Outcomes of anatomic shoulder arthroplasty performed on B2 vs. A1 type glenoids

Glenoid component malpositioning and glenoid component retroversion have been associated with higher rates of radiolucencies, raising concerns about its implications on glenoid loosening and ultimate failure of anatomic total shoulder arthroplasty (TSA). Although there is literature regarding the relative advantages of techniques to address posterior glenoid bone loss, we are not aware of studies comparing outcomes of TSA on these challenging Walch type B2 glenoids vs. more common A1 glenoids. The purpose of this study is to compare outcomes of TSA performed on A1 glenoids and B2 glenoids treated with asymmetric glenoid reaming.

We identified 1045 shoulders that had primary TSAs performed for osteoarthritis in a prospective shoulder arthroplasty registry. Two hundred eighty-nine shoulders met inclusion criteria of a preoperative Walch type A1 (178) or B2 (111) glenoid morphology, treatment with TSA, asymmetric reaming in the B2 group, and a minimum of 2-year clinical and radiographic follow-up. Postoperative radiographs were assessed for lucencies, and patient-reported outcome measures were collected at all follow-up visits.

Follow-up averaged 40 ± 15 months for all patients, and more men presented with a B2 glenoid (80 of 111; 72%) compared with A1 (101 of 178; 57%) (P = .009). Age at surgery (P = .166), dominant-sided surgery (P = .281), body mass index (P = .501), smoking (P = .155), preoperative opioid use (P = .154), and diabetes (P = .331) were not significantly different between groups. Both groups had similar Constant Strength scores preoperatively (A1: 4.7 ± 7.1, and B2: 4.3 ± 7.3) but the B2 group improved significantly more at final follow-up (A1: 10.3 ± 6.2 vs. B2: 12.7 ± 6.7, P = .005). The Total Constant score was also significantly better at follow-up in the B2 glenoid group (P = .039). All other Constant subscales, American Shoulder and Elbow Surgeons Standardized Shoulder Assessment Form (ASES), and Single Assessment Numerical Evaluation (SANE) measures showed significant improvement preoperatively to final follow-up (all P <.001) but there were no significant differences between the A1 and B2 glenoid groups (all P > .05). A similar proportion of patients rated their satisfaction as either very satisfied or satisfied between the A1 (160; 90%) and B2 (100; 90%) (P = .613). Lazarus scores were also similar between the A1 and B2 groups (P = .952) as were the rates of humeral radiolucent lines (P = .749) and humeral osteolysis (P = .507).

Although patients with B2 glenoids may present a more technically challenging anatomic total shoulder arthroplasty, treatment with concurrent asymmetric glenoid reaming produced similar, successful clinical and radiographic early to midterm outcomes for patients undergoing TSA compared with A1 glenoids. Additional follow-up on this cohort will be important to confirm the durability of these early results.