SEISMIC-HF 1: key findings from AHA24 and implications for remote cardiac monitoring

While there is continued progress in developing therapies for patients with heart failure, the condition results in significant morbidity and a sizeable economic impact on our society. Recent advances in wearable sensors combined with machine learning algorithms give hope that heart failure can be better managed remotely and allow for improved clinical outcomes. This is a focused review of the key findings of the SEISMocardiogram In Cardiovascular Monitoring for Heart Failure I (SEISMIC-HF 1) study, presented at the American Heart Association's Scientific Sessions 2024 in Chicago, Illinois. This study showcased the ability of a machine learning algorithm to estimate pulmonary capillary wedge pressure in patients with heart failure with reduced ejection fraction, utilizing seismocardiography, photoplethysmography, and electrocardiography signals obtained non-invasively through a wearable sensor patch (CardioTag) for model input. The authors showed a significant correlation between model-predicted pulmonary capillary wedge pressure and the gold standard pressure measurement obtained from right heart catheterization. Future investigations should assess the implementation of this technology as a part of a treatment strategy for outpatient heart failure care and explore its performance in additional study populations including those with heart failure with preserved ejection fraction and in patients outside of the clinical environment.