MRI thigh measurements predict whole‐body skeletal muscle mass in patients with type 2 diabetes: a comparison with DXA

Abstract

AbstractBackgroundSarcopenia is an age‐related loss of skeletal muscle mass (SMM) and function, associated with falls, frailty, and functional decline. It is more prevalent and often accelerated in people with Type 2 diabetes (T2D), especially when co‐existing with obesity (sarcopenic obesity). Accurate whole‐body SMM measurement, feasible using dual‐energy X‐ray absorptiometry (DXA) or magnetic resonance imaging (MRI), has utility both in clinical practice and in epidemiological and mechanistic research, considering the dual mechanical and metabolic function of skeletal muscle. Compared with MRI, DXA may underestimate age‐related muscle mass by up to 30%, and so direct comparison of DXA/MRI‐derived SMM measurements may be invalid in patients with obesity and T2D, who have potentially even more pronounced sarcopenia/sarcopenic obesity. We aimed to validate single‐slice or multiple‐slice measures of SMM, using MRI, with whole‐body SMM measures derived from DXA scans of appendicular lean soft tissue, specifically in patients with obesity and T2D.MethodsWe undertook a cross‐sectional study in which we performed paired DXA and whole‐body 3.0‐T MRI on 36 people with T2D and obesity (body mass index [BMI] > 30 kg/m2). Single‐slice and seven‐slice estimates of muscle mass, taken from mid‐thigh/thigh MRI images, respectively, were determined, and compared with whole‐body SMM derived from DXA scans of appendicular lean soft tissue, in men and women separately.ResultsSeventeen men (age 48.4 ± 8.2 years, BMI 37.5 ± 5.1 kg/m2, body fat 39.1 ± 6.0%) and 19 women (age 51.8 years ± 9.7, BMI 37.8 ± 5.8 kg/m2, body fat 47.6 ± 5.1%) were recruited. Whole‐body SMM estimated by seven‐slice thigh MRI correlated strongly with DXA‐derived SMM in both men (r = 0.94, P < 0.001) and women (r = 0.94, P < 0.001). Single‐slice thigh estimates of whole‐body SMM by MRI also showed strong relationships to DXA‐derived SMM in both men (r = 0.88, P < 0.001) and women (r = 0.87, P < 0.001). There was no evidence of a statistically significant difference when comparing the performance of using single‐slice vs. multiple‐slices to determine SMM (P = 0.10 for males and P = 0.13 for females).ConclusionsThis study demonstrates that accurate determination of whole‐body SMM is possible using either single‐slice mid‐thigh or seven‐slice thigh MRI in men and women with obesity and T2D in whom there may be significant alterations in body composition. Its application may be of use clinically and in obesity‐related, metabolic, and gerontological research where accurate assessments of body composition add mechanistic insight.