The role of epicardial adipose tissue remodelling in heart failure with preserved ejection fraction

Abstract

Heart failure with preserved ejection fraction (HFpEF) is a growing global health problem characterized by high morbidity and mortality, with limited effective therapies available. Obesity significantly influences haemodynamic and structural changes in the myocardium and vasculature, primarily through the accumulation and action of visceral adipose tissue. Particularly, epicardial adipose tissue (EAT) contributes to HFpEF through inflammation and lipotoxic infiltration of the myocardium. However, the precise signalling pathways leading to diastolic stiffness in HFpEF require further elucidation. This review explores the dynamic role of EAT in health and disease. Drawing upon insights from studies in other conditions, we discuss potential EAT-mediated inflammatory pathways in HFpEF and how they may contribute to functional and structural myocardial and endothelial derangements, including intramyocardial lipid infiltration, fibrosis, endothelial dysfunction, cardiomyocyte stiffening, and left ventricular hypertrophy. Lastly, we propose potential targets for novel therapeutic avenues.

Graphical Abstract

Open in new tabDownload slide

1. Introduction

Heart failure with preserved ejection fraction (HFpEF) is considered one of the greatest unmet needs in contemporary cardiovascular medicine.¹ The incidence of HFpEF is increasing worldwide at an alarming rate as a consequence of an ageing population and an increase in the prevalence of related metabolic and cardiovascular risk factors.² Although HFpEF accounts for half of all heart failure cases, limited effective therapies exist.³

Patients with HFpEF suffer from high morbidity mainly characterized by exertional dyspnoea and impaired exercise tolerance. These symptoms arise due to left ventricular (LV) diastolic dysfunction, atrial and LV stiffness, and subsequent elevated filling pressures in the heart.⁴ Besides, rates of hospitalization are high, paralleled by high healthcare costs, and there is an increased risk of death.⁵

HFpEF is a systemic, heterogeneous disease that exhibits a variety of phenotypes.⁶ It often co-exists with other cardiovascular comorbidities, including obesity, diabetes, hypertension, and atrial fibrillation (AF).^2,7,8 Of these, obesity is considered a key contributor to the development and pathophysiology of HFpEF. Consequently, an obesity-related phenotype of HFpEF has been established.^9,10

Obesity drives haemodynamic and structural transformations of the myocardium and the vasculature. While total body fat, quantified by body mass index (BMI), was traditionally used to describe the extent of obesity and its effects, BMI does not tell the entire story.¹⁰ Adipose tissue (AT) distribution, composition, and inflammatory burden are increasingly being recognized as determining factors of cardiovascular derangements. Notably, visceral AT surrounding the heart has gained traction in the past decades, whereby more visceral AT is related to an increased risk of cardiovascular disease.^11,12 Of importance is the epicardial AT (EAT), a biologically active tissue located underneath the visceral pericardium and contiguous with the surface of the myocardium. There is virtually no separation between the EAT and the myocardium, hence allowing a mechanical and paracrine interaction between both tissues. As a result, it is hypothesized that EAT could exert direct, local influence on functional and structural characteristics of the heart, such as those seen in the obesity-related HFpEF phenotype, including left atrial and ventricular remodelling, increased LV stiffness, and reduced diastolic function.^13,14

Observational studies show an association between EAT expansion and increased disease burden in HFpEF. Of notice are worse haemodynamic impairments and metabolic profiles, worse exercise capacity, more severe pulmonary hypertension, and increased inflammatory markers.^10,15,16 Furthermore, clinical studies investigating the reduction of EAT through bariatric surgery reveal improvements in cardiac structure and functioning.^17,18 However, while associative studies suggest that EAT actively contributes to HFpEF pathophysiology, mechanistic studies are scarce. Consequently, the underlying mechanisms by which EAT influences HFpEF remain to be determined.

Drawing upon insights from studies focusing on other cardiovascular and metabolic diseases, including AF, diabetes mellitus, and coronary artery disease (CAD), this review proposes and discusses potential EAT-mediated inflammatory pathways that might also play a role in HFpEF. We discuss how certain molecules in EAT’s proinflammatory secretome may contribute to functional and structural myocardial and endothelial derangements, and how these, in turn, can lead to alterations that contribute to increased myocardial stiffness. Lastly, we discuss potential targets for the development of novel therapeutic avenues. Based on recent publications on EAT imaging, we have decided to leave this topic outside the scope of the current review.^11,19,20

2. Cardiovascular pathophysiology of HFpEF

Patients with HFpEF have a preserved LV ejection fraction (≥50%) but impaired diastolic function accompanied by high LV end-diastolic pressures. Several processes underlie the development of diastolic dysfunction, including increased pericardial constraint, LV hypertrophy, endothelial dysfunction, and myocardial fibrosis, but also the co-existence of other comorbidities—such as AF, pulmonary hypertension, and diabetes—and deconditioning.

