1
|
Kolk MZH, Ruipérez-Campillo S, Deb B, Bekkers EJ, Allaart CP, Rogers AJ, Van Der Lingen ALCJ, Alvarez Florez L, Isgum I, De Vos BD, Clopton P, Wilde AAM, Knops RE, Narayan SM, Tjong FVY. Optimizing patient selection for primary prevention implantable cardioverter-defibrillator implantation: utilizing multimodal machine learning to assess risk of implantable cardioverter-defibrillator non-benefit. Europace 2023; 25:euad271. [PMID: 37712675 PMCID: PMC10516624 DOI: 10.1093/europace/euad271] [Citation(s) in RCA: 0] [Impact Index Per Article: 0] [Reference Citation Analysis] [Abstract] [Key Words] [MESH Headings] [Grants] [Track Full Text] [Journal Information] [Subscribe] [Scholar Register] [Received: 06/02/2023] [Accepted: 08/07/2023] [Indexed: 09/16/2023] Open
Abstract
AIMS Left ventricular ejection fraction (LVEF) is suboptimal as a sole marker for predicting sudden cardiac death (SCD). Machine learning (ML) provides new opportunities for personalized predictions using complex, multimodal data. This study aimed to determine if risk stratification for implantable cardioverter-defibrillator (ICD) implantation can be improved by ML models that combine clinical variables with 12-lead electrocardiograms (ECG) time-series features. METHODS AND RESULTS A multicentre study of 1010 patients (64.9 ± 10.8 years, 26.8% female) with ischaemic, dilated, or non-ischaemic cardiomyopathy, and LVEF ≤ 35% implanted with an ICD between 2007 and 2021 for primary prevention of SCD in two academic hospitals was performed. For each patient, a raw 12-lead, 10-s ECG was obtained within 90 days before ICD implantation, and clinical details were collected. Supervised ML models were trained and validated on a development cohort (n = 550) from Hospital A to predict ICD non-arrhythmic mortality at three-year follow-up (i.e. mortality without prior appropriate ICD-therapy). Model performance was evaluated on an external patient cohort from Hospital B (n = 460). At three-year follow-up, 16.0% of patients had died, with 72.8% meeting criteria for non-arrhythmic mortality. Extreme gradient boosting models identified patients with non-arrhythmic mortality with an area under the receiver operating characteristic curve (AUROC) of 0.90 [95% confidence intervals (CI) 0.80-1.00] during internal validation. In the external cohort, the AUROC was 0.79 (95% CI 0.75-0.84). CONCLUSIONS ML models combining ECG time-series features and clinical variables were able to predict non-arrhythmic mortality within three years after device implantation in a primary prevention population, with robust performance in an independent cohort.
Collapse
Affiliation(s)
- Maarten Z H Kolk
- Department of Cardiology, Heart Center, Amsterdam UMC location University of Amsterdam, Meibergdreef 9, 1105 AZ Amsterdam, The Netherlands
- Amsterdam Cardiovascular Sciences, Heart Failure and Arrhythmias, Meibergdreef 9, 1105 AZ Amsterdam, The Netherlands
| | - Samuel Ruipérez-Campillo
- Department of Medicine and Cardiovascular Institute, Stanford University, 780 Welch Road, MC 5773, Stanford, CA 94305, USA
- Department of Information Technology and Electrical Engineering, Swiss Federal Institute of Technology Zurich (ETHz), Zurich, Switzerland
| | - Brototo Deb
- Department of Medicine and Cardiovascular Institute, Stanford University, 780 Welch Road, MC 5773, Stanford, CA 94305, USA
| | - Erik J Bekkers
- Faculty of Science, University of Amsterdam, Science Park 904, 1098 XH Amsterdam, the Netherlands
| | - Cornelis P Allaart
- Department of Cardiology, Amsterdam UMC location Vrije Universiteit Amsterdam, De Boelelaan 1117, 1081 HV Amsterdam, The Netherlands
| | - Albert J Rogers
