Valerie A A van Es, Mayke M C J van Leunen, Ignace L J de Lathauwer, Cindy C A G Verstappen, René A Tio, Ruud F Spee, Lu Yuan, Monica Betta, Giacomo Handjaras, Hareld M C Kemps
{"title":"Predicting acute decompensated heart failure using circadian markers from heart rate time series.","authors":"Valerie A A van Es, Mayke M C J van Leunen, Ignace L J de Lathauwer, Cindy C A G Verstappen, René A Tio, Ruud F Spee, Lu Yuan, Monica Betta, Giacomo Handjaras, Hareld M C Kemps","doi":"10.1002/ehf2.15395","DOIUrl":null,"url":null,"abstract":"<p><strong>Aims: </strong>Hospital admissions for acute decompensated heart failure (ADHF) are linked to high readmission rates, emphasizing the need for early intervention. Dysregulation of the circadian rhythm that regulates key physiological processes, such as heart rate (HR), blood pressure and sleep-wake cycles, may precede weight gain and clinical symptoms of worsening heart failure (HF) by weeks, providing a window for timely intervention. This study aims to develop a predictive algorithm for early and accurate ADHF detection.</p><p><strong>Methods and results: </strong>Sixty-five patients discharged after ADHF hospitalization monitored HR with a wrist-worn device for 6 months after reaching stable HF. Circadian parameters (mesor, amplitude and acrophase) were extracted via cosinor analysis and used to train a long short-term memory neural network. The algorithm analysed 21-day periods before an HF event, defined as unplanned outpatient visits for congestion episode, increased diuretics, ADHF hospitalization or sudden cardiac death. Circadian changes appeared in the 21 days preceding HF events, with elevated mesor (70.6 vs. 73.6 b.p.m.; P < 0.001), reduced amplitude (8.3 vs. 4.9 b.p.m.; P = 0.046) and acrophase shifts (11.3 vs. 12.2 h; P = 0.706). The classification algorithm showed 74% sensitivity, 73% specificity and a 74% AUC (P < 0.001). Amplitude was the strongest predictor, contributing 62% to the algorithm's feature importance.</p><p><strong>Conclusions: </strong>Circadian metrics from a wrist-worn device showed progressive alterations over the 3 weeks preceding ADHF, offering potential early detection of HF decompensation with moderate prediction performance. Future research should refine these metrics and results in larger, diverse populations, using various sensor types and explore early interventions.</p>","PeriodicalId":11864,"journal":{"name":"ESC Heart Failure","volume":" ","pages":""},"PeriodicalIF":3.7000,"publicationDate":"2025-08-28","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":"0","resultStr":null,"platform":"Semanticscholar","paperid":null,"PeriodicalName":"ESC Heart Failure","FirstCategoryId":"3","ListUrlMain":"https://doi.org/10.1002/ehf2.15395","RegionNum":2,"RegionCategory":"医学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":null,"EPubDate":"","PubModel":"","JCR":"Q2","JCRName":"CARDIAC & CARDIOVASCULAR SYSTEMS","Score":null,"Total":0}
引用次数: 0
Abstract
Aims: Hospital admissions for acute decompensated heart failure (ADHF) are linked to high readmission rates, emphasizing the need for early intervention. Dysregulation of the circadian rhythm that regulates key physiological processes, such as heart rate (HR), blood pressure and sleep-wake cycles, may precede weight gain and clinical symptoms of worsening heart failure (HF) by weeks, providing a window for timely intervention. This study aims to develop a predictive algorithm for early and accurate ADHF detection.
Methods and results: Sixty-five patients discharged after ADHF hospitalization monitored HR with a wrist-worn device for 6 months after reaching stable HF. Circadian parameters (mesor, amplitude and acrophase) were extracted via cosinor analysis and used to train a long short-term memory neural network. The algorithm analysed 21-day periods before an HF event, defined as unplanned outpatient visits for congestion episode, increased diuretics, ADHF hospitalization or sudden cardiac death. Circadian changes appeared in the 21 days preceding HF events, with elevated mesor (70.6 vs. 73.6 b.p.m.; P < 0.001), reduced amplitude (8.3 vs. 4.9 b.p.m.; P = 0.046) and acrophase shifts (11.3 vs. 12.2 h; P = 0.706). The classification algorithm showed 74% sensitivity, 73% specificity and a 74% AUC (P < 0.001). Amplitude was the strongest predictor, contributing 62% to the algorithm's feature importance.
Conclusions: Circadian metrics from a wrist-worn device showed progressive alterations over the 3 weeks preceding ADHF, offering potential early detection of HF decompensation with moderate prediction performance. Future research should refine these metrics and results in larger, diverse populations, using various sensor types and explore early interventions.
期刊介绍:
ESC Heart Failure is the open access journal of the Heart Failure Association of the European Society of Cardiology dedicated to the advancement of knowledge in the field of heart failure. The journal aims to improve the understanding, prevention, investigation and treatment of heart failure. Molecular and cellular biology, pathology, physiology, electrophysiology, pharmacology, as well as the clinical, social and population sciences all form part of the discipline that is heart failure. Accordingly, submission of manuscripts on basic, translational, clinical and population sciences is invited. Original contributions on nursing, care of the elderly, primary care, health economics and other specialist fields related to heart failure are also welcome, as are case reports that highlight interesting aspects of heart failure care and treatment.