Background: The indications for resurfacing of the glenoid in patients who have osteoarthritis of the shoulder are not clearly defined; some investigators routinely perform hemiarthroplasty whereas others perform total shoulder arthroplasty.
Methods: Forty-seven patients (fifty-one shoulders) who were scheduled to have a shoulder arthroplasty for the treatment of degenerative osteoarthritis were randomly assigned, according to a random-numbers table, to one of two groups: replacement of the humeral head with resurfacing of the glenoid with a polyethylene component with cement (total shoulder arthroplasty [twenty-seven shoulders]) or replacement of the humeral head without resurfacing of the glenoid (hemiarthroplasty [twenty-four shoulders]). All patients received the same type of humeral component, and all operations were performed by or under the direct supervision of the same surgeon. The patients were followed for a mean of thirty-five months (range, twenty-four to seventy-two months) postoperatively. Evaluation was performed with use of the scoring systems of the University of California at Los Angeles and the American Shoulder and Elbow Surgeons.
Results: No difference was observed between the preoperative scores for the two groups of patients. Postoperatively, the mean scores with use of the University of California at Los Angeles system and the American Shoulder and Elbow Surgeons system were 23.2 points (range, 10 to 31 points) and 65.2 points (range, 15 to 94 points), respectively, after hemiarthroplasty and 27.4 points (range, 9 to 34 points) and 77.3 points (range, 3 to 100 points), respectively, after total shoulder arthroplasty. With the numbers available for study, no significant difference was found between the two operative groups with respect to the postoperative score. (Thirty-five subjects per group would be needed, assuming an effect size of 0.60 and a power of 0.80.) Total shoulder arthroplasty provided significantly greater pain relief (p = 0.002) and internal rotation (p = 0.003) than hemiarthroplasty did. Total shoulder arthroplasty also provided superior results in the specific areas of patient satisfaction, function, and strength, although none of these differences were found to be significant, with the numbers available.
Total shoulder arthroplasty was associated with increased cost ($1177), operative time (thirty-five minutes), and blood loss (150 milliliters) per patient compared with hemiarthroplasty. To date, none of the total shoulder arthroplasties in the study group have been revised. Hemiarthroplasty yielded equivalent results for elevation and external rotation. Three of the twenty-five patients who had had a hemiarthroplasty needed a subsequent operation for resurfacing of the glenoid. The mean cost for the revision operations was $15,998.
Conclusions: Total shoulder arthroplasty provided superior pain relief compared with hemiarthroplasty in patients who had glenohumeral osteoarthritis, but it was associated with an increased cost of $1177 per patient.