Bioelectrical impedance techniques for measuring body composition under extreme conditions

Bioelectrical impedance spectroscopy (BIS) measures the spectrum of the body's resistance and reactance to a weak alternating current (5–1000kHz). Cole-Cole modeling of this spectrum yields the resistances of extracellular and intracellular fluid, which can be used to calculate extracellular (ECW), intracellular (ICW), and total body (TBW) water volumes by linear regression.; BIS has been validated for measurement of body composition under conditions of euvolemia and mildly altered ECW:ICW compartmentalization. This dissertation attempts to extend the range of applications by testing the accuracy of BIS under conditions of severely altered water compartmentalization in a total of 71 healthy adult males and females. Interventions included: 44 hr of simulated microgravity (SM) using 6° head-down-tilt bedrest with eucaloric nutrition, 72 hr SM with severe undernutrition (800 kcal/d for 2 d), and acute hypertonic dehydration of 2–5% of body weight.; During eucaloric SM, BIS was accurate with mean differences from dilution techniques ranging from 0–0.1 kg for ECW, 0.3–0.5 for ICW, and 0.4–0.6 kg for TBW. Within-individual changes in compartments were tracked with an error of 2–3.6% of the compartment volume. During SM with undernutrition, BIS accurately and precisely tracked ECW changes, but underestimated the mean 1.4 L acute ICW decrease. Segmental BIS detected the change in trunk ICW missed by whole-body analysis. Segmental BIS of the calf and upper arm also accurately estimated MRI muscle volume (r = 0.96 and r = 0.93) and changes in muscle volume (MRI vs. BIS changes for calf: −49 ± 68 vs. −41 ± 62 cc, for arm: −18 ± 23 vs. −11 ± 24 cc, respectively, P > 0.05 for both).; During acute hypertonic dehydration, the change in ECW from euhydration to dehydration did not differ between BIS (−1.07 ± 0.85 L) and dilution (−1.13 ± 1.07 L). However, BIS did not detect the acute loss in ICW measured by dilution (+1.00 ± 1.74 vs. −1.09 ± 1.26 L, P < 0.05). An ECW:ICW ratio of 0.533 could be used to detect dehydration.; These studies suggest that the Cole-Cole linear model is not valid for the detection of large, acute losses in ICW, particularly from the trunk. Segmental and regional approaches may assist in detection of regional changes undetected by whole-body measurements.