Abstract
Introduction
Stress shielding of bone around the stem components of total shoulder replacement (TSR) implants can result in bone resorption, leading to loosening and failure. Titanium is an ideal biomaterial for implant stems; however, it is much stiffer than bone. Recent advances in additive manufacturing (AM) have enabled the production of parts with complex geometries from titanium alloys, such as hollow or porous stems. The objective of this computational study is to determine if hollow titanium stems can reduce stress shielding at the proximal humerus. We hypothesize that hollow TSR implant stems will reduce stress shielding in comparison with solid stems and the inner wall thickness of the hollow stem will be a design parameter with a direct effect on bone stresses.
Methods
Using a previously developed statistical shape and density model (SSDM) of the humerus based on 75 cadaveric shoulders, a simulated average CT image was created. Using MITK-GEM, the cortical and trabecular bones were segmented from this CT image and meshed with quadratic tetrahedral elements. Trabecular bone was modeled as an isotropic and inhomogeneous material, with the Young's modulus defined element-by-element based on the corresponding CT densities. Cortical bone was assumed isotropic with a uniform Young's modulus of 20 GPa. The Poisson's ratio for all bone was 0.3. The distal humerus was fully constrained. Bone stresses were calculated by performing finite element analyses in ABAQUS with a 320 N force and 2 Nm frictional moment applied to the articular surface of the humeral head, based on an in vivo study during 45 degrees of shoulder abduction. Subsequently, the humeral head was resected and reamed to receive solid- and hollow-stemmed implants with identical external geometries but three different inner wall thicknesses (Figure 1). The identical surrounding bone meshes for the intact and reconstructed bones allowed element-by-element stress comparisons. The volume-weighted average changes in cortical and trabecular bone von Mises stresses were calculated, (wrt the intact humerus), as well as the percentage of bone volume experiencing a relative increase or decrease in stress greater than 10%.
Results
Results for all four implant designs are summarized (Figure 2). The solid stem resulted in the biggest average change in von Mises stresses (4% decrease in cortical and 6% increase in cancellous bone stress). The solid stem also resulted in the largest volume of bone experiencing a decrease in stress. Comparing the hollow stems, the thinnest shell wall resulted in the smallest changes in cortical bone stress, and the lowest volumes of bone experiencing a decrease in stress. Interestingly, this design caused the most cancellous bone to experience an increase in stress.
Discussion
These results suggest a marginal improvement in the bone-implant mechanics of hollow versus solid stems, and that thinner shell walls perform better. That said, the improvements over the solid stem design are minimal. Further increasing the compliance of these stems, e.g. by adding pores, may improve their performance. Future work will focus on optimizing hollow and porous stem designs, and the possibility of leveraging their hollow design for drug delivery.
For any figures or tables, please contact the authors directly.
