Maladaptive lipid partitioning in obesity-related heart failure with preserved ejection fraction: From depot dysfunction to metabolic phenotyping

Abstract

Heart failure with preserved ejection fraction is a heterogeneous syndrome, with obesity defining an increasingly treatable cardiometabolic phenotype. Building on recent work identifying dysfunctional adiposity as an upstream contributor, this minireview focuses on the complementary question of how failure of safe lipid storage redirects substrate across visceral, hepatic, skeletal muscle, epicardial, perivascular, and myocardial compartments, and how this depot-specific routing intersects with myocardial metabolic inflexibility. Human imaging, invasive physiology, myocardial metabolomics, and recent clinical trials support associations between ectopic adiposity, insulin resistance, inflammation, impaired vascular and atrial reserve, congestion, and exercise intolerance, although causal pathways remain incompletely established in humans. We therefore propose an adiposity-aware, hypothesis-generating framework integrating body composition, regional adipose depots, metabolic-inflammatory biomarkers, atrial and right-sided function, and exercise assessment for phenotype-enriched research and future trial design. This framework is not intended to guide current treatment selection. Established treatment should remain evidence based, including exercise, sodium-glucose cotransporter 2 inhibition, mineralocorticoid receptor antagonism, decongestion, and obesity-directed incretin therapy, with attention to lean-mass preservation. Future prospective studies should investigate whether depot quality, metabolic flexibility, sex and ethnic heterogeneity, and body composition provide additional information for phenotypic characterization and risk stratification beyond conventional measures.

Key Words: Heart failure with preserved ejection fraction; Obesity; Lipid partitioning; Epicardial adipose tissue; Metabolic inflexibility; Body composition

Core Tip: Obesity-related heart failure with preserved ejection fraction cannot be adequately characterized by body mass alone. This minireview focuses on the less frequently explored consequences of failed lipid storage: Depot-specific lipid rerouting across visceral, hepatic, skeletal muscle, epicardial/perivascular, and myocardial compartments, together with impaired myocardial substrate flexibility. We distinguish human associations from mechanistic evidence and propose adiposity-aware phenotyping as a research framework rather than a validated treatment algorithm. Recent cardiometabolic trials illustrate how complementary components of this phenotype may be therapeutically modifiable.

INTRODUCTION

Heart failure with preserved ejection fraction (HFpEF) is a heterogeneous clinical syndrome that cannot be adequately characterized by left ventricular ejection fraction alone. Contemporary scientific statements and expert consensus documents characterize HFpEF within a broader multisystem cardiometabolic context shaped by interacting cardiac, vascular, renal, pulmonary, inflammatory, and metabolic abnormalities[1-3]. Within this heterogeneity, obesity-related HFpEF has emerged as a clinically and physiologically recognizable cardiometabolic phenotype. A foundational invasive hemodynamic study demonstrated that patients with obesity-related HFpEF exhibit greater plasma volume expansion, more pronounced concentric left ventricular remodeling, increased right-sided chamber abnormalities, greater epicardial adiposity, and impaired cardiovascular reserve despite comparatively lower natriuretic peptide concentrations[4]. These observations support the concept that obesity modifies the pathophysiology and clinical expression of HFpEF rather than simply coexisting with it.

However, the biologically relevant exposure cannot be inferred from body mass index (BMI) alone because patients with similar BMI may have substantial differences in visceral, hepatic, skeletal muscle, and epicardial adipose tissue (EAT) distributions. Sex and population context further modify adipose distribution and its clinical expression in HFpEF[5,6]. Contemporary guidance therefore supports complementary assessment of adiposity and body composition beyond BMI alone[7].

Maladaptive lipid partitioning provides a mechanistic framework for extending this concept from the amount of adipose tissue to the destination and handling of excess substrate. Recent work has emphasized dysfunctional visceral adiposity as an important upstream contributor to cardiometabolic HFpEF[8]. The present review focuses on the complementary downstream question of what occurs when physiological lipid storage becomes inadequate or dysfunctional. Excess substrate may be redistributed among visceral adipose tissue, liver, skeletal muscle, epicardial and perivascular adipose tissues (PVATs), and the myocardium, where depot-specific abnormalities may coexist with metabolic, inflammatory, vascular, and mechanical stress. Importantly, this framework should not be interpreted as a fully established linear causal pathway in humans. Human myocardial metabolomics instead points to a more complex disturbance in substrate handling. Despite marked obesity and diabetes, the HFpEF myocardium is characterized by lower medium-chain and long-chain acylcarnitines, together with abnormalities in pyruvate handling, tricarboxylic acid cycle intermediates, and branched-chain amino acid metabolism, supporting impaired metabolic flexibility rather than simple myocardial fatty acid excess[9]. These findings shift attention from lipid quantity alone toward the coordination of substrate delivery, storage, oxidation, and metabolic adaptation.

Randomized obesity-directed trials support the clinical modifiability of obesity-related HFpEF, although the mechanisms underlying clinical benefit remain incompletely resolved[10,11]. Against this background, our distinct emphasis is on the less developed downstream link between depot-specific lipid redistribution and altered myocardial substrate handling in humans. Recent work has emphasized dysfunctional adiposity as an important upstream contributor to obesity-related HFpEF[8], while a recent review has highlighted the metabolic dysfunction-associated steatotic liver disease-HFpEF (MASLD-HFpEF) overlap[12]. We examine how failure of physiological lipid storage may be accompanied by depot-specific lipid redistribution, how extracardiac adipose and organ phenotypes intersect with human evidence of myocardial metabolic inflexibility, and which of these abnormalities can currently be measured in clinical or translational studies. We further distinguish established human associations and randomized treatment effects from preclinical mechanistic evidence and consider how measurable adiposity-related characteristics might be incorporated into phenotype-enriched research. The proposed framework is intended to generate testable hypotheses and inform future trial design rather than to serve as a validated clinical treatment-selection algorithm (Figure 1).

Figure 1 Maladaptive lipid-partitioning framework in obesity-related heart failure with preserved ejection fraction. Human studies support depot-specific adiposity, myocardial metabolic abnormalities, and selected clinical associations, whereas several linking mechanisms remain unproven in humans. Solid arrows denote reported human observational or interventional associations and do not imply a single proven causal sequence. Dashed arrows denote hypothesis-generating mechanistic links without proven human causality. VAT: Visceral adipose tissue; MASLD: Metabolic dysfunction-associated steatotic liver disease; EAT: Epicardial adipose tissue; PVAT: Perivascular adipose tissue; TNF-α: Tumor necrosis factor-α; IL-6: Interleukin-6; MCP-1: Monocyte chemoattractant protein-1; DAG: Diacylglycerol; BCAA: Branched-chain amino acid; TCA: Tricarboxylic acid; HFpEF: Heart failure with preserved ejection fraction.

