Pericardial adipose tissue: associations with myocardial fibrosis and risk of adverse outcomes
N Black, F Soltani, C Maldonado-Garcia, J Bradley, L Priestner, L Szabo, S G Williams, M Schmitt, J H Naish, M C Petrie, S E Petersen, Z Raisi-Estabragh, C A MillerAbstract
Background
Pericardial adipose tissue (PAT) has been associated with adverse cardiac remodelling and incident heart failure (HF),[1] potentially through promotion of myocardial fibrosis, but supporting clinical data remain limited.
Purpose
In patients at risk of HF, we investigated the associations between PAT area, cardiovascular magnetic resonance (CMR)-derived measures of myocardial fibrosis (extracellular volume (ECV) and native T1), and adverse cardiovascular outcomes, independent of confounders including body mass index (BMI).
Methods
A total of 2,038 patients at risk of HF were enrolled from a prospective cohort undergoing clinically indicated CMR. Imaging included pre- and post-contrast T1 mapping with same-day haematocrit to measure myocardial ECV. PAT area was quantified at end-diastole from four-chamber cine images using an automated algorithm.[2] Exclusion criteria included hypertrophic or infiltrative cardiomyopathies, congenital heart disease, prior HF hospitalisation, or normal cardiac structure and function without cardiovascular risk factors. Outcome data were obtained from Hospital Episode Statistics ICD-10 codes. The primary composite outcome was time to first HF hospitalisation or cardiovascular (CV) death. Linear and Cox proportional hazards regression were used to assess associations between PAT area, myocardial fibrosis markers, and adverse outcomes. Multivariable models were adjusted for baseline demographics, comorbidities, and CMR measures of cardiac structure and function.
Results
Baseline characteristics are shown in Figure 1A. Greater PAT area and higher BMI were independently associated with lower myocardial ECV, an association observed only in patients with left ventricular ejection fraction (LVEF) ≥40% (Figure 1B). No significant association was found between PAT area and native T1, regardless of LVEF (ß -0.01, 95% CI -0.03 – 0.01, p=0.27). During a median follow-up of 2,676 days (IQR 2,500 – 2,891), the primary outcome occurred in 166 (8%) patients. PAT area was associated with the primary outcome on univariate analysis (Figure 2A). After multivariable adjustment, BMI, but not PAT area, remained independently associated with the primary outcome (Figure 2B), driven by HF hospitalisation (HR 1.56, 95% CI 1.25 – 1.94, p<0.001) rather than CV death (HR 1.04, 95% CI 0.81 – 1.34, p=0.75).
Conclusions
Pericardial adipose tissue does not appear to exert deleterious cardiovascular effects via myocardial fibrosis. Further, after comprehensive multivariable adjustment with deep phenotyping, pericardial fat was not independently predictive of adverse cardiovascular outcomes.For image description, please refer to the figure legend and surrounding text.For image description, please refer to the figure legend and surrounding text.