Phenomapping of patients with heart failure with preserved ejection fraction using machine learning-based unsupervised cluster analysis

Matthew W. Segar, Kershaw V. Patel, Colby Ayers, Mujeeb Basit, W. H.Wilson Tang, Duwayne Willett, Jarett Berry, Justin L. Grodin, Ambarish Pandey

Research output: Contribution to journalArticle

1 Citation (Scopus)

Abstract

Aim: To identify distinct phenotypic subgroups in a highly-dimensional, mixed-data cohort of individuals with heart failure (HF) with preserved ejection fraction (HFpEF) using unsupervised clustering analysis. Methods and results: The study included all Treatment of Preserved Cardiac Function Heart Failure with an Aldosterone Antagonist (TOPCAT) participants from the Americas (n = 1767). In the subset of participants with available echocardiographic data (derivation cohort, n = 654), we characterized three mutually exclusive phenogroups of HFpEF participants using penalized finite mixture model-based clustering analysis on 61 mixed-data phenotypic variables. Phenogroup 1 had higher burden of co-morbidities, natriuretic peptides, and abnormalities in left ventricular structure and function; phenogroup 2 had lower prevalence of cardiovascular and non-cardiac co-morbidities but higher burden of diastolic dysfunction; and phenogroup 3 had lower natriuretic peptide levels, intermediate co-morbidity burden, and the most favourable diastolic function profile. In adjusted Cox models, participants in phenogroup 1 (vs. phenogroup 3) had significantly higher risk for all adverse clinical events including the primary composite endpoint, all-cause mortality, and HF hospitalization. Phenogroup 2 (vs. phenogroup 3) was significantly associated with higher risk of HF hospitalization but a lower risk of atherosclerotic event (myocardial infarction, stroke, or cardiovascular death), and comparable risk of mortality. Similar patterns of association were also observed in the non-echocardiographic TOPCAT cohort (internal validation cohort, n = 1113) and an external cohort of patients with HFpEF [Phosphodiesterase-5 Inhibition to Improve Clinical Status and Exercise Capacity in Heart Failure with Preserved Ejection Fraction (RELAX) trial cohort, n = 198], with the highest risk of adverse outcome noted in phenogroup 1 participants. Conclusions: Machine learning-based cluster analysis can identify phenogroups of patients with HFpEF with distinct clinical characteristics and long-term outcomes.

Original languageEnglish (US)
JournalEuropean Journal of Heart Failure
DOIs
StateAccepted/In press - Jan 1 2019

Fingerprint

Cluster Analysis
Heart Failure
Natriuretic Peptides
Morbidity
Hospitalization
Myocardial Infarction
Type 5 Cyclic Nucleotide Phosphodiesterases
Mineralocorticoid Receptor Antagonists
Mortality
Left Ventricular Function
Proportional Hazards Models
Unsupervised Machine Learning
Exercise
Therapeutics

Keywords

  • Heart failure with preserved ejection fraction
  • Machine learning
  • Outcomes
  • Phenomapping

ASJC Scopus subject areas

  • Cardiology and Cardiovascular Medicine

Cite this

@article{d782bc196f6447e096365e1383caa4db,
title = "Phenomapping of patients with heart failure with preserved ejection fraction using machine learning-based unsupervised cluster analysis",
abstract = "Aim: To identify distinct phenotypic subgroups in a highly-dimensional, mixed-data cohort of individuals with heart failure (HF) with preserved ejection fraction (HFpEF) using unsupervised clustering analysis. Methods and results: The study included all Treatment of Preserved Cardiac Function Heart Failure with an Aldosterone Antagonist (TOPCAT) participants from the Americas (n = 1767). In the subset of participants with available echocardiographic data (derivation cohort, n = 654), we characterized three mutually exclusive phenogroups of HFpEF participants using penalized finite mixture model-based clustering analysis on 61 mixed-data phenotypic variables. Phenogroup 1 had higher burden of co-morbidities, natriuretic peptides, and abnormalities in left ventricular structure and function; phenogroup 2 had lower prevalence of cardiovascular and non-cardiac co-morbidities but higher burden of diastolic dysfunction; and phenogroup 3 had lower natriuretic peptide levels, intermediate co-morbidity burden, and the most favourable diastolic function profile. In adjusted Cox models, participants in phenogroup 1 (vs. phenogroup 3) had significantly higher risk for all adverse clinical events including the primary composite endpoint, all-cause mortality, and HF hospitalization. Phenogroup 2 (vs. phenogroup 3) was significantly associated with higher risk of HF hospitalization but a lower risk of atherosclerotic event (myocardial infarction, stroke, or cardiovascular death), and comparable risk of mortality. Similar patterns of association were also observed in the non-echocardiographic TOPCAT cohort (internal validation cohort, n = 1113) and an external cohort of patients with HFpEF [Phosphodiesterase-5 Inhibition to Improve Clinical Status and Exercise Capacity in Heart Failure with Preserved Ejection Fraction (RELAX) trial cohort, n = 198], with the highest risk of adverse outcome noted in phenogroup 1 participants. Conclusions: Machine learning-based cluster analysis can identify phenogroups of patients with HFpEF with distinct clinical characteristics and long-term outcomes.",
keywords = "Heart failure with preserved ejection fraction, Machine learning, Outcomes, Phenomapping",
author = "Segar, {Matthew W.} and Patel, {Kershaw V.} and Colby Ayers and Mujeeb Basit and Tang, {W. H.Wilson} and Duwayne Willett and Jarett Berry and Grodin, {Justin L.} and Ambarish Pandey",
year = "2019",
month = "1",
day = "1",
doi = "10.1002/ejhf.1621",
language = "English (US)",
journal = "European Journal of Heart Failure",
issn = "1388-9842",
publisher = "Oxford University Press",

}

TY - JOUR

T1 - Phenomapping of patients with heart failure with preserved ejection fraction using machine learning-based unsupervised cluster analysis

AU - Segar, Matthew W.

