The current standard high-resolution peripheral quantitative computed tomography (HRpQCT) based finite element (FE) models of distal radius sections explains 66% of variance in experimental testing results on intact forearms under fall configuration testing. However, this FE model employs a somewhat simplistic modeling approach in that it does not account for variations in mechanical properties within the radius, employs a fixed region of interest, and has been validated using embalmed samples, which can change forearm mechanical properties. Further, the effect of failure criteria on the predictions acquired using this model and their precision error is not known.

The aim of this study was to evaluate two different HR-pQCT-FE modeling approaches of distal radius for predicting failure load of the intact forearm under fall configuration testing. The purpose of study #1 was to evaluate the effect of failure criteria on wrist strength predictions acquired from HR-pQCT-based FE models ex vivo and their precision error in vivo. The purpose of the study #2 was to investigate the effect of using an anatomically standardized region of interest on FE predictions of distal radius failure load.

We acquired in vivo and ex vivo images of the distal radius via HR-pQCT. We performed failure testing on fresh-frozen forearms using a material testing system (MTS Bionix) to determine experimental failure load of forearms under fall configuration. We converted radius images to FE models using manufacturer-provided FE software. For the standard model, we used a single elastic modulus. For the density-based model, we used imaged bone mineral density to define elastic moduli.

In study #1 we derived failure loads for different failure criteria. The density based (EBMD) model explained 91% of variance in failure load when using an energy equivalent stress criteria with critical volume and critical stress limits set to 0.1% and 70 MPa, respectively. CV%RMS was 3.8% for the E-BMD model with highest R².

In study #2, we found that the failure loads were not different between the fixed and anatomically standardized (4% regions) (p > 0.05), and the fixed region and 4% region explained 89% and 87% variance in experimental failure load under fall configuration, respectively.

These findings indicated that the selection of failure criteria can alter the predictions of distal radius strength. Further, our results indicated that both single tissue and density-based HR-pQCT-FE models of distal radius sections can explain high variance (> 0.87) in failure load of intact forearms under fall configuration testing. Finally, the fixed and 4% regions provide similar predictions of distal radius strength for postmenopausal women.