The association between heart rate variability, epicardial adipose tissue, and clinical outcomes in heart failure patients

Background Heart failure (HF) is driven by both autonomic nervous system dysfunction and metabolic dysregulation. Reduced heart rate variability (HRV) indicates sympathetic overactivity, while epicardial adipose tissue (EAT) remodeling (increased volume and density) reflects local inflammation. However, the interplay between this “neuro-adipose axis” and its combined prognostic value in HF remains unclear. Purpose This study investigated the association between Holter-derived HRV parameters and CT-derived EAT phenotypes (volume and density) in HF patients and evaluated their combined impact on HF rehospitalization and all-cause mortality. Methods In this single-center retrospective study, we screened 3,369 patients and included 304 HF patients (Jan 2018–Dec 2023) who underwent both 24-hour Holter monitoring and Coronary CT Angiography (CCTA) within one week. Patients with arrhythmias affecting HRV (e.g., AF, frequent PVCs) were excluded. EAT volume was indexed to body surface area (BSA), and EAT density was measured in Hounsfield Units (HU). The primary endpoint was a composite of HF rehospitalization and all-cause death. Associations were assessed using Kaplan-Meier and Cox regression analyses. Results During the follow-up period, patients with adverse outcomes exhibited significantly distinct neuro-adipose profiles compared to event-free survivors. A significant inverse correlation was observed between HRV metrics (SDNN, SDANN) and EAT inflammatory markers (density and volume). As illustrated in representative cases, patients with adverse events demonstrated higher EAT density (e.g., -75.8 HU vs. -88.5 HU) and larger indexed EAT volume (e.g., 108.5 mm³ vs. 89.6 mm³) compared to stable patients. Kaplan-Meier analysis showed that patients with low SDNN/SDANN combined with high EAT density/volume had the highest cumulative incidence of the composite endpoint. Conversely, the “High HRV + Low EAT” group had the lowest event rates. Mediation analysis revealed that this autonomic-EAT coupling and its prognostic power were more prominent in patients with HFrEF compared to HFpEF. Conclusion Lower 24-hour HRV (SDNN/SDANN) is associated with adverse epicardial fat remodeling (higher density and larger volume) and independently predicts HF rehospitalization or death. This association is particularly strong in HFrEF, suggesting a phenotype-specific neuro-adipose axis. Integrating HRV and EAT assessment provides a novel, non-invasive strategy for risk stratification in heart failure.