Longitudinal agreement of four bioimpedance analyzers for detecting changes in raw bioimpedance during purposeful weight gain with resistance training

Background Due to inherent errors involved in the transformation of raw bioelectrical variables to body fluids or composition estimates, the sole use of resistance (R), reactance (Xc), and phase angle (phi) has been advocated when quantifying longitudinal changes. The aim of this investigation was to assess the ability of four bioimpedance analyzers to detect raw bioimpedance changes induced by purposeful weight gain with resistance training. Methods Twenty-one resistance trained males completed a 6-week lifestyle intervention with the aim of purposeful weight gain. Bioimpedance analysis was performed before and after the intervention using four different analyzers (MFBIA(InBody): InBody 770; MFBIA(SECA): Seca mBCA 515/514; BIS: ImpediMed SFB7; SFBIA: RJL Quantum V) for the quantification of R, Xc, and phi at the 50-kHz frequency. Repeated measures ANOVA and follow up tests were performed. Results Analysis revealed main effects of time and method for R, Xc, and phi (p <= 0.02), without significant time x method interactions (p >= 0.07). Follow up for time main effects indicated that, on average, R decreased by 4.5-5.8%, Xc decreased by 2.3-4.0%, and phi increased by 1.8-2.6% across time for all analyzers combined. However, varying levels of disagreement in absolute values were observed for each bioelectrical variable. Conclusions The differences in absolute bioelectrical values suggests that analyzers should not be used interchangeably, which holds particular importance when reference values are utilized. Despite absolute differences, analyzers with varying characteristics demonstrated similar abilities to detect changes in R, Xc, and phi over time.

Automated summary

The study suggests that different bioimpedance analyzers have absolute differences in detecting bioelectrical changes, which should not be used interchangeably, especially when reference values are utilized. Despite absolute differences, analyzers with varying characteristics demonstrated similar abilities to detect changes in R, Xc, and phi over time.