The investigation presented herein was performed to determine the effect of boundary conditions and architectural parameters on the mechanical properties of bovine cancellous bone. 124 cylindrical cancellous samples (7.5 mm height) were harvested from a total of 9 bovine humeri. Mechanical properties of these samples were determined using compression tests performed at three, quasi-static strain rates, namely: 10−3, 10−2 and 10−1 s−1.The specimen conditions investigated were standard specimens (with marrow, 10 mm diameter), defatted specimens (without marrow, 10 mm diameter), structurally confined specimens (with marrow and a confining collar, 10 mm diameter) and bone confined specimens (with marrow, 20 mm and 28 mm diameters).
Each specimen was scanned using a μCT scanner (Phoenix, voxel size 80 μm, 70 kV, 350 μA, acquisition time of 500 ms per image). The images were used to determine the architectural parameters for each specimen, which were calculated using algorithms developed during the study. These algorithms were validated against existing software (BoneJ) which is available to calculate cancellous bone architectural parameters.
The results of the compression testing showed little dependence of mechanical properties on strain rate. The results of the defatted and structurally confined specimens showed a decrease in scatter with the elimination and reduction, respectively, of flow of the marrow within the trabecular network. This suggests that although marrow does not strengthen bone at quasi-static strain rates, the flow of marrow disrupts the trabecular network.
The bone confinement results showed significantly increased mechanical strength of the inner 10 mm core compared to the whole sample. Apparent modulus was found to be 58% and 60% higher in the central core of the sample for 20 mm and 28 mm samples respectively. This suggests that doubling the diameter effectively removes the edge effects, with any additional diameter increase having no effect. Inner core yield strength was 58% higher in the 20 mm samples, and roughly 96% higher in the 28 mm samples compared to full specimen yield strength. This suggested that post-yield behaviour requires a further increase in overall diameter to sufficiently remove the edge effects due to the boundary condition.
The results of the architectural parameters suggested a linear correlation between the mechanical properties and parameters bone volume versus total volume and trabecular number. An exponential relationship was found to exist between the mechanical properties and mean trabecular separation. No correlation was found between mechanical properties and mean trabecular thickness. It was also concluded that specimen condition affects the relationship between mechanical properties and architectural parameters. Therefore, to effectively predict the response of cancellous bone, specimen condition should be used in conjunction with at least two architectural parameters, preferably bone volume versus total volume and mean trabecular separation.