ELLIPTICAL VERSUS SPHERICAL HUMERAL HEADS IN SHOULDER ARTHROPLASTY: EFFECTS ON CORTICAL BONE LOAD TRANSFER AT THE RESECTION PLANE

The humeral head is known to be non-spherical, with the superior-inferior diameter being up to 6% larger than the anterior-posterior diameter. Most total shoulder arthroplasties use spherical humeral heads, which can result in articular overhang and/or regions of the resection remaining exposed. Elliptical humeral heads can better reproduce native shoulder kinematics [1] and can significantly improve cortical coverage at the resection plane without resulting in articular overhang [2], which is hypothesized to reduce proximal humeral stress shielding. Therefore, the purpose of this computational study was to compare the expected cortical bone response of optimally fitted elliptical and spherical humeral heads to ascertain the effects of humeral head shape on proximal stress shielding. Models of eight cadaveric humeri (68±6 years) were generated from CT data. An orthopaedic surgeon selected the anatomic humeral resection plane and virtually prepared the humerus to accept a short-stem implant (Exactech Equinoxe® Preserve) that was optimally sized and placed centrally in the canal. Both ellipses and circles were then fit to the resection plane of each specimen (Figure 1). Cortical bone was assigned an elastic modulus of 20 GPa and a Poisson's ratio of 0.3. Trabecular bone was assigned varying stiffness based on CT attenuation [3]. A joint reaction force representing 45° and 75° of abduction was applied [4], and the volume of cortical bone with resorbing potential in the region 5 mm beneath the resection plane was determined using a threshold value of 55% [5]. The elliptical humeral heads resulted in 25.1±10.8%. Interestingly, there is a trade-off in cortical bone resorption potential when the results are discretized into the four anatomic quadrants. The elliptical head reduced resorption potential in the lateral and medial, where stress shielding often occurs with short stem humeral TSA implants, quadrants while having the opposite effect in the anterior and posterior quadrants. This was likely a result of the improved cortical coverage of the elliptical heads in the medial and lateral quadrants. Due to the increased load transfer in these locations, the anterior and posterior regions were slightly unloaded. Based on the results of this study, elliptical humeral heads are preferable to circular humeral heads for the overall increase in total cortical bone coverage and the corresponding decrease in bone resorption potential in the medial quadrant. For any figures or tables, please contact the authors directly.