Secondary Logo

Institutional members access full text with Ovid®

Share this article on:

A Simple Model to Identify Risk of Sarcopenia and Physical Disability in HIV-Infected Patients

Farinatti, Paulo1,2; Paes, Lorena1; Harris, Elizabeth A.1; Lopes, Gabriella O.1; Borges, Juliana P.1

Journal of Strength and Conditioning Research: September 2017 - Volume 31 - Issue 9 - p 2542–2551
doi: 10.1519/JSC.0000000000002070
Original Research

Farinatti, P, Paes, L, Harris, EA, Lopes, GO, and Borges, JP. A simple model to identify risk of sarcopenia and physical disability in HIV-infected patients. J Strength Cond Res 31(9): 2542–2551, 2017—Early detection of sarcopenia might help preventing muscle loss and disability in HIV-infected patients. This study proposed a model for estimating appendicular skeletal muscle mass (ASM) to calculate indices to identify “sarcopenia” (SA) and “risk for disability due to sarcopenia” (RSA) in patients with HIV. An equation to estimate ASM was developed in 56 patients (47.2 ± 6.9 years), with a cross-validation sample of 24 patients (48.1 ± 6.6 years). The model validity was determined by calculating, in both samples: (a) Concordance between actual vs. estimated ASM; (b) Correlations between actual/estimated ASM vs. peak torque (PT) and total work (TW) during isokinetic knee extension/flexion; (c) Agreement of patients classified with SA and RSA. The predictive equation was ASM (kg) = 7.77 (sex; F = 0/M = 1) + 0.26 (arm circumference; cm) + 0.38 (thigh circumference; cm) + 0.03 (Body Mass Index; kg·m−2) − 8.94 (R 2 = 0.74; R 2adj = 0.72; SEE = 3.13 kg). Agreement between actual vs. estimated ASM was confirmed in validation (t = 0.081/p = 0.94; R = 0.86/p < 0.0001) and cross-validation (t = 0.12/p = 0.92; R = 0.87/p < 0.0001) samples. Regression characteristics in cross-validation sample (R 2adj = 0.80; SEE = 3.65) and PRESS (R 2 PRESS = 0.69; SEE PRESS = 3.35) were compatible with the original model. Percent agreements for the classification of SA and RSA from indices calculated using actual and estimated ASM were of 87.5% and 77.2% (gamma correlations 0.72–1.0; p < 0.04) in validation, and 95.8% and 75.0% (gamma correlations 0.98–0.97; p < 0.001) in cross-validation sample, respectively. Correlations between actual/estimated ASM vs. PT (range 0.50–0.73, p ≤ 0.05) and TW (range 0.59–0.74, p ≤ 0.05) were similar in both samples. In conclusion, our model correctly estimated ASM to determine indices for identifying SA and RSA in HIV-infected patients.

1Laboratory of Physical Activity and Health Promotion, Institute of Physical Education and Sports, Rio de Janeiro State University, Rio de Janeiro, Brazil; and

2Graduate Program in Sciences of Physical Activity, Salgado de Oliveira University, Niteroi, Brazil

Address correspondence to Paulo Farinatti,

Copyright © 2017 by the National Strength & Conditioning Association.