Prediction and Validation of Total and Regional Skeletal Muscle Mass by Ultrasound in Japanese Adult

Abstract

0506 B-mode ultrasound has been widely employed to index skeletal muscle (SM) mass in vivo. However ultrasound-derived prediction equations for total body SM mass based on direct measurement, such as magnetic resonance imaging (MRI), have not been developed. PURPOSE: To develop regression-based prediction equations for SM mass by ultrasound, and to investigate the validity of these equations in Japanese adults. METHODS: Healthy Japanese men (n = 35) and women (n = 29) aged 18–61 yrs were randomly separated into 2 groups, a model development group (n = 48) and a validation group (n = 16). Total and regional SM mass was measured using magnetic resonance imaging (MRI) 1.5 T-scanners with spin-echo sequence. Contiguous transverse images (about 150 slices) with 1-cm slice thickness were obtained from the first cervical vertebrae to ankle joints. The volume of SM was calculated from the summation of digitized cross-sectional area. SM volume was converted into mass by an assumed SM density of 1.04 kg/l. Muscle thickness (MTH) was measured by B-mode ultrasound (5 MHz scanning head) at 9 sites on anatomical SM belly (lateral forearm, anterior and posterior upper arm, abdomen, subscapula, anterior and posterior thigh, anterior and posterior lower leg). RESULTS: Significant correlations were observed between MTH multiplied height and site-matched SM mass (total, arm, trunk body, thigh, and lower leg) by MRI measurement in the model development group (n = 48, r = 0.90–0.98, p<0.01, SEE = 0.1–1.2kg). When the SM mass prediction equations were applied to the validation group, significant correlations were also observed between the MRImeasured and predicted SM mass (n = 16, r = 0.94–0.98, p<0.01, SEE = 0.2–1.8kg). The predicted total SM mass for the validation group was 20.8 ± 6.9 kg, and was similar to the MRI-measured SM mass of 20.9 ± 7.0 kg. Bland-Altman analysis failed to disclose a bias in prediction of total SM mass for the validation group (n = 16, r = 0.03, NS). CONCLUSIONS: These results indicated that ultrasound-derived prediction equations are a valid method to predict SM mass and an alternative to MRI measurement in Japanese adults.