In Vivo Estimation of Cortical Thickness and Porosity by Axial Transmission: Comparison with High Resolution Computed Tomography

Estimation of key factors of fracture risk, such as cortical bone thickness and porosity, can be obtained using high resolution-peripheral quantitative computed tomography (HR-pQCT). The feasibility of in vivo cortical thickness estimation by analyzing the waveguide response of long bones using axial transmission (AT) has been previously evidenced. The aim of this study is to compare both AT parameters, i.e., cortical thickness Ct. Th and porosity Ct. Po, with HR-pQCT measurements done at both site matched (one-third distal) and conventional (distal) locations. Twenty six patients (17 females and 9 males, 28 to 87 years) underwent measurements with both devices. AT measurements were performed using a 1-MHz prototype (Azalée, Paris, France). Singular value decomposition-based approach, combined with a 2-D transverse isotropic free plate waveguide model, was used to estimate cortical thickness and porosity. Site matched measurements of cortical thickness and volumetric bone mineral density (vBMD g.cm
^-3
) were also obtained using HR-pQCT (Scanco, Brüttisellen, Switzerland). In addition, cortical thickness and vBMD (Dcomp), corresponding to conventional clinical index measured at the distal part of the radius, were automatically provided by the device software. Highly significant Pearson correlations were found between Ct. Th AT estimates and that the values obtained with HRpQCT at the one-third distal radius (R
²
=0.84, RMSE = 0.26 mm) and at the distal part (R
²
=0.57, RMSE = 0.43 mm). A significant correlation between Ct. Po and vBMD was observed at the one-third distal radius (R
²
=0.59, RMSE = 1.8 %), while no correlation existed with Dcomp at the distal site, partly due to thin cortical thickness (<;0.5 mm) limitation. This study shows the potential of AT measurements to provide in vivo cortical thickness and porosity estimates using a portable and non-ionizing device.