EVIDENCE ACQUISITION AND SCOPE

This invited narrative minireview was informed by a targeted PubMed search updated through August 2026 and by manual review of key references from major clinical trials, mechanistic studies, imaging studies, metabolomic analyses, contemporary consensus documents, and relevant reviews. PubMed was the sole bibliographic database searched; EMBASE and the Cochrane Library were not systematically searched.

Search terms combined “HFpEF” or “heart failure with preserved ejection fraction” with “obesity”, “adiposity”, “insulin resistance”, “lipid metabolism”, “lipotoxicity”, “metabolomics”, “epicardial adipose tissue”, “perivascular adipose tissue”, “visceral adiposity”, “metabolic dysfunction-associated steatotic liver disease”, “skeletal muscle fat”, “exercise intolerance”, “sodium-glucose cotransporter 2 inhibitors”, “mineralocorticoid receptor antagonists”, “semaglutide”, “tirzepatide”, and “incretin-based therapy”. During revision, a limited number of additional terms were used to address topics specifically raised during peer review, including “sex”, “central obesity”, “H2FPEF”, “HFA-PEFF”, “finerenone”, “bariatric surgery”, “atrial fibrillation”, and “sleep-disordered breathing”.

Priority was given to original human studies, randomized clinical trials, prospective cohorts, invasive hemodynamic studies, imaging and metabolomic studies, and major contemporary consensus documents relevant to obesity-related HFpEF. Experimental studies were included when they clarified adipose inflammation, lipid partitioning, myocardial substrate handling, mitochondrial dysfunction, or candidate metabolic pathways for which direct human evidence remains limited. Secondary reviews were used primarily for conceptual framing and to identify relevant primary literature; when an original study directly supported a statement, the primary publication was preferentially cited.

No language restriction was applied at the PubMed search stage; however, because this was a targeted narrative minireview, full-text assessment and the final synthesis were limited to English-language publications. This article is a narrative minireview rather than a systematic review, and no formal meta-analysis or risk-of-bias assessment was performed.

OBESITY-RELATED HFPEF AS A DISTINCT CARDIOMETABOLIC PHENOTYPE

The most important conceptual update is that obesity-related HFpEF can be regarded as a distinct cardiometabolic phenotype rather than simply the coincidental overlap of two common conditions[4,13]. Cluster analyses and multi-omics studies have demonstrated that HFpEF populations comprise subgroups with different clinical and molecular profiles, including variable burdens of obesity, diabetes, renal disease, atrial fibrillation, and inflammatory abnormalities[13,14]. Network-medicine analyses have similarly demonstrated that HFpEF is characterized by complex patterns of interacting comorbidities rather than by a single cardiac abnormality[15]. Together, these human data support the concept that obesity-related HFpEF represents a systemic cardiometabolic phenotype with distinct cardiac and extracardiac manifestations.

Clinically relevant adiposity is not adequately characterized by BMI alone. In a predominantly lean East Asian HFpEF cohort, in which the mean BMI was 22.5 kg/m² and obesity (defined by a BMI > 30 kg/m²) was uncommon, lower visceral and subcutaneous adipose tissue measures were associated with underweight, malnutrition, and poorer clinical outcomes[6]. These observations illustrate that BMI-adiposity and adiposity-outcome relationships are population dependent and should not be interpreted using a uniform body-size framework. Consistent with this concept, the 2025 American College of Cardiology Scientific Statement emphasizes limitations of BMI across diverse populations and supports complementary assessment using waist circumference, waist-to-height ratio, or body-composition measures, particularly when the BMI is below 35 kg/m²[7].

Sex represents an additional source of heterogeneity in HFpEF[16,17]. In invasively phenotyped HFpEF, central obesity was more prevalent than BMI-defined general obesity (77% vs 62%) and was associated with higher exercise filling pressures and poorer aerobic capacity, with several adiposity-hemodynamic relationships being more pronounced in women[5]. In the pooled Semaglutide in Patients with HFpEF (STEP-HFpEF) and obesity sex analysis, semaglutide improved the Kansas City Cardiomyopathy Questionnaire Clinical Summary Score similarly in women and men (+7.6 points vs +7.5 points; P for interaction = 0.94), whereas the weight reduction was greater in women (-9.6% vs -7.2%; P for interaction = 0.006)[18]. These data suggest that sex may modify the adiposity-related phenotype and the magnitude of weight loss without eliminating the symptomatic benefit of semaglutide.

The broader multiorgan cardiometabolic context is supported by human imaging and longitudinal studies of adiposity and hepatic disease, together with experimental evidence[19-21]. Human imaging studies associate visceral or ectopic adiposity and adverse skeletal-muscle composition with cardiac remodeling or functional limitation[19,22]. Longitudinal human data associate MASLD with incident heart failure, including HFpEF, whereas experimental nonalcoholic steatohepatitis models provide mechanistic support for hepatic-metabolic contributions to HFpEF-like remodeling[20,21]. Collectively, these observations support broader cardiometabolic involvement across adipose tissue, liver, skeletal muscle, and the heart without establishing a single linear causal pathway.

DEPOT-SPECIFIC ADIPOSE DYSFUNCTION AND MALADAPTIVE LIPID PARTITIONING

Depot-specific adipose abnormalities provide a useful framework for linking systemic cardiometabolic disease to HFpEF physiology (Table 1). Human studies associate local cardiac adipose phenotypes with hemodynamic and functional abnormalities[23,24], whereas randomized invasive hemodynamic evidence demonstrates a modifiable volume-related component of HFpEF[25]. Detailed evidence regarding EAT and PVAT is considered separately below.

Table 1 Depot-specific adipose and organ pathways in obesity-related heart failure with preserved ejection fraction.

