Validity of Standing Posture Eight-electrode Bioelectrical Impedance to Estimate Body Composition in Taiwanese Elderly

Abstract

Standing eight-electrode bioelectrical impedance analysis, which can be used to estimate percentage body fat (BF%) and lean soft tissue (LST) in the whole body and different body segments of elderly adults, is potentially an ideal method for clinical assessment of body composition. In this study, dual-energy X-ray absorptiometry (DXA) was used as a criterion method to validate a standing eight-electrode bioelectrical impedance analysis device BC-418 (hereafter abbreviated as BIA 8 ; Tanita Corp., Tokyo, Japan). LST and BF% were measured in the whole body and various body segments (upper limbs, lower limbs, and trunk) of each participant using BIA 8 and DXA; correlation and differences between the LST and BF% results measured in the whole body and various body segments were compared. A total of 77 individuals, 42 males and 35 females, aged 55.2–76.8 years, were included in the analysis. The impedance indexes ( h 2 / Z ) of hand to foot, and upper and lower limbs of the left side of the body measured by BIA 8 were highly correlated with the LST values in the corresponding body segment measured by DXA ( r = 0.96, r = 0.92, and r = 0.88, respectively; all p < 0.001). LST values of the whole body and various body segments of participants measured using BIA 8 were highly and significantly correlated with the corresponding DXA data (all r > 0.88, p < 0.005); the whole body and segmental BF% measured by BIA 8 and DXA also showed a significant correlation ( r > 0.84, p < 0.005). In addition, the agreement between the results of BIA 8 and DXA was assessed by Bland–Altman analysis; the bias and SD were, respectively, 1.89 kg and −4.25% in limb LST, and 2.18 kg and 4.06% in whole body BF%. The results of this study showed that the impedance index and LST in the whole body, upper limbs, and lower limbs derived from DXA findings were highly correlated. The LST and BF% estimated by BIA 8 in whole body and various body segments were highly correlated with the corresponding DXA results; however, BC-418 overestimates the participants' appendicular LST and underestimates whole body BF%. Therefore, caution is needed when interpreting the results of appendicular LST and whole body BF% estimated for elderly adults.