Macrophage Immunometabolism in Epicardial Adipose Tissue and Obesity-Related Atrial Fibrillation

Abstract

Obesity is a crucial risk factor for atrial fibrillation (AF). The underlying mechanism is partly related to increased epicardial adipose tissue (EAT). EAT directly contacts the adjacent cardiac myocardium where immune and metabolic activities interact. Obesity induces inflammatory and metabolic signals within EAT, which directly influence atrial structure and electrical remodeling. However, the specific immunometabolic mechanisms linking EAT macrophages to atrial arrhythmogenesis remain unclear. We highlight evidence that nutrient excess, hypoxia, oxidative stress, and adipocyte-derived inflammatory signals reprogram glucose, lipid, and amino acid metabolism in EAT macrophages, promoting persistent pro-inflammatory and profibrotic states. Metabolic pathways involving mechanistic target of rapamycin (mTOR)–hypoxia-inducible factor-1α (HIF-1α) signaling, fatty acid and ceramide metabolism, the NOD-, LRR- and pyrin domain-containing protein 3 (NLRP3) inflammasome, and metabolic–epigenetic interactions may sustain macrophage activation and amplify local inflammation. These metabolically reprogrammed macrophages subsequently promote atrial fibrosis, electrical heterogeneity, and autonomic dysfunction, while systemic inflammation, insulin resistance, and neurohumoral activation further reinforce the EAT–atrial pathological circuit. Emerging evidence also suggests that EAT characteristics may improve AF risk stratification and that weight loss and metabolic interventions can attenuate EAT inflammation. More importantly, targeting macrophage immunometabolism may offer a mechanistically distinct approach to interrupt the link between obesity and atrial remodeling rather than treating AF solely after its substrate has developed. Collectively, targeting EAT immunometabolism may shift obesity-related AF management from treating an established arrhythmia toward identifying and modifying the metabolic-inflammatory substrate that promotes its development and recurrence, potentially complementing conventional rhythm-control and ablation strategies. Future studies should establish causal relationships and identify clinically actionable metabolic targets for EAT macrophage-directed therapy.

Graphical Abstract

Introduction

Atrial fibrillation (AF) is the most frequently encountered arrhythmia in clinical practice and remains an important cause of morbidity and mortality through complications including heart failure, stroke, and thromboembolic events.^1,2 The global burden of AF continues to rise, with an estimated 52.55 million individuals affected by AF or atrial flutter in 2021.³ Among its multiple risk factors, obesity has been established by epidemiological and genetic evidence as an independent and causal determinant of AF, and its increasing prevalence is expected to further contribute to the growing burden of AF.^3–10 Initial investigations regarding obesity and AF mainly relied on body mass index (BMI), a global indices of adiposity, to estimate AF risk. However, growing clinical and imaging evidence indicates that the volume, radiodensity, regional distribution, and inflammatory phenotype of epicardial adipose tissue (EAT) are closely associated with AF onset, persistence, and post-ablation recurrence, suggesting distribution and qualitative characteristics of adipose tissue confer greater predictive value than BMI.^11–13

This emerging perspective has highlighted EAT as a potential mechanistic link between obesity and AF. Anatomically contiguous with the atrial myocardium, EAT functions as an active endocrine and paracrine tissue capable of transmitting metabolic and inflammatory signals directly to the adjacent atrium.¹⁴ Inflammation is a main feature of EAT-mediated arrhythmogenicity, of which immune cells, such macrophages and B cells, play an important role.^15–17 In the obese state, this EAT microenvironment promotes immune-cell recruitment and activation, particularly macrophage accumulation and pro-inflammatory polarization.^18–20 Recently, research has further shifted toward immunometabolic reprogramming, emphasizing the reciprocal interplay between intracellular metabolic pathways, such as glycolysis, fatty acid oxidation, and oxidative phosphorylation, and immune effector functions.^21–23 This perspective extends conventional understanding of EAT biology by focusing on how obesity-driven metabolic stress can reprogram macrophage function and thereby sustains local inflammatory and profibrotic activity. However, the mechanisms linking immunometabolic reprogramming of EAT macrophages to atrial remodeling and arrhythmogenesis remain incompletely understood.

The present review synthesizes current knowledge on the role of immunometabolic reprogramming of epicardial adipose tissue in the pathogenesis of obesity-related AF. Specifically, it encompasses: 1) EAT as an endocrine-immunometabolic interface in obesity-related AF; 2) Macrophage immunometabolism in remodeled EAT; 3) From local EAT inflammation to atrial remodeling; 4) Integration with systemic cardiometabolic stress; 5) Translational opportunities: from diagnosis to therapeutic intervention.

It is important to delineate the scope of this review. This review does not exhaustively survey all obesity-related metabolic pathways or the full spectrum of immune cell populations, such as B lymphocytes and mast cells. The interplay between systemic metabolic disorders, including diabetes, and AF is considered only insofar as it intersects with the EAT-macrophage axis. By circumscribing the scope in this manner, we aim to foreground immunometabolic reprogramming of EAT as a main mechanistic conduit, thereby offering a focused and conceptually coherent framework to inform future basic and translational investigations.

This narrative review was based on a targeted literature search conducted in PubMed and Web of Science. The search covered publications from database inception to July 2026. The search was restricted to articles published in English. Search terms were used in different combinations according to the major themes of the review. For example, in PubMed, the search terms were combined according to the specific topic using the Boolean operators AND and OR. The major search terms included “epicardial adipose tissue”, “atrial fibrillation”,“obesity”, “macrophage”, “immunometabolism”, “metabolic reprogramming”, “glycolysis”, “lipid metabolism”, “amino acid metabolism”, “inflammation”, “atrial remodeling”, “epigenetics”, “trained immunity”, “imaging”, “weight loss”, and “bariatric surgery”. Peer-reviewed human studies, clinical imaging and interventional studies, systematic reviews and meta-analyses, as well as key mechanistic studies directly relevant to the EAT-macrophage-AF axis were prioritized. Because direct mechanistic evidence from human EAT remains limited, relevant animal studies, cultured macrophage studies, and studies of non-EAT adipose tissue were also included when they provided important mechanistic insights into macrophage immunometabolism. Study selection was based on relevance to the major mechanistic and translational themes of this narrative review rather than on predefined systematic-review eligibility criteria.