Depot or organ axis	Representative evidence	Proposed mechanism	Clinical measurability	Evidence maturity/key limitation
Subcutaneous adipose tissue	Lower subcutaneous adiposity, together with lower visceral adiposity, was associated with underweight, malnutrition, and poorer outcomes in an East Asian HFpEF cohort[6]; impaired lipid-buffering remains primarily a conceptual framework[8]	Lipid spillover to ectopic depots	Body composition; indirect metabolic markers	Human observational association plus review-level conceptual evidence; direct subcutaneous lipid-buffering capacity has not been established in HFpEF (CT-based East Asian HFpEF cohort, n = 196 + contemporary review)
Visceral adiposity	Visceral or central adiposity is associated with exercise limitation, higher exercise filling pressures, and poorer aerobic capacity[5,22]	Insulin-resistant lipid stress; inflammatory activation	CT/MRI; waist indices; visceral adiposity measures	Human observational and invasive-physiology evidence; causality remains unproven (HFpEF imaging cohort + invasively phenotyped HFpEF cohort)
MASLD/liver	MASLD is associated with incident heart failure, including HFpEF, whereas experimental NASH produces an HFpEF-like phenotype[20,21]	Hepatic-metabolic stress; systemic inflammation	Liver fat/stiffness; liver enzymes; metabolic risk indices	Human longitudinal association plus preclinical mechanistic evidence; human causality remains uncertain (longitudinal human cohort + experimental NASH model)
Skeletal-muscle fat	Intermuscular fat is associated with worse functional limitation, and impaired skeletal muscle mitochondrial respiration is associated with exercise intolerance[22,28]	Impaired peripheral metabolic and oxygen-utilization reserve	CT/MRI muscle composition; grip strength; gait speed	Human imaging/biopsy associations; clinically validated thresholds are lacking (human HFpEF imaging + skeletal biopsy studies)
Epicardial adipose tissue	EAT burden is associated with pericardial constraint and atrial/ventricular dysfunction, whereas lower EAT density is associated with adverse cardiometabolic phenotype and outcomes[23,24,40]	Local mechanical and paracrine amplifier	Echocardiography, CT, CMR; EAT volume/density	Human observational, invasive-physiology, and prospective imaging evidence; EAT density is not a direct measure of inflammation and whether EAT is a mediator or marker remains unresolved (HFpEF observational/invasive-physiology cohorts + prospective imaging cohort, n = 154)
Perivascular adipose tissue	Pericoronary adipose tissue attenuation on CT is associated with HFpEF[41]	Possible vascular-adipose inflammatory and remodeling interface	CT perivascular fat attenuation	Retrospective human imaging association; direct causal and histological validation in HFpEF is lacking (retrospective CCTA/echocardiography cohort, n = 607)

HFpEF: Heart failure with preserved ejection fraction; CT: Computed tomography; MRI: Magnetic resonance imaging; MASLD: Metabolic dysfunction-associated steatotic liver disease; NASH: Non-alcoholic steatohepatitis; EAT: Epicardial adipose tissue; CMR: Cardiovascular magnetic resonance; CCTA: Coronary computed tomographic angiography.

Inflammatory-metabolic abnormalities are also associated with HFpEF and impaired functional reserve in directly phenotyped human cohorts. Among 509 patients with chronic dyspnea undergoing invasive cardiopulmonary exercise testing, higher leptin and C-reactive protein concentrations were independently associated with HFpEF, while adipokine, inflammatory, and insulin-resistance biomarkers were associated with lower peak oxygen uptake and abnormalities in exercise reserve[26]. In a separate cohort of 510 participants, directed lipidomic profiling identified 70 eicosanoid or eicosanoid-related metabolites associated with HFpEF status, with subsets also associated with characteristic exercise abnormalities[27]. These studies support an inflammatory-metabolic component of HFpEF but do not establish that the measured biomarkers are causal mediators or that adipose tissue is their exclusive biological source.

Peripheral and hepatic abnormalities further illustrate the multiorgan nature of maladaptive lipid partitioning in HFpEF[20,21,28]. Magnetic resonance imaging demonstrated greater thigh intermuscular and liver fat in patients with HFpEF than in controls, with greater intramuscular fat associated with worse functional status and shorter 6-minute walk distance[22]. Skeletal-muscle biopsy studies further demonstrated markedly reduced mitochondrial respiratory capacity in patients with HFpEF, with maximal respiratory capacity being associated with peak oxygen uptake, 6-minute walk distance, and physical performance[28]. Analyses of longitudinal human data have revealed associations between MASLD and incident heart failure, including HFpEF, while experimental nonalcoholic steatohepatitis models provide mechanistic support for hepatic-metabolic contributions to HFpEF-like remodeling without establishing causality in humans[20,21]. In addition, a recent state-of-the-art synthesis similarly placed the MASLD-HFpEF overlap within a shared systemic cardiometabolic context characterized by obesity, insulin resistance, inflammation, and fibrotic remodeling[12]. Together, these extracardiac findings support a multiorgan disturbance of substrate storage and metabolic reserve and provide a rationale for examining myocardial substrate handling as a downstream component of the phenotype.

MYOCARDIAL METABOLIC INFLEXIBILITY AND ENERGETIC IMPAIRMENT

Cardiac metabolic flexibility refers to the capacity of the myocardium to adjust substrate utilization and oxidative metabolism in response to changes in fuel availability and energetic demand[29]. Human phosphorus magnetic resonance spectroscopy demonstrates an energetic deficit in HFpEF, with a reduced myocardial phosphocreatine-to-adenosine triphosphate ratio that is associated with impaired exercise reserve and exercise-induced pulmonary congestion[30]. Direct transcardiac metabolomic and lipidomic sampling further indicates altered myocardial substrate utilization in HFpEF, including reduced cardiac fatty acid use and relationships between lipid extraction and invasive hemodynamic abnormalities[29]. Within the lipid-partitioning model, metabolic inflexibility therefore represents the inability of the myocardium to safely accommodate changing substrate delivery rather than a simple increase or decrease in one fuel[9,29,30].

Human myocardial metabolomics provides particularly important evidence for this interpretation. In the study by Hahn et al[9], targeted metabolomics of endomyocardial biopsies and plasma from patients with HFpEF (n = 38), heart failure with reduced ejection fraction (HFrEF) (n = 30), and nonfailing controls (n = 20) showed markedly lower myocardial medium-chain and long-chain acylcarnitines in patients with HFpEF despite a greater burden of obesity and diabetes, together with higher myocardial pyruvate, lower tricarboxylic acid cycle intermediates, and impaired downstream branched-chain amino acid metabolism. Direct transcardiac sampling similarly demonstrated altered myocardial fuel handling rather than uniform fatty acid excess in HFpEF[29]. A more recent integrated analysis of left ventricular biopsies from patients with cardiometabolic HFpEF demonstrated impaired glycolysis and glucose metabolism, preserved fatty acid oxidation, and altered tricarboxylic acid cycle metabolism compared with overweight or obese nonfailing individuals[31]. Together, these human data support a model of disturbed substrate coordination and metabolic adaptability rather than a uniform increase or decrease in myocardial fatty acid metabolism[29,31].

