Trunk muscle quality assessed by computed tomography: Association with adiposity indices and glucose tolerance in men

Background/objectives: Thigh muscle attenuation measured by computed tomography (CT) has been shown to be a reliable and useful index of skeletal muscle fat infiltration. Thigh muscle fat content assessed by CT has been linked to obesity and type 2 diabetes and is a correlate of insulin resistance in sedentary individuals. However, as measurement of mid-thigh fat content requires the assessment of another region of interest beyond the usual abdominal scan required to measure levels of visceral and subcutaneous abdominal adipose tissue, this study aimed at testing the hypothesis that skeletal muscle fat measured from a single abdominal scan (L₄-L₅) would also provide information relevant to the estimation of muscle fat infiltration as it relates to cardiometabolic risk.

Methods: Abdominal (L₄-L₅) and mid-thigh CT scans were performed in a sample of 221 sedentary men covering a wide range of adiposity values. Trunk muscles on the L₄-L₅ scan were classified into 2 groups: 1) psoas and 2) core muscles. The two scans were segmented to calculate muscle areas, mean attenuation values as well as low-attenuation muscle (LAM) areas, the latter being considered as an index of skeletal muscle fat infiltration. Body mass index (BMI), body composition and waist circumference were assessed and a 75 g oral glucose tolerance test (OGTT) was performed.

Results: Mid-thigh, psoas and core LAM areas were all significantly associated with body composition indices (0.46 ≤ r ≤ 0.71, p < 0.0001) whereas trunk muscle indices were more strongly associated with visceral adiposity and waist circumference (0.54 ≤ r ≤ 0.79, p < 0.0001) than were mid-thigh muscle variables (0.44 ≤ r ≤ 0.62, p < 0.0001). Mid-thigh LAM area as well as psoas and core LAM areas were significantly associated with fasting glucose, 2-h plasma glucose levels, the glucose area under the curve and with the HOMA-IR index (mid-thigh LAM area: 0.18 ≤ r ≤ 0.25, p < 0.01; psoas LAM area: 0.27 ≤ r ≤ 0.33, p < 0.0001; core LAM area: 0.24 ≤ r ≤ 0.34, p < 0.01). Multivariable stepwise regression analyses revealed that the associations between trunk muscle indices and glucose tolerance/insulin resistance were no longer significant after controlling for visceral adiposity measured at L₄-L₅.

Conclusion: Our results suggest that CT-imaging derived indices of trunk muscle quality are related to glucose tolerance and visceral adiposity. However, the relationship between skeletal muscle fat and insulin resistance appears to be largely mediated by the concomitant variation in visceral adiposity. Finally, our results suggest that a single CT scan performed at L₄-L₅ is adequate to assess skeletal muscle fat content related to cardiometabolic risk.