients with ischaemic heart disease and heart failure using seismocardiography.

Aims: Knowledge of cardiorespiratory fitness (i.e. VO2peak) is important for determining prognosis and prescribing exercise for patients with heart disease undergoing cardiac rehabilitation (CR) programmes. In this explorative study, we investigated the accuracy of a novel equation using seismocardiography (SCG) at rest for the estimation of VO2peak (SCG eVO2peak) and whether it could detect changes following CR. An interim data analysis was planned after 50% of patients had undergone testing, allowing for adjustment of the SCG eVO2peak. We compared the SCG eVO2peak with a cardiopulmonary exercise test (CPET). Methods and results: We included 125 patients with new-onset ischaemic heart disease (IHD, n = 58) or heart failure with reduced left ventricular ejection fraction (HFrEF, n = 67) from an outpatient CR clinic. Testing included SCG eVO2peak and CPET measurement. The adjusted SCG 4.7_HD was validated in 30 and 34 patients with IHD and HFrEF, respectively. Forty-four out of the 67 patients with HFrEF were tested after completing 12 weeks of CR. A mean absolute percentage error (MAPE) ≤10% was decided for clinical relevance. The SCG 4.7_HD overestimated VO2peak (2.1 mL min-1 kg-1, P = 0.003) with 95% limits of agreement ranging ±10.9 mL min-1 kg-1 when compared with CPET. The standard error of estimation was 6.0 mL min-1 kg-1, and MAPE was 29.1%. No correlation was observed between delta SCG 4.7_HD and CPET after CR for patients with HFrEF. Conclusion: The SCG eVO2peak is not supported for clinical purposes in patients with IHD or HFrEF based on a poor-to-moderate agreement with large estimation errors and the inability to detect changes following CR. Trial registration: The study is registered at ClinicalTrials.gov (NCT05520307).