Assessment of fat-mass loss during weight reduction in obese women

Methods for assessing body fat mass (FM) loss were compared in 32 obese (body mass index [BMI], 29 to 41 kg/m2) premenopausal women before and after a weight loss of 13.0 +/- 3.4 kg (mean +/- SD). A four-component (4C) model was used as the criterion. The other methods were as follows: three-component models (body density with total body water [3W] or bone minerals [3M]), underwater weighing, dual-energy x-ray absorptiometry ([DXA] XR-26, software 2.5.2; Norland, Ft Atkinson, WI), bioelectric impedance analysis (BIA) with an obese-specific equation [Segal et al), skinfolds (Durnin and Womersley), and an equation with BMI (Deurenberg et al). The 3W model (bias +/- SD, 0.5 +/- 0.4 kg), XR-26 (0.6 +/- 2.1 kg), and BMI equation (-0.3 +/- 2.1 kg) gave practically unbiased mean estimations of fat loss. All other methods underestimated fat loss by at least 1.6 kg (range of bias, -2.7 to -1.6 kg). The small bias (0.7 +/- 1.0 kg) between underwater weighing and model 4C before weight reduction indicates that the two-component assumptions were valid in premenopausal, weight-stable obese women. However, particularly the water fraction of the fat-free body component (4C model) was increased after weight reduction (before, 72.9% +/- 1.4%; after, 75.7% +/- 2.2%), making both underwater weighing and the 3M model uncertain for assessment of body composition changes. A general tendency for overestimating FM was seen before and more clearly after weight reduction. However, most methods underestimated fat loss, apparently because of unexpected changes in hydration of the fat-free body component.