EAT as an Endocrine-Immunometabolic Interface in Obesity-Related AF

Obesity-Induced EAT Remodeling

EAT is unevenly distributed and lies adjacent to the atrial myocardium, particularly along the posterior wall of the left atrium. There is no fascia barrier between them, and they share the same microcirculatory network.^24,25 In the early developmental stage, EAT functions as thermogenic brown adipose tissue, and gradually transforms into white adipose tissue to store energy. EAT is not merely a simple lipid storage depot. Instead, it functions as an active endocrine and paracrine organ. It secretes a variety of adipokines, pro-inflammatory cytokines and vasoactive substances, which directly act on the adjacent myocardium and coronary arteries.²⁶ This anatomical continuity facilitates paracrine crosstalk between EAT and the underlying atrial myocardium, allowing metabolic and immune signals generated within EAT to directly influence atrial structure and electrophysiological remodeling.²⁶

Consistent with its anatomical and functional characteristics, clinical observations validate the close association between EAT abnormalities and AF. Patients with AF have significantly greater EAT volume than healthy individuals,²⁷ and EAT enlargement is more pronounced in persistent AF than in paroxysmal AF.²⁸ Additionally, EAT volume can independently predict AF recurrence after catheter ablation.¹¹ However, recent clinical studies have updated this understanding, highlighting that qualitative biological changes of EAT possess superior prognostic value compared with simple volumetric increase. EAT radiodensity, rather than its volume, is more strongly correlated with major adverse cardiovascular events in asymptomatic populations.¹² In this context, EAT with a small volume but severe inflammation can induce atrial fibrosis and myocardial remodeling by releasing proinflammatory cytokines, whereas large-volume EAT with healthy anti-inflammatory properties has limited pathogenic effects.²⁶

Mechanistically, obesity triggers comprehensive and maladaptive EAT remodeling, which involves not only volumetric growth but also biological alterations and hypoxia in the local microenvironment. The increased volume and thickness of EAT in obese individuals primarily stem from adipocyte hyperplasia rather than simple adipocyte hypertrophy.²⁹ Accompanying such structural changes, EAT transforms from a quiescent energy-storing organ into a functionally active inflammatory, endocrine and paracrine tissue. This transition is reflected by the decline in production of anti-inflammatory adiponectin and elevation in pro-inflammatory mediators (leptin, TNF-α, and IL-6) and abundant profibrotic cytokines. Importantly, the local concentration of these mediators within EAT exceeds their systemic levels, creating a steep paracrine gradient that facilitates diffusion of inflammatory signals into the adjacent atrial myocardium.³⁰ In parallel, EAT also releases profibrotic mediators, including activin A, connective tissue growth factor, matrix metalloproteinases (MMPs), and transforming growth factor-β (TGF-β1 and TGF-β2), as well as extracellular vesicles that participate in intercellular communication.^13,31 Fundamentally, the progressive expansion of EAT mass further outpaces the adaptive capacity of local vasculature, leading to insufficient tissue perfusion and impaired oxygen diffusion, which ultimately generates a hypoxic microenvironment.³² This hypoxic condition stabilizes hypoxia-inducible factor-1α (HIF-1α), a key transcription factor that triggers a series of maladaptive molecular responses. Specifically, HIF-1α activation promotes tissue inflammation and fibrosis, disrupts metabolic homeostasis, inhibits adiponectin expression, and facilitates the recruitment of immune cells, thereby exacerbating EAT pathological dysfunction.^33,34 This inflammatory circuit is not limited to macrophages alone. CD8⁺ T cells, which are enriched in obese EAT, contribute to macrophage recruitment and activation and thereby intensify adipose inflammation.^18,19 Their involvement may also extend to thromboinflammatory responses observed in newly diagnosed AF.³⁵ These cellular interactions transform EAT from a relatively quiescent fat depot into an active inflammatory one that perpetuates tissue remodeling. Given the anatomical continuity between EAT and the left atrial myocardium, particularly along the posterior wall, these factors can directly promote atrial fibrosis, conduction heterogeneity, and electrical remodeling.³⁶

EAT Remodeling Mediated Mechanisms of AF

Inflammation is the key feature of EAT. Through sustained release of pro-inflammatory and profibrotic mediators, remodeled EAT promotes fibroblast activation, collagen deposition, and myocardial stiffening.^31,37 Within this process, macrophages play an important role that integrate local metabolic stress signals with inflammatory and fibrotic responses. Importantly, the pathogenic role of EAT macrophages is determined not simply by their abundance, but by their functional and metabolic state within the remodeled adipose microenvironment.

Macrophage Immunometabolism in Remodeled EAT

Increasing evidence indicates that the phenotypic and functional diversity of macrophages in EAT is fundamentally controlled by dynamic metabolic reprogramming under obese conditions.^38,39 Divergent metabolic pathways not only sustain basal cellular activities of EAT macrophages but also serve as core signaling regulators that define macrophage polarization and inflammatory profiles. In this context, the conventional classification of classically activated pro‑inflammatory (M1) macrophages and alternatively activated anti‑inflammatory (M2) macrophages represents two primary metabolic archetypes.⁴⁰ Pro‑inflammatory macrophages rely mainly on glycolysis and anabolic lipid metabolism to maintain inflammatory activation, while reparative macrophages depend on fatty acid oxidation and oxidative phosphorylation (OXPHOS) to support anti‑inflammatory effects and tissue homeostasis.⁴⁰ Although this traditional M1/M2 paradigm cannot fully cover the continuous phenotypic spectrum of tissue-resident macrophages in vivo, it offers a feasible framework to interpret the immunometabolic mechanisms underlying EAT macrophage functional changes in obesity.

The immunometabolic characteristics of EAT macrophages are particularly susceptible to disruption in the obese microenvironment.⁴¹ Obese EAT is characterized by chronic nutrient overload, hypoxia, oxidative stress and sustained adipocyte-derived inflammatory signaling, which collectively reshape substrate utilization and mitochondrial function in resident macrophages. These metabolic alterations drive profound macrophage reprogramming and shift cellular phenotypes toward persistent pro‑inflammatory and profibrotic states, thereby disrupting local immune homeostasis and triggering maladaptive EAT remodeling. Within this pathological microenvironment, glycolysis, oxidative phosphorylation, fatty acid oxidation and lipid metabolism pathways undergo coordinated changes to regulate macrophage behaviors. As the central hub for the metabolism of three major nutrients including carbohydrates, lipids and amino acids, the tricarboxylic acid (TCA) cycle plays a core bridging role in the metabolic reprogramming of EAT macrophages (Figure 1). Glucose through glycolysis, fatty acids via β-oxidation, and amino acids via carbon skeleton catabolism ultimately converge into the TCA cycle as acetyl-CoA or TCA cycle intermediates, enabling the integration of material and energy metabolism and tightly linking systemic metabolic disorders to local inflammatory responses.