Passive myocardial stiffness, a key hallmark of HFpEF, is a multifactorial process involving changes in cardiomyocyte properties and alterations of the extracellular matrix (ECM; Figure 1). A first important determinant is cardiomyocyte stiffening, a process regulated by the sarcomeric protein titin. This protein exists in two isoforms, N2BA and N2B, which differ in elastic properties; N2BA is more compliant, while N2B is stiffer. The ratio of these isoforms influences the viscoelasticity properties of the sarcomere.²¹ Additionally, titin’s role in regulating cardiomyocyte stiffness involves phosphorylation and oxidation of certain regions within the protein. Phosphorylation of specific sites within the N2B region can alter titin’s elastic properties and promote myocardial stiffening. The phosphorylation process is regulated by protein kinase A (PKA) and G (PKG).²¹ Patterns of hypo-phosphorylation of PKG-dependent sites in HFpEF have been identified in animals and humans, leading to titin stiffening, and suggesting their involvement in promoting cardiomyocyte stiffness.^22–24 Furthermore, animal studies have shown that obese rats presented increased titin-based cardiomyocyte stiffness and developed HFpEF, when compared with lean rats.^24,25  In vitro administration of PKA or PKG has been shown to correct cardiomyocyte stiffness, further supporting a causal relationship.²³ Besides, titin’s spring is also comprised of the PEVK region, known to be phosphorylated by protein kinase C (PKC). Indeed, increased PKC-dependent phosphorylation of this region is associated with increased titin-dependent stiffness.²⁶

Endothelial dysfunction is another relevant component in HFpEF. It is related to increased vascular resistance, impaired vascular relaxation, and cardiomyocyte stiffness, processes which can lead to diastolic dysfunction and increased LV filling pressures (Figure 1).²⁷ Heightened oxidative stress and inflammation, exacerbated by cardiovascular risk factors, such as ageing, hypertension, diabetes, and notably obesity, contribute significantly.^28,29 Increased production of reactive oxygen species (ROS) is a hallmark of oxidative stress, directly reducing nitric oxide (NO) bioavailability, a key regulator of vasodilation and endothelial function.^21,30 This reduction in NO bioavailability is a key feature in endothelial dysfunction, further characterized by a proinflammatory, prothrombotic, vasoconstrictive state. Importantly in the context of cardiomyocyte stiffness, decreased NO bioavailability disrupts its downstream cascade, ultimately reducing the levels of cyclic guanosine monophosphate (cGMP). Reduced cGMP levels lead to reduced PKG activity in HFpEF, as cGMP acts as its activator, and thereby diminishing the phosphorylation of titin and exacerbating cardiomyocyte stiffness.^26,30,31 This diminished PKG activity is embedded in a paradigm proposed by Paulus and Tschöpe,²⁸ where reduced cGMP levels and PKG activity are a consequence of a complex pathway involving endothelial dysfunction.

Besides cardiomyocyte stiffening, alterations in the composition and structure of the ECM significantly contribute to myocardial stiffening in HFpEF (Figure 1). Patients with HFpEF exhibit increased extracellular fibrosis because of exacerbated myocardial fibrillar collagen content when compared with controls.^22,26,32 However, the impact of fibrotic remodelling extends beyond total collagen content; the relative proportion of collagen types within the ECM plays an important role. Patients with HFpEF show an increase in the ratio of collagen Type I, known for its stiffening properties, relative to collagen Type III, which confers greater compliance.³³ This imbalance in collagen subtypes directly contributes to the observed myocardial stiffening, exacerbating diastolic dysfunction.^33,34 Additionally, ECM dynamics in HFpEF are accompanied by alterations in profibrotic plasma biomarkers. These biomarkers include inducers of collagen production, such as profibrotic transforming growth factor beta (TGF-β) and soluble ST2 (sST2), and inhibitors of collagen degradation molecules known as matrix metalloproteinases.^26,31,34

The coexistence and interplay of these pathological abnormalities culminate in functional and structural derangements of the myocardium. On the one hand, they disrupt the diastolic properties, rendering the myocardium stiff, causing inadequate LV dilatation during diastole and additional LV hypertrophy over time. Consequently, LV end-diastolic volume is reduced. On the other hand, alongside impaired diastolic filling, there is an elevation of filling pressures, further exacerbating the cardiac dysfunction.

