Between-devices agreement in obtaining raw bioelectrical parameters after a lifestyle intervention targeting weight loss in former athletes

Abstract

It is unclear if different bioelectrical impedance (BI) devices provide similar results regarding raw parameters [Resistance (R), Reactance (Xc), Phase Angle (PhA), and Impedance (Z)] for the same population/individual undergoing a weight loss intervention. The aim was to evaluate the cross-sectional and longitudinal agreement of raw data obtained by two BI devices in former athletes with overweight/obesity. Fifty-nine participants [mean (SD): 43.5 (9.2) years, 30.5 (4.0) kg/m(2), 42% females] were included. All the assessments were performed before and after a 4-months lifestyle intervention targeting weight loss (WL). BI parameters were assessed at 50 kHz by two devices: a BI spectroscopy (Xitron Technologies, 4200B, San Diego, USA) and a phase-sensitive single-frequency device (RJL/Akern Systems, BIA-101, Italy). Cross-sectionally, BIS provided lower mean values for all parameters (0.4% for R, 1.6% for Xc, 1.0% for PhA and 0.4% for Z, p <0.001) compared to SF-BIA. In individuals with a WL >= 2.5% (n =18), no longitudinal differences were found in any of the raw parameters between devices (p >= 0.128) and there was no proportional bias (p >= 0.408). Despite small baseline differences in raw BI parameters, both devices agreed in tracking changes over time at the group level but interpretation should be careful at the individual level.