Clinical assessment of bone quality

Human in vivo bone quality measurements have recently become possible with the release of a new human micro-computed tomography scanner (HR-pQCT) (XtremeCT, Scanco Medical, Switzerland; voxel size 82 μm). Three measurements of bone quality are gaining use with micro-computed tomography (μCT): bone micro-architectural morphology, bone mineral density (BMD), and a prediction of mechanical stiffness using a finite element (FE) analysis.

It was found that bone quality measurements using μCT were able to distinguish low bone quality (reduced bone volume ratio, reaction force to induce 1% strain, and strain energy density) and to monitor a successful drug intervention in a model of osteoarthritis. Human in vivo μCT bone quality measurements using HR-pQCT were found to be accurate for bone mineral measurements (R²=0.69 to 0.80), morphological measurements (R²=0.59 to 0.96), and stiffness predictions (R²=0.73). The short and long term in vivo precisions of HR-pQCT BMD measurements (<1%), morphological measurements (<4.5%) and stiffness predictions (<3.5) may be sufficient for longitudinal measurements. Semi-automated morphological and BMD measurements were found to predict (R²>0.80) apparent stiffness and Young's modulus in the radius and tibia, and will likely function as initial indicators of low bone quality.

Through further investigation of FE analyses, it was found that bone apparent stiffness estimated by a linear FE model was highly correlated with ultimate strength (R²=0.93). However, the direct estimate of ultimate strength was maximized when using plastic FE models (R²=0.95). The use of a scaled tissue modulus consistently improved results compared to a homogeneous tissue modulus, likely due to the intrinsic compensation for partial volume effects.

Bone quality measurements have the potential to increase the accuracy and precision of diagnosis and monitoring of diseases of low bone quality. This study was a first step in determining the potential of bone quality measurements to be applied in a clinical setting. Finite element modeling was found to have unique potential, as it can combine both bone structure and density information into a single prediction of bone quality.