Objective: The aims of this study were to determine the percentage of body fat (%BF) by dual-energy x-ray absorptiometry (DXA) and bioelectrical impedance analysis (BIA) using a standard adult equation and BIA using a standard geriatric equation in a population of older men with prostate cancer and to compare the results from these different methods.
Methods: We conducted a cross-sectional study in 38 men with locally advanced, node-positive, or recurrent prostate cancer and no history of androgen-deprivation therapy. Body composition was evaluated by DXA with the use of a Hologic 4500A densitometer and BIA. BIA %BF was calculated by using standard equations developed for adult and geriatric populations.
Results: %BF by DXA, BIA with the standard adult equation, BIA with the standard geriatric equation, and BIA with the age-appropriate equation were 26.7 +/- 5.3%, 22.5 +/- 5.6%, 38.2 +/- 6.9%, and 35.4 +/- 9.6%, respectively. There were statistically significant differences between %BF by DXA and all BIA estimates. By using the methods described by Bland and Altman (Lancet 1986;1(8476):307), the standard adult equation showed the least bias and variability.
Conclusions: In this group of men with prostate cancer, BIA with the standard adult equation provided a reasonable estimate of %BF compared with DXA, although the differences were statistically significant. BIA with the standard geriatric equation, however, markedly overestimated %BF compared with DXA, even when its use was restricted to elderly men.