Shoulder arthroplasty for the treatment of postinfectious glenohumeral arthritis

BackgroundShoulder arthroplasty after native shoulder infection is an uncommon problem with limited outcomes data. The purpose of this study was to evaluate the rates of reinfection and clinical outcomes after shoulder arthroplasty for the treatment of postinfectious glenohumeral arthritis.MethodsBetween 1977 and 2009, 24 shoulders underwent shoulder arthroplasty for postinfectious glenohumeral arthritis. Twenty-three were monitored for a minimum of 2 years (mean, 8.3 years) or until reoperation. Complications and clinical and radiographic results were documented at the most recent follow-up.ResultsOf the 23 shoulders, 23 had no pain or mild or moderate pain after vigorous activity. Pain scores improved from 4.5 to 2.1 points after shoulder arthroplasty (P < .001). The mean shoulder abduction improved from 62° to 110° (P < .001), and the mean external rotation improved from 14° to 47° (P < .001). Subjectively, the result in 16 of the 23 shoulders was rated as much better or better. Five shoulders required reoperation, with 2 having an infectious cause. The Neer rating was excellent in 2 shoulders, satisfactory or successful in 9, and unsatisfactory or unsuccessful in 12. Radiographic follow-up showed 3 glenoids and 3 humeral components were at risk for loosening.ConclusionsShoulder arthroplasty for the treatment of the sequelae of an infected shoulder can be performed with a low risk of reinfection. The higher-than-expected rate of clinical or radiographic loosening remains concerning for culture negative infection. Although overall pain and motion can be expected to improve, unsatisfactory clinical results are not uncommon and may be secondary to the initial insult of infection.