3. The obesity-related HFpEF phenotype

It is estimated that over 80% of patients with HFpEF are either overweight or obese, presenting a distinct clinical and haemodynamical profile.³⁵ In an elegant study, Obokata et al.⁹ compared obese HFpEF participants, non-obese HFpEF participants, and non-obese controls free of heart failure. Their findings revealed that all HFpEF groups exhibited LV dysfunction and elevated filling pressures, irrespective of BMI. However, obese HFpEF participants additionally presented with greater pericardial constraint, ventricular concentric remodelling, ventricular interaction, and dependence on plasma volume expansion.⁹ Other studies have shown a positive association between BMI and impairments in LV diastolic function and increased diastolic stiffness, as assessed by echocardiography, and independent of other cardiovascular risk factors, hypertension, and diabetes mellitus.^36,37 These results provide compelling evidence for a distinct obesity-related HFpEF phenotype, likely influenced by different pathophysiological mechanisms.

In the context of HFpEF, however, it seems that there is a relationship with obesity beyond BMI. In fact, EAT expansion and its local effects may contribute to a worse haemodynamic profile and are associated with circulating markers of systemic inflammation, insulin resistance, dyslipidaemia, and endothelial dysfunction.^15,16 Obese patients with HFpEF have overall more EAT when compared with non-obese patients with HFpEF and controls, which has been directly related to biventricular hypertrophy in the obese group.⁹ Koepp et al.¹⁰ investigated differences in EAT volume among obese HFpEF participants. They observed that participants with increased EAT deposition experienced greater pericardial constraint, higher cardiac filling pressures, a worse haemodynamic profile, and poorer exercise capacity.¹⁰

However, evidence concerning the relationship between obesity and EAT in HFpEF presents contested perspectives. Patients with HFpEF may exhibit higher EAT volumes compared with healthy controls despite a similar BMI.³⁸ Additionally, one study shows that 20% of non-obese patients with HFpEF exhibited high EAT volume, while 19% of obese patients with HFpEF showed lower EAT volumes.³⁹ However, another possibility is that obese, unhealthy individuals may independently accumulate more EAT and experience adverse myocardial remodelling typical of HFpEF.²⁰ Thus, the question remains whether obesity contributes to HFpEF pathophysiology primarily by general adiposity, directly through EAT, or through a combination of both.

4. EAT: mechanical influence and/or infiltrative lipotoxic?

In the last decade, there has been a significant shift in the understanding of EAT in relation to HFpEF. Initially regarded as a manifestation of obesity, current research supports the hypothesis that EAT plays an active role in the pathophysiology of HFpEF. Two mechanistic pathways have been formulated as a result of compelling evidence from both observational and mechanistic studies.

The first argues that EAT exerts a mechanical compressive force by pericardial constraint. As EAT accumulates, it applies an inwards force on the underlying myocardium, driving up LV intracavitary and filling pressures.^40,41 It is suggested that increased EAT, particularly in patients with an obesity-related HFpEF phenotype, is associated with constrictive haemodynamics.⁴² This can eventually lead to LV eccentricity and ventricular interdependence.⁴³ The latter describes a cardio-mechanical process wherein increased pressures and the restriction of one ventricle, in this case the LV, leads to restriction of the other ventricle, namely the right ventricle.

The second describes the pathological, proinflammatory transformation of EAT, which exerts an ‘infiltrative-lipotoxic’ effect on the myocardium, by direct infiltration of EAT into the myocardium or through paracrine action of proinflammatory adipokines.^44–46 Both mechanisms eventually lead to functional and structural alterations of the myocardium, contributing to the hallmarks of HFpEF, such as LV hypertrophy, increased filling pressures, and ultimately, LV diastolic dysfunction.

Despite the growing body of evidence suggesting a significant role of EAT in HFpEF pathophysiology, it remains to be further elucidated whether causality can be inferred or whether EAT remains a passive bystander in this phenotype and plays no role at all.

5. Location and anatomy of EAT

The human body has various AT depots that vary in anatomy, composition, and function. AT is primarily classified into subcutaneous AT (SAT) and visceral AT (VAT). SAT is composed of AT found directly underneath the skin, while VAT refers to AT surrounding the organs located within the abdominal cavity. Specifically, AT surrounding the heart muscle can be classified into paracardial AT—surrounding the parietal pericardium—and EAT—enclosed within the pericardial sac—and in direct contact with the myocardium. There is thus no anatomical barrier between EAT and the heart, making EAT a unique fat depot able to have a bidirectional paracrine interaction with the myocardium.^47,48

EAT is mainly located in the atrioventricular and interventricular grooves and, to a lesser extent, on the atria, LV free wall, and apex of the heart.^48,49 It is primarily supplied by the branches of the coronary arteries, sharing the same circulation with the myocardium. The direct contact and shared microcirculation suggest that secretory products from EAT could affect cardiomyocyte function through paracrine mechanisms (i.e. passive diffusion), through vasocrine mechanisms (i.e. via the vasa vasorum and microcirculation), or both.

