Abstract
Total shoulder arthroplasty (TSA) is an effective treatment for end-stage glenohumeral arthritis. The use of high modulus uncemented stems causes stress shielding and induces bone resorption of up to 63% of patients following TSA. Shorter length stems with smaller overall dimensions have been studied to reduce stress shielding, however the effect of humeral short stem varus-valgus positioning on bone stress is not known. The purpose of this study was to quantify the effect of humeral short stem varus-valgus angulation on bone stresses after TSA.
Three dimensional models of eight male cadaveric humeri (mean±SD age:68±6 years) were created from computed tomography data using MIMICS (Materialise, Belgium). Separate cortical and trabecular bone sections were created, and the resulting bone models were virtually reconstructed three times by an orthopaedic surgeon using an optimally sized short stem humeral implant (Exactech Preserve) that was placed directly in the center of the humeral canal (STD), as well as rotated varus (VAR) or valgus (VAL) until it was contacting the cortex. Bone was meshed using a custom technique which produced identical bone meshes permitting the direct element-to-element comparison of bone stress. 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. A joint reaction force was then applied to the intact and reconstructed humeri representing 45˚ and 75˚ of abduction. Changes in bone stress, as well as the expected bone response based on change in strain energy density was then compared between the intact and reconstructed states for all implant positions.
Both varus and valgus positioning of the humeral stem altered both the cortical and trabecular bone stresses from the intact states. Valgus positioning had the greatest negative effect in the lateral quadrant for both cortical and trabecular bone, producing greater stress shielding than both the standard and varus positioned implant. Overall, the varus and standard positions produced values that most closely mimicked the intact state.
Surprisingly, valgus positioning produced large amounts of stress shielding in the lateral cortex at both 45˚ and 75˚ of abduction but resulted in a slight decrease in stress shielding in the medial quadrant directly beneath the humeral resection plane. This might have been a result of direct contact between the distal end of the implant and the medial cortex under loading which permitted load transfer, and therefore load-reduction of the lateral cortex during abduction. Conversely, when the implant was placed in the varus angulation, noticeable departures in stress shielding and changes in bones stress were not observed when compared to the optimal STD position. Interestingly, for the varus positioned implant, the deflection of the humerus under load eliminated the distal stem-cortex contact, hence preventing distal load transfer thus precluding the transfer of load.
