Hepatokines are a class of proteins produced by hepatocytes, mainly including fetuin-A, fibroblast growth factor 21 (FGF21), angiopoietin-like proteins (ANGPTLs), leukocyte cell-derived chemotaxin 2 (LECT2), and selenoprotein P (Sep), which have a positive or negative regulatory effect on the pathogenesis of NAFLD by inducing glucose and lipid metabolism, oxidative stress, and systemic inflammation (6).

Adipokines are polypeptides widely present in the adipose tissue, which usually regulate liver fat accumulation, IR, and inflammatory response through autocrine, paracrine and endocrine ways, exerting an important role in the pathogenesis of NAFLD (35). Nowadays, the association between NAFLD and circulating adiponectin and leptin levels has been well-studied, and circulating leptin levels increase, but adiponectin levels decrease with increasing NAFLD severity (35). Notably, the data on the contribution of other adipokines [resistin, retinol binding protein-4 (RBP4), visfatin, chemerin, and irisin] to the occurrence and progression of NAFLD are relatively limited and inconclusive, which needed to be further explored.

Growing evidence disclosed that cytokines were the key mediators of signal transmission among live cells, which may be critically participated in the pathophysiology of NAFLD by inducing hepatic inflammation, IR and fibrosis (60). As an essential part of innate immune system, complements and their cleavage products were mostly produced by hepatocytes, which also played a fundamental role in the pathogenesis of NAFLD (61).

Extracellular vesicles (EVs) are nano-sized vesicles surrounded by a lipid bilayer that can be released by most cells. EVs, mainly including small EVs (sEVs) and microvesicles, have the function of mediating intercellular communication within tissues and organs through transferring proteins, lipids, and RNAs. Excessive lipid overload and accumulation were major characteristics of NAFLD that can trigger hepatocytes to produce and release EVs, and the latter also participated in the initiation and progression of hepatic inflammation, steatosis, fibrosis, and HCC (73, 74). Accumulating data demonstrated that hepatocytes-derived EVs not only activate macrophage-associated inflammation response via TNF-related apoptosis-inducing ligand (TRAIL) in a death receptor (DR5)-dependent manner but also promote endothelial inflammation and atherogenesis via microRNA-1 (75). Hepatocyte-derived sEVs also promoted hepatic fibrosis in NAFLD by directly modulating HSC phenotype and up-regulating the expression of profibrogenic genes, including Col1α1, α-smooth muscle actin (αSMA), and tissue inhibitor of metalloproteinase-2 (TIMP-2) (76). Furthermore, HSC-derived sEVs contained connective tissue growth factor (CCN2), and its concentration increased with HSC activation, which may amplify or fine-tune fibrotic signal transduction and may be considered as a potential candidate biomarker to assess hepatic fibrosis (77). Besides, HSC-derived EVs also contained C-X-C motif ligand 10 (CXCL10), hexokinase 1 (HK1), vascular endothelial growth factor (VEGF), matrix metalloprotein 2 (MMP2) and MMP9, which facilitated tumor growth and transfer, as well as angiogenesis (78). Notably, messenger RNA (mRNA), microRNA (miRNA) and long non-coding RNA (lncRNA) have been found in EVs that elicit a pivotal role in the pathogenesis of NAFLD, especially miR-122, miR-192, miR-21 and miR-7 (79). miR-122 and miR-192 were the most abundant RNAs in hepatocytes, and hepatic miR-122 levels were upregulated during early NAFLD development but decreased gradually with progression towards NASH and hepatic fibrosis (79, 80). Previous studies revealed that miR-122 knockdown in diet-induced NAFLD mouse model could inhibit hepatic lipogenesis via liver kinase B1 (LKB1)/AMPK signaling pathway by targeting Sirt1 (81). Moreover, miR-122 had the capacity to protect against hepatic inflammation and fibrosis by targeting Relb, FN1 and SRF, respectively (79). Therefore, miR-122 may be a potential diagnostic and/or prognostic marker of NAFLD patients. Numerous studies presented that circulating miR-122 had good accuracy in NAFLD diagnosis, with the area under the receiver operating curve of 0.67–0.93 (76). However, miR-192 and miR-21 expression increased with NAFLD progression, and miR-192 had pro-fibrotic properties by enhancing Col1α1, αSMA and TGF-β1 expression in HSC (76). miR-21 induced hepatic steatosis by increasing lipogenesis and promoted hepatic inflammation and fibrosis by inhibiting PPAR-α (82). Similarly, miR-7 levels were significantly elevated in NAFLD mouse model, triggering NLRP3 (NOD-like receptor protein 3) inflammasome activation and coronary microvascular endothelial hyperpermeability through regulation of lysosomal associated membrane protein 1 (LAMP1) and Cathepsin B (83). Therefore, miR-7 acted as a bridge between NAFLD and coronary diseases by targeting the LAMP1/Cathepsin B/NLRP3 signaling pathway. On the whole, EVs are also involved in the pathogenesis of NAFLD and can be acted as novel diagnostic biomarkers and therapeutic targets.

