Background: Recent evidence suggests that vancomycin demonstrates reduced activity against methicillin-resistant Staphylococcus aureus (MRSA) infections when vancomycin MIC values are at the high end of the susceptibility range (> or = 1.5 mg/L). However, scant research exists on factors predictive of high vancomycin MICs (> or = 1.5 mg/L) among MRSA bacteraemic patients. Empirical therapy decisions would greatly benefit from such information.
Objectives: To identify the parameters predictive of high vancomycin MICs (> or = 1.5 mg/L) among MRSA bacteraemic patients and to develop an evidence-based clinical prediction tool.
Methods: This observational cohort study included adult patients with MRSA bloodstream infections between January 2005 and May 2007. Demographics, co-morbid conditions, and microbiology and antibiotic exposure data were collected. Vancomycin MICs were determined by Etest. Stepwise logistic regression was used to identify independent predictors of high vancomycin MICs.
Results: Of the 105 patients who met the inclusion criteria, 77 patients (73.3%) exhibited a high vancomycin MIC (> or = 1.5 mg/L). In the bivariate analysis, prior vancomycin exposure within 30 days of index culture collection [15 patients (19.5%) versus 1 patient (3.6%), P = 0.05] and residence in an intensive care unit (ICU) at the onset of infection [27 patients (35.1%) versus 3 patients (10.7%), P = 0.02] were both significantly associated with a high vancomycin MIC value and both were independent predictors of high MICs in the logistic regression.
Conclusions: Patients with MRSA bloodstream infections in the ICU or with a history of vancomycin exposure should be considered at high risk of infection with strains for which vancomycin MICs are elevated. Appropriate and aggressive empirical therapy is required for these patients.