Context: Bioelectrical impedance spectroscopy (BIS) and skinfold anthropometry (SKF) have been used to monitor body composition among patients with HIV-wasting, however, validation of these techniques during recombinant human growth hormone (rhGH) treatment has not been performed.

Objective: To evaluate the degree of agreement between criterion measurements of dual-energy x-ray absorptiometry (DXA) and those of BIS and SKF in patients with HIV- wasting treated with rhGH.

Design and Setting: Randomized, double-blinded, placebo-controlled, two-period crossover trial conducted at the University of Toronto and Mount Sinai Hospital (Toronto, Canada).

Patients: Referred sample of 27 community-dwelling men with HIV-associated weight loss (10% over preceding 12 months) despite optimal antiretroviral therapy.

Intervention: One daily injection of rhGH (6 mg) or placebo self-administered for 3 months in a cross-over fashion with a 3-month washout.

Main Outcome Measures: Fat-free mass (FFM) and fat mass (FM) were measured by BIS, SKF, and DXA before and after rhGH and placebo treatment.

Results: FFM_BIS was not significantly different than FFM_DXA after rhGH treatment (P = 0.10). Mean differences (bias±SD) according to Bland-Altman analysis were smaller for SKF than for BIS (P < 0.05) at all time points yet treatment-induced change in FM was better detected with BIS than with SKF. BIS estimates of FFM and FM showed better agreement with those of DXA following rhGH treatment (1.6 ± 4.6 kg and -2.1 ± 3.9 kg) compared with baseline (3.8 ± 3.5 kg and -4.1 ± 3.6 kg) and placebo (2.7 ± 4.4 kg and -3.1 ± 4.6) (P < 0.05). BIS overestimated, and SKF underestimated, the treatment-induced changes in FFM and FM.

Conclusions: SKF is more accurate than BIS when measuring body composition in patients with HIV-wasting before and after rhGH treatment, nonetheless, the accuracy of BIS increases following treatment. Change in FM due to treatment is not accurately assessed with SKF.