The liver interacts with the gut through the portal vein, and the liver is able to remove a large number of bacterial products, toxins and metabolites from the gut. However, NAFLD patients have excessive bacterial overgrowth, intestinal dysbiosis and metabolic product alterations, as well as increased intestinal permeability, which facilitates the translocation of pathogen-associated molecular pattern (PAMP)/microbial-associated molecular pattern (MAMP) to the liver, inducing hepatic and systemic inflammation and immune responses (84, 85). In particular, increasing research focus on the association between NAFLD and microbial metabolites, such as trimethylamine N-oxide (TMAO), lipopolysaccharide (LPS), bile acids, short-chain fatty acids (SCFAs), and N,N,N-trimethyl-5-aminovaleric acid (TMAVA) (86–89). Recent meta-analyses showed that NAFLD patients had significantly higher concentrations of circulating TMAO (SMD = 0.66, 95% CI: −0.12 to 1.21), and plasma TMAO levels were related to all-cause mortality in NAFLD subjects, independently of traditional risk factors (90, 91). TMAO participates in the initiation of NAFLD through a variety of mechanisms. On the one hand, TMAO promotes IR by regulating glucose metabolism and increasing the levels of CCL2 and inflammatory cytokine C-C motif chemokine ligand 2; on the other hand, it exacerbates hepatic steatosis by reducing the conversion of cholesterol into bile acids and inhibiting farnesoid X receptor (FXR) signaling (86). However, TMAO attenuates liver fibrosis by inhibiting ATP1B1 expression in vascular endothelial cells (88). LPS, also named as endotoxin, was identified as a diagnostic biomarker for NAFLD, which can interact with its receptor TLR4 in inflammatory cells (e.g., macrophages and platelets) to initiate inflammatory processes and augment NAFLD susceptibility (92). Higher endotoxin levels are indeed more prevalent in NAFLD patients versus liver-healthy controls, and its levels are associated with circulating C-reactive protein (CRP) concentrations (93). However, intestine-derived HDL3 can restrain liver injure by preventing LPS binding to hepatic macrophages and inflammatory activation through the portal vein (94). Bile acids are crucial regulators of insulin sensitivity, glucolipid metabolism and immune response, which contribute to the pathogenesis of NAFLD by regulating FXR and Takeda G-protein receptor 5 (TGR5) (3). FXR activation alleviates the progression of NAFLD through reduced lipogenesis and lipid absorption mediated by bile acids, increased FFA oxidation, affected cholesterol transportation, and inhibited NF-κB-mediated inflammatory signaling pathways (92). TGR5 activation not only modulates liver inflammation through TGR5-cAMP dependent pathways, but also induces the release of glucagon-like peptide-1 (GLP-1) and represses the activation of NLRP3 inflammasome to regulate glucose homeostasis (86). SCFAs, mainly include acetate, propionate, and butyrate, which activates the G-protein-coupled receptors (GPCRs) GPR41 and GPR43, and the latter stimulates the secretion of GLP-1 and peptide-YY in the gut enteroendocrine L cells, therefore enhancing lipid and FFA oxidation, increasing insulin sensitivity, as well as promoting hepatic lipogenesis through AMPK-dependent pathway (86, 95). In addition, SCFAs also reduces hepatic inflammation and steatosis by increasing c-Jun N-terminal kinase (JNK) and NF-κB phosphorylation, inhibiting the secretion of proinflammatory cytokines, and suppressing the activity of histone acetyltransferases to reduce the generation of regulatory T cells (96). Zhao et al. indicated that plasma TMAVA levels were increased in patients with liver steatosis compared with healthy controls, and TMAVA treatment exacerbated hepatic steatosis and caused gut microbial imbalance in mice (89). Furthermore, intestinal microbes metabolized trimethyllysine to TMAVA, which inhibited carnitine biosynthesis by competitively the binding of γ-BB to γ-butyrobetaine hydroxylase (BBOX) and reduced hepatic fatty acid oxidation (FAO), thereby promoting hepatic steatosis (89). Besides, TMAVA was a key determinant for the pathogenesis of cardiac hypertrophy through disturbed cardiac energy metabolism and altered mitochondrial ultrastructure, resulting in inhibition of FAO, myocardial lipid accumulation and carnitine synthesis (97). In summary, gut-derived factors are believed to play an indispensable role in the onset of NAFLD. TMAO has the properties of aggravating hepatic steatosis and alleviating liver fibrosis. SCFAs have beneficial roles in reducing liver inflammation and steatosis. LPS-TLR4, FXR, TGR5-cAMP, AMPK signaling may be the promising pathways to regulate the progression of NAFLD. TMAVA exacerbates liver steatosis and cardiac hypertrophy by inhibiting carnitine synthesis and hepatic fatty acid oxidation.

