BACKGROUND: Cardiac resynchronization therapy (CRT) has significant nonresponse rates. We assessed whether machine learning (ML) could predict CRT response beyond current guidelines. METHODS: We analyzed CRT patients from Cleveland Clinic and Johns Hopkins. A training cohort was created from all Johns Hopkins patients and an equal number of randomly sampled Cleveland Clinic patients. All remaining patients comprised the testing cohort. Response was defined as ≥10% increase in left ventricular ejection fraction. ML models were developed to predict CRT response using different combinations of classification algorithms and clinical variable sets on the training cohort. The model with the highest area under the curve was evaluated on the testing cohort. Probability of response was used to predict survival free from a composite end point of death, heart transplant, or placement of left ventricular assist device. Predictions were compared with current guidelines. RESULTS: Nine hundred twenty-five patients were included. On the training cohort (n=470: 235, Johns Hopkins; 235, Cleveland Clinic), the best ML model was a naive Bayes classifier including 9 variables (QRS morphology, QRS duration, New York Heart Association classification, left ventricular ejection fraction and end-diastolic diameter, sex, ischemic cardiomyopathy, atrial fibrillation, and epicardial left ventricular lead). On the testing cohort (n=455, Cleveland Clinic), ML demonstrated better response prediction than guidelines (area under the curve, 0.70 versus 0.65; P=0.012) and greater discrimination of event-free survival (concordance index, 0.61 versus 0.56; P<0.001). The fourth quartile of the ML model had the greatest risk of reaching the composite end point, whereas the first quartile had the least (hazard ratio, 0.34; P<0.001). CONCLUSIONS: ML with 9 variables incrementally improved prediction of echocardiographic CRT response and survival beyond guidelines. Performance was not improved by incorporating more variables. The model offers potential for improved shared decision-making in CRT (online calculator: http://riskcalc.org:3838/CRTResponseScore ). Significant remaining limitations confirm the need to identify better variables to predict CRT response.
BACKGROUND: Cardiac resynchronization therapy (CRT) has significant nonresponse rates. We assessed whether machine learning (ML) could predict CRT response beyond current guidelines. METHODS: We analyzed CRT patients from Cleveland Clinic and Johns Hopkins. A training cohort was created from all Johns Hopkins patients and an equal number of randomly sampled Cleveland Clinic patients. All remaining patients comprised the testing cohort. Response was defined as ≥10% increase in left ventricular ejection fraction. ML models were developed to predict CRT response using different combinations of classification algorithms and clinical variable sets on the training cohort. The model with the highest area under the curve was evaluated on the testing cohort. Probability of response was used to predict survival free from a composite end point of death, heart transplant, or placement of left ventricular assist device. Predictions were compared with current guidelines. RESULTS: Nine hundred twenty-five patients were included. On the training cohort (n=470: 235, Johns Hopkins; 235, Cleveland Clinic), the best ML model was a naive Bayes classifier including 9 variables (QRS morphology, QRS duration, New York Heart Association classification, left ventricular ejection fraction and end-diastolic diameter, sex, ischemic cardiomyopathy, atrial fibrillation, and epicardial left ventricular lead). On the testing cohort (n=455, Cleveland Clinic), ML demonstrated better response prediction than guidelines (area under the curve, 0.70 versus 0.65; P=0.012) and greater discrimination of event-free survival (concordance index, 0.61 versus 0.56; P<0.001). The fourth quartile of the ML model had the greatest risk of reaching the composite end point, whereas the first quartile had the least (hazard ratio, 0.34; P<0.001). CONCLUSIONS:ML with 9 variables incrementally improved prediction of echocardiographic CRT response and survival beyond guidelines. Performance was not improved by incorporating more variables. The model offers potential for improved shared decision-making in CRT (online calculator: http://riskcalc.org:3838/CRTResponseScore ). Significant remaining limitations confirm the need to identify better variables to predict CRT response.