6. Physiological role of EAT

In healthy conditions, EAT is an active contributor to the functioning and structure of the heart, conferring an overall protective role to the myocardium and coronary arteries. It does so by playing an important role in energy metabolism, acting as an endocrine organ, and providing mechanical support (Figure 2).^49,50 It must be noted, however, that the physiological roles of EAT have been predominantly observed in experimental and animal models and have not always been corroborated in human studies.

Due to its proximity, EAT forms an external protective layer on the myocardium and surrounds the coronary arteries, through which EAT could provide structural support and offer mechanical protection against any type of shock and cardiac contraction.

EAT is constituted by both white and brown adipocytes, resulting in a beige AT phenotype.⁴⁹ While white adipocytes are mainly responsible for energy storage, brown adipocytes are involved in heat production.

White adipocytes serve as a local energy storer and supplier. Upon increases in energy demand, white adipocytes stimulate lipid metabolism and release free fatty acids (FFAs) into the circulation, acting as a major energy source for the myocardium. Additionally, animal studies have shown that EAT protects the myocardium from excess FFA concentrations acting as a local storer and buffer for FFAs.⁵¹ In situations of excessive FFAs, lipotoxic effects on the myocardium are prevented due to the high basal rate of FFAs incorporation by EAT, which has been shown to be almost twice as high as that of other (cardiac) AT depots.⁵¹ FFAs are extracted from coronary arterial blood and transported to the myocardium by fatty-acid-binding protein 4 (FABP4), which has been shown to be highly expressed by EAT in humans, especially in obesity.⁵²

This process is tightly regulated by the (indirect) actions of leptin and insulin. Under physiological conditions, the pancreas secretes insulin in response to increased postprandial circulating glucose levels. Insulin stimulates glucose uptake and metabolism by the myocardium, a process that has also been shown to be mediated by EAT through glucose transporter 4 (GLUT-4). Simultaneously, animal studies have shown that increased insulin promotes FFAs storage in AT, preventing them from being an energy source for the mycoardium.⁵³ However, when there is an increase in energy demand, white adipocytes will secrete leptin, which stimulates lipolysis, releasing FFAs into the circulation. These FFAs serve as a crucial metabolic substrate for the myocardium, especially during moments of high energy demand.⁵¹

Brown adipocytes are involved in non-shivering thermogenesis, a process that combats hypothermia as well as releasing excess energy in the form of heat. Studies on the transcriptomic signature of human EAT show that the expression of mitochondria uncoupling protein 1 responsible for non-shivering thermogenesis in brown AT—is highly expressed in EAT.^50,54,55 While existing studies provide preliminary insights, more comprehensive and mechanistic research is warranted to clarify their (patho)physiological role within EAT.

In physiological conditions, EAT secretes cardioprotective adipokines, such as adiponectin, known for its insulin-sensitizing, antioxidant, and an anti-atherogenic effects on the myocardium. Adiponectin contributes to insulin sensitivity by inhibiting gluconeogenesis by the liver and stimulating glucose uptake and metabolism by the myocardium.⁵⁶ Furthermore, adiponectin has been shown to enhance AMP-activated protein kinase (AMPK) activity, a key regulator of energy homeostasis.^57,58 By activating AMPK, adiponectin stimulates FFA oxidation and consequent utilization by the myocardium, as well as decreasing lipid deposition in the myocardium. Regarding its antioxidant role, adiponectin promotes endothelial NO synthase phosphorylation, favouring NO synthesis, and stimulating vasodilation.^59,60 Furthermore, upon an increase in ROS and oxidative stress, the myocardium promotes adiponectin synthesis in EAT, which can then exert its antioxidant, cardioprotective effects.⁵⁸ In terms of its anti-atherogenic effect, adiponectin inhibits key inflammatory signalling pathways, such as the nuclear factor-κB (NF-κB) pathway, by suppressing the upstream cytokines like tumour necrosis factor alpha (TNF-α).⁶¹

7. Pathological transformation of EAT in obesity

Excessive accumulation of EAT in obesity can trigger a low-grade, chronic inflammatory state. While EAT’s physiological role is primarily cardioprotective, excessive EAT deposition and systemic inflammatory state can trigger a pathological transformation of EAT.⁶² This transformation could render EAT a hypertrophic, proinflammatory, profibrotic, and proarrhythmogenic fat depot.

