Tracking Body Composition Change with Weight Loss by BIA and DXA in Female Adolescent Runners: A Validation Study

Abstract

ABSTRACT Introduction/Purpose Monitoring percent body fat (%fat) is important for optimizing nutrition for balanced health and performance in adolescent female runners. We studied the effectiveness of segmental multifrequency bioelectrical impedance analysis (S-MFBIA) for tracking seasonal changes in body composition among competitive female adolescent runners. Methods We compared the body compositions of 25 female runners (15.0 ± 0.2 yr old) analyzed using S-MFBIA, using dual-energy x-ray absorptiometry (DXA) as a reference, in preparatory (PRE) and competitive (CMP) seasons. Pearson’s correlation coefficients were used to calculate associations between S-MFBIA and DXA. Paired t-test was used to compare PRE with CMP. Alpha level was corrected to 0.0023 using the Bonferroni method. Bland–Altman analysis was used to evaluate agreement between the methods. Results The runners gained a significant amount of fat-free mass (FFM) but lost %fat and weight from PRE to CMP (37.3–39.1 kg, 19.7%–12.7%, and 47.1–44.8 kg, respectively). Body composition variables measured by S-MFBIA and DXA were significantly correlated (r = 0.61–0.96) with respect to PRE, CMP, and longitudinal changes from PRE to CMP. S-MFBIA underestimated %fat (−1.7; 95% confidence interval (CI), −2.7 to −0.7 percentage points (pp)) and fat mass (FM; −0.7; 95% CI, −1.2 to −0.3 kg), but overestimated FFM (1.1; 95% CI, 0.6 to 1.5 kg) against DXA in PRE. No systematic errors were detected in CMP. In longitudinal evaluation, S-MFBIA underestimated decreases in %fat (1.2; 95% CI, 0.3 to 2.1 pp) and FM (0.5; 95% CI, 0.1 to 0.9 kg) and increases in FFM (−1.0; 95% CI, −1.4 to −0.6 kg). The limits of agreement were −3.0 to 5.4 pp, −1.4 to 2.4 kg, and −3.0 to 1.0 kg, respectively. Conclusions Although small systematic errors might not hinder the use of S-MFBIA for group-based analysis, large random errors relative to the size of measurement limit its ability to accurately monitor the individual body composition of competitive female adolescent runners over a weight loss period.