Non-invasive assessments of the calcaneus, by techniques such as quantitative ultrasound (QUS) and dual x-ray absorptiometry (DXA), have been used as predictors of fracture risk assessment and to monitor the response to mechanical loading and treatments for osteoporosis. However, these current devices lack the sensitivity to measure microstructural changes in the bone. High-resolution peripheral quantitative computer tomography (HR-pQCT) has been specifically developed to provide volumetric bone mineral density (vBMD), geometry, trabecular and cortical microstructure measurements, revealing DXA-independent predictors of fracture risk and compartment-specific adaptations to environmental stimuli.

This thesis describes the development of a procedure to quantitate calcaneal vBMD and trabecular microstructure using HR-pQCT (first generation XtremeCT), in order to overcome limitations with previous quantitative devices.

Studies using human cadaveric feet and participants in vivo were used to develop the procedure and determine measurement accuracy and precision. Regional differences in calcaneus trabecular bone were found, with ~2-fold higher trabecular vBMD at the superior compared to the inferior region of the bone in vivo. The superior region of the calcaneus had better HR-pQCT measurement accuracy compared to the middle and inferior regions, due to fewer tissues surrounding the region and a higher trabecular density. Increasing HR-pQCT scans integration time (100 to 200ms) improved trabecular microstructure accuracy ex vivo;​ acceptable quality images could be obtained with a 200ms integration time in vivo. Calcaneus vBMD and trabecular microstructure could be measured with good relative precision in premenopausal women (vBMD, 0.9-1.9%;​ trabecular microstructure 1.4-1.8%), where it was necessary to correct for positional errors using image registration.

In conclusion, calcaneus vBMD and trabecular microstructure can be quantified using HR-pQCT in vivo. Continued development of the positioning apparatus would benefit the future application of the method, to enable greater flexibility with the region measured and enable image acquisition at a 200ms scan integration time.