Reproducibility of fluid status measured by bioelectrical impedance analysis in healthy volunteers: a key requirement to monitor fluid status in the intensive care unit.

Little is known about the use of bioelectrical impedance analysis (BIA) in critically ill patients. The objective of this study was to evaluate the reproducibility of BIA measurements by comparing non-dominant versus dominant body-side measurements at 2 separate time points in healthy volunteers in order to extrapolate key elements that may be of relevance in critically ill patients. A prospective observational validation experiment was carried out in healthy volunteers. Full-body and segmental multiple frequency BIA measurements were carried out at the non-dominant and the dominant body side, consecutively, and on 2 separate occasions within 1 week. Parameters of interest were both raw data (impedance and phase angle) at the individual frequencies (5-50-100-200 kHz) and body fluid compartment volume estimations (total body water, extracellular water volume, intracellular water volume, volume excess). A total of 42 measurements were performed in 22 volunteers. Median (interquartile range) age and time between measurements was 26 years (24; 35) and 2.07 days (1.00; 2.99), respectively. The intraclass-correlation coefficients (ICCs) for body fluid compartment volumes estimated by full-body BIA, were greatly above 90%, showing excellent agreement, except for volume excess which showed moderate agreement. Full-body raw impedance and phase angle measurements showed highly variable and much lower ICCs. For both estimated body fluid compartment volumes and raw measurements, segmental BIA showed also highly variable and low ICCs. Overall this study showed that in healthy volunteers, BIA-derived fluid parameters are reproducible, and differences can be attributed to the changes in clinical status.