Key result
Supra-normal left ventricular ejection fraction and its genetic risk score were associated with an increased risk of all-cause and cardiovascular mortality.
Answers are grounded in this paper and cite their sources.
This editorial highlights that heart failure with supra-normal ejection fraction (snLVEF >65%) is a distinct clinical and polygenic entity associated with increased mortality and likely underdiagnosed.
This article refers to ‘Genetic and phenotypic profiling of supranormal ejection fraction reveals decreased survival and underdiagnosed heart failure’ by I.S. Forrest et al., published in this issue on pages 2118–2127. In the last years, a fourth category of heart failure (HF), characterized by supra-normal left ventricular ejection fraction (LVEF) >65% (snLVEF), termed HF with snLVEF (HFsnEF), made its debut in the clinical arena. The first, population-based, insight into this unexplored clinical entity was established in the milestone paper of Wehner et al.,1 elucidating an increased risk of all-cause mortality for LVEF values above the normal range (LVEF 55–65%). While drafting this editorial, PubMed lists only 38 works on snLVEF, most of them dating after 2020. This means that, actually, snLVEF is a category of patients poorly characterized by scientific reports and not recognized by HF international guidelines,2, 3 although, apparently, associated with a similar prognosis. The U-shaped relationship between mortality and LVEF, with LVEF of 60–65% at the nadir of the curve, also suggests that patients with preserved LVEF should not be considered as a single entity.1, 4 Further indirect evidence comes from the results of the EMPEROR-Preserved trial, where a lower benefit of empagliflozin was demonstrated in the subgroup with snLVEF.5 Moreover, the increased risk of all-cause mortality of individuals with snLVEF, confirmed also after adjusting for age, sex, height, body mass index, and multiple comorbidities, was independent from HF diagnosis and comparable to patients with reduced LVEF.1 In that work, however, authors hypothesized a likely underdiagnosis of HF.1 All these reasons justify and support the primary necessity of the scientific community to deeply investigate the phenotypic manifestation and genetic background of patients with snLVEF, and their roles in defining prognosis. In this issue of the Journal, Forrest and colleagues begin to shed some light on this field.6 Common disorders with a major public health impact, such as HF, share a complex aetiology in which environmental factors combine with predisposing genetic background in determining clinical phenotype. Genome-wide association studies (GWAS) on large biobanks elucidated the genetic loci associated with most of these complex diseases (HF, coronary artery disease [CAD], cancer, atrial fibrillation, and so forth), unravelling their polygenic inheritance. Indeed, this form of inheritance has been recognized as the most prevalent, playing an epidemiological impact greater than rare monogenic mutations.7, 8 Since a unique, common, single nucleotide polymorphism (SNP) is not sufficient to determine the disease of interest, this inheritance involves the combination of multiple common SNPs (from dozen to millions). A weighted sum of all SNPs (as risk alleles) detected in a patient will generate its polygenic risk score, also called genetic risk score (GRS).8 The higher the number of risk alleles inherited, the higher the risk of developing the disease. In addition, loci identified with GWAS could help in discovering the Mendelian bases of the disease of interest, since usually they are enriched for regions of DNA containing Mendelian genes directly causing the phenotype.9, 10 Furthermore, GWAS findings on complex cardiovascular diseases are deeply interconnected. For example, CAD-GRS, created for detecting the genetic susceptibility to develop CAD, results to be highly and significantly associated with susceptibility to common CAD risk factors such as hypertension and hypercholesterolaemia.11 In a large meta-analysis of GWAS for HF phenotype, the multiple loci identified resulted to be linked with the principal HF predisposing phenotypes, such as atrial fibrillation, diastolic blood pressure, high body mass index, dilated cardiomyopathy.12 The same could be said for dilated and hypertrophic cardiomyopathies, in which opposite risk alleles in shared genetic loci confer the risk for the two opposite phenotypes.10, 13 In other words, similar diseases share similar genetic backgrounds between each other and between their predisposing phenotypes, whereas a unique genetic footprint is highly suggestive for a distinct disease. Forrest and colleagues exploited all this previous knowledge on polygenic heritability of common cardiac disease to perform the first GWAS on the new category of snLVEF patients.6 These analyses were performed across two large population-based cohorts (UK Biobank – mainly healthy volunteers – and BioMe Biobank – mainly outpatients). Firstly, the authors generated a GRS from a GWAS of snLVEF on a subpopulation with available magnetic resonance imaging or echocardiographic data (30 557 individuals, of which 7548 with snLVEF >70%). This allowed to identify 16 specific, significant and independent lead SNPs associated with snLVEF that were incorporated in the