As adiposity increases, EAT becomes predominantly dedicated to lipid storage. It is suggested that, as with ageing and advanced CAD, the adipocyte profile undergoes a brown-to-white transformation resulting in more white adipocytes, an increase in adipocyte size, and eventually EAT hypertrophy.^11,63 Furthermore, EAT expansion is, at least partly, influenced by an epithelial-mesenchymal transition resulting in atrial epicardial progenitor cells differentiating into adipocytes, both in vitro and in vivo.⁶⁴

As EAT expands, its lipid storage capacity increases, accompanied by enhanced lipolysis, resulting in a greater release of FFAs into the circulation.⁵¹ This increase manifests two main effects related to the metabolic homeostasis of the myocardium. First, the increased circulating FFAs interfere with insulin signalling pathways, promoting cellular desensitization to insulin.^65,66 Consequently, insulin resistance promotes uptake and utilization of FFAs while diminishing cellular uptake of glucose, resulting in impaired glucose metabolism. Given that cardiomyocytes rely on glucose as their second main energy substrate, this disruption can significantly hamper their functioning.⁶⁷

Secondly, excessive circulating FFA levels and oxidation can lead to lipotoxic effects when the amount of FFA exceeds the storage capacity of the EAT. This excess contributes to mitochondrial dysfunction, heightened oxidative stress, production of ROS, and apoptosis within the myocardium and neighbouring tissues.^68,69 The enhancement of oxidative stress contributes to the low-grade inflammatory state observed in obesity. Additionally, the pathological level of circulating FFAs leads to intramyocardial lipid accumulation that might contribute to cardiac hypertrophy, contractile dysfunction, and lipid-induced inflammation in the myocardium.

Increased EAT volume in obesity is also closely related to leptin resistance. As fat mass increases, so does leptin secretion. However, prolonged elevated leptin levels in obesity can desensitize the brain to leptin’s signals, compromising appetite control and leading to excessive calorie intake, exacerbating weight gain and EAT expansion. Leptin resistance is tightly associated with insulin resistance, where the one enhances the other and vice versa, and both contribute to increased oxidative stress and favour a low-grade inflammatory state.

In pathological conditions, EAT shifts towards a proinflammatory fat depot which is infiltrated by numerous inflammatory cells, including neutrophils, lymphocytes, T cells, mast cells, and proinflammatory Type 1 macrophages.^57,70 Subsequently, through the activation of the NF-κB pathway, there is an up-regulation in proinflammatory cytokine secretion, such as TNF-α), interleukin (IL)-1β, IL-18, and IL-6, and cell surface adhesion molecules like intracellular adhesion molecule-1 (ICAM-1), which promote pathological myocardial remodelling.^71–74 The complex immune response seen in pathological AT remodelling is detailed elsewhere.⁷⁵ In addition to cytokines, EAT can stimulate the production and secretion of ROS, hydrogen sulphide, lipid metabolites, and miRNAs from adipocytes, which can exert proinflammatory effects on the underlying myocardium. Consequently, the combined secretome of EAT contributes to a low-grade inflammatory state in HFpEF which is predominantly orchestrated by activation of toll-like receptor 4 (TLR-4), abundantly expressed in cardiomyocytes.⁷⁶ Furthermore, animal studies indicate that inhibition of the TLR-4 pathway reduces the production of proinflammatory cytokines, leading to the reversal of myocardial remodelling and improvement in cardiomyocyte function.^71,77 The up-regulation of the inflammatory state of EAT could therefore mediate, at least partly, lipid infiltration and subsequent LV remodelling, as well as myocardial fibrosis in HFpEF, while also contributing to the progression of associated comorbidities, such as AF and CAD.

8. Potential inflammatory effect of EAT on derangements of the myocardium in HFpEF

As part of its pathological transformation, EAT has been shown to undergo a shift in its secretome, favouring the secretion of profibrotic and proinflammatory adipokines.^78,79 This proinflammatory secretome not only disrupts the balance of adipokines that act on the myocardium, but additionally contributes to local and systemic inflammation. Consequently, these adipokines could trigger metabolic, structural, and functional alterations in the myocardium through the activation of various pathways, contributing to the development of the characteristic hallmarks of HFpEF. It is of notice that the relationship between a proinflammatory EAT secretome and the myocardium is bidirectional; while EAT can exert a local inflammatory effect on the myocardium and cardiomyocytes, inflammation within the myocardium can reciprocally influence the pathological transformation of EAT.¹¹ Nonetheless, the complex bidirectional communication between EAT and the myocardium/vasculature in HFpEF warrants further investigation.

