RESUMO
BACKGROUND: In 2015, our center replaced vancomycin with linezolid for the postoperative empirical treatment of osteoarticular infections (OAI). OBJECTIVES: To assess the bacteriological relevance of linezolid for orthopedic postoperative probabilistic antibiotic therapy. METHODS: Analysis of an observational cohort of patients empirically treated with a combination of linezolid and piperacillin/tazobactam during the immediate postoperative stage for an OAI between July 1st 2015 and July 1st 2016, in a French reference center. RESULTS: Seventy-seven of 126 patients who received a probabilistic postoperative combination of linezolid with piperacillin/tazobactam had microbiological proof of infection. Sixty-six of 77 OAI involved material, including an osteosynthesis in 45 cases (68%) and prosthesis in 21 cases (32%). Infection was due to Gram-positive bacteria in 62 cases (80.5%), mostly S. aureus (n=32, 41.6%), and S. epidermidis (n=14, 18.2%) accounting for 74.2% of Gram-positive bacteria. Among 14 OAI due to S. epidermidis, 11 (78.6%) were due to methicillin-resistant strains. All the S. aureus and S. epidermidis strains were susceptible to linezolid (MICs ≤ 4 mg/L), except in one patient previously treated with linezolid who was infected with a linezolid-resistant S. epidermidis strain (MIC > 256 mg/L). CONCLUSION: Linezolid can be used empirically in postoperative antibiotic therapy of OAI before obtaining definitive microbial results. Although linezolid resistance is rare in this population, previous oxazolidinone treatment should be documented before initiation of probabilistic postoperative treatment to highlight potential linezolid resistance.