Cm. Modlesky et al., COMPARISON OF BODY-COMPOSITION AND BONE-MINERAL MEASUREMENTS FROM 2 DXA INSTRUMENTS IN YOUNG MEN, The American journal of clinical nutrition, 64(5), 1996, pp. 669-676
The purpose of this study was to compare body-composition and whole-bo
dy, lumbar spine (LS), femoral:neck (FN), trochanter, and Ward's trian
gle (WT) bone mineral measurements by using the Hologic QDR 1000W (DXA
(H)) and the Lunar DPX-L (DXA(L)) dual-energy X-ray absorptiometry ins
truments. In addition, the ability of conversion equations to predict
DXA(H) data from DXA(L) data were tested. Thirteen healthy young adult
males (aged 22.2 +/- 3.6 y, 177.4 +/- 5.3 cm in height, and 72.7 +/-
9.6 kg in weight) were scanned on the same day by using DXA(H) and DXA
(L). Whereas measured body mass was not different (P > 0.05) between m
achines, whole-body fat mass [DXA(H) - DXA(L) (DXA(diff)) = 1152 +/- 1
395 g], percentage fat (DXA(diff) = 1.5 +/- 1.7% of body mass), bone m
ineral density (BMC; DXA(diff) = 0.016 +/- 0.023 g/cm(2)), and bone mi
neral content (BMC; DXA(diff) = 316 +/- 50 g) were lower and whole-bod
y fat-free soft tissue (FFST; DXA(diff) = 1781 +/- 1859 g) was higher
with DXA(H) than with DXA(L). Lower fat mass (DXA(diff) = 2145 +/- 855
g) and BMC (DXA(diff) = 216 +/- 36 g) and higher FFST (DXA(diff) = 19
66 +/- 943 g) in the trunk were primarily responsible for the whole-bo
dy differences. Lower BMD and BMC values were found for LS (DXA(diff)
= 0.145 +/- 0.038 g/cm(2) and 3 +/- 2 g, respectively), trochanter (DX
A(diff) = 0.100 +/- 0.044 g/cm(2) and 1.7 +/- 1.0 g), and WT (DXA(diff
) = 0.195 +/- 0.061 g/cm(2) and 1.93 +/- 0.51 g) with DXA(H) compared
with DXA(L). DXA(H) BMD of FN was also lower (DXA(diff) = 0.141 +/- 0.
032 g/cm(2)) than with DXA(L). Only DXA(H) whole-body BMC and LS BMD w
ere accurately predicted from DXA(L) with conversion equations. Predic
ted DXA(H) FN BMD was significantly lower than the actual DXA(H) value
(P less than or equal to 0.05), whereas the discrepancy between DXA(H
) and DXA(H) actually increased for whole-body percentage fat and BMD
(DXA(diff) = 6.6 +/- 1.3% body mass and 0.020 +/- 0.025 g/cm(2)). In c
onclusion, lower whole-body fat mass, percentage fat, and BMC, and hig
her whole-body FFST with DXA(H) were due primarily to measurement diff
erences in the trunk. Whereas conversion equations accurately predicte
d DXA(H) whole-body BMC and LS BMD from DXA(L) measurements in young a
dult males, they did not accurately predict DXA(H), percentage fat, an
d BMD of the whole body and FN BMD.