Ultrasound predicts skeletal muscle fat infiltration in healthy middle-aged and young adults: Validation against MRI

This study developed an ultrasound prediction equation for measuring fat fraction in lower limb muscles of healthy middle-aged and young participants and applied it in a randomised controlled trial (RCT) to assess the relationship between fat fraction and strength in exercising and non-exercising females. Twenty-eight participants were recruited into <40 years (10m, 5f, age 28.9 ± 5.3 years, body mass index (BMI) 23.6 ± 2.3 kg m^-2) and >50 years (8m, 5f, age 60.4 ± 5.7 years, BMI 24.8 ± 3.6 kg m^-2) groups. T1-weighted magnetic resonance imaging (MRI)-Dixon fat fraction in gastrocnemius medialis (GM) and vastus lateralis (VL) and ¹H-magnetic resonance spectroscopy for intramyocellular (IMCL) and extramyocellular (EMCL) lipid content in GM was used to validate ultrasound measures of muscle, subcutaneous fat thickness (SFT) and echointensity to generate a multivariable model of fat infiltration. For the RCT, 72 females aged 40-60 years were recruited to either a low impact resistance exercise (4-5 sessions/week for 12 weeks) or a control group, and muscle strength was tested using isokinetic dynamometry. Multiple linear regression models for ultrasound measurement of fat fraction were validated for GM (r² = 0.71) and VL (r² = 0.91). SFT was the dominant variable in the older group in GM (77%) and VL (92%), but of near equal proportion, with echointensity, in GM (52%) and VL (51%) in the younger group. Strength was negatively correlated with fat fraction (r = -0.44). Strength increased in the exercising group, but fat fraction remained unchanged. A multivariable model using ultrasound measurements of SFT, muscle thickness, echointensity and age demonstrates that ultrasound is capable of predicting lower limb muscle fat infiltration.