Figure 1.

Metabolic reprogramming of EAT macrophages in obesity-related AF. Obese microenvironment induces comprehensive metabolic reprogramming of epicardial adipose tissue (EAT) macrophages, driving pro-inflammatory and profibrotic phenotypes via dysregulated glucose, lipid, and amino acid metabolism. Cluster of differentiation 8-positive (CD8⁺) T cell-derived C-C Motif Chemokine ligand 5 (CCL5) drives the recruitment of macrophages to EAT. Obese EAT macrophages present an atypical signature with elevated aerobic glycolysis and retained oxidative phosphorylation (OXPHOS), controlled by the mechanistic target of rapamycin (mTOR)–hypoxia-inducible factor 1 alpha (HIF-1α)/AMP-activated protein kinase (AMPK) axis and suppressed AMPK activity. Epigenetic modifications including DNA methyltransferase 3 beta (DNMT3B)-mediated deoxyribonucleic acid (DNA) methylation and altered histone acetylation and methylation establish innate immune memory, stably locking this pro-inflammatory glycolytic-OXPHOS metabolic state and sustaining long-term macrophage hyperactivation. Excessive palmitic acid and ceramide trigger lipotoxicity-mediated Toll-like receptor 4 (TLR4)–nuclear factor kappa B (NF-κB)/mitogen-activated protein kinase (MAPK) and NOD-, LRR- and pyrin domain-containing protein 3 (NLRP3) inflammasome activation, thereby amplifying pro-inflammatory cytokine secretion and stabilizing macrophage pro-inflammatory polarization via epigenetic priming. For amino acid metabolism, dysregulated arginine and glutamine metabolism closely correlates with EAT macrophage functional abnormalities and local inflammatory disorders, and bidirectional crosstalk between amino acid metabolic dysfunction and epigenetic remodeling further disrupts macrophage immune homeostasis.

When such metabolic reprogramming remains persistent and imbalanced, EAT macrophages fail to maintain tissue-stabilizing functions and instead propagate chronic inflammation and fibrotic progression. This process gradually promotes the formation of arrhythmogenic atrial substrates, which drives the initiation and progression of obesity-related atrial fibrillation.

The following sections focus on the key metabolic circuits of glucose, lipid, and amino acid metabolism that modulate EAT macrophage bioactivities and further determine how these immune cells affect atrial remodeling processes.

Glucose Metabolism

In obese EAT, macrophages undergo specific glucose metabolic reprogramming that differs from the metabolic pattern of macrophages activated in acute inflammatory responses. In classical acute inflammation, M1 macrophages rely exclusively on aerobic glycolysis with suppressed mitochondrial OXPHOS, a metabolic pattern defined as the Warburg effect. By contrast, EAT macrophages in obese conditions exhibit a mixed metabolic state with elevated aerobic glycolysis and retained OXPHOS activity.^42,43 This abnormal metabolic feature is induced by multiple adverse factors in obese EAT, including chronic low-grade inflammation, excessive lipid deposition and sustained pseudo-hypoxic stress. Studies on obese mouse models have confirmed that adipose tissue macrophages display enhanced glycolytic flux and increased mitochondrial oxidative metabolism simultaneously. This special glucose metabolic phenotype supports continuous inflammatory activation of EAT macrophages, establishing a metabolic basis for the crosstalk between obesity-induced metabolic disturbance and atrial inflammatory remodeling.⁴³

The mTOR–HIF-1α axis is a key regulator of this metabolic phenotype. Nutrient excess and persistent inflammatory stimulation activate mechanistic target of rapamycin complex 1 (mTORC1), whereas reduced AMP-activated protein kinase (AMPK) activity weakens its inhibitory effect on the mTORC1–HIF-1α pathway.⁴⁴ HIF-1α enhances glucose uptake and glycolytic gene expression, while mTOR signaling contributes to the maintenance of mitochondrial oxidative metabolism, resulting in the coexistence of enhanced glycolysis and retained OXPHOS.^42,45 Such glucose metabolic rewiring of EAT macrophages can be stably locked by epigenetic modifications and innate immune training, sustaining long-term pro-inflammatory metabolic phenotypes. Epigenetic regulation mediates persistent transcriptional changes without altering deoxyribonucleic acid (DNA) sequences and affects AF chronicity.⁴⁶ In obese adipose tissue, DNA methyltransferase 3 beta (DNMT3B)-mediated DNA methylation is closely involved in the regulation of macrophage glycolytic profiles. Increased DNMT3B expression in M1-like adipose tissue macrophages maintains the transcription of glycolytic and pro-inflammatory genes.^47,48 Histone acetylation further reinforces glycolytic reprogramming; elevated histone H3 lysine 9 and histone H3 lysine 14 acetylation relaxes chromatin accessibility, facilitating the expression of HIF-1α-targeted glycolytic genes and sustaining the dual glycolysis-OXPHOS metabolic signature.^49,50 Furthermore, glucose-driven metabolic stress induces stable epigenetic remodeling characterized by elevated trimethylation of histone H3 lysine 4 and acetylation of histone h3 lysine 27 at inflammatory gene promoters. This epigenetic priming establishes innate immune memory, allowing EAT macrophages to retain hyper-glycolytic features even after metabolic stress relief and exhibit amplified glycolytic responses upon secondary stimulation, which continuously fuels atrial inflammatory remodeling.^51,52

Lipid Metabolism

Disordered lipid metabolism represents another key metabolic alteration of EAT macrophages in obesity. Obesity promotes excessive lipolysis in EAT, leading to massive release of free fatty acids. Macrophages exhibit elevated expression of lipid transporters including cluster of differentiation 36 and fatty acid transport proteins on their plasma membrane. Despite systemic lipid metabolic disturbance and partial suppression of β‑oxidation, a fraction of fatty acids can still translocate into mitochondria normally and undergo continuous catabolism to produce acetyl‑CoA, which directly fuels the TCA cycle. Palmitic acid, one of the major saturated fatty acids in EAT, acts not only as an energy substrate but also as an effective inflammatory trigger.^53,54 Accumulated palmitic acid directly activates macrophage inflammatory responses and induces the secretion of pro-inflammatory cytokines including interleukin-6 (IL-6) and tumor necrosis factor-α (TNF-α).⁵⁵ Under persistent lipid overload, impaired fatty acid β-oxidation further promotes intracellular ceramide accumulation.⁵⁶

