Machine learning-based prediction of in-hospital death for patients with takotsubo syndrome: The InterTAK-ML model
Language English Country England, Great Britain Media print-electronic
Document type Journal Article
PubMed
37522520
DOI
10.1002/ejhf.2983
Knihovny.cz E-resources
- Keywords
- Artificial intelligence, Machine learning, Mortality prediction, Outcome, Takotsubo syndrome,
- MeSH
- Humans MeSH
- Hospital Mortality MeSH
- Prognosis MeSH
- Heart Failure * complications MeSH
- Machine Learning MeSH
- Takotsubo Cardiomyopathy * diagnosis complications MeSH
- Check Tag
- Humans MeSH
- Publication type
- Journal Article MeSH
AIMS: Takotsubo syndrome (TTS) is associated with a substantial rate of adverse events. We sought to design a machine learning (ML)-based model to predict the risk of in-hospital death and to perform a clustering of TTS patients to identify different risk profiles. METHODS AND RESULTS: A ridge logistic regression-based ML model for predicting in-hospital death was developed on 3482 TTS patients from the International Takotsubo (InterTAK) Registry, randomly split in a train and an internal validation cohort (75% and 25% of the sample size, respectively) and evaluated in an external validation cohort (1037 patients). Thirty-one clinically relevant variables were included in the prediction model. Model performance represented the primary endpoint and was assessed according to area under the curve (AUC), sensitivity and specificity. As secondary endpoint, a K-medoids clustering algorithm was designed to stratify patients into phenotypic groups based on the 10 most relevant features emerging from the main model. The overall incidence of in-hospital death was 5.2%. The InterTAK-ML model showed an AUC of 0.89 (0.85-0.92), a sensitivity of 0.85 (0.78-0.95) and a specificity of 0.76 (0.74-0.79) in the internal validation cohort and an AUC of 0.82 (0.73-0.91), a sensitivity of 0.74 (0.61-0.87) and a specificity of 0.79 (0.77-0.81) in the external cohort for in-hospital death prediction. By exploiting the 10 variables showing the highest feature importance, TTS patients were clustered into six groups associated with different risks of in-hospital death (28.8% vs. 15.5% vs. 5.4% vs. 1.0.8% vs. 0.5%) which were consistent also in the external cohort. CONCLUSION: A ML-based approach for the identification of TTS patients at risk of adverse short-term prognosis is feasible and effective. The InterTAK-ML model showed unprecedented discriminative capability for the prediction of in-hospital death.
1st Department of Cardiology Medical University of Gdansk Gdansk Poland
Berlin Institute of Health Berlin Germany
Center for Cardiology Cardiology 1 University Medical Center Mainz Mainz Germany
Center for Molecular Cardiology Schlieren Campus University of Zurich Zurich Switzerland
Centro Cardiologico Monzino IRCCS Milan Italy
Clinic for Cardiology and Pneumology Georg August University Goettingen Goettingen Germany
Department of Cardiology and Angiology Hannover Medical School Hannover Germany
Department of Cardiology and Cardiac Imaging Center University Hospital of Rangueil Toulouse France
Department of Cardiology and Internal Medicine B University Medicine Greifswald Greifswald Germany
Department of Cardiology Centro Hospitalar Universitário de São João Porto Portugal
Department of Cardiology Charité Campus Rudolf Virchow Berlin Germany
Department of Cardiology Chiba Emergency Medical Center Chiba Japan
Department of Cardiology Christchurch Hospital Christchurch New Zealand
Department of Cardiology Heidelberg University Hospital Heidelberg Germany
Department of Cardiology John Radcliffe Hospital Oxford University Hospitals Oxford UK
Department of Cardiology Kantonsspital Frauenfeld Frauenfeld Switzerland
Department of Cardiology Kantonsspital Lucerne Lucerne Switzerland
Department of Cardiology Kantonsspital St Gallen St Gallen Switzerland
Department of Cardiology Kantonsspital Winterthur Winterthur Switzerland
Department of Cardiology King's College Hospital London UK