Lipotoxic intermediates may contribute to this metabolic phenotype, but their biological roles should be distinguished from their value as circulating biomarkers[32]. In patients with HFpEF, a lower plasma ceramide C24:0/C16:0 ratio was independently associated with adverse cardiovascular outcomes and heart failure hospitalization, supporting a prognostic association but not establishing myocardial ceramide accumulation as a causal mechanism[32]. Metabolomic studies of human myocardium also demonstrate that plasma metabolic profiles do not necessarily reflect myocardial abnormalities, emphasizing the importance of distinguishing circulating signatures from tissue-level metabolism[9]. Ceramide species should therefore be considered candidate lipotoxic mediators and biomarkers whose tissue-specific causal roles in human obesity-related HFpEF remain to be established[32].

Within this framework, mitochondrial stress and impaired quality control can be interpreted as downstream consequences of a mismatch among substrate delivery, storage capacity, oxidative demand, and metabolic adaptation[33,34]. Experimental HFpEF studies implicate impaired adenosine monophosphate-activated protein kinase (AMPK) signaling in maladaptive cardiac stress responses, including atrial remodeling and susceptibility to atrial fibrillation[33]. In an experimental HFpEF model, mitochondrial respiratory dysfunction, increased reactive oxygen species production, and blunted stress-induced mitophagy were observed, whereas restoration of mitophagy associated with fatty acid oxidation improved features of the HFpEF phenotype[34]. These observations provide mechanistic plausibility for AMPK and mitophagy as components of metabolic stress adaptation, although direct validation of these pathways in human obesity-related HFpEF myocardium remains limited[31,34].

Ketone metabolism illustrates the need to distinguish metabolic rewiring from a validated therapeutic mechanism[9,35,36]. Metabolomic analysis of human myocardium showed increased ketone abundance in HFrEF but not in HFpEF, arguing against simple extrapolation of ketone adaptation across heart failure phenotypes[9]. In an experimental HFpEF model, myocardial ketone oxidation was impaired, and increasing ketone availability restored ketone oxidation without improving cardiac function[35]. In the randomized crossover ketogenic-HFpEF trial, acute ketone ester administration in 20 symptomatic patients altered systemic substrate utilization and reduced estimated filling pressures but did not improve peak oxygen consumption or exercise endurance[36]. Therefore, current evidence supports ketone handling as a component of metabolic adaptability but does not establish ketone oxidation as a validated therapeutic target in patients with obesity-related HFpEF[35,36].

EPICARDIAL AND PVAT: FROM QUANTITY TO BIOLOGICAL QUALITY

Human tissue studies demonstrate that obesity and diabetes are associated with greater inflammatory activity within EAT, a biologically active adipose depot adjacent to the myocardium and coronary vasculature[37]. Human mechanistic studies further suggest that EAT can communicate with adjacent cardiac tissue through locally released mediators, including microRNAs and extracellular vesicles, although these studies were not conducted specifically in HFpEF populations and therefore provide mechanistic plausibility rather than HFpEF-specific causal evidence[38,39]. In this review, the “quality” of EAT therefore refers to imaging-level or tissue-level characteristics beyond depot size and should not be interpreted as synonymous with directly measured inflammation[37].

EAT quantity nevertheless remains clinically relevant in HFpEF. In multicenter cohorts, EAT thickness was greater in HFpEF than in HFrEF or heart failure with mildly reduced ejection fraction (HFmrEF) and was associated with worse left atrial and left ventricular function in patients with HFpEF[24]. In a separate invasively characterized HFpEF cohort, greater EAT thickness was associated with hemodynamic signs of pericardial constraint[23]. These observational findings establish an association between EAT burden and HFpEF physiology but do not demonstrate that increased EAT volume or thickness is itself a sufficient causal mechanism.

Beyond quantity, computed tomography (CT)-derived EAT density may provide complementary information about the adipose tissue phenotype. In a prospective cohort of 154 patients with HFpEF, lower EAT density was associated with a more adverse cardiometabolic profile after adjustment for BMI and EAT volume, showed greater discriminatory performance than EAT volume for the presence and severity of metabolic syndrome, and was associated with higher rates of heart failure readmission and the composite clinical endpoint during follow-up. However, EAT density was not validated against direct histological measures of inflammation in that study and should therefore be interpreted as an imaging phenotype that may reflect differences in tissue composition and remodeling rather than as a direct measurement of inflammatory activity[40].

PVAT extends the depot-specific framework to the vascular interface, but HFpEF-specific human evidence remains substantially less mature than that for EAT. In a retrospective study of 607 patients undergoing coronary CT angiography and echocardiography, higher pericoronary adipose tissue attenuation was independently associated with the presence of HFpEF[41]. Because this evidence is observational and imaging-based, pericoronary adipose tissue attenuation should currently be regarded as an imaging marker associated with a local vascular-adipose phenotype rather than proof that PVAT inflammation causes HFpEF. Current EAT density and PVAT attenuation studies provide phenotype-level associations but do not establish externally validated thresholds for routine treatment selection, supporting the need for prospective studies linking standardized imaging measurements to clinical outcomes and, where feasible, direct tissue-level biological validation.

ADIPOSITY-AWARE PHENOTYPING

Adiposity-aware phenotyping should supplement rather than replace established diagnostic approaches to HFpEF. The 2026 American College of Cardiology Expert Consensus Decision Pathway discusses the H₂FPEF score, the Heart Failure Association-PEFF (HFA-PEFF) algorithm, and the HFpEF-ABA score as complementary approaches for estimating the likelihood that HFpEF underlies otherwise unexplained dyspnea or congestion[3,42-44]. These tools address diagnostic probability, whereas adiposity-aware phenotyping addresses a complementary question: How body composition, regional adipose distribution, ectopic fat, and associated cardiometabolic abnormalities shape the clinical expression of suspected or established HFpEF. Consistent with the 2025 American College of Cardiology Scientific Statement on obesity and heart failure, this framework therefore extends conventional HFpEF assessment rather than creating an alternative diagnostic standard and recognizes that BMI alone incompletely characterizes excess adiposity and its distribution[7].