GRS, providing the first important result of this study. Some of the selected SNPs were in loci already described as associated with (i) subclinical changes in left ventricular structure and function and increased risk of HF development, such as BAG3, TTN and SLC6A6,13-15 or (ii) associated with cardiac growth (GATA4, MITF). However, the most interesting SNPs are the ones in loci not previously reported in other cardiac diseases, such as SEC23IP and XKR6. The SNP-based heritability was 30%. In conclusion, the results of GWAS strongly validate snLVEF as a distinct cardiac disease, with its own genetic footprint, paving the way for new specific research. Moreover, when snLVEF GRS was finally applied to the entire population of both biobanks (486 754 individuals) and compared with an already published GRS for reduced LVEF,13 an increased risk of all-cause mortality and cardiovascular mortality was detected for both supra-normal and reduced GRS, confirming the different genetic bases of the U-shape of mortality.1 As the authors stated, this is a strong point of the study, since the genetic background is less likely influenced by confounding factors, compared to phenotype. In this study,6 clinical phenotypes were obtained from electronic records. The authors acknowledged this as the main limitation, which is intrinsic to epidemiological studies. Considering this, other important results of this work are represented by the insights provided into the phenotype(s) of this population. Both snLVEF patients and snLVEF-GRS were associated with HF symptoms and medications, and elevated HF biomarkers, but not with HF diagnosis, supporting the hypothesis of its underdiagnosis. These findings provide clues on a possible prevalent phenotype. Other unknown features, however, may characterize patients with snLVEF, as HF medications, when present, only mildly attenuate the association with mortality.6 Clinical cardiologists in fact may not feel themselves at ease when trying to identify these patients among those normally encountered in a HF clinic. Clinicians are used to consider snLVEF not associated with cardiac disease, and, consequently, HF may be underdiagnosed, and these patients are not (or rarely) referred to cardiologists, nor frequently treated with HF medications.16, 17 When thinking about ‘classical’ and recognized clinical models of snLVEF, an expert physician may firstly consider mitral insufficiency, hypertrophic cardiomyopathy, amyloidosis, aortic stenosis with low gradient, hypertension (in older women), or high-output states, or alternatively acute illness such as sepsis or severe anaemia.12 However, the growing amount of evidence clearly shows, actually, that these diseases are not associated with, or at least do not represent the majority of, snLVEF phenotypes and snLVEF-GRS.1, 4, 6 Coming back to physiology, the mechanisms underlining snLVEF are still only superficially explored and clarified. Recent investigation showed an interesting U-curve existing between LVEF and coronary flow reserve, with a higher incidence of reduced coronary flow reserve in LVEF <55% and in snLVEF patients compared to patients with preserved LVEF, both related to poor prognosis in terms of major adverse cardiovascular events (MACE).18 Shah et al.4 just published on higher risk of MACE detected also in a different large cohort of snLVEF outpatients without previous cardiovascular disease, particularly in those with a lower indexed stroke volume. These findings point out how the pathophysiology of this entity could be strongly related to a disproportionately small left ventricular cavity, which forces the heart to squeeze above normal value during systole, and thereby, possibly, increasing its metabolic demand.19 In this view, the associations with hypertension and obesity described by Forrest et al.,6 together with the genetic footprint involving genes regulatory for cardiac growth and hypertrophy, could be considered confirmatory. Finally, it is not negligible that the association of snLVEF with mortality (cardiovascular) is identified also in patients without (known) hypertension or (known) HF. Hence, for cardiologists involved in cardiomyopathies and arrhythmias, it may be speculated also on a possible relationship between snLVEF and arrhythmic death. Currently, no association between snLVEF and arrhythmias has been investigated, but according to all these new evidences, this question, among others, may arise spontaneously. Do patients with snLVEF really exist as a separate clinical entity? Which are the main causes of mortality in these individuals? What is their prevalent phenotype? Is it HFsnEF? Definitely worth investigating this exciting population with a deep phenotypic characterization. Conflict of interest: none declared.
No takes yet. Share an insight, caveat, or question.
Sinagra et al. (2022) conducted an editorial in Heart failure with supra-normal left ventricular ejection fraction (n=486,754). Supra-normal left ventricular ejection fraction (snLVEF) vs. Normal or reduced LVEF was evaluated on All-cause and cardiovascular mortality. Supra-normal left ventricular ejection fraction and its genetic risk score were associated with an increased risk of all-cause and cardiovascular mortality.
Synapse has enriched 10 related papers. Consider them for comparative context: