Open Access Open Badges Research

Development and validation of anthropometric equations to estimate appendicular muscle mass in elderly women

Piettra Moura Galvão Pereira12*, Giselma Alcântara da Silva12, Gilberto Moreira Santos12, Edio Luiz Petroski3 and Amandio Aristides Rihan Geraldes12

Author Affiliations

1 Laboratory of Physical Fitness, Performance and Health, Center for Physical Education and Sports Education Center, Federal University of Alagoas, Maceió-Alagoas, Brazil

2 Graduate Program in Nutrition, Federal University of Alagoas, Maceió-Alagoas, Brazil

3 Sports Center, Center for Research in Kinanthropometry and Human Performance, Federal University of Santa Catarina, Florianópolis, Santa Catarina, Brazil

For all author emails, please log on.

Nutrition Journal 2013, 12:92  doi:10.1186/1475-2891-12-92

Published: 2 July 2013



This study aimed to examine the cross validity of two anthropometric equations commonly used and propose simple anthropometric equations to estimate appendicular muscle mass (AMM) in elderly women.


Among 234 physically active and functionally independent elderly women, 101 (60 to 89 years) were selected through simple drawing to compose the study sample. The paired t test and the Pearson correlation coefficient were used to perform cross-validation and concordance was verified by intraclass correction coefficient (ICC) and by the Bland and Altman technique. To propose predictive models, multiple linear regression analysis, anthropometric measures of body mass (BM), height, girth, skinfolds, body mass index (BMI) were used, and muscle perimeters were included in the analysis as independent variables. Dual-Energy X-ray Absorptiometry (AMMDXA) was used as criterion measurement. The sample power calculations were carried out by Post Hoc Compute Achieved Power. Sample power values from 0.88 to 0.91 were observed.


When compared, the two equations tested differed significantly from the AMMDXA (p <0.001 and p = 0.001). Ten population / specific anthropometric equations were developed to estimate AMM, among them, three equations achieved all validation criteria used: AMM (E2) = 4.150 +0.251 [bodymass (BM)] - 0.411 [bodymass index (BMI)] + 0.011 [Right forearm perimeter (PANTd) 2]; AMM (E3) = 4.087 + 0.255 (BM) - 0.371 (BMI) + 0.011 (PANTd) 2 - 0.035 [thigh skinfold (DCCO)]; MMA (E6) = 2.855 + 0.298 (BM) + 0.019 (Age) - 0,082 [hip circumference (PQUAD)] + 0.400 (PANTd) - 0.332 (BMI). The equations estimated the criterion method (p = 0.056 p = 0.158), and explained from 0.69% to 0.74% of variations observed in AMMDXA with low standard errors of the estimate (1.36 to 1.55 kg) and high concordance (ICC between 0,90 and 0.91 and concordance limits from -2,93 to 2,33 kg).


The equations tested were not valid for use in physically active and functionally independent elderly women. The simple anthropometric equations developed in this study showed good practical applicability and high validity to estimate AMM in elderly women.

Aging; Sarcopenia; Skeletal muscle mass; Anthropometry; Body composition