A Novel Anthropometry‐Based Model to Estimate Appendicular Muscle Mass in Brazilian Older Women

<b>Background:</b> The assessment of appendicular skeletal muscle mass (ASM) is central to the diagnosis of sarcopenia (SA). We developed an anthropometric model for estimating ASM and tested its validity to identify SA and associated risk of disability (RSA) in older women. <b>Methods:</b> The equation was developed with 89 women (60-88 years, 72 ± 6 years), with a cross-validation sample of 12 women (60-84 years, 67 ± 5 years). Validity was determined through concordance between actual versus estimated ASMs, correlations between actual/estimated ASM versus peak torque (PT) and total work (TW) during isokinetic knee extension/flexion and handgrip strength, and agreement of patients classified with SA and RSA. <b>Results:</b> The predictive equation was ASM (kg) = 0.177 (body mass, kg)-0.075 (arm circumference, cm) + 0.020 (thigh circumference, cm) + 5.376 (<i>R</i> = 0.905; <i>R</i> ² = 0.819; <i>R</i> ²ad = 0.809; <i>F</i> = 86.96; <i>p</i> < 0.0001; SEE = 1.35 kg). Agreement between actual and estimated ASMs was confirmed by validation (ICC = 0.81; <i>p</i> < 0.0001) and cross-validation samples (ICC = 0.72, <i>p</i> < 0.035). Regression characteristics in PRESS statistics (<i>R</i> ² PRESS = 0.79; SEE-PRESS = 1.61) were compatible with the original model. Percent agreements for the classification of SA and RSA from indices calculated using actual/estimated ASM were 98% (gamma = 0.98, <i>p</i> = 0.015) and 68% (gamma = 0.89, <i>p</i> < 0.0001) in validation and 67% (gamma = 1.0, <i>p</i> = 0.032) and 70% (gamma = 0.84, <i>p</i> < 0.001) in cross-validation samples. Correlations between actual/estimated ASM versus PT (range 0.57-0.76, <i>p</i> < 0.05), TW (range 0.51-0.75, <i>p</i> < 0.05), and handgrip (range 0.67-0.74, <i>p</i> < 0.001) were theoretically consistent, being moderate and similar in both samples. <b>Conclusion:</b> This new anthropometric model has satisfactory prediction qualities and could be applied as a simple and practical method for estimating ASM in Brazilian older women.