Redefining β-blocker response in heart failure patients with sinus rhythm and atrial fibrillation: a machine learning cluster analysis

Andreas Karwath; Karina V Bunting; Simrat K Gill; Otilia Tica; Samantha Pendleton; Furqan Aziz; Andrey D Barsky; Saisakul Chernbumroong; Jinming Duan; Alastair R Mobley; Victor Roth Cardoso; Luke Slater; John A Williams; Emma-jane Bruce; Xiaoxia Wang; Marcus D Flather; Andrew J S Coats; Georgios V Gkoutos; Dipak Kotecha

doi:10.1016/s0140-6736(21)01638-x

Redefining β-blocker response in heart failure patients with sinus rhythm and atrial fibrillation: a machine learning cluster analysis

Andreas Karwath, Karina V Bunting, Simrat K Gill, Otilia Tica, Samantha Pendleton, Furqan Aziz, Andrey D Barsky, Saisakul Chernbumroong, Jinming Duan, Alastair R Mobley, Victor Roth Cardoso, Luke Slater, John A Williams, Emma-jane Bruce, Xiaoxia Wang, Marcus D Flather, Andrew J S Coats, Georgios V Gkoutos, Dipak Kotecha

Research output: Contribution to journal › Article › peer-review

168 Downloads (Pure)

Abstract

Mortality remains unacceptably high in patients with heart failure and reduced left ventricular ejection fraction (LVEF) despite advances in therapeutics. We hypothesised that a novel artificial intelligence approach could better assess multiple and higher-dimension interactions of comorbidities, and define clusters of β-blocker efficacy in patients with sinus rhythm and atrial fibrillation. Neural network-based variational autoencoders and hierarchical clustering were applied to pooled individual patient data from nine double-blind, randomised, placebo-controlled trials of β blockers. All-cause mortality during median 1·3 years of follow-up was assessed by intention to treat, stratified by electrocardiographic heart rhythm. The number of clusters and dimensions was determined objectively, with results validated using a leave-one-trial-out approach. This study was prospectively registered with ClinicalTrials.gov (NCT00832442) and the PROSPERO database of systematic reviews (CRD42014010012). 15 659 patients with heart failure and LVEF of less than 50% were included, with median age 65 years (IQR 56-72) and LVEF 27% (IQR 21-33). 3708 (24%) patients were women. In sinus rhythm (n=12 822), most clusters demonstrated a consistent overall mortality benefit from β blockers, with odds ratios (ORs) ranging from 0·54 to 0·74. One cluster in sinus rhythm of older patients with less severe symptoms showed no significant efficacy (OR 0·86, 95% CI 0·67-1·10; p=0·22). In atrial fibrillation (n=2837), four of five clusters were consistent with the overall neutral effect of β blockers versus placebo (OR 0·92, 0·77-1·10; p=0·37). One cluster of younger atrial fibrillation patients at lower mortality risk but similar LVEF to average had a statistically significant reduction in mortality with β blockers (OR 0·57, 0·35-0·93; p=0·023). The robustness and consistency of clustering was confirmed for all models (p

Original language	English
Pages (from-to)	1-9
Number of pages	9
Journal	The Lancet
Volume	2021
Issue number	10309
Early online date	30 Aug 2021
DOIs	https://doi.org/10.1016/s0140-6736(21)01638-x
Publication status	E-pub ahead of print - 30 Aug 2021

Keywords

Artificial intelligence
Autoencoders
Clustering
Beta-blockers

ASJC Scopus subject areas

Artificial Intelligence
Cardiology and Cardiovascular Medicine

Access to Document

10.1016/s0140-6736(21)01638-xLicence: Creative Commons: Attribution (CC BY)

KarwathA2021RedefiningFinal published version, 1.42 MBLicence: Creative Commons: Attribution (CC BY)

1 Finished
1 Active

BHF Research Accelerator Award
Bicknell, R., Watson, S., Rainger, E., Senis, Y., Fabritz, L. & Kirchhof, P.
British Heart Foundation
1/04/19 → 30/09/24
Project: Research
Andreas Karwath - Semantic-based secure infrastructure for interoperable, in-depth biomedical data and analytics
Cazier, J., Karwath, A., Yau, C. & Gkoutos, G.
Medical Research Council
14/02/18 → 13/02/21
Project: Research Councils

