Anatomic total shoulder arthroplasty using a short humeral stem and a non-augmented, minimally cemented all-polyethylene glenoid: minimum 2-year outcome and predictors of clinical failure

Anatomic total shoulder arthroplasty (aTSA) using standard (non-augmented) glenoid implants has demonstrated excellent clinical results, but concerns have been raised regarding their use in patients with posterior glenoid wear. The primary purpose of this study is to investigate short-term patient-reported outcomes (PROs) and overall active range of motion in patients who underwent aTSA using short humeral stems and standard glenoid implants for glenohumeral arthritis with minimal or concentric wear and with posterior glenoid wear. The secondary purpose was to the determine predictors of an inferior outcome or reoperation following aTSA using this construct. Patients in this study underwent aTSA with minimal glenoid reaming to achieve >90% contact of a pegged, all-polyethylene standard glenoid implant, inserted with minimal cement. Plain radiographs, active range of motion, and PROs—comprising American Shoulder and Elbow Surgeons (ASES), simple shoulder test, and visual analog scale pain scores—were obtained preoperatively and at final follow-up. One hundred twenty-eight consecutive aTSAs in 122 patients (75 males) with mean age of 66 years (range 48-84) were reviewed. Thirteen patients were lost to follow-up, 6 patients were deceased or had cognitive impairment precluding evaluation, and 5 patients underwent early revision for subscapularis failure, leaving 104 aTSAs for evaluation at minimum 2-year follow-up (mean 3 years, range 2-5.6 years). Twenty-four shoulders had undergone previous surgery. Fifty-nine patients had posterior glenoid wear (Walch B2 or B3) and 47 had concentric or minimal glenoid wear (Walch A1, A2, or B1). Forward elevation improved from 101° ± 25° to 147° ± 17°, abduction from 87° ± 26° to 145° ± 23°, external rotation from 21° ± 15° to 45° ± 13°, and internal rotation from the sacrum to thoracolumbar junction ( P < .001 for all). Overall, ASES improved from 41 ± 20 to 89 ± 13, simple shoulder test from 4.5 ± 2.9 to 10.2 ± 2.5, and visual analog scale pain from 5.5 ± 2.5 to 0.7 ± 1.3 ( P < .001). Ten patients had a final ASES score <70, which was associated with a history of previous surgery ( P = .03). The degree of posterior glenoid wear did not influence any outcome score. Eight patients underwent subsequent ipsilateral shoulder surgery including 5 (4%) undergoing revision to reverse shoulder arthroplasty for subscapularis insufficiency (2 following traumatic injury). No revisions for prosthetic joint infection or humeral or glenoid implant loosening were performed. aTSA using a minimally cemented standard glenoid implant following partial version correction for Walch B2 and B3 glenoids leads to marked improvements in ROM and PROs, regardless of glenoid wear pattern. No reoperations were performed for glenoid or humeral loosening or infection at most recent follow-up. A history of previous surgery predicted an inferior clinical outcome.