Assessment of the degree of abdominal myosteatosis by magnetic resonance imaging in subjects with diabetes, prediabetes and healthy controls from the general population.

ABSTRACT

OBJECTIVES: Intra- and intermyocellular lipid deposition and adipose tissue are part of glucose homeostasis and insulin resistance; however, their role in type 2 diabetes mellitus (T2DM) remains unclear. We assessed differences in the degree of abdominal myosteatosis among subjects with T2DM and prediabetes. MATERIALS AND METHODS: Asymptomatic subjects from the general population were classified as subjects with T2DM, prediabetes or healthy controls and underwent multi-echo Dixon magnetic resonance imaging (MRI) (TR 8.90 ms, six echo times, flip-angle 4°). Abdominal myosteatosis was quantified as proton-density fat-fraction (PDFFmuscle) by a standardized segmentation-algorithm. Cardiometabolic risk factors were prospectively obtained in a comprehensive health assessment and visceral and subcutaneous adipose tissue (VAT and SAT) were quantified semi-automatically. Uni- and multivariate quantile regression were used to examine associations. RESULTS: Among 349 included subjects (mean age 56.0 ±â€¯8.0years, 56.7% males), 45 were classified as subjects with T2DM and 84 with prediabetes (12.9% and 24.1%; respectively). Median PDFFmuscle was significantly higher in subjects with T2DM and prediabetes compared to healthy controls (13.1% (IQR10.5-16.6%); 11.1% (IQR8.9-15.0%) and 10.1% (IQR7.5-13.3%); respectively, p < 0.001). The observed differences were independent of age and gender (all p < 0.002) but attenuated after adjustment for BMI (ß -0.02, 95%CI -1.49 to 1.44, p = 0.974; ß 0.47, 95%CI -0.91 to 1.86, p = 0.506; prediabetes and T2DM, respectively). This effect was attributable to VAT, which remained independently associated with PDFFmuscle after full adjustment (ß 0.01, 95%CI 0.01-0.02, p = 0.002). CONCLUSIONS: There are significant differences in the degree of abdominal myosteatosis between subjects with T2DM, prediabetes and healthy controls, that may be confounded by VAT. However, abdominal myosteatosis by MRI might serve as a cardiometabolic imaging-biomarker, specifically in the setting of impaired glucose metabolism.