Use of Relative vs Fixed Offset Distance to Define Region of Interest at the Distal Radius and Tibia in High-Resolution Peripheral Quantitative Computed Tomography.

Although the region of interest in high-resolution peripheral quantitative computed tomography, defined based on the manufacturer's protocol for in vivo scanning, provides consistency and is practically convenient, it does not take into account possible variation in morphology in the regions adjacent to the measurement site. This study aimed at compare the morphologic variation in measurements using the standard fixed offset distance to define the distal starting slice against those obtained by using a relative measurement position scaled to the individual bone length at the distal radius and tibia in normal healthy adult subjects. A total of 40 healthy adult subjects (median height, 175.3 cm; range: 150.0-196.0 cm) were included in the study. High-resolution peripheral quantitative computed tomography at the distal radius and tibia was performed in all subjects, the region of interest defined by, first, the standard measurement protocol, where the most distal CT slice was 9.5 mm and 22.5 mm from the end plate of the radius and tibia, respectively, and second, the relative measurement method, where the most distal CT slice was at 4% and 7% of the radial and tibial lengths, respectively. Volumetric densities and microarchitectural parameters were compared between the 2 methods. Measurements of the total and cortical volumetric density and cortical thickness at the radius and tibia and cortical porosity, trabecular volumetric density, and trabecular number at the tibia were significantly different between the 2 methods (all p < 0.001). The predicted morphologic variation with varying measurement position was substantial at both the radius (up to 34%) and the tibia (up to 36%). A lack of consideration to height (and in turn the bone lengths) in the standard patient protocol could lead to the introduction of systematic errors in radial and tibial measurements. Although this may not be of particular significance in longitudinal studies in the same individual, it potentially assumes critical importance in cross-sectional studies.