Authors: Michael R Bristow; Leslie A Saxon; John Boehmer; Steven Krueger; David A Kass; Teresa De Marco; Peter Carson; Lorenzo DiCarlo; David DeMets; Bill G White; Dale W DeVries; Arthur M Feldman Journal: N Engl J Med Date: 2004-05-20 Impact factor: 91.245
Authors: Matthew M Kalscheur; Ryan T Kipp; Matthew C Tattersall; Chaoqun Mei; Kevin A Buhr; David L DeMets; Michael E Field; Lee L Eckhardt; C David Page Journal: Circ Arrhythm Electrophysiol Date: 2018-01
Authors: Aysha Arshad; Arthur J Moss; Elyse Foster; Luigi Padeletti; Alon Barsheshet; Ilan Goldenberg; Henry Greenberg; W Jackson Hall; Scott McNitt; Wojciech Zareba; Scott Solomon; Jonathan S Steinberg Journal: J Am Coll Cardiol Date: 2011-02-15 Impact factor: 24.094
Authors: John G F Cleland; Jean-Claude Daubert; Erland Erdmann; Nick Freemantle; Daniel Gras; Lukas Kappenberger; Luigi Tavazzi Journal: N Engl J Med Date: 2005-03-07 Impact factor: 91.245
Authors: Rutger J van Bommel; Jeroen J Bax; William T Abraham; Eugene S Chung; Luis A Pires; Luigi Tavazzi; Peter J Zimetbaum; Bart Gerritse; Nina Kristiansen; Stefano Ghio Journal: Eur Heart J Date: 2009-08-30 Impact factor: 29.983
Authors: Albert K Feeny; Mina K Chung; Anant Madabhushi; Zachi I Attia; Maja Cikes; Marjan Firouznia; Paul A Friedman; Matthew M Kalscheur; Suraj Kapa; Sanjiv M Narayan; Peter A Noseworthy; Rod S Passman; Marco V Perez; Nicholas S Peters; Jonathan P Piccini; Khaldoun G Tarakji; Suma A Thomas; Natalia A Trayanova; Mintu P Turakhia; Paul J Wang Journal: Circ Arrhythm Electrophysiol Date: 2020-07-06
Authors: Mina K Chung; Angela Fagerlin; Paul J Wang; Tinuola B Ajayi; Larry A Allen; Tina Baykaner; Emelia J Benjamin; Megan Branda; Kerri L Cavanaugh; Lin Y Chen; George H Crossley; Rebecca K Delaney; Lee L Eckhardt; Kathleen L Grady; Ian G Hargraves; Mellanie True Hills; Matthew M Kalscheur; Daniel B Kramer; Marleen Kunneman; Rachel Lampert; Aisha T Langford; Krystina B Lewis; Ying Lu; John M Mandrola; Kathryn Martinez; Daniel D Matlock; Sarah R McCarthy; Victor M Montori; Peter A Noseworthy; Kate M Orland; Elissa Ozanne; Rod Passman; Krishna Pundi; Dan M Roden; Elizabeth V Saarel; Monika M Schmidt; Samuel F Sears; Dawn Stacey; Randall S Stafford; Benjamin A Steinberg; Sojin Youn Wass; Jennifer M Wright Journal: Circ Arrhythm Electrophysiol Date: 2021-12-06
Authors: Albert K Feeny; John Rickard; Kevin M Trulock; Divyang Patel; Saleem Toro; Laurie Ann Moennich; Niraj Varma; Mark J Niebauer; Eiran Z Gorodeski; Richard A Grimm; John Barnard; Anant Madabhushi; Mina K Chung Journal: Circ Arrhythm Electrophysiol Date: 2020-06-14
Authors: Albert J Rogers; Anojan Selvalingam; Mahmood I Alhusseini; David E Krummen; Cesare Corrado; Firas Abuzaid; Tina Baykaner; Christian Meyer; Paul Clopton; Wayne Giles; Peter Bailis; Steven Niederer; Paul J Wang; Wouter-Jan Rappel; Matei Zaharia; Sanjiv M Narayan Journal: Circ Res Date: 2020-11-10 Impact factor: 17.367
Authors: Niraj Varma; Robert C Bourge; Lynne Warner Stevenson; Maria Rosa Costanzo; David Shavelle; Philip B Adamson; Greg Ginn; John Henderson; William T Abraham Journal: J Am Heart Assoc Date: 2021-02-25 Impact factor: 5.501