The H₂FPEF score was derived and validated against invasive hemodynamic assessment during exercise and appropriately recognizes obesity as an important diagnostic feature, with BMI > 30 kg/m² contributing 2 points[42]. However, BMI within the score functions as a binary probability marker rather than a direct measure of visceral or ectopic adipose biology. Substantial weight loss can therefore change the BMI component of the score without necessarily establishing that residual depot-specific or hemodynamic abnormalities during exercise have resolved[42]. Adiposity-aware assessment may consequently provide complementary biological characterization after conventional diagnostic probability has been considered.

Sleep-disordered breathing (SDB) is common in patients with HFpEF and intersects strongly with obesity. In a recent prospective HFpEF cohort, moderate-to-severe SDB was identified in 42% of patients, with obstructive sleep apnea (OSA) being more frequent than central sleep apnea; higher BMI was independently associated with OSA[45]. The prevalence and severity of OSA have also been shown to be associated with higher H₂FPEF scores, although this relationship was largely attenuated after adjustment for BMI, underscoring obesity as an important shared determinant[46]. Screening should therefore be considered when clinical features suggest OSA, but SDB should be treated as an important comorbidity rather than as a direct surrogate for lipid-partitioning abnormalities.

In cases where diagnostic uncertainty persists, the HFA-PEFF algorithm provides a complementary stepwise assessment incorporating clinical evaluation, echocardiographic findings, natriuretic peptides, and functional testing[43]. Importantly, normal natriuretic peptide concentrations do not exclude HFpEF[43], a limitation that is particularly relevant in obesity. In I-PRESERVE, 808 of 3480 patients (23.2%) with symptomatic heart failure and left ventricular ejection fraction ≥ 45% had N-terminal pro-B-type natriuretic peptide concentrations < 125 pg/mL, and obesity was more prevalent in this subgroup than among patients with higher concentrations (48.4% vs 38.7%)[47]. Although cardiovascular event rates were lower in the low-natriuretic-peptide subgroup, health status remained similarly impaired[47]. Thus, a low natriuretic peptide concentration should not by itself exclude clinically meaningful HFpEF when the clinical phenotype remains compatible with HFpEF. The HFpEF-ABA score, based on age, BMI, and atrial fibrillation, offers a simpler screening approach that can guide subsequent echocardiographic and biomarker assessment[44]. When resting findings remain discordant with symptoms, functional testing should be performed in accordance with established diagnostic pathways[43].

Atrial and right-sided phenotyping deserves greater emphasis[48-50]. Left atrial reservoir strain, right ventricular-pulmonary arterial coupling, and global longitudinal strain are prognostically informative in HFpEF[49]. Mechanical atrial dysfunction predicts outcomes in HFpEF[48], and comprehensive cardiovascular magnetic resonance (CMR) in obese patients with diabetes reveals that atrial dysfunction may serve as a contrast-free marker of HFpEF when conventional assessment is confounded by obesity and diabetes[50]. These measures may complement adiposity-aware characterization when atrial or right-sided abnormalities are prominent, but they should be interpreted as phenotype descriptors rather than direct evidence of a specific adipose-mediated mechanism.

Biomarkers should be interpreted as panels rather than as isolated measures, because obesity-related HFpEF reflects overlapping inflammatory, metabolic, and matrix-remodeling pathways. In human cohorts undergoing direct phenotyping, leptin, C-reactive protein, and broader adipokine, inflammatory, and insulin resistance profiles were associated with HFpEF and impaired exercise reserve[26]. These circulating signals may help identify an inflammatory-metabolic phenotype, but they do not establish whether an individual marker is a causal mediator, a consequence of metabolic stress, or an adaptive response. Biomarkers should therefore complement imaging and functional assessment for phenotype enrichment rather than serve as stand-alone diagnostic criteria or treatment-selection thresholds.

Imaging can operationalize the obesity-lipid-heart model. Echocardiography remains the most accessible tool for assessing cardiac structure, estimating filling pressures, and assessing strain, while CMR can provide complementary characterization of cardiac structure and atrial function[50]. CT can quantify the EAT burden, and automated approaches may improve reproducibility of depot measurement[51,52]. Importantly, the imaging-derived depot quantity should not be equated with biological activity; human tissue evidence demonstrates inflammatory abnormalities within EAT in obesity and diabetes, but depot size itself is not a direct measure of inflammation[37]. Standardized thresholds for adipose tissue imaging measures have not yet been established for routine treatment selection. Consistent with the 2025 American College of Cardiology Scientific Statement, BMI may be supplemented by waist circumference, waist-to-height ratio, or body composition assessment, particularly when BMI is < 35 kg/m²[7]. Within this context, regional adipose imaging, atrial and right-sided functional indices, exercise capacity, and selected metabolic and inflammatory markers may refine phenotypic characterization when clinically available. The purpose is not to create an alternative diagnostic algorithm, but to characterize adipose-metabolic heterogeneity within established HFpEF diagnostic categories and to support phenotype-enriched translational studies.

Routine assessment may include BMI together with waist indices, evaluation of congestion, natriuretic peptides interpreted in the context of obesity, echocardiography, rhythm status, relevant cardiometabolic comorbidities, and functional capacity. When additional phenotypic characterization is warranted and resources permit, body-composition assessment, evaluation of MASLD, regional adipose imaging, selected metabolic or inflammatory biomarkers, atrial and right-sided functional measures, and exercise testing may provide complementary information. More advanced approaches, such as positron emission tomography-based metabolic or inflammatory imaging, detailed adipose-tissue imaging analyses, and tissue-omics, should currently be considered primarily as research tools because standardized thresholds and their utility in routine clinical practice require further validation.

THERAPEUTIC IMPLICATIONS

Therapeutic implications should be interpreted according to the maturity of the evidence, separating established clinical outcome evidence from mechanistic human studies and hypothesis-generating experimental work. Current treatment should remain guided by established HFpEF, obesity, and comorbidity indications, while the clinical utility of depot-specific phenotypes for treatment selection requires prospective validation. Lifestyle intervention remains foundational, but its goal should focus on functional reserve, body composition, and lean-mass preservation rather than weight loss alone. The goal is not simply to lose kilograms, but to reduce visceral and ectopic adiposity, preserve skeletal muscle, and restore functional reserve[53,54]. In a multicenter randomized trial, combined endurance and resistance exercise training did not significantly improve the primary composite endpoint but improved peak oxygen uptake and New York Heart Association functional class as secondary outcomes[53]. Calorie restriction combined with exercise may improve body composition and exercise capacity in older patients with obesity and HFpEF, although a small long-term follow-up suggested that some of these gains may diminish after the supervised intervention period[54]. Resistance exercise is therefore particularly relevant when substantial weight loss is pursued, especially in older or frail patients, because preservation of muscle and functional reserve is a therapeutic objective rather than a secondary consideration.

