268. PLURIMICROBIAL INFECTION OF TOTAL HIP ARTHROPLASTY: 68 CASES

Abstract

Purpose of the study: Plurimicrobial infections account for 4 to37% of all infections of total hip arthroplasties (THA). According to data in the literature, they are the source of failure, contraindicating one-phase replacement procedures. The purpose of our study was to evaluate the results of our management practices in this group of patients and also to identify factors of risk associated with multimicrobial infection.

Material and methods: A prospective study included 116 patients with an infected THA from November 2002 to December 2006. Sixteen patients (13.8%), mean age 68±12.7 years had a plurimicrobial infection defined by having at last two interoperative bacteriological samples positive for two or more germs. Surgical treatment consisted in a single-phase replacement in seven cases, a two-phase replacement in seven, resection of the head and neck in one, and wash-out resection in one. Mean duration of the antibiotic therapy ws 91±6 days, including 46±14 days intravenously. Anaerobic germs were isolated more commonly in plurimicrobial infections than monomicrobial infections (50% versus 11%). Patients were assessed with prospectively collected data. Mean follow-up was 34±13 months, with none lost to follow-up. The main outcome was apparent cure rate of the initial infection at minimum two years follow-up, defined by the absence of clinical, biological and radiographic signs of infection, and absence of death attributable to infection or its treatment. If infection was suspected, a hip puncture or intraoperative samples confirmed the relapse (same germs) or reinfection (different germs).

Results: The cure rate was 100% for plurimicrobial infections and 97% for monomicrobial infections. There were however four reinfections in the monomicrobial group. In this series, the risk factor statistically associated with plurimicrobial infections was the presence of a fistula with an odds ratio of 5.4.

Discussion: A larger number of patients would probably enable identification of other risk factors associated with plurimicrobial infections.

Conclusion: The cure rate of plurimicrobial infections was higher than reported in the literature but for a small group of patients. The presence of a fistula was strongly associated with these plurimicrobial infections.