- Department of Medicine and Cardiovascular Institute, Stanford University, 780 Welch Road, MC 5773, Stanford, CA 94305, USA
| | - Anne-Lotte C J Van Der Lingen
- Department of Cardiology, Amsterdam UMC location Vrije Universiteit Amsterdam, De Boelelaan 1117, 1081 HV Amsterdam, The Netherlands
| | - Laura Alvarez Florez
- Department of Cardiology, Heart Center, Amsterdam UMC location University of Amsterdam, Meibergdreef 9, 1105 AZ Amsterdam, The Netherlands
- Amsterdam Cardiovascular Sciences, Heart Failure and Arrhythmias, Meibergdreef 9, 1105 AZ Amsterdam, The Netherlands
- Department of Biomedical Engineering and Physics, Amsterdam UMC location University of Amsterdam, Meibergdreef 9, 1105 AZ Amsterdam, The Netherlands
| | - Ivana Isgum
- Faculty of Science, University of Amsterdam, Science Park 904, 1098 XH Amsterdam, the Netherlands
- Department of Biomedical Engineering and Physics, Amsterdam UMC location University of Amsterdam, Meibergdreef 9, 1105 AZ Amsterdam, The Netherlands
- Department of Radiology and Nuclear Medicine, Amsterdam UMC location University of Amsterdam, Meibergdreef 9, 1105 AZ Amsterdam, The Netherlands
| | - Bob D De Vos
- Department of Biomedical Engineering and Physics, Amsterdam UMC location University of Amsterdam, Meibergdreef 9, 1105 AZ Amsterdam, The Netherlands
| | - Paul Clopton
- Department of Medicine and Cardiovascular Institute, Stanford University, 780 Welch Road, MC 5773, Stanford, CA 94305, USA
| | - Arthur A M Wilde
- Department of Cardiology, Heart Center, Amsterdam UMC location University of Amsterdam, Meibergdreef 9, 1105 AZ Amsterdam, The Netherlands
- Amsterdam Cardiovascular Sciences, Heart Failure and Arrhythmias, Meibergdreef 9, 1105 AZ Amsterdam, The Netherlands
| | - Reinoud E Knops
- Department of Cardiology, Heart Center, Amsterdam UMC location University of Amsterdam, Meibergdreef 9, 1105 AZ Amsterdam, The Netherlands
- Amsterdam Cardiovascular Sciences, Heart Failure and Arrhythmias, Meibergdreef 9, 1105 AZ Amsterdam, The Netherlands
| | - Sanjiv M Narayan
- Department of Medicine and Cardiovascular Institute, Stanford University, 780 Welch Road, MC 5773, Stanford, CA 94305, USA
| | - Fleur V Y Tjong
- Department of Cardiology, Heart Center, Amsterdam UMC location University of Amsterdam, Meibergdreef 9, 1105 AZ Amsterdam, The Netherlands
- Amsterdam Cardiovascular Sciences, Heart Failure and Arrhythmias, Meibergdreef 9, 1105 AZ Amsterdam, The Netherlands
- Department of Medicine and Cardiovascular Institute, Stanford University, 780 Welch Road, MC 5773, Stanford, CA 94305, USA
| |
Collapse
|
2
|
Farouq M, Rorsman C, Marinko S, Mörtsell D, Chaudhry U, Wang L, Platonov PG, Borgquist R. Age-stratified comparison of prognosis in cardiac resynchronization therapy with or without prophylactic defibrillator for nonischemic cardiomyopathy-a nationwide cohort study. Europace 2023; 25:euad187. [PMID: 37392462 PMCID: PMC10368447 DOI: 10.1093/europace/euad187] [Citation(s) in RCA: 0] [Impact Index Per Article: 0] [Reference Citation Analysis] [Abstract] [Key Words] [MESH Headings] [Grants] [Track Full Text] [Journal Information] [Subscribe] [Scholar Register] [Received: 03/04/2023] [Accepted: 06/06/2023] [Indexed: 07/03/2023] Open
Abstract