A bidirectional association between hypertension and NAFLD has been proposed, and recent systemic reviews and meta-analyses have indicated that NAFLD is significantly associated with an increased risk of developing hypertension (98). Several prospective studies disclosed that the presence of NAFLD independently predicted the risk of incident hypertension, and the incidence rate of hypertension increased with the severity of NAFLD (99). A large amount of evidence delineated that systemic inflammation, IR, renin-angiotensin system (RAAS) and sympathetic nervous system (SNS) activation may be the potential pathophysiological mechanisms of hypertension in NAFLD patients. Systemic inflammation induced by NAFLD can not only activate RAAS and SNS by elevated circulating cytokines and complements levels (e.g., IL-6, TNF-α, and CCL2) but also alter adipokine profiles characterized by increased leptin levels and reduced adiponectin concentrations (99). Moreover, IL-6 and TNF-α also promote the production of RAAS components, especially angiotensinogen, further contributing to systemic and local angiotensin II formation and hypertension development (100). In addition, higher leptin levels are involved in the pathogenesis of hypertension by sex-specific mechanisms that increase sympathetic activation in males and aldosterone production in females, respectively, while lower adiponectin concentrations regulate hypertension development by increasing SNS activity and induing endothelial dysfunction (101). On the other hand, other altered hepatokine and adipokine profiles in NAFLD subjects, including fetuin-A, SeP, RBP4, and FGF21, have been increasingly recognized to induce systemic IR by impairing insulin signaling, activating proinflammatory signaling and affecting lipid and glucose metabolism disorders. Hyperinsulinemia and hyperglycemia caused by IR are strongly associated with sodium and water retention and vasoconstriction, which play a vital role in the development of hypertension. Until now, very few studies have explored the relationship between EVs, gut-derived factors and hypertension in NAFLD patients. Limited evidence presented that miR-122 may cause the depression of 3′-untranslated region (UTR) variant in SLC7A1 gene expression, further leading to endothelial dysfunction and increased susceptibility to hypertension (102). SCFAs had hypotensive properties by inducing arterial vasodilation and renin secretion via GPCRs pathways, and regulating anti-inflammatory immune cells, especially regulatory T cells (103). On the contrary, TMAO had a detrimental impact on blood pressure via promoting inflammatory responses and activating RAAS, and there was a positive dose-dependent association between plasma TMAO levels and the prevalence of hypertension (104, 105). Together, substantial evidence indicated that altered hepatokine (fetuin-A, SeP, and FGF21), adipokine (leptin, adiponectin, and RBP4), cytokines (IL-6 and TNF-α), EVs (miR-122), and gut-derived factors (SCFAs and TMAO) in NAFLD patients participated in the development of hypertension through inducing inflammation, IR, endothelial dysfunction, and RAAS and SNS activation. However, whether these metabolic organ-secreted factors can be used as diagnostic and prognostic markers or therapeutic targets for hypertension in NAFLD subjects is not yet known and remains to be further explained.

