Effects of diet and exercise on the density and composition of the fat-free mass in obese women.

PURPOSE

The purpose of this study was to determine whether the density (D(FFM)) and composition of the fat-free mass (FFM) and the accuracy of estimates of body composition from body density (%Fat(d)) are affected by diet and exercise.

METHODS

Twenty-nine obese women (body mass index (BMI) = 25.0-43.7 kg x m(-2) and %Fat(d) = 35.7-47.1%) were assigned to one of three groups: diet only (DO, N = 9); diet and aerobic exercise (DE, N = 9); or control (C, N = 11). Measures of body density by hydrostatic weighing, body water by deuterium dilution, and bone mineral by whole-body dual-energy x-ray absorptiometry, and estimates of body composition from body density and from a four-component model were obtained before and after a 16-wk diet and exercise intervention.

RESULTS

Mean (+/- SD) changes in body mass were -7.2 +/- 7.4, -3.9 +/- 3.3, and +1.2 +/- 2.8 kg for the DO, DE, and C, respectively. The density and composition of the FFM did not change significantly (P > 0.05) in any of the groups. Individual changes in D(FFM) (-0.011 to +0.011 g x mL(-1)), and differences between changes in %Fat estimated using a four-component model and %Fat(d) (-2.1 to +2.7% body mass) were not related to changes in body mass (r = -0.08). Individual changes in D(FFM) were most strongly related to changes in water fraction (r = -0.95) and protein fraction (r = +0.88), and were unrelated to changes in the mineral fraction (r = +0.04) of the FFM.

CONCLUSIONS

We conclude that in obese women, the density and composition of the FFM are unaltered and densitometry correctly assesses group mean changes in body composition with moderate weight loss induced by diet or diet and aerobic exercise. However, individual deviations in D(FFM) from the assumed value of 1.1 g x mL(-1) are substantial, and a multi-component model in which body water is measured is needed to accurately assess individual body composition changes resulting from diet and exercise.