Lower-limb and whole-body tissue composition assessment in healthy active older women

AIM: The study evaluated, in active elderly women, the accuracy and bias of anthropometry and bioelectrical impedance analysis (BIA) for lower-limb and whole-body tissue composition measures using dual-energy X-ray absorptiometry (DXA) as the criterion method. METHODS: Nineteen individuals (66.1 +/- 4.2 years) participated in the study. Whole-body fat mass (FM) and fat-free mass (FFM) were measured by anthropometry, BIA and DXA. Lower-limb volume (LLV) and lower-limb FFM (LLFFM) were assessed by anthropometry and DXA. RESULTS: LLV and LLFFM were significantly overestimated by anthropometry vs. DXA (p < 0.05 and p < 0.001, respectively) but significant relationships were observed [coefficient of determination (R(2)) > 0.25, p < 0.05]. No significant difference was observed between FM(A) (where (A) stands for anthropometry) vs. FM(DXA) and FFM(A) vs. FFM(DXA) and significant relationships were observed [R(2) = 0.93, p < 0.001, coefficient of variation (CV) = 7.3%; and R(2) = 0.85, p < 0.001, CV = 4.4%, respectively]. No significant difference was observed between FM(BIA) and FM(DXA) and a significant relationship was observed (R(2) = 0.80, p < 0.001, CV = 11.6%). FFM was significantly underestimated by BIA vs. DXA (p < 0.01). CONCLUSIONS: In active elderly women, (i) compared with DXA, anthropometry overestimates LLV and LLFFM; (ii) anthropometry can be an accurate method for assessing whole-body composition; and (iii) despite a non-significant bias for the FM measurement, the BIA tends to overestimate FM and underestimate FFM.