Results of cemented total shoulder replacement with a minimum follow-up of ten years.

BACKGROUND: The purpose of this study was to assess the clinical and radiographic long-term outcomes of patients treated with a third-generation cemented total shoulder replacement and followed for at least ten years.

METHODS: The results of thirty-nine arthroplasties were analyzed clinically with use of the Constant score and on radiographs in two projections, with special regard to glenoid component loosening, at a mean of eleven years (range, ten to fifteen years) postoperatively.

RESULTS: The mean Constant score was 27 points (range, 11 to 54 points) preoperatively and 61 points (range, 21 to 86 points) postoperatively (p < 0.0001). Mean shoulder flexion increased from 84° (range, 40° to 150°) preoperatively to 133° (range, 40° to 180°) postoperatively; mean abduction, from 77° (range, 40° to 110°) to 123° (range, 40° to 180°); and mean external rotation, from 11° (range, -20° to 40°) to 35° (range, 0° to 60°). No humeral components but 36% of the glenoid components were radiographically loose at the time of follow-up. Kaplan-Meier survivorship of the glenoid component was 100% after thirteen years with revision as the end point, whereas survivorship with radiographic loosening as the end point was only 48% after thirteen years. We found no correlation between glenoid loosening on radiographs and clinical findings such as the Constant score. Cranial migration of the humerus was seen in 69% of the cases.

CONCLUSIONS: With the implants, cementation, and surgical technique utilized in this group of patients with primary glenohumeral osteoarthritis, radiographic loosening of the glenoid component and rotator cuff deficiency were very common at the ten to fifteen-year review. The follow-up was not long enough to fully identify the clinical sequelae of these findings.