Palmitic acid and accumulated ceramide jointly amplify EAT inflammation through lipotoxic signaling. These lipids activate the toll-like receptor 4/cluster of differentiation 14 (TLR4/CD14) receptor complex and downstream myeloid differentiation primary response 88-dependent nuclear factor kappa B (NF-κB) and mitogen-activated protein kinase (MAPK) pathways, thereby promoting pro-inflammatory gene transcription.^56,57 Ceramide accumulation and excessive lipid uptake also induce endoplasmic reticulum stress and mitochondrial dysfunction, leading to increased reactive oxygen species (ROS) production.^58,59 Mitochondrial damage further promotes the release of mitochondrial DNA (mtDNA) into the cytoplasm, which acts as a damage-associated molecular pattern and, together with ROS, facilitates NOD-, LRR-, and Pyrin Domain-Containing Protein 3 (NLRP3) inflammasome activation.^60,61 Subsequent caspase-1 activation promotes interleukin-1β (IL-1β) maturation and sustains the pro-inflammatory phenotype of EAT macrophages.⁶¹ The resulting inflammatory factors and lipotoxic mediators can act on adjacent atrial cardiomyocytes through paracrine pathways, contributing to atrial electrical remodeling and interstitial fibrosis and ultimately promoting obesity-related AF.

Epigenetic modifications and trained immunity further sustain lipotoxicity-induced inflammatory activation in EAT macrophages. Lipotoxic stimulation alters histone modification patterns and stabilizes pro-inflammatory transcriptional programs, thereby maintaining inflammatory signaling and NLRP3 inflammasome responsiveness.^49,50 Persistent lipid overload may also impair histone deacetylase 3-associated anti-inflammatory transcriptional reprogramming, further favoring a pro-inflammatory state.⁶² In addition, lipid-induced metabolic stress can establish durable epigenetic immune memory, allowing EAT macrophages to remain highly responsive to subsequent lipotoxic stimulation even after the initial metabolic stress has subsided.⁶³

Amino Acid Metabolism

Amino acid metabolism participates in the modulation of macrophage immune function and metabolic homeostasis, and its dysregulation contributes to inflammatory changes in obese EAT. Among various amino acids, arginine and glutamine metabolism are closely associated with macrophage polarization and effector function changes.⁶⁴ Under inflammatory and nutrient-excess conditions associated with obesity, glutamine uptake and catabolism are markedly enhanced in macrophages, thereby supplying carbon substrates to sustain TCA cycle flux. Dysregulation of these two amino acids can induce abnormal immune responses and support chronic inflammatory progression in adipose tissues.⁶⁴

Existing evidence^64,65 indicates that amino acid-related immunometabolic changes in macrophages are also under epigenetic regulation, which provides a potential mechanism for persistent inflammatory disturbance. Epigenetic remodeling can reshape the transcriptional profiles of amino acid metabolic enzymes and transporters, leading to sustained disorder of arginine and glutamine metabolism in obese EAT macrophages. Such metabolic abnormalities further stabilize pro-inflammatory epigenetic signatures and form a bidirectional regulatory loop between amino acid metabolism and epigenetic modification. Combined with innate immune memory, this regulatory pattern may contribute to the long-term imbalance of macrophage immune homeostasis, exacerbating chronic EAT inflammation and atrial pathological damage in obesity.^51,63

From Local EAT Inflammation to Atrial Remodeling

AF is initiated by focal ectopic activity, while its persistence is sustained by reentrant and ongoing focal firing.⁶⁶ Sustained reentry requires a vulnerable atrial substrate, which is formed by atrial remodeling, involving both structural and functional changes such as atrial enlargement, fibrosis, and shortened refractoriness (Figure 2). Following immunometabolic reprogramming, EAT macrophages act as major effector cells that translate local metabolic stress into atrial structural, electrical, and autonomic remodeling.

Figure 2.

EAT macrophage-cardiac crosstalk mediates atrial remodeling in atrial fibrillation. Epicardial adipose tissue (EAT) macrophages translate local pathological stress into atrial structural and electrical remodeling. Mechanistically, macrophage-derived paracrine factors, predominantly transforming growth factor beta (TGF-β), activate the canonical TGF-β/Smad signaling axis, driving atrial fibroblast-myofibroblast transformation, excessive collagen deposition and progressive atrial fibrosis. Additionally, pro-inflammatory macrophage cytokines including tumor necrosis factor alpha (TNF-α) and interleukin-1 beta (IL-1β) disrupt cardiomyocyte ion channel expression and calcium homeostasis, impair gap junction function, and induce electrical heterogeneity. The synergistic structural and electrical remodeling establishes vulnerable atrial substrates to initiate and sustain atrial fibrillation (AF).

Atrial Structural Remodeling

Atrial structural remodeling mainly refers to pathological changes in atrial tissue morphology and interstitial composition, and atrial fibrosis serves as its major pathological manifestation. Under pathological conditions of obesity, large numbers of infiltrating and activated macrophages accumulate within both atrial tissue and EAT. These cells release large amounts of cytokines, growth factors and inflammatory mediators into the myocardial microenvironment, and they act directly on atrial fibroblasts to trigger the transformation of fibroblasts into myofibroblasts.⁶⁷ In this biological process, transforming growth factor-β (TGF-β) acts as the main paracrine mediator.⁶⁸ Macrophages synthesize and secrete TGF-β, and this molecule binds to TGF-β type II receptors on the membrane of atrial fibroblasts. The binding event induces the phosphorylation and activation of Smad2/3, and activated Smad2/3 then forms heterotrimers with Smad4. The complex translocates into the cell nucleus and functions as a transcriptional complex to increase the expression of fibrosis-related genes, including α-smooth muscle actin, type I collagen and type III collagen. Under the regulation of the classical TGF-β/Smad signaling pathway, atrial fibroblasts convert into the myofibroblast phenotype. The transformed cells exhibit greater contractile ability and produce more collagen, and these changes lead to excessive deposition of extracellular matrix, as well as atrial interstitial fibrosis and scar formation. Damaged or stressed cardiomyocytes secrete IL-1β, interleukin-18, monocyte chemoattractant protein-1 (MCP-1) and other molecules.⁶⁹ These molecules recruit and activate more macrophages, and they also increase the expression of TGF-β and its receptors.⁷⁰ In this way, a self-amplifying loop between inflammation and structural remodeling takes shape.

