A prospective functional outcome study of shoulder arthroplasty for osteoarthritis with an intact rotator cuff

Prospective pain relief and functional outcome data were obtained on 65 shoulder arthroplasties in 55 patients. Thirty-seven total shoulder arthroplasties (TSAs) and twenty-eight hemiarthroplasties were followed up for a mean of 4.3 years (range, 2-8 years). The mean age was 63 years. Pain scores on a visual analog scale improved from a preoperative mean of 64 to 12 postoperatively. TSA and hemiarthroplasty postoperative scores did not differ significantly, but TSA patients started with a worse preoperative score and improved more. Five other visual analog scale scores improved, including function and quality of life. There were similar relationships between TSAs and hemiarthroplasties in each instance (poorer preoperative scores and more improvement). The American Shoulder and Elbow Surgeons Shoulder Score Index improved from 39 to 88 with surgery. The hemiarthroplasty group improved from a mean of 42 to 84, and the TSA group improved from 37 to 91. Although the numerical differences were small, the superiority of TSAs with regard to final score and rate of improvement was statistically significant. Patients with the poorest preoperative scores improved the most, regardless of prosthesis type. Mean active forward elevation improved from 100° to 147°, external rotation improved from a mean of 7° to 39°, and internal rotation improved by a mean of 3 spinal segments. These measures did not differ between TSA and hemiarthroplasty patients. One implant failure was treated with revision from TSA to hemiarthroplasty. Radiographic parameters characterizing component position and offset were measured, but none was found to predict outcome. The results suggest a modest superiority of TSA over hemiarthroplasty in the medium term. Because both TSA and hemiarthroplasty provide considerable and nearly comparable improvement, the long-term risks of glenoid wear and loosening need to be clearly defined before a definitive conclusion can be reached regarding the differential indications for these two procedures.