Limited evidence exists evaluating pharmacokinetic thresholds for vancomycin efficacy and nephrotoxicity using non-Bayesian methods. The objective of this study was to evaluate the 24-h steady-state vancomycin area under the concentration-time curve (AUC24) thresholds for efficacy and nephrotoxicity in patients with methicillin-resistant Staphylococcus aureus bacteraemia (MRSA-B) after implementing two-point pharmacokinetic therapeutic drug monitoring. A single-centre, retrospective cohort study was performed including adult patients admitted between 1 June 2016 and 1 January 2018 with MRSA-B treated with vancomycin for ≥72 h. The AUC24 was calculated using peak and trough vancomycin serum concentrations. Clinical success was defined as defervescence and blood culture sterilisation by Day 7. Nephrotoxicity was defined as an increase in serum creatinine of >0.5 mg/dL (or ≥50%) from baseline. Classification and regression tree (CART) analyses were performed to identify AUC24 thresholds for efficacy and nephrotoxicity. Forty-six patients were included in the study. Clinical success and nephrotoxicity were observed in 81.8% and 13.0%, respectively. The CART-derived vancomycin AUC24 thresholds for clinical success and nephrotoxicity were ≥297 mg·h/L and ≥710 mg·h/L, respectively. Patients with an AUC24 ≥297 mg·h/L had a >2.7-fold increase in clinical success compared with those who did not (89.5% vs. 33.3%, respectively; P = 0.01), and patients with an AUC24 ≥710 mg·h/L had a >7-fold increase in nephrotoxicity compared with those with an AUC24 <710 mg·h/L (66.7% vs. 9.3%, respectively; P = 0.04). This study supports current recommendations to target vancomycin AUC24 values of 400-600 mg·h/L when calculated using two-point pharmacokinetics, although a wider range may exist.
Keywords: AUC; Area under the concentration–time curve; Methicillin-resistant Staphylococcus aureus; Pharmacokinetics; Therapeutic drug monitoring; Vancomycin.
Copyright © 2018 Elsevier B.V. and International Society of Chemotherapy. All rights reserved.