High-Resolution Cone-Beam Computed Tomography is a Fast and Promising Technique to Quantify Bone Microstructure and Mechanics of the Distal Radius

Obtaining high-resolution scans of bones and joints for clinical applications is challenging. HR-pQCT is considered the best technology to acquire high-resolution images of the peripheral skeleton in vivo, but a breakthrough for widespread clinical applications is still lacking. Recently, we showed on trapezia that CBCT is a promising alternative providing a larger FOV at a shorter scanning time. The goals of this study were to evaluate the accuracy of CBCT in quantifying trabecular bone microstructural and predicted mechanical parameters of the distal radius, the most often investigated skeletal site with HR-pQCT, and to compare it with HR-pQCT. Nineteen radii were scanned with four scanners: (1) HR-pQCT (XtremeCT, Scanco Medical AG, @ (voxel size) 82 mu m), (2) HR-pQCT (XtremeCT-II, Scanco, @60.7 mu m), (3) CBCT (NewTom 5G, Cefla, @75 mu m) reconstructed and segmented using in-house developed software and (4) microCT (VivaCT40, Scanco, @19 mu m-gold standard). The following parameters were evaluated: predicted stiffness, strength, bone volume fraction (BV/TV) and trabecular thickness (Tb.Th), separation (Tb.Sp) and number (Tb.N). The overall accuracy of CBCT with in-house optimized algorithms in quantifying bone microstructural parameters was comparable (R-2 = 0.79) to XtremeCT (R-2 = 0.76) and slightly worse than XtremeCT-II (R-2 = 0.86) which were both processed with the standard manufacturer's technique. CBCT had higher accuracy for BV/TV and Tb.Th but lower for Tb.Sp and Tb.N compared to XtremeCT. Regarding the mechanical parameters, all scanners had high accuracy (R-2 >= 0.96). While HR-pQCT is optimized for research, the fast scanning time and good accuracy renders CBCT a promising technique for high-resolution clinical scanning.