Increasing Vancomycin Serum Trough Concentrations and Incidence of Nephrotoxicity

BACKGROUND: Conflicting evidence regarding the association of vancomycin serum concentrations with efficacy and toxicity has resulted in controversy regarding optimal target concentrations. Recent publications recommend attaining higher vancomycin trough concentrations of 15 to 20 mg/L for target infections, yet limited research is available assessing the correlation of vancomycin serum concentrations with toxicity. The aim of this study was to evaluate the association between vancomycin serum trough concentrations and nephrotoxicity. METHODS: A 2-phase retrospective analysis was completed. Phase 1 evaluated 2493 courses of vancomycin completed between January 2003 and December 2007. The analysis describes a 5-year trend in vancomycin prescribing practices and assesses the association of nephrotoxicity with baseline serum creatinine, vancomycin serum trough concentrations, and duration of vancomycin therapy. Phase 2 examined patients receiving vancomycin therapy during 2007 to evaluate specific risk factors for development of nephrotoxicity. RESULTS: The proportion of vancomycin serum trough concentrations >= 15 mg/L and >= 20 mg/L increased significantly over time. Statistical analysis identified vancomycin serum trough concentrations >= 14 mg/L, duration of vancomycin therapy >= 7 days, and baseline serum creatinine levels >= 1.7 mg/dL as independent predictors of nephrotoxicity. Phase 2 analysis again implicated mean vancomycin serum trough concentration as a significant predictor of nephrotoxicity. Nephrotoxicity resolved in 81% (17/21) of cases evaluated. CONCLUSIONS: A higher vancomycin serum trough concentration and prolonged vancomycin therapy are associated with an increased risk of nephrotoxicity. The decision to target increased vancomycin trough concentrations should be based on an assessment of the severity of the infection and must consider the nephrotoxicity risk associated with increased vancomycin levels. (C) 2010 Elsevier Inc. All rights reserved. The American Journal of Medicine (2010) 123, 1143-1149