Department of Cardiology Leiden University Medical Centre Leiden The Netherlands
Department of Cardiology Medical University of Warsaw Warsaw Poland
Department of Cardiology National University Heart Centre Singapore Singapore
Department of Cardiology University Hospital Basel Basel Switzerland
Department of Cardiovascular Diseases Mayo Clinic Rochester MN USA
Department of Cardiovascular Medicine Chiba University Graduate School of Medicine Chiba Japan
Department of Cardiovascular Medicine Nippon Medical School Tokyo Japan
Department of Intensive Care Medicine University Medical Center Hamburg Eppendorf Hamburg Germany
Department of Internal Medicine 1 Cardiology University Hospital Olomouc Olomouc Czech Republic
Department of Internal Medicine 2 Cardiology Medical Center University of Ulm Ulm Germany
Department of Internal Medicine 3 Heart Center University of Cologne Cologne Germany
Department of Internal Medicine Cardiology and Angiology Magdeburg University Magdeburg Germany
Department of Internal Medicine Cardiology Heart Center Leipzig University Hospital Leipzig Germany
Department of Medical Sciences University of Turin Turin Italy
Department of Medicine Surgery and Dentistry University of Salerno Baronissi Italy
Department of Medicine Surgery and Pharmacy University of Sassari Sassari Italy
Department of Vascular Physiopathology IRCCS Neuromed Pozzilli Italy
Deutsches Herzzentrum München Technische Universität München Munich Germany
Division of Cardiology 'Antonio Cardarelli' Hospital Naples Italy
Division of Cardiology A O U San Luigi Gonzaga Turin Italy
Division of Cardiology Heart and Vascular Center University of Iowa Iowa City IA USA
Division of Cardiology Kimitsu Central Hospital Kisarazu Japan
Division of Cardiology Medical University of Graz Graz Austria
Dorset Heart Centre Royal Bournemouth Hospital Bournemouth UK
DZHK Partner Site Greifswald Greifswald Germany
DZHK Partner Site Hamburg Kiel Luebeck Hamburg Germany
DZHK Partner Site Heidelberg Mannheim Mannheim Germany
DZHK Partner Site Munich Heart Alliance Munich Germany
Heart and Vascular Centre Bad Bevensen Bad Bevensen Germany
Heart Center Turku University Hospital University of Turku Turku Finland
Intensive Coronary Care Unit Moscow City Hospital No 1 named after N Pirogov Moscow Russia
Keck School of Medicine University of Southern California Los Angeles CA USA
Klinik für Innere Medizin 3 Universitätsklinikum des Saarlandes Homburg Saar Germany
Klinik und Poliklinik für Innere Medizin 2 Universitätsklinikum Regensburg Regensburg Germany
Krankenhaus 'Maria Hilf' Medizinische Klinik Stadtlohn Germany
Local Health Unit n 8 Cardiology Unit Vicenza Italy
Royal Brompton and Harefield Hospitals Trust and Imperial College and Kings College London UK
Service de Cardiologie Hôpitaux Universitaires de Genève Geneva Switzerland
Structural Interventional Cardiology Careggi University Hospital Florence Italy
TJ Health Partners Heart and Vascular Glasgow KY USA
University Hospital for Internal Medicine 3 Medical University Innsbruck Innsbruck Austria
See more in PubMed
Sato H. Tako-tsubo-like left ventricular dysfunction due to multivessel coronary spasm. In: Kodama K, Haze K, Hori M, eds. Clinical aspect of myocardial injury: From ischemia to heart failure. Tokyo: Kagakuhyoronsha Publishing Co; 1990. p56-64.
Templin C, Ghadri JR, Diekmann J, Napp LC, Bataiosu DR, Jaguszewski M, et al. Clinical features and outcomes of Takotsubo (stress) cardiomyopathy. N Engl J Med 2015;373:929-938. https://doi.org/10.1056/NEJMoa1406761
Ghadri JR, Kato K, Cammann VL, Gili S, Jurisic S, Di Vece D, et al. Long-term prognosis of patients with Takotsubo syndrome. J Am Coll Cardiol 2018;72:874-882. https://doi.org/10.1016/j.jacc.2018.06.016
Kim H, Senecal C, Lewis B, Prasad A, Rajiv G, Lerman LO, et al. Natural history and predictors of mortality of patients with Takotsubo syndrome. Int J Cardiol 2018;267:22-27. https://doi.org/10.1016/j.ijcard.2018.04.139
