The effect of humeral head positioning and incomplete backside contact on bone stresses following total shoulder arthroplasty with a short humeral stem

The use of uncemented humeral stems in total shoulder arthroplasty (TSA) is known to be associated with stress shielding. This may be decreased with smaller stems that are well-aligned and do not fill the intramedullary canal; however, the effect of humeral head positioning and incomplete head backside contact has not yet been investigated. The purpose of this study was to quantify the effect of changes in humeral head position and incomplete head backside contact on bone stresses and expected bone response following reconstruction. Three-dimensional finite element models of 8 cadaveric humeri were generated, which were then virtually reconstructed with a short-stem implant. An optimally sized humeral head was then positioned in both a superolateral and inferomedial position for each specimen that was in full contact with the humeral resection plane. Additionally, for the inferomedial position, 2 incomplete humeral head backside contact conditions were simulated whereby contact was defined between only the superior or inferior half of the backside of the humeral head and the resection plane. Trabecular properties were assigned based on computed tomography attenuation and cortical bone was applied uniform properties. Loads representing 45° and 75° of abduction were then applied, and the resulting differentials in bone stress versus the corresponding intact state and the expected time-zero bone response were determined and compared. The superolateral position reduced resorbing potential in the lateral cortex and increased resorbing potential in the lateral trabecular bone, while the inferomedial position produced the same effects but in the medial quadrant. For the inferomedial position, full backside contact with the resection plane was best in terms of changes in bone stress and expected bone response, although a small region of the medial cortex did experience no load transfer. The implant-bone load transfer of the inferior contact condition was concentrated at the midline of the backside of the humeral head, leaving the medial aspect largely unloaded as a result of the lack of lateral backside support. This study shows that inferomedial humeral head positioning loads the medial cortex at the cost of unloading the medial trabecular bone, with the same occurring for the superolateral position except that the lateral cortex is loaded at the cost of unloading the lateral trabecular bone. Inferomedial positioned heads also were predisposed to humeral head lif