Multiple levels of evidence indicated that subclinical atherosclerosis and atherosclerotic CVD were more prevalent in NAFLD patients, and the prevalence of carotid atherosclerosis and coronary heart disease (CHD) in NAFLD subjects were found to be 35.02% and 44.6%, respectively (106, 107). Meanwhile, individuals with NAFLD had a significantly higher risk of carotid atherosclerosis (OR = 3.2, 95% CI: 2.37–4.32), CHD (OR = 1.33, 95% CI: 1.21–1.45) and acute myocardial infarction (HR = 1.17, 95% CI:1.05–1.30) than non-NAFLD controls, and NAFLD had been proved to be an independent risk factor for atherosclerotic CVD (106–108). Several studies reported that NAFLD was also strongly associated with the markers of subclinical atherosclerosis, such as carotid intima-media thickness (CIMT), carotid plaque, arterial stiffness and coronary artery calcium (CAC) (107, 109). Although NAFLD can promote and aggravate atherosclerosis development, the exact pathogenesis is still poorly unclear. The potential mechanisms for accelerating atherosclerotic CVD in NAFLD patients were complex and might be related to lipid disturbances, IR, chronic inflammation, and endothelial dysfunction. Patients with NAFLD had a pro-atherosclerotic lipid profile characterized by high triglyceride, elevated very low-density lipoprotein (VLDL), high apolipoprotein B (ApoB) to apolipoprotein AI (ApoAI) ratio, and low levels of HDL, which lead to an increased CVD risk (110). IR was a pivotal characteristic of NAFLD pathogenesis that contributed to increased FFA concentrations, caused lipotoxicity, damaged endothelium-dependent vasodilation, enhanced oxidative stress, and had cardiotoxic effects. In addition, in subjects with NAFLD, inflammation appeared as increased levels of cytokines and chemokines in peripheral blood, such as IL-6, TNF-α, CRP, and MCP-1, which further accelerated atherosclerotic CVD by increasing plasminogen activator inhibitor-1 (PAI-1) and adhesion molecules levels in endothelial cells, reducing nitric oxide production, promoting the influx of low-density lipoprotein (LDL) into macrophages, and stimulating vascular smooth muscle cell proliferation (110). Besides, endothelial dysfunction and impaired endothelial flow-mediated vasodilatation observed in NAFLD patients could cause chronic inflammation, increased vasoconstriction, and augmented production of prothrombotic factors, resulting in an increased risk of atherosclerotic cardiovascular events (111). Furthermore, endothelial dysfunction was regarded as an independent risk factor for the occurrence and development of CHD (111). In short, lipid disturbances, IR, inflammation, and endothelial dysfunction are strongly associated with a higher risk of atherosclerotic CVD in NAFLD patients.

Recent studies focused on the potential role of hepatokines (fetuin-A, FGF21, LECT2, and ANGPTLs), adipokines (adiponectin, leptin, resistin and chemerin) and gut-derived factors (TMAO, bile acids and SCFAs) in the development of atherosclerotic CVD in NAFLD patients. These hepatokines and adipokines not only directly act on vascular cells but also regulate other metabolic components, such as IR, systemic inflammation and dyslipidemia, to involve in the pathogenesis of CVD. Among them, elevated fetuin-A levels in NAFLD promote an inflammatory response in vascular endothelial cells, macrophage foam formation, smooth muscle cell proliferation and collagen deposition, contributing to an increased risk of atherosclerotic disease (112). LECT2 may enhance atherosclerotic inflammatory response in endothelial cells through CD209 receptor-mediated JNK phosphorylation (113). ANGPT3, ANGPTL4 and ANGPTL8 can induce endothelial dysfunction and promote dyslipidemia by suppressing the LPL activity, which is associated with a higher risk of atherosclerosis (114). Contrary to other hepatokines, FGF21 has anti-atherosclerotic properties by inducing adiponectin production in adipocytes and by reducing cholesterol biosynthesis through inhibiting hepatic SREBP2 expression (115). Adiponectin has insulin-sensitizing and anti-inflammation effects, which can stimulate vascular endothelial nitric oxide synthase expression, improve atherogenic lipid profile, selectively inhibits endothelial cell apoptosis, and regulate the NF-κB signaling pathway, thereby slowing the progression of atherosclerosis and cardiovascular events (116). Additionally, elevated leptin, resistin, and chemerin concentrations are common in NAFLD patients that also play a crucial role in the aggravation of NAFLD and the development of coronary atherosclerosis, and the potential mechanisms are described below (116, 117). Hyperleptinemia results in selective leptin resistance that has a synergistic effect with IR, which can reduce adiponectin levels and nitric oxide synthase, promote increased IMT, and is considered a predictor of CVD (116). Resistin can induce vascular endothelial dysfunction by triggering vascular cell adhesion molecule-1 (VCAM-1) and intercellular adhesion molecule-1 (ICAM-1) release, promote the proliferation and migration of vascular smooth muscle cells by activating various intracellular signal transduction and enhance oxidative stress, vascular inflammation and lipid accumulation (116, 