ABSTRACT
BACKGROUND: There is little information on whether bioelectrical impedance analysis (BIA) accurately predicts changes in body composition associated with energy restriction, exercise, or both. OBJECTIVE: We had 2 objectives: to determine the validity of the leg-to-leg BIA system in 1) estimating body composition in obese and nonobese women, with a cross-sectional design, and 2) assessing changes in body composition in obese women in response to 12 wk of energy restriction, exercise training, or both. DESIGN: Subjects were 98 moderately obese women (43.2 +/- 0.6% body fat, 45.0 +/- 1.1 y of age) and 27 nonobese control subjects (24.0 +/- 1.5% body fat, 43.5 +/- 2.5 y of age). Obese subjects were randomly divided into 1 of 4 groups, with fat-free mass, fat mass, and percentage body fat estimated with BIA and underwater weighing before and after 12 wk of intervention. The 4 groups were diet only (4.19-5.44 MJ/d), exercise only (five, 45-min sessions/wk at 78.5 +/- 0.5% of maximum heart rate), both exercise and diet, and control (no diet or exercise). RESULTS: No significant difference was found between underwater weighing and BIA in estimating the fat-free mass of the obese and nonobese women (all subjects combined, r = 0.78, P < 0.001, SEE = 3.7 kg) or in estimating decreases in fat mass during 12 wk of energy restriction, exercise, or both in obese subjects (F[3.85] = 1.45, P = 0.233). CONCLUSIONS: The leg-to-leg BIA system accurately assessed fat-free mass in obese and nonobese women, and changes in fat mass with diet alone or when combined with exercise.