Article
Micro finite element analysis of the effect of bone density and anisotropy in periprosthetic bone and cement stresses after anatomical glenoid replacement
Search Medline for
Authors
Published: | October 23, 2017 |
---|
Outline
Text
Objectives: Glenoid loosening was suggested to relate not only to glenohumeral conformity, cementing techniques, orientation, design or eccentric loads occurring with shoulder instability but also to patient-specific factors. In this study, the hypothesis that bone quality in the periprosthetic region, trabecular anisotropy and cortical thickness can influence stresses within the surrounding bone tissue and cement mantle was tested with micro finite element (μFE) models of the replaced joint depicting realistic bone microstructure.
Methods: The 82 μm high resolution peripheral quantitative computer tomography (HRpQCT) images of isolated cadaveric human scapulae were processed to allow virtual implantation of two types of anatomical glenoid replacements (UHMWPE keeled and pegged fixations, Exactech). The resulting images were converted to μFE models consisting of nearly 38 million hexahedral elements with bone tissue, cement layer and implant regions assigned isotropic elastic properties based on literature. Decreases in the glenohumeral conformity were simulated by decreasing contact area on the glenoid articular surface and instability was simulated by shifting superior and inferiorly the contact position for the lower conformity case. The models were solved for internal stresses within the structures with a parallel solver (parFE, ETH Zurich) on a Linux Cluster (Figure 1 (a) [Fig. 1]). Peak von Mises stresses were compared by design for different loading conditions and related to bone volume fraction (BV/TV), degree of anisotropy and cortical thickness of the specimens.
Results and Conclusion: The peak cement stresses were achieved near the cement-bone interface in all loading scenarios. Higher stresses within trabecular bone tissue and cement mantle were obtained within specimens of lower bone volume fraction and in regions of low anisotropy, increasing with decreasing glenohumeral conformity and reaching their maximum below the keeled design when the load is shifted superiorly (Figure 1 (b) [Fig. 1]).
Our μFE analyses confirm the combined influences of eccentric load shifts with reduced bone volume fraction and anisotropy on increasing periprosthetic stresses. They also suggest that improving fixation of glenoid replacements must reduce internal cement and bone tissue stresses, particularly in glenoids of low bone density and heterogeneity.