Electrophysiological Remodeling

Atrial electrical remodeling acts as the core pathological basis for the onset and persistence of atrial fibrillation. Activated macrophages, especially pro‑inflammatory M1 subtypes, secrete large amounts of pro‑inflammatory cytokines including TNF‑α, IL‑1β and IL‑6.⁷¹ These cytokines regulate the expression and function of key ion channels on cardiomyocyte membranes. TNF‑α suppresses T‑type calcium channels and affects the activity of sarcoplasmic/endoplasmic reticulum Ca²⁺-ATPase 2a (SERCA2a),⁷² while IL‑1β lowers the expression and current density of L‑type calcium channels.⁷³ Both changes disrupt intracellular calcium homeostasis and trigger abnormal calcium handling as well as delayed afterdepolarization.⁷⁴ Inflammatory cytokines also reduce potassium currents such as the transient outward potassium current, and this effect prolongs action potential duration and raises electrical heterogeneity.⁷⁵

In obese guinea pigs, higher expression of IK shortens action potential duration, and such changes link to atrial fibrillation progression.⁷⁶ Diet‑induced obese mice show elevated atrial oxidative stress alongside ion channel remodeling. The downregulation of sodium current represents a major plasma membrane current change that drives cell atrial fibrillation phenotypes in obese animal models.⁷⁷ Connexin proteins form major components of cardiomyocyte gap junctions and support cardiac electrical conduction. Pro‑inflammatory cytokines including TNF‑α, IL‑1β and macrophage migration inhibitory factor reduce connexin 40 and connexin 43 expression, and thus affect atrial electrical activity and conduction.⁷⁸ Accordingly, immune cells and macrophages can induce atrial electrical remodeling through direct and indirect pathways, and further drive atrial fibrillation development.

Integration with Systemic Cardiometabolic Stress

EAT is also a functional medium that responds to systemic metabolic stress. It receives circulating nutrient signals, endocrine molecules and inflammatory stimuli, and converts these systemic changes into localized pathological remodeling in the atrium. Multiple systemic metabolic changes in obesity can continuously regulate and strengthen these local immune activities, such as chronic low-grade inflammation, oxidative stress, insulin resistance, and neurohumoral activation⁷⁹ (Figure 3). The arrhythmogenic effects of EAT thus originate from the crosstalk between systemic metabolic stress and local immunometabolic remodeling, instead of simple local adipose inflammation.

Figure 3.

Systemic endocrine stressors drive EAT-mediated atrial remodeling and atrial fibrillation susceptibility. Light purple sector: Obesity-induced visceral adipose tissue inflammation promotes pro-inflammatory immune cell activation and excessive reactive oxygen species (ROS) production, resulting in systemic chronic low-grade inflammation and oxidative stress. These systemic stimuli further impair epicardial adipose tissue (EAT) adipocyte function, facilitate pro-inflammatory polarization of EAT-resident immune cells, and amplify local inflammatory and oxidative injury. Light blue sector: Increased visceral fat releases free fatty acids and inflammatory cytokines that exacerbates insulin resistance and contributes to hyperglycemia. Hyperglycemia-induced advanced glycation end products (AGEs) further enhance oxidative stress and activate transforming growth factor beta (TGF-β)/Smad signaling, accelerating extracellular matrix deposition and atrial fibrosis. Light yellow sector: Sustained inflammatory and hormonal stimulation activates EAT-resident ganglionated plexuses (GPs), inducing sympathetic hyperinnervation and autonomic remodeling. Simultaneously, EAT-derived angiotensinogen, angiotensin, and angiotensin-converting enzyme (ACE) promote local renin-angiotensin-aldosterone system (RAAS) activation, enhancing angiotensin II (Ang-II)-mediated pathological changes.

Chronic Low-Grade Inflammation and Oxidative Stress

Visceral adipose tissue in obesity exhibits robust activation of inflammatory signaling pathways.⁷⁹ As a result, immune cells in adipose tissue, especially macrophages, but also T cells and neutrophils, adopt pro-inflammatory phenotypes and produce inflammatory cytokines, ultimately contributing to systemic chronic low-grade inflammation and oxidative stress.⁷⁹ Systemic inflammatory activation and excessive ROS accumulation can further deteriorate the functional status of EAT adipocytes, facilitate proinflammatory polarization of EAT immune cells, and amplify local inflammatory and oxidative damage.⁸⁰

Consistent with this EAT-mediated pathogenic cascade, experimental studies on obese mice have confirmed that obesity elevates atrial oxidative stress in diet-induced obese mice. These mice exhibit higher levels of mitochondrial ROS in atrial tissue compared with control animals,⁷⁷ along with increased protein expression of NADPH oxidase 2 (NOX2). NOX2 has been closely linked to AF pathophysiology in recent studies, and its upregulation correlates with AF development. Increased NOX2 expression promotes angiotensin II–induced cardiac hypertrophy, myocardial fibrosis and ROS generation.^81,82 Transgenic mice with NOX2 overexpression show a higher incidence of AF after rapid atrial pacing. In contrast, inhibition of mitochondrial oxidative stress by mitochondrial catalase overexpression reduces AF occurrence.⁸³ NOX2-derived ROS and the subsequent shift in intracellular redox status modify key proteins involved in excitation-contraction coupling, including ryanodine receptor 2, calcium/calmodulin-dependent protein kinase II (CaMKII), phospholamban and SERCA2a, through oxidative modification. Such alterations cause abnormal calcium release, myocardial fibrosis and arrhythmia progression.^84,85

