Investigation of the mechanical interaction of the trabecular core with an external shell using rapid prototype and finite element models

The mechanical properties of vertebral bone have been widely studied with the ultimate goal of improving fracture risk prediction. However, the mechanical interaction between the cortical shell and the trabecular core is not well understood. The objective of this study was to investigate this interaction and to determine what effect it has on the ultimate strength of the whole bone. This objective was achieved by compression testing rapid prototype (RP) models of cylindrical trabecular bone cores, with and without an integral surrounding shell and incorporating increasing levels of artificially induced bone loss. Corresponding finite element (FE) models were generated and the load sharing of the shell and trabecular core was analysed under linear elastic loading conditions. The results of the physical RP model tests and corresponding FE analyses indicated that there was a reinforcing effect between the cortical shell and the trabecular core for all models tested and that the reinforcing effect became relatively more important to the ultimate strength of the whole bone as the bone volume fraction of the trabecular core decreased. It was found that two mechanisms contributed to the reinforcing effect: (i) load transfer from the highly stressed shell into the connecting outer trabeculae of the core for the shelled model. This did not occur for the un-shelled model where the load dropped off at the outer unsupported trabeculae; (ii) the stiffening effect on the shell due to the support provided by the connecting struts of the trabecular core, which serves to inhibit bending and buckling behaviour in the shell under compression loading. It was found that the stiffening on the shell was the more dominant contributor to the overall reinforcing effect between the shell and the trabecular core. (C) 2009 Elsevier Ltd. All rights reserved.This work is part of the Bone for Life project, funded by the Programme for Research in Third Level Institutions (PRTLI), administered by the Higher Education Authority in Ireland. The authors wish to acknowledge the SFI/HEA Irish Centre for High-End Computing (ICHEC) for the provision of computational facilities and support