This review explores several adipokines selected from existing evidence that highlights their role in cardiovascular disease and their potential implication in HFpEF. It is important to note that knowledge of EAT’s secretome in the context of HFpEF is limited, primarily derived from observational and associative studies, with mechanistic studies being scarce. Therefore, we draw insights from EAT in other pathological conditions and propose a selection of adipokines that may play a role in the pathogenesis of this disease (Figure 3). However, this selection is not exhaustive, and the associations remain speculative, pending future mechanistic studies.

8.1 Leptin

As EAT expands, there is an increase in leptin production and secretion by EAT adipocytes. Leptin is recognized to activate the NF-κΒ pathway and stimulate ROS activity, promoting oxidative stress on the myocardium (Figure 3). Additionally, leptin promotes cardiac fibrosis by augmenting collagen deposition, stimulating the synthesis of profibrotic mediators, such as TGF-β, and exacerbating oxidative stress through activation of the Akt pathway^80,81 (Figure 3). Furthermore, observational studies show that increased circulating levels of leptin are associated with LV hypertrophy and worse exercise capacity in HFpEF.^15,82,83

8.2 Activin-A

Activin-A is an adipokine of the TGF-β superfamily that is highly expressed in human EAT and is known for its profibrotic effects. It exerts its influence by binding to activin receptor-like kinase 4 (ALK4), thereby initiating a downstream signalling cascade that promotes collagen accumulation and subsequent myocardial fibrosis⁸⁴ (Figure 3). Histological analyses show significant interstitial fibrosis on EAT’s neighbouring myocardium, suggesting a paracrine effect of EAT’s secretome on the myocardium.⁸⁴ Notably, in patients with AF, there is an up-regulated expression of ALK4 and Activin-A, which correlates positively with atrial fibrosis.⁸⁵ Moreover, induced ALK4 deficiency attenuates cardiac fibrosis and reduces structural and electrophysiological changes in mice, as well as cardiac dysfunction, underscoring its implication in fibrotic remodelling.^85,86 This pathological mechanism might not only be present in AF, as increased Activin-A serum levels have also been found in heart failure patients.⁸⁷ Extrapolating this evidence to the obesity-related HFpEF phenotype, Activin-A secreted by EAT might be an intermediate marker mediating increased fibrosis in these patients, ultimately affecting myocardial stiffening and thus impaired diastolic relaxation.

Besides a role in myocardial fibrosis, Activin-A can also impair insulin signalling in cardiomyocytes, contributing to insulin resistance and impaired cardiomyocyte contractile function.^88,89 In the context of HFpEF, this may favour lipotoxicity and heightened oxidative stress, further contributing to diastolic dysfunction.

8.3 RBP4

In both obesity and HFpEF, levels of the proinflammatory factor retinol-binding protein 4 (RBP4) are elevated.^57,90 It has been established that EAT is a significant source of RBP4, and that in pathological conditions, its expression is elevated.⁹¹ Animal studies suggest that RBP4 mediates the vicious cycle between insulin resistance and HFpEF by activating the TLR-4 pathway, involved in impaired insulin signalling, inflammation, and hypertrophic myocardial remodelling.⁹⁰ In this context, RBP4 promotes the production of ROS, inducing cardiomyocyte hypertrophy, as well as reducing the expression of GLUT-4 in cardiomyocytes, impairing insulin-mediated glucose uptake and metabolism⁹⁰ (Figure 3). This shift leads to reduced cardiac efficiency and increases metabolic stress on cardiomyocytes.⁹² Notably, the TLR-4 pathway also plays a role in adipocytes, where RBP4 promotes the expression of proinflammatory cytokines and macrophages in mice and humans, further impairing insulin signalling.⁹³ Additionally, considering that GLUT-4 is highly expressed in adipocytes, EAT may directly contribute to insulin resistance in HFpEF through the action of RBP4.

