Concordance between appendicular skeletal muscle mass measured with DXA and estimated with mathematical models in...

Appendicular skeletal muscle mass (ASMM) is a measure of body muscle content, and it correlates with nutrition and physical status. Estimation of ASMM using anthropometric models is a well-established strategy to overcome issues related to the restricted availability of sophisticated techniques in measuring ASMM. This study aimed to assess the validity of four selected anthropometric models in estimating ASMM in middle-aged women in Sri Lanka.

A group of women (n = 165) aged 30-60 years underwent a series of anthropometric measurements such as body weight, height, circumferences, and skin fold thickness at specific sites. The limb circumferences were corrected for subcutaneous adipose tissue thickness. Two models developed by Lee et al. (ASM 1, ASM2) and two models developed by Wen et al. (ASM3, ASM4) were validated using ASMM measured by dual energy x-ray absorptiometry (ASMM_DXA) as the reference standard.

Mean (SD) age of the study group was 49.0 (8.2) years. Mean (SD) ASMM_DXA and ASMM estimated by the four models were ASMM_DXA = 15.39 (2.75) kg, ASM1 = 18.36 (3.27) kg, ASM2 = 16.46 (3.01) kg, ASM3 = 15.44 (2.40) kg, and ASM4 = 14.44 (2.45) kg. Correlations of ASMM_DXA with ASMM estimated by the models were as follows: ASM1, r = 0.68, R² = 0.46, SEE = 2.02 kg; ASM2, r = 0.90, R² = 0.81, SEE = 1.18 kg; ASM3, r = 0.90, R² = 0.81, SEE = 1.17 kg; and ASM4, r = 0.91, R² = 0.82, SEE = 1.14 kg. ASMM estimated by ASM3 was not significantly different (P > 0.05) from ASMM_DXA with mean difference of - 0.05 (range, 0.12 to - 0.23). Bland and Altman plot revealed satisfactory measurement agreements between ASM3 and ASMM_DXA. The ASMM estimated by the other three models was significantly different from the ASMM_DXA (P < 0.05).

The ASM3 model introduced by Wen et al. met all validation criteria and can be recommended for the estimation of ASMM in middle-aged women in Sri Lanka.