• BACKGROUND AND PURPOSE
    •  It is unclear whether total shoulder arthroplasty (TSA) results in better outcomes in patients with degenerative shoulder diseases compared with hemiarthroplasty (HA). This randomized controlled trial (NCT01288066) is an international, multicenter study with the primary aim to assess pain and shoulder joint function at 5-year follow-up in adults surgically treated with TSA or HA.
  • METHODS
    •  The inclusion criteria were patients aged 18 and older with primary or secondary osteoarthritis, with a functionally intact rotator cuff and glenoid morphology of Walch type A1, A2, or B1. Randomization to either TSA or HA occurred intraoperatively after full surgical preparation for glenoid prosthetic implantation in all patients. The primary outcome measure was the Constant-Murley scores (CMS) at 5-year follow-up. Secondary outcomes were patient-reported outcomes (Shoulder Pain and Disability Index [SPADI], EQ-5D, and EQ VAS), adverse events, and implant survivorship at 5-year follow-up.
  • RESULTS
    •  There were 79 patients eligible, of whom 40 and 39 patients were randomized to receive TSA or HA, respectively. The dropout rate at 5 years was 34% due to 27 of 79 patients withdrawing consent or being lost to follow-up. The mean CMS improved from preoperative to 5 years' follow-up for both TSA and HA treatment groups. At 5 years, the TSA group had a significantly better mean CMS (77, 95% confidence interval [CI] 72-82) than the HA group (65, CI 57-73). The mean treatment difference was 12 (Cl 2.8-22; P = 0.01). The adverse event rate and relative risk of revision were not significantly different between the groups.
  • CONCLUSION
    •  In patients with glenohumeral osteoarthritis randomized to either TSA or HA, TSA was the favorable approach based on less pain and better joint function 5 years after surgery.