Insulin Resistance and Hyperglycemia

Increased body fat, particularly visceral fat, is a primary driver of insulin resistance, as it releases free fatty acids and inflammatory cytokines that impair insulin signaling.⁸⁶ Lipotoxicity, resulting from elevated fatty acid levels, further exacerbates insulin resistance and contributes to hyperglycemia.⁸⁷ Under obese, insulin-resistant and hyperglycemic conditions, EAT undergoes profound phenotypic switching from a metabolically quiescent state to an activated pro-inflammatory and profibrotic phenotype.⁸⁸ This transformation is directly triggered by circulating free fatty acids, excess glucose and systemic inflammatory factors derived from visceral fat and impaired glucose metabolism.⁸⁸ Functionally, activated EAT exacerbates cardiac injury via multiple paracrine pathways. First, EAT continuously secretes large quantities of free fatty acids, which aggravate local lipotoxicity in atrial cardiomyocytes and endothelial cells, synergizing with systemic lipotoxicity to disrupt cellular metabolism. Second, EAT serves as a major local source of pro-inflammatory mediators. It abundantly produces TNF-α, IL-1β, IL-6 and chemokine MCP-1, which superimpose on the inflammatory milieu induced by hyperglycemia. The infiltrated immune cells further secrete profibrotic factors such as TGF-β to activate cardiac fibroblasts and initiate fibrotic progression.^89,90 Third, activated EAT itself releases a panel of profibrotic cytokines including TGF-β, which collaborates with hyperglycemia-induced advanced glycation end products (AGEs). AGEs exert profibrotic effects through the AGE receptor pathway and oxidative stress, while the downstream TGF-β/Smad signaling pathway accelerates extracellular matrix deposition and atrial fibrosis.⁹¹ Obesity, insulin resistance and hyperglycemia interact with each other to maintain EAT-related inflammation, which further triggers atrial fibrosis and electrical remodeling, and thus raises the susceptibility to AF.^89,92

Neurohumoral Activation

Both clinical observations and animal experiments support the presence of autonomic nerve remodeling and renin-angiotensin-aldosterone system activation under obese conditions. Clinical data show increased sympathetic activity in obese individuals.⁹³ Animal studies using high‑fat diet induced obese rats also exhibit sympathetic hyperinnervation and nerve hyperplasia, along with reduced basal acetylcholine release in the atrium.⁹⁴ EAT is enriched in cardiac ganglionated plexuses (GPs), where sympathetic and vagal nerves are densely distributed. Under obese metabolic stress, sustained inflammatory and hormonal stimulation activates EAT-resident GPs, inducing excessive sympathetic neuronal proliferation and hyperactivity. Sympathetic activation shortens the atrial effective refractory period and increases the amplitude of myocardial calcium transients, and eventually promotes AF progression.^95,96

EAT serves as an inflammatory adipose tissue depot in obese individuals with abundant pro-inflammatory cytokines such as TNF-α. This cytokine can further upregulate angiotensinogen protein levels and enhance angiotensin II secretion in adipocytes.⁹⁷ Enhanced angiotensin-converting enzyme 2 (ACE2) expression is detected in regions with thickened extracellular matrix in fresh EAT samples collected from patients with AF. The anatomical proximity of EAT to the myocardium, combined with its inflammatory secretory profile and ACE2 expression, enables EAT to mediate myocardial inflammation and cytokine storm.⁹⁸ Angiotensin II triggers multiple pathological cascades conducive to AF. It stimulates NADPH oxidase-mediated ROS production and elevates the levels of oxidized CaMKII,⁸² and these responses trigger disordered intracellular calcium release to affect AF susceptibility. Furthermore, angiotensin II suppresses matrix metalloproteinase-1 expression in cardiac fibroblasts and promotes type I collagen synthesis, which leads to atrial fibrosis.⁹⁹

Translational Opportunities: From Diagnosis to Therapeutic Intervention

Imaging-Based Risk Stratification and Diagnosis

EAT volume is widely validated as a robust cardiovascular imaging biomarker, with patients exhibiting AF showing higher volumes than healthy controls, and persistent AF patients having greater volumes than those with paroxysmal AF.^11,100 Importantly, EAT volume predicts recurrence after catheter ablation, particularly when quantified regionally around the left atrium (LA), where it exhibits stronger and more consistent associations than total cardiac fat.^101,102 Large, low-density EAT reflecting increased water content and inflammatory cell infiltration represents a high-risk phenotype.¹³

Computed tomography (CT) attenuation, expressed in Hounsfield units (HU), enables noninvasive assessment of EAT quality. Lower HU values indicate metabolically active, inflamed tissue, and composite indices such as the fat attenuation index have been used to quantify epicardial inflammation, stratifying obese AF patients at higher risk for post-ablation recurrence.^103,104 Cardiac magnetic resonance (CMR) mapping techniques further provide tissue characterization: T2 mapping sensitively detects myocardial edema and inflammation, while native T1 and extracellular volume reflect fibrosis. Elevated T1 and T2 values in AF patients correlate with adverse outcomes and identify patients with heightened structural remodeling.^105,106 Metabolic imaging with ¹⁸F-fluorodeoxyglucose positron emission tomography/computed tomography (¹⁸F-FDG PET/CT) offers complementary insights by capturing glucose uptake in EAT and atrial tissue. Elevated FDG uptake indicates active macrophage-mediated inflammation and correlates with AF burden and recurrence risk. Persistent AF patients exhibit higher right atrial FDG uptake than paroxysmal AF patients, and FDG positivity independently predicts adverse outcomes, including stroke.^107–110 Integrating anatomical volume, tissue quality, and metabolic activity provides a multidimensional assessment of AF susceptibility and procedural risk.

These imaging biomarkers have direct implications for catheter ablation. Larger LA EAT volumes correlate with higher recurrence rates post-ablation, with each 1 mL increase in volume increasing recurrence risk by approximately 6% in obese populations.¹¹¹ Automated segmentation and indexing to body surface area enhance predictive accuracy and enable personalized procedural planning. Pre-procedural identification of high-risk patients allows clinicians to optimize metabolic status and consider weight management strategies, which reduce EAT volume and inflammation, potentially improving ablation outcomes. Furthermore, recognition of regional EAT heterogeneity may inform lesion set design, particularly in areas such as the posterior atrial wall, to mitigate localized recurrence risk.^28,101,102

Therapeutic Approaches on EAT and Macrophage Immunometabolism

Therapeutic strategies targeting EAT macrophage immunometabolism can be broadly divided into indirect and direct approaches. Indirect strategies primarily reduce adiposity and systemic metabolic stress, whereas direct strategies target macrophage immunometabolic pathways.

