Time-dependent antibacterial effects of linezolid in experimental rabbit endocarditis.

Abstract

were >100 in the present study, although statistical comparisons did not confirm a correlation with significant antibacterial effects. Studies in standard animal models of lung and thigh infection have shown that the efficacy of β-lactams and macrolides does not correlate with AUC/MIC, but does correlate with the percentage of time that the drug levels were maintained above the MIC for the pathogen used. Preliminary evidence shows a positive correlation between antibacterial effects and maintenance of linezolid plasma levels above the MIC for ∼40% of the dosing interval in a neutropenic mouse model of thigh infection. 5 In the present endocarditis study, linezolid trough plasma levels were maintained above the MIC for this MRSA strain for the entire dosing interval on each day except day 1. Other rabbit endocarditis studies have stressed the importance of maintaining plasma levels above the MIC to achieve efficacy. 6 It is important to consider that the pharmacokinetic requirements for efficacy in experimental endocarditis are more stringent compared with those for lung or thigh infections. Bacterial endocarditis is a deep-seated infection that is difficult to cure for several reasons. The vegetation is avascular, and is comprised of layers of platelets and fibrin, which reduces both neutrophil host-defence mechanisms and penetration of antimicrobial agents. In addition, the metabolic state of the bacteria within the vegetation may be reduced, lowering the effectiveness of drugs that target actively growing bacteria. In this setting, antibacterial agents such as β-lactams that have slow killing rates and no post-antibiotic effect must be administered at short intervals in order to maintain drug levels above the MIC throughout the entire dosing interval. The present study with linezolid demonstrates that maintaining plasma and valve vegetation drug levels above the MIC for the entire dosing interval over multiple days results in progressive reduction in bacterial counts in infected valves. While regional linezolid concentrations within the valve vegetation were not determined, the antibacterial effects and substantial drug concentration in valve homogenates suggest sufficient penetration of the drug.