8.4 Adiponectin

In pathological states, such as obesity, hypertension, CAD, and HFpEF, the production of adiponectin by EAT is diminished, and its cardioprotective effect is lost.^83,94,95 The decline in adiponectin levels reduces its capacity to inhibit inflammatory pathways and stimulate NO synthesis, contributing towards increased oxidative stress and inflammation within the myocardium.^59,60 These effects can lead to down-regulation of PKG activity, a process associated with cardiomyocyte stiffness and hypertrophy³¹ (Figure 3). Additionally, reduced adiponectin levels lead to down-regulation of AMPK activity, reducing FFA oxidation and increasing lipid deposition in the myocardium^57,58 (Figure 3). Decreased adiponectin levels significantly contributed to heart failure progression in an animal model of volume overload heart failure, in part through diminished AMPK phosphorylation.⁹⁶

8.5 FABP4

Heightened expression of FABP4 in EAT occurs in obesity and is implicated in both EAT expansion and hypertrophy, and myocardial dysfunction.^52,97 On one hand, FABP4 potentially mediates these effects by regulating metabolic substrate utilization of cardiomyocytes, favouring FFA oxidation while reducing glucose utilization.⁹⁷ This can promote insulin resistance, impacting myocardial structure and function through altered metabolism, increased inflammation, oxidative stress, and over-activation of the neurohormonal system.

On the other hand, increased FABP4 can have cardiodepressive effects on cardiomyocytes, suppressing their contractility, and mediating atherosclerosis progression, suggesting a paracrine role of FABP4.⁹⁸ Overall, the evidence points to heightened FABP4 expression in obesity contributing to pathological functional changes in HFpEF. As such, increased FABP4 expression has been associated with adverse LV hypertrophy and dysfunction, correlating with increased mortality and hospital admission in this population.⁹⁹

8.6 Resistin

Resistin is another adipokine released by EAT identified as a potential link between inflammation, obesity, and cardiovascular disease.^100,101 Resistin is involved in insulin resistance and impaired glucose tolerance, as well as in impaired vascular endothelial cell function.¹⁰² Resistin triggers the up-regulation of key vascular endothelial cell markers, such as vascular cellular adhesion molecule-1 (VCAM-1) and ICAM-1 through the NF-κB pathway.^103,104 This activation of adhesion molecules initiates atherogenesis by promoting macrophage recruitment and infiltration, eventually leading to endothelial dysfunction. Notably, resistin’s actions on vascular endothelial cells are countered by levels of adiponectin, underscoring the balance between these adipokines in vascular homeostasis.¹⁰³ In HFpEF, elevated resistin is associated with worse exercise capacity but not with HFpEF status or prognosis.^83,105 However, the mechanistic involvement of resistin has not been explored in the context of HFpEF. Given resistin is released by EAT, it is of interest to explore whether resistin can play a (in)direct role in HFpEF pathophysiology.

8.7 Omentin-1

Omentin-1 is an antioxidant and anti-inflammatory molecule that is highly expressed in EAT.^50,106 However, omentin-1 levels decline during obesity, which has been linked to insulin resistance, reduced glucose uptake, and heightened inflammatory activity characterized by increased oxidative stress and TNF-α-induced VCAM-1 expression.^57,107,108 Furthermore, reduced omentin-1 is associated with the development of atherosclerosis, CAD, and adverse outcomes in heart failure patients.^106,109 Therefore, exploring the potential role of omentin-1 as a therapeutic target to ameliorate heightened inflammation in the obesity-related HFpEF phenotype is warranted.

9. EAT as a therapeutic target in HFpEF

The phenotypic diversity of HFpEF and the interplay with different comorbidities warrant a tailored treatment approach. In the context of the obesity-related HFpEF phenotype, targeting the pathological transformation and underlying inflammatory pathways mediated by EAT may present new therapeutic avenues. We therefore present therapeutic options that pose a clinical benefit that arise from changes to EAT or EAT-mediated pathways.

Sodium-glucose cotransporter-2 inhibitors (SGLT2i), like dapagliflozin and empagliflozin, are the first type of pharmacological drugs to reduce the risk of heart failure–related hospitalization in HFpEF trials.^110,111 Given the low expression of SGLT2 in cardiomyocytes, it is likely that SGLT2i mediate their positive effects in HFpEF by off-target mechanisms in the myocardium.¹¹² At least in part, these effects might be mediated by EAT. Upon SGLT2i administration, EAT volume decreases and glucose uptake by EAT is enhanced, suggesting that these types of drugs might improve EAT’s secretome and may improve metabolic regulation associated with insulin resistance in obesity.^113,114 Furthermore, SGLT2i-treated patients show a reduction of ECM and cardiomyocyte volume, along with a decrease in inflammatory biomarkers, such as FABP4, IL-1, IL-6, and TGF-β, indicating a role in improving myocardial inflammation and fibrosis.^114,115 Additionally, SGLT2i show a mild, sustained diuretic effect that can improve volume overload and congestion seen in HFpEF, while reducing the need for diuretics use.¹¹⁶

