Accuracy of peripheral quantitative computed tomography (pQCT) for assessing area and density of mouse cortical bone

Peripheral quantitative computed tomography (pQCT) is increasingly used for measurement of cortical bone geometry and density in mice. We evaluated the accuracy of pQCT for area and density measurements of thin-walled aluminum phantoms and mouse femora. Aluminum tubes with varying wall thicknesses and femora from 1- to 6-month-old C3H/HeJ (C3H) and C57B1/6J (136) mice (average cortical thickness 0.14-0.29 mm) were scanned at 70- or 90-mum resolution. pQCT values of area were compared to optical values determined after sectioning, while pQCT density (vBMD) was compared to solid aluminum density or correlated to bone ash content. For the aluminum phantoms, the error in pQCT area and density depended strongly on wall thickness, and density was consistently underestimated. For mouse femora, threshold values were found that produced zero error in bone area for each strain and age group, although the optimal threshold differed between groups. pQCT vBMD correlated strongly with ash content (r(2) = 0.7), although the regression equations differed between strains and the magnitude of the inter-strain difference in vBMD was fourfold greater than the difference in ash content. This finding suggests that pQCT can overestimate the differences in volumetric mineral density between inbred mouse strains whose bones are of different thickness (e.g., C3H vs. B6). In conclusion, both area and density values obtained by pQCT depend strongly on specimen thickness, consistent with a partial volume averaging artifact. Investigators using pQCT to assess cortical bones in mice should be aware of the potential for cortical thickness-dependent errors.