A New Microarchitecture-Based Parameter to Predict the Micromechanical Properties of Bone Allografts

Abstract

Scaffolds are an essential component of bone tissue engineering. They provide support and create a physiological environment for cells to proliferate and differentiate. Bone allografts extracted from human donors are promising scaffolds due to their mechanical and structural characteristics. Bone microarchitecture is well known to be an important determinant of macroscopic mechanical properties, but its role at the microscopic, i.e., the trabeculae level is still poorly understood. The present study investigated linear correlations between microarchitectural parameters obtained from X-ray computed tomography (micro-CT) images of bone allografts, such as bone volume fraction (BV/TV), degree of anisotropy (DA), or ellipsoid factor (EF), and micromechanical parameters derived from micro-finite element calculations, such as mean axial strain (ε_z) and strain energy density (W_e). DAEF, a new parameter based on a linear combination of the two microarchitectural parameters DA and EF, showed a strong linear correlation with the bone mechanical characteristics at the microscopic scale. Our results concluded that the spatial distribution and the plate-and-rod structure of trabecular bone are the main determinants of the mechanical properties of bone at the microscopic level. The DAEF parameter could, therefore, be used as a tool to predict the level of mechanical stimulation at the local scale, a key parameter to better understand and optimize the mechanism of osteogenesis in bone tissue engineering.

Keywords: bone allografts; finite element analysis; microarchitectural parameters; micromechanical parameters.

Figure 1

Methods overview: Step (1) Cylindrical samples (10 mm-height and 6.9 mm-diameter) were extracted from commercial cubic blocks obtained from a bone bank; Step (2) Cylindrical samples were scanned with micro-CT and 3D images were reconstructed; Step (3) A cylindrical region of interest (8 mm-height and 6 mm height) was selected for each sample, and the threshold of gray levels were calculated to binarize images; Step (4) For each sample, microarchitectural parameters were measured using CTAn software, except for EF, which was measured using Fiji software; Step (5) Micro-finite element meshes were generated for the region of interest for each sample using Avizo software; Step (6) Micro-finite element analysis was performed for each sample under 0.016 mm displacement, corresponding to a 0.2% uniaxial compressive strain, using FEBioStudio software; Step (7) For each sample, micromechanical parameters were calculated as the averaged values of all the micro-finite elements; Step (8) The relationship between the microarchitectural and micromechanical parameters obtained in Step 4 and Step 7, respectively, was studied.

Figure 2

Finite element meshing process performed in Avizo^®: (a) filtered image; (b) binarization and removal of unconnected component; (c) originally generated meshed surface; (d) simplified meshed surface.

Figure 3

Box-whisker plots of microarchitectural parameters derived from CTAn, except for EF derived from Bone J: (A) Bone volume fraction (BV/TV); (B) Bone surface-to-volume ratio (BS/BV); (C) Trabecular thickness (Tb.Th); (D) Trabecular number (Tb.N); (E) Trabecular separation (Tb.Sp); (F) Structure model index (SMI); (G) Degree of anisotropy (DA); (H) Ellipsoid factor (EF). Box center lines, bounds of boxes, and whiskers indicate median, first and third quartiles, and minima and maxima within a 1.5 times interquartile range (IQR), respectively, points in the graph are outliers, N = 29.

Figure 4

Box-whisker plots of micromechanical parameters derived from micro-finite element calculations: (A) Mean axial strain (ε_z); (B) Mean axial stress (σ_z); (C) Mean von Mises stress (σ_e); (D) Mean strain energy density (W_e). Box center lines, bounds of boxes and whiskers indicate median, first and third quartiles and minima and maxima within a 1.5 times interquartile range (IQR), respectively, points in the graph are outliers, N = 29.

Figure 5

Linear regression plots between micromechanical parameters and ellipsoid factor (column 1), degree of anisotropy (column 2), and DAEF (column 3) for the 29 samples: (A) ε_z-EF; (B) ε_z-DA; (C) ε_z-DAEF; (D) σ_z-EF; (E) σ_z-DA; (F) σ_z-DAEF; (G)σ_e -EF; (H) σ_e-DA; (I) σ_e-DAEF; (J) W_e-EF; (K) W_e-DA; (L) W_e- DAEF.