Glucagon-like peptide 1 agonists (GLP-1a), such as semaglutide and liraglutide, are a class of anti-diabetic drugs known for their glucose-lowering effects and insulin sensitization, which show a rapid and significant reduction of EAT thickness independent of body weight loss.¹¹⁷ Clinical trials in patients with HFpEF show symptom reduction and exercise capacity improvement after GLP-1a administration, across the entire obesity spectrum.¹¹⁸ It is suggested that part of the effect of GLP-1a might be mediated by EAT, given that this fat depot expresses specific GLP-1 receptors.¹¹⁹ While GLP-1a primarily enhances insulin sensitivity, in EAT, it induces adipocyte browning through activation of the AMPK pathway and reduces local adipogenesis by improving FFA oxidation.^120,121 However, a reduction in EAT is likely accompanied by simultaneous reductions in other fat depots. GLP-1a reduces both VAT and SAT in equal proportions, and these reductions correspond with overall weight loss.¹²² However, due to the presence of specific GLP-1 receptors on EAT, these agonists may directly remodel EAT into a more cardioprotective fat depot, and therefore have a direct effect mediated by EAT, independent of overall weight loss.¹⁹

Bariatric surgery causes drastic and often lasting weight loss, triggering favourable metabolic changes in obese patients.¹²³ Significant weight loss is typically accompanied by a reduction of abdominal and visceral fat depots, including EAT.^124,125 Although robust and longitudinal data in patients with HFpEF is lacking, significant weight loss may improve haemodynamics, exercise capacity, and symptoms.¹²⁶ This effect could be mediated, at least in part, by reducing EAT-mediated pericardial constraint, which may alleviate pressure on the LV and improve its (diastolic) function.

Statins, particularly atorvastatin, show improved cholesterol profiles and reduction in EAT.¹²⁷ Known for their lipid-lowering and anti-inflammatory effects, statins could influence EAT expansion and hamper the secretion of proinflammatory molecules, contributing to improvements in cardiovascular health. While statins are not recommended in the guidelines for HFpEF treatment, they are often prescribed for patients with HFpEF to address cardiovascular comorbidities, such as hypertension, CAD, and dyslipidaemia.³ Furthermore, some suggest that early initiation of statin use is associated with improved outcomes in HFpEF, although this is a topic of current debate.¹²⁸

10. Future research perspectives

Despite increasing efforts to understand the role of EAT in HFpEF, several critical areas require further investigation. First, the impact of EAT's mechanistic and inflammatory effects on patient outcomes remains unclear. While associations with worsening heart failure, diabetes, and arrhythmias exist, EAT’s relation to disease progression, quality of life, and mortality in HFpEF is limited. Secondly, given EAT’s proximity to the myocardium and vasculature, its effects are likely bidirectional. Future studies should explore whether pathological transformation of EAT directly drives remodelling of the myocardium and vasculature, whether cardiac remodelling influences EAT’s pathological transformation, or whether it is a combination of both. Thirdly, it is important to determine whether EAT functions as a whole or whether regional EAT depots differently affect the myocardium and diastolic function. Although some studies find no correlation between EAT localization and cardiac function, others contest these findings.^20,129 Therefore, clinical and preclinical studies should clearly distinguish between different EAT localizations to better understand its role. Lastly, technical limitations inherent to EAT research currently hinder our understanding of its role in HFpEF pathophysiology and may influence the interpretation of existing findings. These limitations include the difficulty of obtaining high-quality EAT biopsies from patients with HFpEF, limited availability of tissue for mechanistic studies, tissue heterogeneity, and the ethical and methodological constraints associated with retrieving EAT biopsies from control populations.

11. Conclusion

The incidence of HFpEF and obesity is rising globally, making it imperative to understand the obesity-related HFpEF phenotype. The pathological transformation of EAT in this phenotype is a potential contributor to hallmarks of HFpEF. EAT is thought to cause structural and functional myocardial abnormalities by promoting dysregulated lipid metabolism, cardiac fibrosis, oxidative stress, a proinflammatory state, and endothelial dysfunction. Consequently, these alterations may stiffen the myocardium, leading to LV hypertrophy and decreased diastolic function. Understanding how EAT mediates these alterations is crucial for developing novel personalized therapies to reverse haemodynamic derangements in HFpEF. Consequently, mechanistic studies investigating the composition of EAT and its secretome, and their interaction with the myocardium in the context of HFpEF, are warranted to advance our knowledge and treatment strategies.

Funding

M.L.H. is financially supported by the Hartstichting (2020T058).

Data availability

No new data were generated or analysed in support of this research.

References

Author notes