An analysis of factors affecting the long-term results of total shoulder arthroplasty in inflammatory arthritis.

The authors studied 36 patients undergoing 50 total shoulder arthroplasties for inflammatory arthritis with an average follow-up period of 5 years (range, 2-9 years). There were 7 men and 29 women. Twenty-nine patients (38 shoulders) had rheumatoid arthritis, 6 patients (11 shoulders) had systemic lupus erythematosus, and 1 patient (1 shoulder) had psoriatic arthritis. Twenty-four of the procedures were performed on the dominant hand. Twenty-one patients (32 shoulders) were receiving steroids, chemotherapeutic agents, or Gold therapy at the time of arthroplasty. Using the Hospital for Special Surgery scoring system, 22 shoulders had excellent, 18 good, 1 fair, and 9 poor results. Forty-eight of 50 shoulders had satisfactory pain relief. The average shoulder score was 85 points (range, 50-100 points). There were no reoperations, infections, or persistent nerve palsies and no symptomatically loose prostheses, although there were three cases of glenoid component shift. Eighteen glenoid components and 4 humeral components had radiolucent lines and in two cases metaphyseal bone atrophy was noted about the proximal humerus with firm cement fixation and bone hypertrophy about the stem tip. Total shoulder arthroplasty provides a predictable reconstructive alternative for patients with inflammatory arthritis. Overall functional results are related to the biologic condition and motivation of the patient, the status of the rotator cuff, and prosthesis alignment.