Association between epicardial adipose tissue and incident heart failure mediating by alteration of natriuretic peptide and myocardial strain

BACKGROUND: Epicardial adipose tissue (EAT) has been suggested to exert deleterious effects on myocardium and cardiovascular disease (CVD) consequence. We evaluated the associations of EAT thickness with adverse outcomes and its potential mediators in the community. METHODS: Participants without heart failure (HF) who had undergone cardiac magnetic resonance (CMR) to measure EAT thickness over the right ventricular free wall from the Framingham Heart Study were included. The correlation of EAT thickness with 85 circulating biomarkers and cardiometric parameters was assessed in linear regression models. The occurrence of HF, atrial fibrillation, coronary heart disease (CHD), and other adverse events was tracked since CMR was implemented. Their associations with EAT thickness and the mediators were evaluated using Cox regression and causal mediation analysis. RESULTS: Of 1554 participants, 53.0% were females. Mean age, body mass index, and EAT thickness were 63.3 years, 28.1 kg/m(2), and 9.8 mm, respectively. After fully adjusting, EAT thickness positively correlated with CRP, LEP, GDF15, MMP8, MMP9, ORM1, ANGPTL3, and SERPINE1 and negatively correlated with N-terminal pro-B-type natriuretic peptide (NT-proBNP), IGFBP1, IGFBP2, AGER, CNTN1, and MCAM. Increasing EAT thickness was associated with smaller left ventricular end-diastolic dimension, thicker left ventricular wall thickness, and worse global longitudinal strain (GLS). During a median follow-up of 12.7 years, 101 incident HF occurred. Per 1-standard deviation increment of EAT thickness was associated with a higher risk of HF (adjusted hazard ratio [HR] 1.43, 95% confidence interval [CI] 1.19–1.72, P < 0.001) and the composite outcome consisting of myocardial infarction, ischemic stroke, HF, and death from CVD (adjusted HR [95% CI], 1.23 [1.07–1.40], P = 0.003). Mediation effect in the association between thicker EAT and higher risk of HF was observed with NT-proBNP (HR [95% CI], 0.95 [0.92–0.98], P = 0.011) and GLS (HR [95% CI], 1.04 [1.01–1.07], P = 0.032). CONCLUSIONS: EAT thickness was correlated with inflammation and fibrosis-related circulating biomarkers, cardiac concentric change, myocardial strain impairment, incident HF risk, and overall CVD risk. NT-proBNP and GLS might partially mediate the effect of thickened EAT on the risk of HF. EAT could refine the assessment of CVD risk and become a new therapeutic target of cardiometabolic diseases. TRIAL REGISTRATION: URL: https://clinicaltrials.gov. Identifier: NCT00005121. SUPPLEMENTARY INFORMATION: The online version contains supplementary material available at 10.1186/s12916-023-02836-4.