Clinical and preclinical studies have consistently demonstrated that weight reduction, achieved either through structured lifestyle interventions or bariatric surgery, can reverse atrial remodeling and reduce AF burden.^112,113 Bariatric procedures decrease insulin resistance, systolic blood pressure, and EAT volume, correlating with a lower risk of AF recurrence.^112,114

Weight loss via pharmacological inhibition with glucagon-like peptide-1 receptor agonists (GLP-1RAs) has also shown efficacy in obesity-related AF.^115–118 Real-world evidence demonstrates that GLP-1RA therapy in obese AF patients undergoing catheter ablation is associated with reduced recurrence risk, lower progression to permanent AF, and decreased all-cause mortality and cardiovascular hospitalizations.^119,120

The sodium-glucose cotransporter 2 inhibitors (SGLT2i) have emerged as significant agents in managing obesity and atrial fibrillation (AF), particularly in patients with type 2 diabetes and heart failure. These medications not only promote weight loss but also exhibit cardioprotective effects that may reduce the incidence and severity of AF.⁴⁶ The cardioprotective effects partly stem from the reduction of EAT volume and thickness and alleviation of inflammation. Clinical data indicate that SGLT2i can reduce EAT volume by an average of 6.57 mL and decrease EAT thickness by 1.55 mm.¹²¹ A relevant meta-analysis reported a standardized mean difference of −1.79, which confirms the ability of SGLT2i to shrink EAT mass.¹²²

Apart from morphological improvement, SGLT2i also exert anti-inflammatory effects. They restrain NLRP3 inflammasome activity to slow atrial fibrosis and suppress arrhythmia development.¹²³ In addition, it is reported that SGLT2i reduce inflammation in EAT by modulating pathways that affect macrophage polarization. Specifically, SGLT2i raise ketone body content to reduce the release of inflammatory molecules, and promote macrophage polarization toward an M2-like phenotype.¹²⁴ Direct modulation of macrophage immunometabolism may provide a complementary therapeutic strategy for obesity-related AF. mTOR signaling has been implicated in AF in both human and experimental studies, and pharmacological inhibition of AKT/mTOR signaling reduces atrial immune-cell recruitment, inflammation, and structural remodeling in murine models.^125–128 Conversely, AMPK activation may counteract pro-inflammatory metabolic programming through suppression of mTORC1 activity.¹²⁹ NLRP3 inhibition is also a promising strategy, as genetic or pharmacological suppression of NLRP3 has been shown to prevent obesity-induced AF and ameliorate atrial electrophysiological remodeling and calcium-handling abnormalities in experimental models.^130–132 Together, these findings support the AMPK–mTOR axis and NLRP3 inflammasome as potential immunometabolic targets, although the available evidence remains predominantly mechanistic and preclinical. Metformin, an AMPK-activating agent, may further improve metabolic homeostasis, reduce oxidative stress and inflammation, and reverse Warburg-like metabolic reprogramming, with available evidence suggesting potential benefits for AF prevention.¹³³ Table 1 depicts evidence-tiered endocrine and metabolic targets for EAT macrophage-directed therapy in obesity-related AF.

Table 1.

Evidence-Tiered Endocrine and Metabolic Targets for EAT Macrophage-Directed Therapy in Obesity-Related AF

Intervention	Endocrine/Metabolic Target	EAT	Evidence Level	Macrophage Polarization	Evidence Level	AF	Evidence Level
Weight loss	Adiposity	Reduce EAT volume.134	1b	Modulates macrophage polarization and inflammation.33	2a	Reduce AF inducibility.135,136	2a
Bariatric surgery	Adiposity	Reduce EAT volume.134,137	2a	Modulates macrophage polarization and inflammation.138,139	2b	Reduce AF incidence.135,140	2a
GLP-1RA (Semaglutide)	Incretin signaling	Reduce EAT volume.141,142	2b	Modulates macrophage polarization and inflammation.143,144	3b	Reduce AF recurrence.119	2b
SGLT2 inhibitor (Empagliflozin)	Glucose disposal and caloric loss	Reduce EAT volume.145,146	2b	Modulates macrophage polarization and inflammation.147	3b	Reduce AF recurrence148 and incidence.149	2a
AMPK indirect activators (Metformin)	AMPK signaling	Reduce EAT volume.150	3a	Modulates macrophage polarization and inflammation.151	5	Reduce AF incidence.152	3a
NLRP3 inhibitor (MCC950, Colchicine)	NLRP3 inflammasome	None	4	Modulates macrophage polarization and inflammation.153	5	Reduce AF recurrence.154,155	4
mTOR inhibitor (Sirolimus)	mTOR signaling	None	5	Modulates macrophage polarization and inflammation.156	5	None	5

Notes: Evidence levels were classified according to clinical trial and translational research grading standards. Level 1a (Highest): Systematic reviews/meta-analyses of high-quality Randomized Controlled Trials (RCTs), providing the most reliable, unbiased evidence. Level 1b: Individual high-quality RCTs. Level 2a: Systematic reviews of cohort studies. Level 2b: Individual high-quality cohort studies or low-quality RCTs. Level 3a: Systematic reviews of case-control studies. Level 3b: Individual high-quality case-control studies. Level 4: Case series, case-control, or cohort studies with high risk of bias. Level 5 (Lowest): Expert opinion, bench research, or mechanistic studies without direct clinical evidence.

Abbreviations: EAT, epicardial adipose tissue; AF, atrial fibrillation; GLP-1RA, GLP-1 receptor agonist; SGLT2i, SGLT2 inhibitor.

Discussion

Current evidence supports EAT as an important immunometabolic interface linking obesity to atrial remodeling. The pathogenic relevance of EAT appears to depend not only on its volume but also on its inflammatory and metabolic phenotype. In particular, obesity-induced metabolic stress can reprogram EAT macrophages toward persistent pro-inflammatory and profibrotic states, thereby contributing to local inflammation and atrial structural and electrical remodeling. In EAT macrophages, nutrient excess, lipid accumulation and hypoxia suppress AMPK while activating mTORC1–HIF-1α, promoting glycolytic reprogramming and metabolic dysfunction. Lipotoxicity and mitochondrial stress increase ROS, which, together with mitochondrial DNA, activates NLRP3, whereas TLR4-mediated signaling activates NF-κB, amplifying inflammatory cytokine production. ROS and inflammation further impair mitochondrial and metabolic homeostasis, creating a self-reinforcing feedback loop. These signals promote TGF-β/Smad activation in atrial fibroblasts, driving myofibroblast differentiation and extracellular matrix deposition, while inflammatory mediators and ROS disrupt cardiomyocyte calcium handling and ion-channel function. Thus, AMPK/mTOR/HIF-1α, ROS/NLRP3/NF-κB, and TGF-β constitute interconnected nodes linking metabolic stress to persistent atrial inflammation, fibrosis, and electrical remodeling (Figure 4). These findings suggest that macrophage immunometabolism provides a mechanistic framework linking systemic metabolic dysfunction to the local arrhythmogenic substrate. Nevertheless, several uncertainties limit interpretation of the current evidence. The relative importance of EAT quantity versus its inflammatory and metabolic phenotype remains unresolved, partly because imaging methods, regional EAT definitions, and AF populations vary across studies. In addition, the conventional M1/M2 framework does not fully capture the heterogeneity of human EAT macrophages. Much of the mechanistic evidence linking glycolytic reprogramming, lipotoxicity, AMPK/mTOR/HIF-1α signaling, ROS, NLRP3 activation, and trained immunity to macrophage dysfunction derives from animal models, cultured macrophages, or non-EAT adipose tissues, limiting direct extrapolation to human AF.