Cite this

Karwath, A., Bunting, K. V., Gill, S. K., Tica, O., Pendleton, S., Aziz, F., Barsky, A. D., Chernbumroong, S., Duan, J., Mobley, A. R., Cardoso, V. R., Slater, L., Williams, J. A., Bruce, E., Wang, X., Flather, M. D., Coats, A. J. S., Gkoutos, G. V., & Kotecha, D. (2021). Redefining β-blocker response in heart failure patients with sinus rhythm and atrial fibrillation: a machine learning cluster analysis. The Lancet, 2021(10309), 1-9. Advance online publication. https://doi.org/10.1016/s0140-6736(21)01638-x

@article{7ce02ae32f2a4b1d8ab40d388d4004ec,

title = "Redefining β-blocker response in heart failure patients with sinus rhythm and atrial fibrillation: a machine learning cluster analysis",

abstract = "Mortality remains unacceptably high in patients with heart failure and reduced left ventricular ejection fraction (LVEF) despite advances in therapeutics. We hypothesised that a novel artificial intelligence approach could better assess multiple and higher-dimension interactions of comorbidities, and define clusters of β-blocker efficacy in patients with sinus rhythm and atrial fibrillation. Neural network-based variational autoencoders and hierarchical clustering were applied to pooled individual patient data from nine double-blind, randomised, placebo-controlled trials of β blockers. All-cause mortality during median 1·3 years of follow-up was assessed by intention to treat, stratified by electrocardiographic heart rhythm. The number of clusters and dimensions was determined objectively, with results validated using a leave-one-trial-out approach. This study was prospectively registered with ClinicalTrials.gov (NCT00832442) and the PROSPERO database of systematic reviews (CRD42014010012). 15 659 patients with heart failure and LVEF of less than 50% were included, with median age 65 years (IQR 56-72) and LVEF 27% (IQR 21-33). 3708 (24%) patients were women. In sinus rhythm (n=12 822), most clusters demonstrated a consistent overall mortality benefit from β blockers, with odds ratios (ORs) ranging from 0·54 to 0·74. One cluster in sinus rhythm of older patients with less severe symptoms showed no significant efficacy (OR 0·86, 95% CI 0·67-1·10; p=0·22). In atrial fibrillation (n=2837), four of five clusters were consistent with the overall neutral effect of β blockers versus placebo (OR 0·92, 0·77-1·10; p=0·37). One cluster of younger atrial fibrillation patients at lower mortality risk but similar LVEF to average had a statistically significant reduction in mortality with β blockers (OR 0·57, 0·35-0·93; p=0·023). The robustness and consistency of clustering was confirmed for all models (p",

keywords = "Artificial intelligence, Autoencoders, Clustering, Beta-blockers",

author = "Andreas Karwath and Bunting, {Karina V} and Gill, {Simrat K} and Otilia Tica and Samantha Pendleton and Furqan Aziz and Barsky, {Andrey D} and Saisakul Chernbumroong and Jinming Duan and Mobley, {Alastair R} and Cardoso, {Victor Roth} and Luke Slater and Williams, {John A} and Emma-jane Bruce and Xiaoxia Wang and Flather, {Marcus D} and Coats, {Andrew J S} and Gkoutos, {Georgios V} and Dipak Kotecha",

year = "2021",

month = aug,

day = "30",

doi = "10.1016/s0140-6736(21)01638-x",

language = "English",

volume = "2021",

pages = "1--9",

journal = "The Lancet",

issn = "0140-6736",

publisher = "Elsevier",

number = "10309",

}

Karwath, A , Bunting, KV, Gill, SK, Tica, O, Pendleton, S, Aziz, F, Barsky, AD, Chernbumroong, S, Duan, J , Mobley, AR , Cardoso, VR , Slater, L , Williams, JA, Bruce, E, Wang, X, Flather, MD, Coats, AJS, Gkoutos, GV & Kotecha, D 2021, 'Redefining β-blocker response in heart failure patients with sinus rhythm and atrial fibrillation: a machine learning cluster analysis', The Lancet, vol. 2021, no. 10309, pp. 1-9. https://doi.org/10.1016/s0140-6736(21)01638-x

TY - JOUR

T1 - Redefining β-blocker response in heart failure patients with sinus rhythm and atrial fibrillation