Among established HFpEF pharmacotherapies, sodium-glucose cotransporter 2 inhibitors have robust outcome-trial evidence. In Empagliflozin Outcome Trial in Patients with Chronic Heart Failure with Preserved Ejection Fraction, empagliflozin reduced the primary composite of cardiovascular death or heart failure hospitalization [hazard ratio: 0.79; 95% confidence interval (CI): 0.69-0.90][55]. Dapagliflozin Evaluation to Improve the Lives of Patients with Preserved Ejection Fraction Heart Failure trial subsequently randomized 6263 patients with left ventricular ejection fraction > 40% and demonstrated that treatment with dapagliflozin reduced the risk of worsening heart failure or cardiovascular death[56]. Moreover, in Cardiac and Metabolic Effects of Dapagliflozin in Heart Failure with Preserved Ejection Fraction, treatment with dapagliflozin reduced pulmonary capillary wedge pressure at rest and during exercise, plasma volume, and body weight in patients with HFpEF[25]. This sequence distinguishes clinical outcome evidence from invasive hemodynamic evidence. Mechanistic human tissue and preclinical data suggest additional effects on lipid handling, inflammatory and oxidative pathways, and adipose biology[57,58], but these observations remain mechanistic and should not be presented as the established basis for the outcome benefit of sodium-glucose cotransporter 2 inhibition. Decongestion also remains an important component of evidence-based HFpEF care when clinically indicated and should be titrated according to congestion, renal function, blood pressure, and treatment tolerance rather than inferred from adiposity phenotype alone[3].

Mineralocorticoid receptor antagonism is an evidence-based therapeutic option in HFpEF and is mechanistically relevant to this cardiometabolic framework because aldosterone-sensitive pathways intersect with sodium retention, vascular inflammation, and fibrosis. In Finerenone Trial to Investigate Efficacy and Safety Superior to Placebo in Patients with Heart Failure, a randomized trial in patients with mildly reduced or preserved ejection fraction (left ventricular ejection fraction ≥ 40%), treatment with finerenone reduced the rate of the primary composite of total worsening heart failure events and cardiovascular death compared with placebo (rate ratio: 0.84; 95%CI: 0.74-0.95)[59]. In the prespecified obesity analysis, the treatment effect was consistent across categorical BMI groups (P for interaction = 0.32), whereas modeling BMI as a continuous variable suggested a potentially greater treatment effect at higher BMI (P for interaction = 0.005)[60]. Finerenone should therefore be regarded as a heart failure therapy supported by clinical outcome data rather than an adiposity-specific intervention; the exploratory BMI gradient and the accompanying waist-based anthropometric analyses do not yet establish a validated marker for treatment selection, while renal function and serum potassium levels remain important safety considerations[59,60].

Incretin-based obesity therapy has provided strong evidence that obesity-related HFpEF is therapeutically modifiable. Semaglutide improved heart failure-related symptoms, physical limitations, body weight, and exercise function in obesity-related HFpEF[10], whereas in SUMMIT, treatment with tirzepatide reduced the risk of cardiovascular death or worsening heart failure and improved health status[11]. Mechanistic secondary analyses should be separated from these clinical efficacy findings. In SUMMIT, the clinical benefits of tirzepatide were accompanied by reductions in systolic blood pressure, estimated blood volume, C-reactive protein, and troponin T, together with kidney-related changes[61]. In the SUMMIT CMR substudy, treatment with tirzepatide also reduced left ventricular mass and paracardiac adipose tissue compared with placebo. Although the reduction in left ventricular mass paralleled weight loss, these imaging findings do not establish that modification of a specific adipose depot mediates the clinical benefit[62]. These multidomain effects make it difficult to disentangle the contributions of weight loss, decongestion, blood pressure reduction, modification of inflammation, and direct tissue metabolic effects. Clinically meaningful pharmacological weight loss also includes loss of both fat mass and lean mass. In the SURMOUNT-1 dual-energy X-ray absorptiometry substudy of adults with obesity or overweight, approximately 75% of weight lost with tirzepatide was fat mass and 25% was lean mass[63]. A 2026 meta-analysis found that the proportion of weight loss attributable to lean mass varied across incretin agents and studies, indicating that there is no fixed proportion[64]. These findings reinforce the importance of resistance exercise, adequate nutrition, and monitoring of physical function and body composition in older or frail patients, without implying that incretin therapy inevitably causes sarcopenia. The obesity-atrial fibrillation-EAT axis may represent an additional component of this phenotype. A STEP-HFpEF analysis showed clinical improvement in patients with and without atrial fibrillation treated with semaglutide[65]. Given the proximity of EAT to atrial tissue and human evidence of local EAT-derived signaling[38,39], EAT may contribute to the link between obesity, atrial remodeling, and atrial fibrillation; however, whether a reduction in EAT directly mediates rhythm or heart failure outcomes remains uncertain.

Metabolic and bariatric surgery warrants separate consideration because HFpEF-specific evidence remains limited and is not based on randomized outcome trials. In the HFpEF-specific mechanistic study by Mikhalkova et al[66], women with obesity and HFpEF experienced improved symptoms, reduced cardiac mass and liver fat, and improved echocardiographic relaxation after bariatric surgery; however, myocardial fat was unchanged and changes in circulating lipid species did not correlate with improvement in cardiac function. This discordance argues against a simple model in which reduction in cardiac fat is required for improvement in diastolic physiology. Although bariatric surgery may therefore be appropriate for selected patients according to established obesity indications and multidisciplinary assessment, current evidence does not justify presenting it as an HFpEF-specific treatment or as proof that clinical benefit is mediated by reversal of myocardial lipid accumulation.

Metabolic combination therapy is likely to become an important clinical and trial-design frontier. Retrospective real-world data in patients with overweight or obesity, diabetes, and HFpEF receiving sodium-glucose cotransporter 2 inhibitors revealed that glucagon-like peptide-1 receptor agonist use was associated with fewer exacerbations of heart failure and fewer emergency visits or hospitalizations[67]. However, this observation requires confirmation in randomized trials and should not be interpreted as evidence that a specific drug combination is superior or that benefit is mediated through correction of lipid partitioning. Future combination trials should prespecify clinical endpoints together with mechanistic measures that can distinguish changes in congestion, body composition, metabolic and inflammatory status, and functional reserve.