117). Chemerin increases atherosclerotic CVD risk by affecting the adhesion and migration of human endothelial progenitor cells and exacerbating lipid accumulation in atherosclerotic plaques through p38 MAPK signaling pathway (118). TMAO is considered as a proatherogenic metabolite, which is involved in atherosclerotic CVD by inhibiting reverse cholesterol transport, facilitating migration of macrophage foam cells to the vascular wall, promoting platelet reactivity and thrombosis, and exacerbating inflammatory gene expression through activation of MAPK/ERK and NF-κB pathways (103, 119). Bile acids mainly modulate inflammation and lipid and glucose metabolism through liver FXR and G protein-coupled bile acid receptor 1 (GPBAR1), especially VLDL and triglyceride production, de novo fatty acid synthesis and IR, ultimately regulating the progression of CVD in NAFLD patients (120). Notably, SCFAs play the protective effects on atherosclerotic CVD by decreasing pro-inflammatory cytokines (IL-6, IL-8) and VCAM-1 expression, suppressing histone deacetylases and activation of NF-κB inflammatory pathways, inhibiting endothelial ROS formation, and improving pro-atherogenic plasma lipid profile (121). Collectively, the development of subclinical atherosclerosis and atherosclerotic CVD in NAFLD patients may also be partly attributable to metabolic organ-secreted factors. In addition to the above-mentioned factors, the role of other metabolic organ-secreted factors in the pathogenesis of atherosclerotic CVD in NAFLD patients still needs further exploration.

In recent years, a number of studies focused on the relationship between NAFLD and cardiac arrhythmia, especially atrial fibrillation (AF). Several systemic reviews and meta-analyses suggested that NAFLD was associated with high risk of AF (OR = 1.27, 95% CI: 1.18–1.37), prolonged QT interval (OR = 2.86, 95% CI: 1.64–4.99), heart block (HR = 2.65, 95% CI: 1.88–3.72), premature atrial/ventricular contraction (OR = 2.53, 95% CI: 1.70–3.78), and cardiac conduction defect (OR = 5.17, 95% CI: 1.34–20.01) (122–124). Accumulated studies have demonstrated that NAFLD is often accompanied by impaired lipid and glucose metabolism, IR, oxidative stress, systemic inflammation and epicardial adipose tissue (EAT) accumulation, resulting in structural, functional, and electrical remodeling of the heart and predisposing patients to the onset of cardiac arrhythmia (125). Notably, the elevated release of proinflammatory, pro-fibrotic and prooxidant molecules in NAFLD patients such as cytokines, hepatokines, adipokines and gut-derived factors, were also able to change myocardial electrical properties and structural substrates, increasing the vulnerability of arrhythmias (126). For instance, proinflammatory cytokines, namely TNF-α, IL-1 and IL-6, can induce myocardial and atrial remodeling by modulating potassium and calcium homeostasis, prolonging action potential duration and QTc interval duration and reducing the expression of sarco-endoplasmic reticulum Ca2+ ATPase 2a (SERCA2a) and connexin-40 (125, 126). Additionally, proinflammatory cytokines also promote myolysis, cardiomyocyte apoptosis and myocardial fibrosis, and affect the expression and function of specific connexin-formed channels that possibly contribute to the development of re-entry arrhythmias. Conflicting results on the association between adipokines and AF were often reported, but a recent meta-analysis of 34 studies by Agbaedeng et al. found that adipokines, principally adiponectin and resistin, were associated with the risk of AF (127). Moreover, a Mendelian randomization study revealed that resistin levels was positively correlated with AF risk (OR = 1.09, 95% CI: 1.04–1.13) (128). Mechanistically, adipokines promoted systemic oxidative stress and inflammatory states that were involved in the occurrence and development of AF (127). Among gut-derived factors, TMAO not only enhanced susceptibility to AF by exacerbating cardiac autonomic nerves activity and promoting inflammatory responses, but also promoted ventricular arrhythmia via activating cardiac SNS (129). Furthermore, higher TMAO levels were associated with increased risk of adverse cardiovascular events in AF subjects (130). LPS had shown to up-regulate NLRP3 inflammasome expression and increase atrial fibrosis, contributing to the development of AF. Zhang et al. revealed that a selective inhibitor of the NLRP3 inflammasome (MCC950) can reduce atrial fibrosis and AF susceptibility by inhibiting inflammasome (131). Contrary to other factors, emerging lines of evidence showed that FGF21 had a beneficial effect on CVD that can protect against atrial remodeling by reducing oxidative stress and preventing atrial fibrosis, which provide new strategies for AF (132). Furthermore, SCFAs and bile acids also alleviated AF development via regulating GPR43/NLRP3 signaling pathway and promoting apoptosis of atrial myocytes, respectively (133). Strikingly, in subjects with NAFLD, the study on the role of hepatokines, adipokines and gut-derived factors in cardiac arrhythmia is scarce, and the above evidence is mainly based on general population or diabetes mellitus patients, thereby further research is needed to verify these findings.

