UC San Diego

Objectives

To determine whether severity of non-alcoholic fatty liver disease (NAFLD) and liver fibrosis quantitatively assessed in individuals with diabetes mellitus (DM)-2 correlate with increased coronary heart disease (CHD) risk using non-invasive markers.

Methods

We conducted a single-centre, prospective, cross-sectional study in 100 consecutive diabetic individuals without known CHD recruited between March 2013 and September 2014. History, physical examination, serum markers, cardiac computed tomography (CT), magnetic resonance (MR) imaging-estimated proton density fat fraction (PDFF) and MR elastography (MRE) were obtained for 95 participants. Written informed consent was provided. Institutional review board approved this study. Spearman rank correlation was performed to assess for correlations. Multiple linear regression model determined independent predictors of epicardial adipose tissue (EAT) volume.

Results

A p value < 0.05 determined statistical significance. The EAT volume was higher in the NAFLD group, defined as MR-imaging PDFF ≥ 5 %, compared to the non-NAFLD group (126.5 ml (IQR 80.9) versus 85.4 ml (IQR 44.7), p=0.002). MR imaging-PDFF correlated with EAT (r=0.42, p < 0.0001). MR imaging-PDFF and liver fibrosis were independently associated with EAT.

Conclusions

Higher liver fat content and liver fibrosis may portend worse cardiovascular risk in diabetics.

Key points

• EAT volume is higher in diabetic individuals with NAFLD. • Liver fat content is positively correlated with EAT. • Liver fat content and liver fibrosis were independently associated with EAT. • Higher liver fat content and fibrosis may adversely affect cardiovascular risk.