Figure 4.

Integrated metabolic-inflammatory signaling network in obesity-associated atrial remodeling. Obesity-related nutrient excess, lipid overload, and hypoxia dysregulate the adenosine monophosphate-activated protein kinase (AMPK)-mechanistic target of rapamycin (mTOR)-hypoxia-inducible factor-1α (HIF-1α) axis and increase glycolytic activity. Lipotoxicity and mitochondrial dysfunction promote reactive oxygen species (ROS) accumulation, NOD-, LRR- and pyrin domain-containing protein 3 (NLRP3) inflammasome activation, and nuclear factor kappa B (NF-κB)-mediated inflammation. These pathways converge on cardiomyocytes and fibroblasts, where inflammatory and transforming growth factor-β (TGF-β)/Smad signaling drive electrical remodeling and fibrosis, ultimately increasing susceptibility to atrial fibrillation.

Multimodal imaging modalities, including CT, CMR, and ¹⁸F-FDG PET/CT, have emerged as valuable tools for quantifying EAT burden, inflammatory activity, and myocardial tissue alterations, providing mechanistically informed risk stratification that can guide both procedural planning and targeted therapeutic interventions. Current preclinical and clinical evidence further suggests the regulatory effects of multiple metabolic and inflammatory interventions on EAT homeostasis, macrophage polarization, and atrial fibrillation progression in the context of obesity. Lifestyle intervention via weight loss serves as the most well-supported strategy, possessing high-level clinical evidence to reduce EAT volume, normalize aberrant macrophage polarization and adipose inflammation, and mitigate multiple AF-related phenotypes, including AF inducibility, burden, duration, and incidence. Similarly, bariatric surgery, GLP-1 receptor agonists, and SGLT2 inhibitors exert protective effects against obesity-associated AF through modulating EAT morphology and macrophage inflammatory status, with moderate clinical evidence confirming their efficacy in lowering AF incidence and recurrence. In contrast, targeted pharmacological agents including AMPK activators, NLRP3 inflammasome inhibitors, and mTOR inhibitors primarily regulate macrophage polarization and inflammation without consistent impacts on EAT volume. Notably, the overall body of existing evidence exhibits prominent hierarchical disparities and limitations. Most intervention strategies only provide indirect correlative evidence linking improved EAT immunometabolism and suppressed AF progression, while high-quality, direct causal evidence verifying that EAT macrophage polarization mediates the anti-AF effects of these interventions remains scarce. Specifically, low-grade evidence and insufficient translational studies still constrain the definitive mechanistic interpretation of NLRP3 and mTOR inhibitors in obesity-related AF.

These evidence gaps are compounded by the predominance of cross-sectional imaging studies, heterogeneous patient populations, variable definitions and quantification of EAT, and limited longitudinal data directly linking macrophage metabolic states to clinical outcomes. Future research should prioritize longitudinal, multi-center studies integrating high-resolution imaging, tissue-level phenotyping, and functional assays of macrophage metabolism to establish causal relationships and clarify temporal dynamics. The development of standardized imaging and biochemical biomarkers to quantify EAT inflammation and metabolic activity will be crucial for translating mechanistic insights into patient-specific therapeutic strategies. Additionally, single-cell RNA sequencing and spatial transcriptomics will be particularly valuable for resolving macrophage heterogeneity and defining their microanatomical interactions with adipocytes, fibroblasts, vascular cells, nerves, and adjacent atrial myocardium. Integration with human tissue biobank studies and multi-omics datasets will enable mechanistic mapping of immunometabolic networks, while interventional trials targeting these pathways can validate causal links and therapeutic potential. Overall, EAT macrophage immunometabolism represents a promising therapeutic target for obesity-related AF by providing a potential means to modify the metabolic-inflammatory substrate underlying atrial remodeling. However, direct causal clinical evidence remains limited, and well-designed human mechanistic and interventional studies are required before this strategy can be translated into routine clinical practice.

Funding Statement

This work was supported by grants from the National Natural Science Foundation of China (82274414), National Natural Science Foundation of Sichuan (2024NSFSC2118), Key Laboratory of Acupuncture and Medicine Research (Nanjing University of Chinese Medicine) and Key Laboratory of Acupuncture for Senile Disease (Chengdu University of TCM), Ministry of Education (zykf202501), Chengdu University of TCM (QJRC2022037 and QJJJ2024011) for cultivation of young sci-tech talents.

Data Sharing Statement

No datasets were generated or analyzed during the preparation of this review. All information discussed in this article is derived from previously published studies, which are appropriately cited in the reference list.

Author Contributions

Xin Cao: Conceptualization, Methodology, Writing-review and editing, Supervision, Project administration, Funding acquisition. Zefei Jiang: Conceptualization, Methodology, Writing-review and editing, Supervision. Linrui Xia: Methodology, Investigation, Formal analysis, Visualization, Writing-original draft, Writing-review and editing. Xiaoxiang Sun: Investigation, Formal analysis, Writing-review and editing. Luyi Xie: Investigation, Formal analysis, Writing-review and editing. Shuang Liu: Investigation, Writing-review and editing. Teva Phanaksri: Investigation, Formal analysis, Writing-review and editing. Zhi Yu: Visualization, Writing-review and editing. All authors gave final approval of the version to be published; have agreed on the journal to which the article has been submitted; and agree to be accountable for all aspects of the work.

Disclosure

Xin Cao reports grants or contracts supporting this work from the National Natural Science Foundation of Sichuan (2024NSFSC2118), Key Laboratory of Acupuncture and Medicine Research (Nanjing University of Chinese Medicine) and Key Laboratory of Acupuncture for Senile Disease (Chengdu University of TCM), Ministry of Education (zykf202501), Chengdu University of TCM (QJRC2022037 and QJJJ2024011) for cultivation of young sci-tech talents. Luyi Xie reports grants or contracts from Fund and laboratory outside of this work.The authors report no other conflicts of interest in this work.