BACKGROUND: Lean body mass is an important predictor of survival and functional status in patients with AIDS wasting. The bias between different techniques for assessing body composition in AIDS wasting is not known. DESIGN: We compared total body potassium (TBK) with fat-free mass (FFM) determined by dual-energy X-ray absorptiometry (DXA), bioelectrical impedance analysis (BIA), and skinfold-thickness measurement (SKF) in 132 patients (63 men, 69 women) with AIDS wasting (weight < 90% of ideal body weight, or weight loss > 10% of original, or both). None of the subjects exhibited clinical lipodystrophy. Comparisons were made by using different BIA equations. RESULTS: Lean body mass determined by DXA was highly correlated with TBK in men (r = 0.79, P: < 0.0001) and women (r = 0.84, P: < 0.0001). FFM(BIA) and FFM(DXA) were significantly different (P: < 0.01 in men and P: < 0.0001 in women). The difference between FFM(DXA) and FFM(BIA) was significantly greater with greater weight and body fat, particularly in HIV-infected women (r = -0.39, P: = 0.001 for weight; r = -0.60, P: < 0.0001 for fat). The comparability of FFM and fat mass determined by DXA and BIA was dependent on the specific BIA equation used. Among men, no single BIA equation was more highly predictive of fat mass and FFM in comparison with DXA. CONCLUSIONS: The differences between DXA, BIA, and SKF in the determination of fat mass and FFM are significant in patients with AIDS wasting. BIA overestimates FFM compared with DXA in those with greater body fat. Standard BIA equations may not accurately estimate FFM and fat mass in men and women with AIDS wasting.
BACKGROUND: Lean body mass is an important predictor of survival and functional status in patients with AIDS wasting. The bias between different techniques for assessing body composition in AIDS wasting is not known. DESIGN: We compared total body potassium (TBK) with fat-free mass (FFM) determined by dual-energy X-ray absorptiometry (DXA), bioelectrical impedance analysis (BIA), and skinfold-thickness measurement (SKF) in 132 patients (63 men, 69 women) with AIDS wasting (weight < 90% of ideal body weight, or weight loss > 10% of original, or both). None of the subjects exhibited clinical lipodystrophy. Comparisons were made by using different BIA equations. RESULTS: Lean body mass determined by DXA was highly correlated with TBK in men (r = 0.79, P: < 0.0001) and women (r = 0.84, P: < 0.0001). FFM(BIA) and FFM(DXA) were significantly different (P: < 0.01 in men and P: < 0.0001 in women). The difference between FFM(DXA) and FFM(BIA) was significantly greater with greater weight and body fat, particularly in HIV-infectedwomen (r = -0.39, P: = 0.001 for weight; r = -0.60, P: < 0.0001 for fat). The comparability of FFM and fat mass determined by DXA and BIA was dependent on the specific BIA equation used. Among men, no single BIA equation was more highly predictive of fat mass and FFM in comparison with DXA. CONCLUSIONS: The differences between DXA, BIA, and SKF in the determination of fat mass and FFM are significant in patients with AIDS wasting. BIA overestimates FFM compared with DXA in those with greater body fat. Standard BIA equations may not accurately estimate FFM and fat mass in men and women with AIDS wasting.