Heart failure (HF), the end-stage of CVD, is a not uncommon but life-threatening manifestation of cardiac involvement. Growing evidence supported that NAFLD and HF often coexisted as they shared similar pathophysiological features and a variety of cardiometabolic risk factors (134). Moreover, there was a strong association between NAFLD and increased risk of new-onset HF, and this risk increased with the severity of NAFLD, especially in the fibrotic stage (134). A meta-analysis of 11 longitudinal cohort studies with a median follow-up of 10 years disclosed that NAFLD was associated with a 1.5-fold increased risk of new-onset HF, regardless of the presence of diabetes, hypertension, adiposity and other common cardiovascular risk factors (135). Simultaneously, in-hospital mortality of primary HF with preserved ejection fraction (HFpEF) and reduced ejection fraction (HFrEF) was higher in patients with NAFLD compared with those without NAFLD (136). In addition to the comorbidity of cardiovascular risk factors, several putative mechanisms that contribute to the relationship between NAFLD and increased risk of HF have been proposed. First, IR, disturbance of glucose and lipid metabolism were the crucial pathophysiological characteristics of NAFLD, which may eventually lead to the decrease of myocardial energy metabolism, cardiac dysfunction and remodeling (126, 134). Second, the activation of chronic systemic inflammation and mitochondrial dysfunction facilitated the production of higher levels of ROS and proinflammatory cytokines (e.g., IL-1β, IL-6 and TNF-α), which were the critical mediators of HF progression (126). Third, increased activity of RAAS and SNS, imbalance with angiotensin-converting enzyme 2 (ACE2)/Ang 1–7 axis, and altered expression of gut microbiota-derived metabolites were also related to HF development in NAFLD patients (82). It was worth mentioning that expression change of adipokines, hepatokines and gut-derived factors also were involved in the development of HF. For example, resistin, a new metabolic marker, was associated with the incident and severity of HF and can predict HF prognosis. Furthermore, for each standard deviation (7.45 ng/ml) increased in resistin, the risk of developing new-set HF elevated by 26% (137). These relationships may be induced by IR, abnormal glucose metabolism, inflammation, and depressed cardiomyocyte contractility. A prospective observational study demonstrated that higher Sep levels were also connected with hepatic hypoperfusion and predicted adverse outcomes in HF (138). A recent Mendelian randomization study indicated that RBP4 levels were causally associated with an increased risk of HF (OR = 1.14, 95% CI: 1.02–1.27), but the underlying pathogenesis still needs to be further explored (128). Elevated TMAO concentrations enhanced the risk of HF by increasing myocardial contractility and calcium flux in a dose-dependent manner, exacerbating myocardial fibrosis and accompanying diastolic dysfunction (139). Moreover, high TMAO levels were related to disease severity and poor prognosis in patients with HF (140). Notably, although there are few studies focusing on the association between SCFAs and HF, limited evidence suggests that SCFAs have protective effects on intestinal mucosal barrier. The depletion of SCFAs would lead to intestinal barrier disruption, which facilitates LPS and proinflammatory cytokines transport into systemic circulation and aggravates cardiac remodeling by modulating CD4+ T cells, ultimately resulting in HF (141). Altogether, the above-mentioned results indicate that metabolic organ-secreted factors expression levels are correlated well with the risk and prognosis of HF, but the exact mechanism requires further elucidation.

It is worth mentioning that altered expression levels of metabolic organ-secreted factors are associated with the pathogenesis of NAFLD and CVD, but it is relatively difficult to determine the total effect of these metabolic organ-secreted factors in the occurrence and development of cardiovascular events, given the interactions of multiple organs throughout the body. Secretome analysis and system biology approach may provide better insights into the role of metabolic organ-secreted factors, which still need further study.