Emerging therapies that directly target cardiac energetics or lipid handling remain investigational. In the two-part, open-label phase 2a IMPROVE-DiCE study, ninerafaxstat was evaluated in two small cohorts of patients with obesity and type 2 diabetes: One with HFpEF and one without. Part 1 enrolled 21 participants without HFpEF and showed an improvement in the myocardial phosphocreatine-to-adenosine triphosphate (PCr/ATP) ratio, together with a 34% reduction in myocardial triglyceride content. In part 2, which enrolled 21 patients with symptomatic HFpEF, preliminary improvements were observed in the PCr/ATP ratio, systolic augmentation during stress, 6-minute walk distance, and New York Heart Association class. Because both cohorts were small and lacked a placebo control, these findings should be considered early human mechanistic evidence and remain hypothesis-generating rather than efficacy-defining[68]. In contrast, approaches targeting lipid handling and stress adaptation, including AdipoRon-mediated modulation of lipid accumulation, spliced X-box-binding protein 1-endoplasmic reticulum degradation-enhancing alpha-mannosidase-like protein 2 signaling, and fibroblast growth factor 21-mediated mitochondrial bioenergetics, are supported only by preclinical or mechanistic evidence[69-71]. These approaches should be viewed as hypothesis-generating strategies for phenotype-enriched trials rather than as established therapies for obesity-related HFpEF.

KNOWLEDGE GAPS AND FUTURE DIRECTIONS

Taken together, the available evidence supports a conceptual but not yet fully causal model in which disordered lipid storage and metabolic adaptation across adipose tissue, the liver, skeletal muscle, the vasculature, and myocardium interact with hemodynamic and inflammatory stress.

The first translational priority is phenotyping. Future studies should determine which adiposity-related measures provide reproducible phenotypic or prognostic information beyond established HFpEF assessment. Particular priorities are to distinguish depot quantity from biologically informative tissue characteristics, standardize acquisition and reporting, define reproducible thresholds, and determine whether candidate measures remain informative across sex, ethnicity, body size, and clinical settings. Advanced imaging and tissue-omics should be used primarily for biological validation of candidate phenotypes until their incremental clinical value and reproducibility are established.

The second translational priority is to determine prospectively whether prespecified phenotypic characteristics modify treatment response. Treatment-effect heterogeneity should be tested rather than assumed. Future studies could evaluate volume expansion and pericardial restraint as candidate modifiers of response to decongestive strategies, while EAT-related measures could be incorporated as research phenotyping variables rather than treatment-selection criteria.

Similarly, inflammatory-metabolic phenotypes could be prospectively evaluated for differential responses to sustained weight loss, incretin-based therapy, and exercise. Frailty or sarcopenic obesity could be specifically examined in studies incorporating nutritional supervision, resistance training, and cautious weight-loss strategies. Trials enrolling patients with diabetes, renal stress, or vascular dysfunction could prespecify analyses of sodium-glucose cotransporter 2 (SGLT2) inhibition and mineralocorticoid receptor antagonism when these therapies are otherwise indicated, while also monitoring renal function and volume status. Serial phenotypic assessment should be used to characterize treatment-associated changes. At the same time, clinical improvement should not be taken as proof that a therapy has directly corrected lipid partitioning. Mechanistic studies of incretin-based and other metabolic therapies should, where feasible, use weight-matched, blood pressure-matched, or decongestion-matched comparators together with serial body composition assessment, regional adipose imaging, inflammatory markers, and exercise hemodynamics to separate indirect hemodynamic or weight-mediated effects from potentially direct metabolic effects.

To test these hypotheses rigorously, the third priority is trial design (Table 2). Phenotype enrichment should be evaluated prospectively rather than assumed to improve treatment selection. Multicenter cohorts and trials should use standardized baseline and serial measurements, predefined clinical and functional endpoints, and prespecified treatment-interaction analyses. Stratification by EAT burden or density, visceral adiposity, MASLD, triglyceride-glucose index and homeostatic model assessment of insulin resistance, inflammatory biomarkers, atrial dysfunction, and exercise reserve could be used to evaluate treatment-response heterogeneity across prespecified phenotypic subgroups. Endpoints should include patient-reported symptoms, exercise performance, hemodynamics, body composition, lean mass, adipose depot activity, hepatic and renal markers, and cardiac remodeling.

Table 2 Proposed phenotype-enriched research and trial-design framework for obesity-related heart failure with preserved ejection fraction.

Candidate phenotype axis	Phenotype-enrichment variables	Candidate intervention domain for prospective testing	Key trial monitoring variables	Evidence maturity; key limitation (study design/population/model)
Circulatory overload	Congestion; elevated filling pressure; plasma-volume expansion; exercise PCWP elevation	Decongestion strategies; SGLT2 inhibitors; MRA/finerenone according to established clinical indications	Renal function; blood pressure; potassium; volume depletion	Tier I: EMPEROR-Preserved[55]; DELIVER[56]; FINEARTS-HF[59]. Tier II: CAMEO-DAPA[25]. Randomized outcome trials in heart failure with mildly reduced or preserved ejection fraction; randomized placebo-controlled HFpEF hemodynamic study
EAT/pericardial restraint	High EAT burden and/or low computed tomography-derived EAT density; right-sided congestion; pericardial restraint; atrial enlargement	Decongestion; obesity-directed strategies for prospective testing; serial EAT imaging as a research measure	Imaging availability; overdiuresis risk; symptom-volume mismatch	Tier II: SUMMIT CMR substudy[62]. Tier III: Crum et al[23]; Jin et al[24]; Liu et al[40]. HFpEF-obesity randomized-trial CMR substudy; invasive-physiology HFpEF cohort; observational HFpEF imaging cohort; prospective HFpEF imaging cohort, n = 154
Inflammatory-metabolic phenotype	Elevated CRP/IL-6; leptin/resistin imbalance; TyG/HOMA-IR; diabetes/metabolic syndrome	Incretin-based obesity therapy; exercise; SGLT2 inhibitors according to established indications; nutritional intervention	GI intolerance; renal-volume status; lean-mass preservation	Tier I: STEP-HFpEF[10]; SUMMIT[11]. Tier III: Ramirez et al[26]; Lau et al[27]. Obesity-HFpEF randomized outcome trials; invasive-CPET biomarker cohort, n = 509; lipidomic cohort, n = 510
MASLD-dominant phenotype	Liver fat/stiffness; elevated metabolic risk; visceral adiposity	Weight loss strategies; incretin-based therapy; SGLT2 inhibitors according to established indications; hepatometabolic risk control	Liver markers; nutrition; sarcopenia risk	Tier III: Chang et al[20]. Tier IV: Kucsera et al[21]. Longitudinal human cohort; experimental NASH model
Frailty/sarcopenic obesity	Low muscle mass; poor grip strength or gait speed; low exercise reserve	Resistance training; protein/nutrition support; cautious weight-loss strategies	Lean-mass loss; falls; undernutrition	Tier II: Edelmann et al[53]; SURMOUNT-1 DXA substudy[63]. Tier III: Upadhya et al[54]. HFpEF randomized exercise trial with secondary functional outcomes; randomized DXA substudy in adults with obesity or overweight, not an HFpEF-specific population and therefore indirect body-composition evidence; small long-term HFpEF follow-up
Atrial-myopathy dominant phenotype	Low LA strain; AF; EAT proximity; LA enlargement; right-sided congestion	AF/rhythm management; decongestion according to established indications; obesity-directed weight loss strategies; EAT-aware assessment as a research measure	Atrial reserve; anticoagulation eligibility; renal function	Tier II: STEP-HFpEF atrial-fibrillation analysis[65]. Tier III: Weerts et al[48]; Wang et al[49]. Pooled randomized-trial secondary analysis; HFpEF observational atrial-function/imaging cohorts

