Osteoporosis is a skeletal disorder characterised by compromised bone strength and increased fracture risk. Fractures of the distal radius (Colles fractures) occur earlier in life than other osteoporotic fractures. High resolution peripheral quantitative computed tomography (HR-pQCT) enables the in-vivo assessment not only of local bone mineral density (BMD) but also of 3D morphometric indices such as trabecular orientation (fabric). The aim of the current work was to validate experimentally a fast patient-specific homogenised finite element (hFE) model of the human distal radius and to develop a fully automated diagnostic tool for the in-vivo prediction of Colles fracture load for the last generation HR-pQCT scanner (XtremeCT II, SCANCO Medical). For this purpose, 12 pairs of cadaveric human forearms (mean age =77.2±8.5 years) were scanned intact using the high (61 μm) and the low (82 μm) resolution protocols to replicate the in-vivo scanning conditions. The 20 mm most distal radius sections were dissected out of the forearms and scanned at a resolution of 16.4 μm on a μCT 100 (SCANCO Medical) used to calibrate BMD and fabric obtained from the XtremeCT II reconstructions. These radial sections were tested in compression to assess failure loads. The tests were simulated by non-linear hFE and linear microFE analyses based on XtremeCT II images with a priori determined material constants. The computed fracture loads were strongly correlated with the experiments: R2=0.95 and R2=0.94 for the hFE and R2=0.93 and R2=0.94 for the microFE models based on the high and low resolution protocols, respectively. Computation of fracture load with hFE was around 3 times faster than with microFE for a similar accuracy. This study delivers an extensive validation for the in-vivo use of an accurate and fast hFE diagnostic tool to improve fracture risk, prediction, and treatment follow-up of patients with osteoporosis.
Disclosure: The authors declared no competing interests. The project was funded by the Swiss Commission for Technology and Innovation (CTI) with the grant number: CTI P No: 14311.1 PFLS-LS.