Posterior-approach total shoulder arthroplasty: a case series with a minimum of 2-year outcomes.

Background: A posterior approach to shoulder arthroplasty has been described that may improve exposure to the retroverted glenoid and spare the rotator cuff. The purpose of this study was to evaluate posterior total shoulder arthroplasty (PTSA) at a minimum of 2 years.

Methods: Between September 2016 and October 2017, 28 patients (30 shoulders) who underwent PTSA were retrospectively identified. Patients were monitored at baseline and at 6 weeks, 3 months, 6 months, 1 year, and a minimum of 2 years after surgery for range of motion, American Shoulder and Elbow Surgeons (ASES) and Simple Shoulder Test scores, radiographic outcomes, complications, and patient satisfaction. Radiographs were assessed for head height, neck-shaft angle, humeral head centering, and the presence or absence of glenoid-component cement lines. Repeated-measures analysis-of-variance models were implemented to compare the means for each outcome variable over time.

Results: Of the 28 patients (30 shoulders), 24 patients (26 shoulders) (87%) were available for follow-up at a minimum of 2 years (average, 31.2 months; range, 23-40 months). At latest follow-up, the average ASES pain score was 45.2 (standard deviation [SD], 10.9; range, 20-50; 95% confidence interval [CI], 40.8-49.6); average ASES function score, 42.3 (SD, 8.9; range, 21.7-50; 95% CI, 38.7-45.8); average ASES total score, 87.5 (SD, 18.8; range, 31.7-100; 95% CI, 79.8-95.1); and average Simple Shoulder Test score, 10.2 (SD, 2.0; range, 5-12; 95% CI, 9.4-11.0). Average forward flexion was 164° (SD, 13.9°; range, 130°-180°; 95% CI, 158.3°-170.4°); average external rotation, 78° (SD, 9.2°; range, 55°-90°; 95% CI, 74.3°-82.3°); and average internal rotation, T10 (SD, 2; range, L3 to T6; 95% CI, T10 to T8). The average head height was 0.86 mm above the greater tuberosity (range, -6.5 to 6.9 mm), the average neck-shaft angle was 128° (range, 113°-148°), and the average implant placement was 0.12 mm superior to the center of the neck cut (range, -3.3 to 3.5 mm). Six implants were considered malpositioned based on the criterion of >5 mm of head height above or below the tuberosity or >2 mm off-center. Glenoid cement lines were present in 7 shoulders. Two patients experienced complications: 1 patient had symptomatic glenoid loosening 29 months postoperatively that required revision to an anterior reverse total shoulder arthroplasty, and 1 patient had postoperative weakness that resolved by 6 months.

Conclusion: PTSA is safe and efficacious at 2 years, spares the rotator cuff, and may improve early postoperative function. PTSA is more easily performed with stemless arthroplasty. Challenges with creation of the humeral head osteotomy and identification of the center of the cut humeral head surface can affect the accuracy of implant positioning. Further improvements in technique will likely improve the feasibility of this approach.