We report our experience of staged revision surgery for the treatment of infected total elbow arthroplasty (TEA). Between 1998 and 2010 a consecutive series of 33 patients (34 TEAs) underwent a first-stage procedure with the intention to proceed to second-stage procedure when the infection had been controlled. A single first-stage procedure with removal of the components and cement was undertaken for 29 TEAs (85%), followed by the insertion of antibiotic-impregnated cement beads, and five (15%) required two or more first-stage procedures. The most common organism isolated was coagulase-negative Staphylococcus in 21 TEAs (62%).

A second-stage procedure was performed for 26 TEAs (76%); seven patients (seven TEAs, 21%) had a functional resection arthroplasty with antibiotic beads in situ and had no further surgery, one had a persistent discharge preventing further surgery.

There were three recurrent infections (11.5%) in those patients who underwent a second-stage procedure. The infection presented at a mean of eight months (5 to 10) post-operatively. The mean Mayo Elbow Performance Score (MEPS) in those who underwent a second stage revision without recurrent infection was 81.1 (65 to 95).

Staged revision surgery is successful in the treatment of patients with an infected TEA and is associated with a low rate of recurrent infection. However, when infection does occur, this study would suggest that it becomes apparent within ten months of the second stage procedure.

Cite this article: Bone Joint J 2013;95-B:1681–6.

At our institution between 1994 and 2003 a total of 36 revision total elbow Arthroplasties were performed in 34 patients. We clinically reviewed 25 patients and reviewed the notes and x-rays of all of them. Of eleven who were not reviewed clinically seven had died from an unrelated cause and four were unable to attend because of illness but we were able to include them as sufficient data were available in the notes. There were 24 female and 12 male, Average age was 67 years and twelve had elbow Arthroplasty in a non-dominant side. The average follow up was 6 years (range 5–13 years). The mean period between the primary and revision surgery was sixty three months (range 3–240 months). The indication for surgery was mainly for aseptic loosening in 15 cases, followed by septic loosening in twelve. All cases of septic loosening had two stage revisions. Other reasons for revision in this series include unstable elbows, implant fracture and peri-prosthetic fractures. Twelve of these revisions had a further revision for a variety of reasons at an average period of twenty eight months. Seven patients had thirteen complications in this series, two radial nerve palsies (one recovered), one distal humeral fracture, five cortical perforations and five triceps weakness. Most of the patients are satisfied with their elbows. The mean Mayo elbow Performance Score was 79 points. We conclude that revision Elbow Arthroplasty is a specialized surgery which is technically demanding, with high risk of complications and high re-revision rate and therefore, should be done in a specialised centres.

We present a case of idiopathic osteonecrosis of the humeral capitellum in a 44-year-old female in the absence of any associated risk factors. Arthroscopy was undertaken to remove the loose bodies and debride the capitellum, with a satisfactory outcome.