Peripheral quantitative computed tomography: optimization of reproducibility measures of bone density, geometry, and...

The aim of this study was to determine the reproducibility for in vivo measurements at the radius and tibia for trabecular and cortical parameters, bone geometry, and bone strength indices with the peripheral quantitative computed tomography (pQCT) XCT 3000. We performed 3 repeated scans within 2mo at the radius (N=18) and tibia (N=16) on healthy, premenopausal women, aged 22-35 yrs and report precision measures including %coefficient of variation (%CV) and least significant changes (LSCs). For the radius, we studied 2 sections (4% and 33% of total length) and for the tibia, 3 sections (4%, 38%, and 66% of total length). Reproducibility for radius at 33% and tibia at every site was good (%CV ranged from 0.02% to 2.19%). The precision error for the distal 4% radius was, however, higher. The reproducibility at the distal radius improved when we considered only the scans with a change of ± 10mm(2) in the radius total area at this site (%CV from 0.87% to 2.25%). This study showed that, when follow-up measurements are carefully obtained, pQCT yields excellent reproducibility at both the radius and tibia. These precision errors, in conjunction with changes in LSC for the pQCT measures, are useful for research and potential clinical applications.