Assessment of bioelectrical impedance analysis for the prediction of total body water in cystic fibrosis

The aim of this study was to compare the measurement of total body water (TBW) by deuterium (2H2O) dilution and bioelectrical impedance analysis (BIA) in patients with cystic fibrosis (CF) and healthy controls. Thirty-six clinically stable patients with CF (age 25.4+/-5.6 yrs) and 42 healthy controls (age 25.4+/-4.8) were recruited into this study. TBW was measured by 2H2O dilution and predicted by BIA in patients and controls. The TBW predicted from BIA was significantly different from TBW as measured using 2H2O in patients (P<0.05) but not in controls. Mean (+/-SD) values for predicted and measured TBW differed by 5.6 (+/-9.1) L in patients and 0.4 (+/-3.6)L in controls. This bias was consistent for all controls but not for patients. In CF, BIA over predicted TBW determined by 2H2O dilution to an increasing extent at larger TBW volumes. There was a strong correlation between height2/impedance and TBW in patients with CF (r=0.90; y=0.67x+2.50) and in controls (r=0.81; y=0.57x+9.60). The slope of the regression lines was similar for both groups, however the y intercepts were significantly different (P<0.05). BIA overestimates TBW in patients with CF, possibly due to invalid factory installed regression equations within BIA instrumentation. Future studies employing BIA as a measure of TBW or FFM in CF should use alternative predictive equations to those that have been developed for healthy individuals. A large scale study to develop specific regression equations for use in CF is warranted.