AIMS Prior studies have suggested that the benefit from primary preventive defibrillator treatment for patients with nonischemic cardiomyopathyy, treated with cardiac resynchronization therapy, may be age-dependent. We aimed to compare age-stratified mortality rates and mode of death in patients with nonischemic cardiomyopathy who are treated with either primary preventive cardiac resynchronization therapy with defibrillator (CRT-D) or CRT with pacemaker (CRT-P). METHODS AND RESULTS All patients with nonischemic cardiomyopathy and CRT-P or primary preventive CRT-D who were implanted in Sweden during the period 2005-2020 were included. Propensity scoring was used to create a matched cohort. Primary outcome was all-cause mortality within 5 years. In all, 4027 patients were included: 2334 with CRT-P and 1693 with CRT-D. Crude 5-year mortality was 635 (27%) vs. 246 (15%), P < 0.001. In Cox regression analysis, adjusted for clinically relevant covariables, CRT-D was independently associated with higher 5-year survival [0.72 (0.61-0.85), P < 0.001]. Cardiovascular mortality was similar between groups (62 vs. 64%, P = 0.64), but death from heart failure was more common in the CRT-D group (46 vs. 36%, P = 0.007). In the matched cohort (n = 2414), 5-year mortality was 21% (24 vs. 16%, P < 0.001). In age-stratified analyses, CRT-P was associated with higher mortality in age groups <60 years and 70-79 years, but there was no difference in age groups 60-69 years or 80-89 years. CONCLUSION In this nationwide registry-based study, patients with CRT-D had better 5-year survival compared to patients with CRT-P. The interaction between age and mortality reduction was not consistent, but patients with CRT-D aged <60 years had the largest absolute mortality reduction.
Collapse
Affiliation(s)
- Maiwand Farouq
- Cardiology, Department of Clinical Sciences, Lund University, Entrégatan 7, Lund 22185, Sweden
- Arrhythmia Section, Skane University Hospital, Träslövsvägen 68, 432 37 Varberg, Sweden
| | - Cecilia Rorsman
- Internal Medicine Department, Varberg Hospital, Varberg, Sweden
| | - Sofia Marinko
- Cardiology, Department of Clinical Sciences, Lund University, Entrégatan 7, Lund 22185, Sweden
- Arrhythmia Section, Skane University Hospital, Träslövsvägen 68, 432 37 Varberg, Sweden
| | - David Mörtsell
- Cardiology, Department of Clinical Sciences, Lund University, Entrégatan 7, Lund 22185, Sweden
- Arrhythmia Section, Skane University Hospital, Träslövsvägen 68, 432 37 Varberg, Sweden
| | - Uzma Chaudhry
- Cardiology, Department of Clinical Sciences, Lund University, Entrégatan 7, Lund 22185, Sweden
- Arrhythmia Section, Skane University Hospital, Träslövsvägen 68, 432 37 Varberg, Sweden
| | - Lingwei Wang
- Cardiology, Department of Clinical Sciences, Lund University, Entrégatan 7, Lund 22185, Sweden
- Arrhythmia Section, Skane University Hospital, Träslövsvägen 68, 432 37 Varberg, Sweden
| | - Pyotr G Platonov
- Cardiology, Department of Clinical Sciences, Lund University, Entrégatan 7, Lund 22185, Sweden
- Arrhythmia Section, Skane University Hospital, Träslövsvägen 68, 432 37 Varberg, Sweden
| | - Rasmus Borgquist
- Cardiology, Department of Clinical Sciences, Lund University, Entrégatan 7, Lund 22185, Sweden
- Arrhythmia Section, Skane University Hospital, Träslövsvägen 68, 432 37 Varberg, Sweden
| |
Collapse
|
3
|
Nielsen JC, Dagres N. How can we assess the risk for sudden cardiac death to decide for primary prophylactic implantable cardioverter-defibrillator in patients with heart failure in 2022? Europace 2022; 24:1199-1200. [PMID: 35574673 DOI: 10.1093/europace/euac061] [Citation(s) in RCA: 0] [Impact Index Per Article: 0] [Reference Citation Analysis] [Track Full Text] [Journal Information] [Subscribe] [Scholar Register] [Indexed: 11/12/2022] Open
Affiliation(s)
- Jens Cosedis Nielsen
- Department of Cardiology, Aarhus University Hospital, Aarhus, Denmark.,Department of Clinical Medicine, Aarhus University, Aarhus, Denmark
| | - Nikolaos Dagres
- Department of Electrophysiology, Heart Center Leipzig at University of Leipzig, Leipzig, Germany
| |
Collapse
|