A longitudinal comparison of body composition by total body water and bioelectrical impedance in adolescent girls.

Previous studies have assessed the ability of bioelectrical impedance analysis (BIA) to estimate body composition cross-sectionally, but less is known about the ability of BIA to detect changes in body composition longitudinally over the adolescent growth period. Body composition was assessed by isotopic dilution of H(2)(18)O and BIA in 196 initially nonobese girls enrolled in a longitudinal study. Two prediction equations for use in our population of girls were developed, one for use premenarcheally and one for use postmenarcheally. We compared estimates from our equation with those derived from several published equations. Using longitudinal data analysis techniques, we estimated changes in fat-free mass (FFM) and percentage body fat (%BF) over time from BIA, compared with changes in FFM and % BF estimated by H(2)(18)O. A total of 422 measurements from 196 girls were available for analysis. Of the participants, 26% had one measurement of body composition, 43% had two measurements of body composition and 31% had three or more measurements of body composition. By either H(2)(18)O or BIA, the mean %BF at study entry was 23% (n = 196) and the mean %BF at 4 y postmenarche was 27% (n = 133). In our cohort, the best predictive equations to estimate FFM by BIA were: PREMENARCHE: FFM = -5.508 + (0.420 x height(2)/resistance) + (0.209 x weight) + (0.08593 x height) + (0.515 x black race) - (0.02273 x other race). POSTMENARCHE: FFM = -11.937 + (0.389 x height(2)/resistance) + (0.285 x weight) + (0.124 x height) + (0.543 x black race) + (0.393 x other race). Overall, we found that BIA provided accurate estimates of the change in both FFM and %BF over time.