Soriano Janet-Mae P, Ioannidou Elli, Wang Jack, Thornton John C, Horlick Mary N, Gallagher Dympna, Heymsfield Steven B, Pierson Richard N
Body Composition Unit and New York Obesity Research Center, St. Luke's-Roosevelt Hospital, Columbia University College of Physicians and Surgeons, New York, NY 10025, USA.
J Clin Densitom. 2004 Fall;7(3):281-9. doi: 10.1385/jcd:7:3:281.
We compared bone mineral density (BMD) and body fat percentage (%fat) between two pencil-beam (Lunar DPX and DPX-L) and two fan-beam (Lunar Prodigy, Hologic Delphi A) dual-energy X-ray absorptiometry (DXA) systems. We examined these values in the total-body, spine, femur, and forearm scans in 78 healthy adults across these four DXA systems. BMD and %fat values were highly correlated among the four instruments. DPX-L gave the lowest mean %fat and Prodigy gave the highest mean %fat for both sexes. The means were system dependent for %fat estimates across the four DXA machines. There was a significant difference detected in BMD estimates across manufacturers, with the Delphi A providing systematically lower values than the Lunar systems in the whole body, spine, and femur sites but higher values than the Lunar systems in the forearm. The present study results show that both %fat and bone mineral estimates between pencil-beam and fan-beam systems are highly correlated, but vary by system. Significant differences exist between the instruments, especially between different manufacturers, and most of the comparisons are sex dependent. We conclude that longitudinal studies should always be evaluated on the same system when possible, and translation models should be used to assess cross-instrument differences.
我们比较了两种笔形束（Lunar DPX和DPX-L）和两种扇形束（Lunar Prodigy、Hologic Delphi A）双能X线吸收法（DXA）系统之间的骨密度（BMD）和体脂百分比（%fat）。我们在78名健康成年人中，通过这四种DXA系统对全身、脊柱、股骨和前臂扫描的这些数值进行了检测。四种仪器之间的BMD和%fat值高度相关。对于男女两性，DPX-L得出的平均%fat最低，而Prodigy得出的平均%fat最高。在这四种DXA仪器上，%fat估计值的均值取决于系统。不同制造商的BMD估计值存在显著差异，Delphi A在全身、脊柱和股骨部位得出的数值系统地低于Lunar系统，但在前臂部位高于Lunar系统。本研究结果表明，笔形束和扇形束系统之间的%fat和骨矿物质估计值高度相关，但因系统而异。仪器之间存在显著差异，尤其是不同制造商之间的差异，并且大多数比较都取决于性别。我们得出结论，纵向研究应尽可能始终在同一系统上进行评估，并且应使用转换模型来评估不同仪器之间的差异。
