Impact of epicardial adipose tissue volume upon left ventricular dysfunction in patients with mild-to-moderate aortic stenosis: A post-hoc analysis

Abstract

Background: Aortic stenosis (AS) may lead to diastolic dysfunction and later on heart failure (HF) with preserved left ventricular ejection fraction (HFpEF) via increased afterload and left-ventricular (LV) hypertrophy. Since epicardial adipose tissue (EAT) is a metabolically active fat depot that is adjacent to the myocardium and can influence cardiomyocytes and LV function via secretion of proinflammatory cytokines, we hypothesized that high amounts of EAT, as assessed by computed tomography (CT), may aggravate the development and severity of LV hypertrophy and diastolic dysfunction in the context of AS.

Methods: We studied 50 patients (mean age 71 ± 9 years; 9 women) in this preliminary study with mild or moderate AS and mild to severe LV diastolic dysfunction (LVDD), diagnosed by echocardiography, who underwent non-contrast cardiac CT and echocardiography. EAT parameters were measured on 2nd generation dual source CT. Conventional two-dimensional echocardiography and Tissue Doppler Imaging (TDI) was performed to assess LV function and to derive myocardial straining parameter. All patients had a preserved LV ejection fraction > 50%. Data was analysed using Pearson's correlation.

Results: Only weak correlation was found between EAT volume or density and E/é ratio as LVDD marker (r = -.113 p = .433 and r = .260, p = .068 respectively). Also, EAT volume or density were independent from Global Strain Parameters (r = 0.058 p = .688 and r = -0.207 p = .239). E/é ratio was strongly associated with LVDD (r = .761 p≤0.0001) and Strain Parameters were moderately associated with LV Ejection Fraction (r = -.669 p≤0.001 and r = -.454 P≤0.005).

Conclusions: In this preliminary study in patients with AS, the EAT volume and density as assessed by CT correlated only weakly with LVDD, as expressed by the commonly used E/é ratio, and with LV strain function. Hence, measuring EAT volume and density may neither contribute to the prediction nor upon the severity of LVDD, respectively.

Fig 1. Correlation between EAT volume & density and E/é ratio.

Scatterplots for the weak correlation between EAT volume and EAT density with E/é ratio (r = -.113, p = .433 or r = -.260, p = .068).

Fig 2. Correlation between EAT volume and body weight, BMI, LV pressure & LV EF and GCS, GLS.

Correlation scatterplot between EAT volume and (A) body weight (r = .514, p = ≤.0001) (B) BMI (r = .492, p = ≤.0001) (C) LV end-systolic pressure (r = .350, p = .016) (D) LV end-diastolic pressure (r = .290, p = .045). Correlation scatterplot between LV ejection fraction and (E) Global Longitudinal Strain (r = .058, p = .688) (F) Global Circumferential Strain (r = .207, p = .239).