This table shows a proposed, hypothesis-generating framework for phenotype-enriched research and trial design. It has not been prospectively validated for clinical treatment selection and should not be interpreted as guidance for assigning therapy to individual patients. Candidate intervention domains are included for prospective testing only. Evidence tiers apply to the individually cited studies rather than to the proposed phenotype-enriched research framework, which remains unvalidated in all rows. Tier I: Primary randomized clinical outcome evidence. Tier II: Randomized evidence from secondary, subgroup, substudy, mechanistic, or hemodynamic analyses or studies. Tier III: Nonrandomized human observational, prospective, longitudinal, imaging, biomarker, or other associative evidence. Tier IV: Preclinical or experimental mechanistic evidence. PCWP: Pulmonary capillary wedge pressure; SGLT2: Sodium-glucose cotransporter 2; EMPEROR-Preserved: Empagliflozin Outcome Trial in Patients with Chronic Heart Failure with Preserved Ejection Fraction; FINEARTS-HF: Finerenone Trial to Investigate Efficacy and Safety Superior to Placebo in Patients with Heart Failure; CAMEO-DAPA: Cardiac and Metabolic Effects of Dapagliflozin in Heart Failure with Preserved Ejection Fraction; EAT: Epicardial adipose tissue; CMR: Cardiovascular magnetic resonance; CRP: C-reactive protein; IL-6: Interleukin-6; TyG: Triglyceride-glucose index; HOMA-IR: Homeostatic model assessment of insulin resistance; GI: Glycemic index; SUMMIT: A study of tirzepatide (LY3298176) in Participants with Heart Failure with Preserved Ejection Fraction and Obesity; STEP-HFpEF: Semaglutide treatment effect in people with obesity and heart failure with preserved ejection fraction; MASLD: Metabolic dysfunction-associated steatotic liver disease; NASH: Non-alcoholic steatohepatitis; SURMOUNT-1: A study of tirzepatide (LY3298176) in Participants With Obesity or Overweight; DXA: Dual-energy X-ray absorptiometry; LA: Left atrial; AF: Atrial fibrillation.

Beyond treatment-effect heterogeneity, establishing mechanistic causality will require longitudinal, multimodal studies linking myocardial and adipose tissue biology, plasma biomarkers, and imaging within the same participants, ideally with repeated assessment before and after treatment. Biopsy-linked metabolomics combined with transcriptomic or proteomic profiling, repeated measurement of candidate lipid intermediates, mediation analyses, genetic or other causal-inference approaches, and assessment of tissue-plasma concordance could help distinguish causal or mediating pathways from adaptive responses and downstream markers. Such studies are needed to determine which components of the lipid-dysregulation model are causal, prognostic, and therapeutically modifiable in humans.

LIMITATIONS OF THE EVIDENCE BASE AND REVIEW

This narrative minireview has several limitations. Evidence acquisition was targeted rather than systematic, PubMed was the sole bibliographic database searched, and full-text assessment and final synthesis were limited to English-language publications. The included studies are heterogeneous in HFpEF definitions, populations, adiposity measures, imaging methods, and endpoints. Most depot-specific human evidence is observational and remains susceptible to residual confounding. Standardized thresholds for EAT/PVAT and skeletal-muscle imaging measures are not sufficiently established for routine clinical use, and human myocardial tissue data remain sparse. In addition, several intracellular mechanisms are supported more strongly by preclinical evidence than by direct human HFpEF data. Accordingly, the proposed framework should be regarded as hypothesis-generating rather than as a basis for clinical treatment selection.

CONCLUSION

Obesity-related HFpEF is best viewed as a cardiometabolic syndrome in which abnormal adipose storage, ectopic lipid partitioning, and impaired metabolic reserve interact with hemodynamic and inflammatory stress. The relevant clinical question is not only whether obesity is present, but where lipid is stored, how adipose depots behave, and whether the heart, vasculature, and skeletal muscle can adapt to metabolic and exercise stress.

This view has practical consequences. Adiposity-aware phenotyping may extend assessment beyond BMI and resting biomarkers by integrating body composition, regional fat depots, MASLD, metabolic and inflammatory markers, atrial and right-sided mechanics, and functional capacity. Current treatment should remain guided by established evidence and clinical indications for HFpEF, obesity, and comorbidity, including decongestion, SGLT2 inhibition, mineralocorticoid receptor antagonism, incretin-based obesity therapy, exercise, nutrition, and lean mass preservation. Adiposity-aware phenotyping may instead provide a framework for prospective research examining whether specific depot, metabolic, or functional characteristics are associated with differential treatment responses. Prospective studies are needed to determine whether measures of depot-specific adiposity and metabolic flexibility provide additional prognostic information beyond established HFpEF diagnostic and therapeutic frameworks.