Influence of preoperative humeral head morphology on clinical and radiographic outcomes following total shoulder arthroplasty

Background: Aspherical humeral head morphology has previously been shown to be associated with progression of glenohumeral osteoarthritis. Habermeyer et al classified humeral head morphology as either spherical or aspherical. Aspherical head morphology was associated with increased incidence of glenoid deformity (B2) as well as decentering of the humeral head, but the authors did not correlate these findings to clinical outcomes. The purpose of this investigation is to apply these findings to the evaluation of clinical outcomes following anatomic total shoulder arthroplasty (aTSA) and to further define preoperative humeral head pathomorphology. Methods: We performed a retrospective review of all patients who underwent primary aTSA who had adequate pre- and postoperative radiographs and 2-year follow-up. Preoperative radiographs were analyzed and humeral heads were classified as either spherical, mild aspherical, or major aspherical. Axillary images were evaluated to quantify centering of the humeral head and glenoid morphology. Postoperative radiographs were analyzed to quantify implant center of rotation (COR) utilizing the best fit circle technique. Patient-reported outcomes, range of motion and strength at 2-years postoperatively were recorded. Statistical significance was evaluated using chi-square and paired t-test statistics for both regression analysis and univariate analyses. Results: A total of 259 patients met inclusion criteria for this study. Of these, 158 patients (61%) were classified as spherical, 64 patients (25%) mild aspherical, and 37 patients (14%) major aspherical. Males were higher represented among patients with mild and major aspherical humeral heads. Aspherical humeral heads had a significantly higher incidence of B2 glenoid morphology, while spherical heads had a significantly higher incidence of A1 glenoids. Patients with aspherical humeral heads were found to have significantly worse external rotation preoperatively compared to those with spherical heads. There were no significant differences in postoperative range of motion. However, postoperative improvements in external rotational strength and belly press strength were significantly higher in patients with aspherical humeral heads. Preoperative humeral head morphology did not have any influence on patient-reported outcomes at 2-year follow-up or on radiographic postoperative implant restoration of COR. Discussion/Conclusion: Aspherical head morphology was associated with a higher incidence of male gender and B2 glenoid morphology; however, humeral head deformity alone was not shown to lead to diminished outcomes or worse COR restoration following TSA. Patients with aspherical preoperative humeral head morphology were not found to have worse clinical or radiographic outcomes compared to patients with spherical heads; however, patients with aspherical humeral heads did make significant improvements in rotational strength at 2-year follow-up.