Influence of the MBC/MIC ratio on the antibacterial activity of vancomycin versus linezolid against methicillin-resistant Staphylococcus aureus isolates in a pharmacodynamic model simulating serum and soft tissue interstitial fluid concentrations reported in diabetic patients.

OBJECTIVES To explore serum and tissue pharmacodynamics of linezolid versus vancomycin against methicillin-resistant Staphylococcus aureus (MRSA) clinical isolates with different MBC/MIC ratios. METHODS Five strains (vancomycin MIC/MBCs, mg/L) were used: TOL-1 (2/≥64), TOL-2 (1/16), LT-1 and LT-2 (1/8) and NT (1/2). The linezolid MIC/MBC for all strains was 2/≥64 mg/L. A two-compartment dynamic computerized device was used (inocula 10(7) cfu/mL). Free concentrations obtained in serum and interstitial fluid with twice-daily regimens of 1 g of vancomycin or 600 mg of linezolid were simulated over 48 h. ABBCs (differences between control growth curves and killing curves of bacteria exposed to antibiotics; log10 cfu × h/mL) and log10 reductions in initial inocula were calculated. RESULTS In serum simulations, vancomycin (AUC0-24/MIC = 251.8 for TOL-1 and 503.6 for the remaining strains) was bacteriostatic against strains with MBC/MIC ≥8, but bactericidal against NT. In interstitial fluid simulations (AUC0-24/MIC = 54.6 for TOL-1 and 109.2 for the remaining strains), initial inocula grew in all cases. Linezolid, both in serum (AUC0-24/MIC = 87.0) and in interstitial fluid (AUC0-24/MIC = 130.6) simulations, reduced initial inocula ≥2.2 log10 for all strains (apart from LT-1 in serum simulations that showed a bacteriostatic profile). ABBCs were similar in serum and interstitial fluid with linezolid, but significantly lower in interstitial fluid simulations with vancomycin. CONCLUSIONS From the pharmacodynamic perspective (serum concentrations), vancomycin tolerance should include MBC/MIC ≥8 since strains exhibiting this ratio showed bacteriostatic profiles similar to those obtained with isolates with MBC/MIC ratios of 16 or 32. Insufficient concentrations of vancomycin at the simulated infected site were linked to bacteriological failure. Free concentrations of linezolid at the infection site pharmacodynamically covered MRSA.