Long-term results of unconstrained Roper-Tuke total elbow arthroplasty in patients with rheumatoid arthritis

We evaluated the long-term results of 12 unconstrained Roper-Tuke total elbow replacements that were performed in 12 patients with rheumatoid arthritis from 1983 to 1989. The mean follow-up period was 9.5 years (range 8 to 13 years). We used the Ewald elbow-scoring system to chart results. This showed that the scores for the 12 elbows had improved from an average preoperative score of 39 points (range 17 to 72 points) to an average postoperative score of 80 points (range 45 to 97 points). The greatest improvements were in terms of pain relief, function, and range of motion. Eight elbows were free, of pain by the end of follow-up. Average elbow flexion increased from 115° before operation to 140° after operation, and pronation and supination increased from 52° to 61° and 42° to 71°, respectively. Radiographs of the 12 elbows showed constant wear of the ulnar polyethylene with loosening of 2 ulnar components. Revision of the prosthesis was necessary in 2 elbows because of aseptic loosening. Complications included 1 subluxation, 1 supracondylar fracture, and 2 ulnar neuropathies. Despite some excellent clinical results with a follow-up of over 10 years, the authors no longer recommend the use of this kind of elbow prosthesis in patients with rheumatoid arthritis because of the high complication rate and the impossibility of adapting this implant in the event of bone loss. The authors propose a new classification of humeral bone loss that will allow for better planning of primary and revision total elbow arthroplasties.