Stacey J Howell1, Tim Stivland2, Kenneth Stein2, Kenneth A Ellenbogen3, Larisa G Tereshchenko4. 1. Knight Cardiovascular Institute, Oregon Health and Science University, Portland, Oregon, USA. 2. Boston Scientific, Marlborough, Massachusetts, USA. 3. Medical College of Virginia/Virginia Commonwealth University School of Medicine, Richmond, Virginia, USA. 4. Knight Cardiovascular Institute, Oregon Health and Science University, Portland, Oregon, USA; Cardiovascular Division, Department of Medicine, Johns Hopkins School of Medicine, Baltimore, Maryland, USA. Electronic address: tereshch@ohsu.edu.
Abstract
OBJECTIVES: This study aimed to apply machine learning (ML) to develop a prediction model for short-term cardiac resynchronization therapy (CRT) response to identifying CRT candidates for early multidisciplinary CRT heart failure (HF) care. BACKGROUND: Multidisciplinary optimization of cardiac resynchronization therapy (CRT) delivery can improve long-term CRT outcomes but requires substantial staff resources. METHODS: Participants from the SMART-AV (SmartDelay-Determined AV Optimization: Comparison of AV Optimization Methods Used in Cardiac Resynchronization Therapy [CRT]) trial (n = 741; age: 66 ± 11 years; 33% female; 100% New York Heart Association HF functional class III-IV; 100% ejection fraction ≤35%) were randomly split into training/testing (80%; n = 593) and validation (20%; n = 148) samples. Baseline clinical, electrocardiographic, echocardiographic, and biomarker characteristics, and left ventricular (LV) lead position (43 variables) were included in 8 ML models (random forests, convolutional neural network, lasso, adaptive lasso, plugin lasso, elastic net, ridge, and logistic regression). A composite of freedom from death and HF hospitalization and a >15% reduction in LV end-systolic volume index at 6 months after CRT was the end point. RESULTS: The primary end point was met by 337 patients (45.5%). The adaptive lasso model was the most more accurate (area under the receiver operating characteristic curve: 0.759; 95% CI: 0.678-0.840), well calibrated, and parsimonious (19 predictors; nearly half potentially modifiable). Participants in the 5th quintile compared with those in the 1st quintile of the prediction model had 14-fold higher odds of composite CRT response (odds ratio: 14.0; 95% CI: 8.0-14.4). The model predicted CRT response with 70% accuracy, 70% sensitivity, and 70% specificity, and should be further validated in prospective studies. CONCLUSIONS: ML predicts short-term CRT response and thus may help with CRT procedure and early post-CRT care planning. (SmartDelay-Determined AV Optimization: A Comparison of AV Optimization Methods Used in Cardiac Resynchronization Therapy [CRT] [SMART-AV]; NCT00677014).
OBJECTIVES: This study aimed to apply machine learning (ML) to develop a prediction model for short-term cardiac resynchronization therapy (CRT) response to identifying CRT candidates for early multidisciplinary CRT heart failure (HF) care. BACKGROUND: Multidisciplinary optimization of cardiac resynchronization therapy (CRT) delivery can improve long-term CRT outcomes but requires substantial staff resources. METHODS: Participants from the SMART-AV (SmartDelay-Determined AV Optimization: Comparison of AV Optimization Methods Used in Cardiac Resynchronization Therapy [CRT]) trial (n = 741; age: 66 ± 11 years; 33% female; 100% New York Heart Association HF functional class III-IV; 100% ejection fraction ≤35%) were randomly split into training/testing (80%; n = 593) and validation (20%; n = 148) samples. Baseline clinical, electrocardiographic, echocardiographic, and biomarker characteristics, and left ventricular (LV) lead position (43 variables) were included in 8 ML models (random forests, convolutional neural network, lasso, adaptive lasso, plugin lasso, elastic net, ridge, and logistic regression). A composite of freedom from death and HF hospitalization and a >15% reduction in LV end-systolic volume index at 6 months after CRT was the end point. RESULTS: The primary end point was met by 337 patients (45.5%). The adaptive lasso model was the most more accurate (area under the receiver operating characteristic curve: 0.759; 95% CI: 0.678-0.840), well calibrated, and parsimonious (19 predictors; nearly half potentially modifiable). Participants in the 5th quintile compared with those in the 1st quintile of the prediction model had 14-fold higher odds of composite CRT response (odds ratio: 14.0; 95% CI: 8.0-14.4). The model predicted CRT response with 70% accuracy, 70% sensitivity, and 70% specificity, and should be further validated in prospective studies. CONCLUSIONS: ML predicts short-term CRT response and thus may help with CRT procedure and early post-CRT care planning. (SmartDelay-Determined AV Optimization: A Comparison of AV Optimization Methods Used in Cardiac Resynchronization Therapy [CRT] [SMART-AV]; NCT00677014).
Authors: Kenneth A Ellenbogen; Michael R Gold; Timothy E Meyer; Ignacio Fernndez Lozano; Suneet Mittal; Alan D Waggoner; Bernd Lemke; Jagmeet P Singh; Francis G Spinale; Jennifer E Van Eyk; Jeffrey Whitehill; Stanislav Weiner; Maninder Bedi; Joshua Rapkin; Kenneth M Stein Journal: Circulation Date: 2010-11-15 Impact factor: 29.690
Authors: Robert K Altman; Kimberly A Parks; Christopher L Schlett; Mary Orencole; Mi-Young Park; Quynh A Truong; Peerawut Deeprasertkul; Stephanie A Moore; Conor D Barrett; Gregory D Lewis; Saumya Das; Gaurav A Upadhyay; E Kevin Heist; Michael H Picard; Jagmeet P Singh Journal: Eur Heart J Date: 2012-05-21 Impact factor: 29.983
Authors: Albert K Feeny; John Rickard; Divyang Patel; Saleem Toro; Kevin M Trulock; Carolyn J Park; Michael A LaBarbera; Niraj Varma; Mark J Niebauer; Sunil Sinha; Eiran Z Gorodeski; Richard A Grimm; Xinge Ji; John Barnard; Anant Madabhushi; David D Spragg; Mina K Chung Journal: Circ Arrhythm Electrophysiol Date: 2019-06-20
Authors: Jagmeet P Singh; Dali Fan; E Kevin Heist; Chrisfouad R Alabiad; Cynthia Taub; Vivek Reddy; Moussa Mansour; Michael H Picard; Jeremy N Ruskin; Theofanie Mela Journal: Heart Rhythm Date: 2006-08-10 Impact factor: 6.343
Authors: Wilfried Mullens; Richard A Grimm; Tanya Verga; Thomas Dresing; Randall C Starling; Bruce L Wilkoff; W H Wilson Tang Journal: J Am Coll Cardiol Date: 2009-03-03 Impact factor: 24.094
Authors: Michael R Gold; Yinghong Yu; Jagmeet P Singh; Ulrika Birgersdotter-Green; Kenneth M Stein; Nicholas Wold; Timothy E Meyer; Kenneth A Ellenbogen Journal: Circ Arrhythm Electrophysiol Date: 2018-08
Authors: Bartosz Krzowski; Jakub Rokicki; Renata Główczyńska; Nikola Fajkis-Zajączkowska; Katarzyna Barczewska; Mariusz Mąsior; Marcin Grabowski; Paweł Balsam Journal: J Cardiovasc Dev Dis Date: 2022-01-10