Uribarri A, Nunez-Gil IJ, Conty DA, Vedia O, Almendro-Delia M, Duran Cambra A, et al. Short- and long-term prognosis of patients with Takotsubo syndrome based on different triggers: Importance of the physical nature. J Am Heart Assoc 2019;8:e013701. https://doi.org/10.1161/JAHA.119.013701
Parodi G, Scudiero F, Citro R, Silverio A, Bellandi B, Zito C, et al. Risk stratification using the CHA2DS2-VASc score in Takotsubo syndrome: Data from the Takotsubo Italian Network. J Am Heart Assoc 2017;6:e006065. https://doi.org/10.1161/JAHA.117.006065
Santoro F, Nunez Gil IJ, Stiermaier T, El-Battrawy I, Guerra F, Novo G, et al. Assessment of the German and Italian stress cardiomyopathy score for risk stratification for in-hospital complications in patients with Takotsubo syndrome. JAMA Cardiol 2019;4:892-899. https://doi.org/10.1001/jamacardio.2019.2597
Motwani M, Dey D, Berman DS, Germano G, Achenbach S, Al-Mallah MH, et al. Machine learning for prediction of all-cause mortality in patients with suspected coronary artery disease: A 5-year multicentre prospective registry analysis. Eur Heart J 2017;38:500-507. https://doi.org/10.1093/eurheartj/ehw188
D'Ascenzo F, De Filippo O, Gallone G, Mittone G, Deriu MA, Iannaccone M, et al. Machine learning-based prediction of adverse events following an acute coronary syndrome (PRAISE): A modelling study of pooled datasets. Lancet 2021;397:199-207. https://doi.org/10.1016/S0140-6736(20)32519-8
Ghadri JR, Cammann VL, Templin C. The International Takotsubo Registry: Rationale, design, objectives, and first results. Heart Fail Clin 2016;12:597-603. https://doi.org/10.1016/j.hfc.2016.06.010
Cammann VL, Wurdinger M, Ghadri JR, Templin C. Takotsubo syndrome: Uncovering myths and misconceptions. Curr Atheroscler Rep 2021;23:53. https://doi.org/10.1007/s11883-021-00946-z
Ghadri JR, Wittstein IS, Prasad A, Sharkey S, Dote K, Akashi YJ, et al. International expert consensus document on Takotsubo syndrome (part I): Clinical characteristics, diagnostic criteria, and pathophysiology. Eur Heart J 2018;39:2032-2046. https://doi.org/10.1093/eurheartj/ehy076
Cammann VL, Szawan KA, Stähli BE, Kato K, Budnik M, Wischnewsky M, et al. Age-related variations in Takotsubo syndrome. J Am Coll Cardiol 2020;75:1869-1877. https://doi.org/10.1016/j.jacc.2020.02.057
Cammann VL, Scheitz JF, von Rennenberg R, Jancke L, Nolte CH, Szawan KA, et al. Clinical correlates and prognostic impact of neurologic disorders in Takotsubo syndrome. Sci Rep 2021;11:23555. https://doi.org/10.1038/s41598-021-01496-9
Di Vece D, Citro R, Cammann VL, Kato K, Gili S, Szawan KA, et al. Outcomes associated with cardiogenic shock in Takotsubo syndrome. Circulation 2019;139:413-415. https://doi.org/10.1161/CIRCULATIONAHA.118.036164
Jurisic S, Gili S, Cammann VL, Kato K, Szawan KA, D'Ascenzo F, et al. Clinical predictors and prognostic impact of recovery of wall motion abnormalities in Takotsubo syndrome: Results from the International Takotsubo Registry. J Am Heart Assoc 2019;8:e011194. https://doi.org/10.1161/JAHA.118.011194
Gili S, Cammann VL, Schlossbauer SA, Kato K, D'Ascenzo F, Vece DD, et al. Cardiac arrest in takotsubo syndrome: Results from the InterTAK Registry. Eur Heart J 2019;40:2142-2151. https://doi.org/10.1093/eurheartj/ehz170
Hartigan JA, Wong MA. Algorithm AS 136: A K-means clustering algorithm. J R Stat Soc C 1979;28(1):100-108. https://doi.org/10.2307/2346830
Citro R, Radano I, Parodi G, di Vece D, Zito C, Novo G, et al. Long-term outcome in patients with Takotsubo syndrome presenting with severely reduced left ventricular ejection fraction. Eur J Heart Fail 2019;21:781-778. https://doi.org/10.1002/ejhf.1373
Wischnewsky MB, Candreva A, Bacchi B, Cammann VL, Kato K, Szawan KA, et al. Prediction of short- and long-term mortality in takotsubo syndrome: The InterTAK prognostic score. Eur J Heart Fail 2019;21:1469-1472. https://doi.org/10.1002/ejhf.1561
Akashi YJ, Goldstein DS, Barbaro G, Ueyama T. Takotsubo cardiomyopathy: A new form of acute, reversible heart failure. Circulation 2008;118:2754-2762. https://doi.org/10.1161/CIRCULATIONAHA.108.767012
