Factors governing the healing of Staphylococcus aureus infections following hip and knee prosthesis implantation: a retrospective study of 95 patients.

INTRODUCTION: The prognostic factors for total hip arthroplasty (THA) and total knee arthroplasty (TKA) Staphylococcus aureus prosthetic joint infections are poorly known, notably because of the heterogeneous management in terms of both antibiotic administration and adopted surgical strategy. Uniform treatment regimens would make it easier to define the outcome of these S. aureus infections.

PATIENTS AND METHODS: Between 2001 and 2006, 95 patients with a S. aureus joint infection after THA or TKA were treated, strictly following a standardized protocol according to the recommendations of Zimmerli et al. The patients' mean age was 65.7 years, 71 with THA and 28 with TKA (four patients had two infected joints). These 95 patients presented 120 infectious episodes, all of whom had surgical treatment: 53 lavages (44.1%), 17 one-stage prosthesis revisions (14.2%), 29 two-stage prosthesis revisions (24.2%), and 21 prostheses removed (17.5%). On the intraoperative samples taken, methicillin-sensitive S. aureus (MSSA) was isolated in 88 patients (73.3%) and methicillin-resistant S. aureus (MRSA) in 18 patients (15%); finally 14 patients were included because of the positive results of preoperative samples taken. Twenty-seven infections (22.5%) were multibacterial, including at least S. aureus and 93 were single S. aureus bacteria. Success was defined at a minimum 12 months of follow-up by the association of the following parameters: normal erythrocyte sedimentation rate (ESR) and/or C-reactive protein (CRP) results, noninflammatory scar with no fistula, no surgical revision, and no antibiotic treatment.

RESULTS: At a mean follow-up of 38±24.9 months, 81 of the 120 infectious episodes were resolved (67.5%) and 77 of the 95 patients were healed (81%). Six parameters significantly influenced the healing of the infection: initial cementless fixation, THA, preoperative knowledge of the bacterium, immediate postoperative antibiotic therapy adapted to the microbiological data, changing the prosthesis, and monobacterial infection. Only the latter two were independent, with an odds ratio of 5 (1.6-14.9) and 2.9 (1.1-7.7) respectively. However, resistance to methicillin did not appear to be a factor of failure.

DISCUSSION: Management of prosthetic S. aureus infections according to a uniform protocol in a specialized medical center healed the infection in 81% of the patients. Treatment improvement requires knowledge of the bacterium, adaptation of immediate probabilistic antibiotic therapy, and preference for changing the prosthesis over simple lavage.