Abstract
Background: Monitoring of vancomycin using the area under the curve (AUC) to minimum inhibitory concentration (MIC) ratio is now preferred for serious methicillin-resistant Staphylococcus aureus infections. Vancomycin AUC/MIC monitoring is being investigated but is not yet well elucidated with other bacterial pathogens. Methods: A retrospective cross-sectional study was conducted assessing patients with streptococcal bacteremia treated with vancomycin definitive therapy. AUC was calculated using a Bayesian approach, and classification and regression tree analysis was used to identify a vancomycin AUC threshold predictive of clinical failure. Results: Eleven patients had a vancomycin AUC < 329 of which 8 (73%) experienced clinical failure, while 35 patients had a vancomycin AUC ≥ 329 of which 12 (34%) experienced clinical failure (P = .04). Hospital length of stay was longer in the AUC ≥ 329 group (15 vs 8 days, P = .05), whereas time to bacteremia clearance (29 [22-45] vs 25 [20-29] hours, P = .15) and toxicity incidence (13% vs 4%, P = 1) were similar between groups. Conclusions: This study identified a VAN AUC threshold of <329 to be predictive of clinical failure in patients with streptococcal bacteremia which should be interpreted as hypothesis-generating. Studies evaluating VAN AUC-based monitoring for streptococcal bloodstream infections along with other infection types are needed before implementation into clinical practice can be recommended.