AU - Patel, Kershaw V.

AU - Ayers, Colby

AU - Basit, Mujeeb

AU - Tang, W. H.Wilson

AU - Willett, Duwayne

AU - Berry, Jarett

AU - Grodin, Justin L.

AU - Pandey, Ambarish

PY - 2019/1/1

Y1 - 2019/1/1

N2 - Aim: To identify distinct phenotypic subgroups in a highly-dimensional, mixed-data cohort of individuals with heart failure (HF) with preserved ejection fraction (HFpEF) using unsupervised clustering analysis. Methods and results: The study included all Treatment of Preserved Cardiac Function Heart Failure with an Aldosterone Antagonist (TOPCAT) participants from the Americas (n = 1767). In the subset of participants with available echocardiographic data (derivation cohort, n = 654), we characterized three mutually exclusive phenogroups of HFpEF participants using penalized finite mixture model-based clustering analysis on 61 mixed-data phenotypic variables. Phenogroup 1 had higher burden of co-morbidities, natriuretic peptides, and abnormalities in left ventricular structure and function; phenogroup 2 had lower prevalence of cardiovascular and non-cardiac co-morbidities but higher burden of diastolic dysfunction; and phenogroup 3 had lower natriuretic peptide levels, intermediate co-morbidity burden, and the most favourable diastolic function profile. In adjusted Cox models, participants in phenogroup 1 (vs. phenogroup 3) had significantly higher risk for all adverse clinical events including the primary composite endpoint, all-cause mortality, and HF hospitalization. Phenogroup 2 (vs. phenogroup 3) was significantly associated with higher risk of HF hospitalization but a lower risk of atherosclerotic event (myocardial infarction, stroke, or cardiovascular death), and comparable risk of mortality. Similar patterns of association were also observed in the non-echocardiographic TOPCAT cohort (internal validation cohort, n = 1113) and an external cohort of patients with HFpEF [Phosphodiesterase-5 Inhibition to Improve Clinical Status and Exercise Capacity in Heart Failure with Preserved Ejection Fraction (RELAX) trial cohort, n = 198], with the highest risk of adverse outcome noted in phenogroup 1 participants. Conclusions: Machine learning-based cluster analysis can identify phenogroups of patients with HFpEF with distinct clinical characteristics and long-term outcomes.

AB - Aim: To identify distinct phenotypic subgroups in a highly-dimensional, mixed-data cohort of individuals with heart failure (HF) with preserved ejection fraction (HFpEF) using unsupervised clustering analysis. Methods and results: The study included all Treatment of Preserved Cardiac Function Heart Failure with an Aldosterone Antagonist (TOPCAT) participants from the Americas (n = 1767). In the subset of participants with available echocardiographic data (derivation cohort, n = 654), we characterized three mutually exclusive phenogroups of HFpEF participants using penalized finite mixture model-based clustering analysis on 61 mixed-data phenotypic variables. Phenogroup 1 had higher burden of co-morbidities, natriuretic peptides, and abnormalities in left ventricular structure and function; phenogroup 2 had lower prevalence of cardiovascular and non-cardiac co-morbidities but higher burden of diastolic dysfunction; and phenogroup 3 had lower natriuretic peptide levels, intermediate co-morbidity burden, and the most favourable diastolic function profile. In adjusted Cox models, participants in phenogroup 1 (vs. phenogroup 3) had significantly higher risk for all adverse clinical events including the primary composite endpoint, all-cause mortality, and HF hospitalization. Phenogroup 2 (vs. phenogroup 3) was significantly associated with higher risk of HF hospitalization but a lower risk of atherosclerotic event (myocardial infarction, stroke, or cardiovascular death), and comparable risk of mortality. Similar patterns of association were also observed in the non-echocardiographic TOPCAT cohort (internal validation cohort, n = 1113) and an external cohort of patients with HFpEF [Phosphodiesterase-5 Inhibition to Improve Clinical Status and Exercise Capacity in Heart Failure with Preserved Ejection Fraction (RELAX) trial cohort, n = 198], with the highest risk of adverse outcome noted in phenogroup 1 participants. Conclusions: Machine learning-based cluster analysis can identify phenogroups of patients with HFpEF with distinct clinical characteristics and long-term outcomes.

KW - Heart failure with preserved ejection fraction

KW - Machine learning

KW - Outcomes

KW - Phenomapping

UR - http://www.scopus.com/inward/record.url?scp=85074328619&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=85074328619&partnerID=8YFLogxK

U2 - 10.1002/ejhf.1621

DO - 10.1002/ejhf.1621

M3 - Article

C2 - 31637815

AN - SCOPUS:85074328619

JO - European Journal of Heart Failure

JF - European Journal of Heart Failure

SN - 1388-9842

ER -