PATIENT OUTCOME WITH REINFECTION FOLLOWING REIMPLANTATION FOR THE INFECTED TOTAL KNEE ARTHROPLASTY

From 1976 to 1992, 24 knees (24 patients) were treated for reinfection after reimplantation of a prosthesis to treat an infected total knee arthroplasty. Including the index total knee arthroplasty, the average number of total procedures for the affected knee was 9.3 procedures ( range, 5-23 procedures), and surgical procedures used for the affected knee after reinfection averaged 3.7 procedures (range, 1-12 procedure s). The final outcome included 10 knees with a successful knee arthrod eses, 5 patients with infected prostheses who were prescribed suppress ive oral antibiotic treatment, 4 above the knee amputations, 3 persist ent pseudarthroses of the knee joint, 1 resection arthroplasty, and 1 uninfected total knee prosthesis. A poor prognosis was associated with use of a hinged knee design: 3 of the 4 amputations were done after a hinge knee prosthesis was implanted; 2 patients had a persistent pseu darthrosis after a failed attempt at knee arthrodesis; 1 patient had a resection arthroplasty; and 1 patient had a retained infected prosthe sis. A successful arthrodesis in which the initial attempt included us e of an external fixation device was more likely for prostheses withou t stems (75%) when compared with cemented stemmed prostheses (40%). Lo ng intramedullary arthrodesis was successful in all 3 attempts. Aspira tion for diagnosis followed by oral antibiotic suppression was univers ally unsuccessful in 4 patients, whereas early aggressive open debride ment facilitated retention of the prosthesis in 4 of 6 patients. The i ncreased difficulty in achieving a healed wound, a successful knee art hrodesis, and eradication of infection with subsequent nonprosthetic s alvage procedures as well as the attendant morbidity associated with r einfection must be considered carefully and discussed with the patient before the reimplantation prosthesis is inserted to treat the infecte d total knee arthroplasty.