Defining the Phenotypes for Heart Failure With Preserved Ejection Fraction.
Dane RuckerJacob JosephPublished in: Current heart failure reports (2022)
A few studies have used machine learning methods to uncover sub-phenotypes within HFpEF in an unbiased manner based on clinical features, echocardiographic findings, and biomarker levels. We synthesized the literature and propose three broad phenotypes: (1) young, with few comorbidities, usually obese and with low natriuretic peptide levels, (2) obese with substantive cardiometabolic burden and comorbidities and impaired ventricular relaxation, (3) old, multimorbid, with high rates of atrial fibrillation, renal and coronary artery disease, chronic obstructive pulmonary disease, and left ventricular hypertrophy. We also propose potential therapeutic strategies for these phenotypes.
Keyphrases
- left ventricular
- chronic obstructive pulmonary disease
- machine learning
- coronary artery disease
- heart failure
- left atrial
- atrial fibrillation
- adipose tissue
- weight loss
- metabolic syndrome
- type diabetes
- catheter ablation
- systematic review
- mitral valve
- percutaneous coronary intervention
- acute myocardial infarction
- hypertrophic cardiomyopathy
- bariatric surgery
- aortic stenosis
- risk factors
- acute coronary syndrome
- middle aged
- deep learning
- climate change
- oxide nanoparticles
- aortic valve