T2 - a machine learning cluster analysis

AU - Karwath, Andreas

AU - Bunting, Karina V

AU - Gill, Simrat K

AU - Tica, Otilia

AU - Pendleton, Samantha

AU - Aziz, Furqan

AU - Barsky, Andrey D

AU - Chernbumroong, Saisakul

AU - Duan, Jinming

AU - Mobley, Alastair R

AU - Cardoso, Victor Roth

AU - Slater, Luke

AU - Williams, John A

AU - Bruce, Emma-jane

AU - Wang, Xiaoxia

AU - Flather, Marcus D

AU - Coats, Andrew J S

AU - Gkoutos, Georgios V

AU - Kotecha, Dipak

PY - 2021/8/30

Y1 - 2021/8/30

N2 - Mortality remains unacceptably high in patients with heart failure and reduced left ventricular ejection fraction (LVEF) despite advances in therapeutics. We hypothesised that a novel artificial intelligence approach could better assess multiple and higher-dimension interactions of comorbidities, and define clusters of β-blocker efficacy in patients with sinus rhythm and atrial fibrillation. Neural network-based variational autoencoders and hierarchical clustering were applied to pooled individual patient data from nine double-blind, randomised, placebo-controlled trials of β blockers. All-cause mortality during median 1·3 years of follow-up was assessed by intention to treat, stratified by electrocardiographic heart rhythm. The number of clusters and dimensions was determined objectively, with results validated using a leave-one-trial-out approach. This study was prospectively registered with ClinicalTrials.gov (NCT00832442) and the PROSPERO database of systematic reviews (CRD42014010012). 15 659 patients with heart failure and LVEF of less than 50% were included, with median age 65 years (IQR 56-72) and LVEF 27% (IQR 21-33). 3708 (24%) patients were women. In sinus rhythm (n=12 822), most clusters demonstrated a consistent overall mortality benefit from β blockers, with odds ratios (ORs) ranging from 0·54 to 0·74. One cluster in sinus rhythm of older patients with less severe symptoms showed no significant efficacy (OR 0·86, 95% CI 0·67-1·10; p=0·22). In atrial fibrillation (n=2837), four of five clusters were consistent with the overall neutral effect of β blockers versus placebo (OR 0·92, 0·77-1·10; p=0·37). One cluster of younger atrial fibrillation patients at lower mortality risk but similar LVEF to average had a statistically significant reduction in mortality with β blockers (OR 0·57, 0·35-0·93; p=0·023). The robustness and consistency of clustering was confirmed for all models (p

AB - Mortality remains unacceptably high in patients with heart failure and reduced left ventricular ejection fraction (LVEF) despite advances in therapeutics. We hypothesised that a novel artificial intelligence approach could better assess multiple and higher-dimension interactions of comorbidities, and define clusters of β-blocker efficacy in patients with sinus rhythm and atrial fibrillation. Neural network-based variational autoencoders and hierarchical clustering were applied to pooled individual patient data from nine double-blind, randomised, placebo-controlled trials of β blockers. All-cause mortality during median 1·3 years of follow-up was assessed by intention to treat, stratified by electrocardiographic heart rhythm. The number of clusters and dimensions was determined objectively, with results validated using a leave-one-trial-out approach. This study was prospectively registered with ClinicalTrials.gov (NCT00832442) and the PROSPERO database of systematic reviews (CRD42014010012). 15 659 patients with heart failure and LVEF of less than 50% were included, with median age 65 years (IQR 56-72) and LVEF 27% (IQR 21-33). 3708 (24%) patients were women. In sinus rhythm (n=12 822), most clusters demonstrated a consistent overall mortality benefit from β blockers, with odds ratios (ORs) ranging from 0·54 to 0·74. One cluster in sinus rhythm of older patients with less severe symptoms showed no significant efficacy (OR 0·86, 95% CI 0·67-1·10; p=0·22). In atrial fibrillation (n=2837), four of five clusters were consistent with the overall neutral effect of β blockers versus placebo (OR 0·92, 0·77-1·10; p=0·37). One cluster of younger atrial fibrillation patients at lower mortality risk but similar LVEF to average had a statistically significant reduction in mortality with β blockers (OR 0·57, 0·35-0·93; p=0·023). The robustness and consistency of clustering was confirmed for all models (p

KW - Artificial intelligence

KW - Autoencoders

KW - Clustering

KW - Beta-blockers

UR - http://www.scopus.com/inward/record.url?scp=85117131049&partnerID=8YFLogxK

U2 - 10.1016/s0140-6736(21)01638-x

DO - 10.1016/s0140-6736(21)01638-x

M3 - Article

C2 - 34474011

SN - 0140-6736

VL - 2021

SP - 1

EP - 9

JO - The Lancet

JF - The Lancet

IS - 10309

ER -

Redefining β-blocker response in heart failure patients with sinus rhythm and atrial fibrillation: a machine learning cluster analysis

Abstract

Keywords

ASJC Scopus subject areas

Access to Document

Fingerprint

Projects

BHF Research Accelerator Award

Andreas Karwath - Semantic-based secure infrastructure for interoperable, in-depth biomedical data and analytics

Cite this