Kurisu S, Sato H, Kawagoe T, Ishihara M, Shimatani Y, Nishioka K, et al. Tako-tsubo-like left ventricular dysfunction with ST-segment elevation: A novel cardiac syndrome mimicking acute myocardial infarction. Am Heart J 2002;143:448-455. https://doi.org/10.1067/mhj.2002.120403
Tsuchihashi K, Ueshima K, Uchida T, Oh-mura N, Kimura K, Owa M, et al. Transient left ventricular apical ballooning without coronary artery stenosis: A novel heart syndrome mimicking acute myocardial infarction. Angina Pectoris-Myocardial Infarction Investigations in Japan. J Am Coll Cardiol 2001;38:11-18. https://doi.org/10.1016/s0735-1097(01)01316-x
Lyon AR, Bossone E, Schneider B, Sechtem U, Citro R, Underwood SR, et al. Current state of knowledge on Takotsubo syndrome: A position statement from the Taskforce on Takotsubo Syndrome of the Heart Failure Association of the European Society of Cardiology. Eur J Heart Fail 2016;18:8-27. https://doi.org/10.1002/ejhf.424
Krittanawong C, Zhang H, Wang Z, Aydar M, Kitai T. Artificial intelligence in precision cardiovascular medicine. J Am Coll Cardiol 2017;69:2657-2664. https://doi.org/10.1016/j.jacc.2017.03.571
Ghadri JR, Cammann VL, Napp LC, Jurisic S, Diekmann J, Bataiosu DR, et al.; International Takotsubo (InterTAK) Registry. Differences in the clinical profile and outcomes of typical and atypical Takotsubo syndrome: Data from the International Takotsubo Registry. JAMA Cardiol 2016;1:335-340. https://doi.org/10.1001/jamacardio.2016.0225
Bohm M, Cammann VL, Ghadri JR, Ukena C, Gili S, Di Vece D, et al. Interaction of systolic blood pressure and resting heart rate with clinical outcomes in takotsubo syndrome: Insights from the International Takotsubo Registry. Eur J Heart Fail 2018;20:1021-1030. https://doi.org/10.1002/ejhf.1162
D'Ascenzo F, Gili S, Bertaina M, Iannaccone M, Cammann VL, Di Vece D, et al. Impact of aspirin on takotsubo syndrome: A propensity score-based analysis of the InterTAK Registry. Eur J Heart Fail 2020;22:330-337. https://doi.org/10.1002/ejhf.1698
Almendro-Delia M, Núñez-Gil IJ, Lobo M, Andrés M, Vedia O, Sionis A, et al.; RETAKO Investigators. Short- and long-term prognostic relevance of cardiogenic shock in Takotsubo syndrome: Results from the RETAKO registry. JACC Heart Fail 2018;6:928-936. https://doi.org/10.1016/j.jchf.2018.05.015
Kato K, Lyon AR, Ghadri JR, Templin C. Takotsubo syndrome: Aetiology, presentation and treatment. Heart 2017;103:1461-1469. https://doi.org/10.1136/heartjnl-2016-309783
Schwartz PJ, De Ferrari GM. Sympathetic-parasympathetic interaction in health and disease: Abnormalities and relevance in heart failure. Heart Fail Rev 2011;16:101-107. https://doi.org/10.1007/s10741-010-9179-1
Li P, Wang Y, Liang J, Zuo X, Li Q, Sherif AA, et al. Takotsubo syndrome and respiratory diseases: A systematic review. Eur Heart J Open 2022;24:oeac009. https://doi.org/10.1093/ehjopen/oeac009
Scally C, Abbas H, Ahearn T, Srinivasan J, Mezincescu A, Rudd A, et al. Myocardial and systemic inflammation in acute stress-induced (Takotsubo) cardiomyopathy. Circulation 2019;139:1581-1592. https://doi.org/10.1161/CIRCULATIONAHA.118.037975
Dusi V, Ghidoni A, Ravera A, De Ferrari GM, Calvillo L. Chemokines and heart disease: A network connecting cardiovascular biology to immune and autonomic nervous systems. Mediators Inflamm 2016;2016:5902947. https://doi.org/10.1155/2016/5902947
Li P, Dai Q, Cai P, Teng C, Pan S, Dixon RAF, et al. Identifying different phenotypes in takotsubo cardiomyopathy by latent class analysis. ESC Heart Fail 2021;8:555-565. https://doi.org/10.1002/ehf2.13117
Galiuto L, Crea F. Primary and secondary takotsubo syndrome: Pathophysiological determinant and prognosis. Eur Heart J Acute Cardiovasc Care 2020;9:690-693. https://doi.org/10.1177/2048872620963493
Nunez-Gil IJ, Almendro-Delia M, Andres M, Sionis A, Martin A, Bastante T, et al. Secondary forms of Takotsubo cardiomyopathy: A whole different prognosis. Eur Heart J Acute Cardiovasc Care 2016;5:308-316. https://doi.org/10.1177/2048872615589512
