Implementation of a system-wide vancomycin AUC-based Bayesian dosing software on incidence of acute kidney injury

PURPOSE: In 2020, the Infectious Diseases Society of America (ISDA) consensus guideline recommended use of area under the curve (AUC)-based dosing in preference to trough-based dosing of vancomycin. AUC-based dosing has been shown to minimize the risk of vancomycin-induced acute kidney injury. The purpose of this study was to analyze the impact of system-wide implementation of vancomycin AUC-based Bayesian dosing software on the incidence of acute kidney injury, time to therapeutic range, and hospital length of stay. METHODS: This study was a retrospective, observational study in Houston, Texas from January 2019 to July 2019 and January 2022 to July 2022. Patients included were age 18 years and older with intravenous (IV) vancomycin use for at least 96 hours and admitted to the Memorial Hermann Health System. No interventions were performed. The primary outcome was a comparison of vancomycin-induced acute kidney injury of trough-based vs AUC-based dosing and non-pharmacy-to-dose vs pharmacy-to-dose strategies. Secondary outcomes included time to therapeutic range and hospital length of stay. RESULTS: Vancomycin-induced acute kidney injury did not significantly change between the pre- and post-implementation groups (p=1.000). Additionally, in multivariate logistic regression, there was no difference in acute kidney injury between the two study groups (p=0.801), controlling for number of nephrotoxin categories administered and vancomycin indication. Time to therapeutic range in days mean (SD) in the pre-implementation group was 1.74 (1.36) days and 0.95 (1.47) in the postimplementation group (p<0.001). Hospital length of stay in days, mean (SD), in the preimplementation group was 13.2 (16.5) and 13.7 (10.8) in the post-implementation group (p=0.7567). CONCLUSION: Implementation of an AUC-based Bayesian dosing software in conjunction with pharmacy-to-dose vancomycin consults did not impact acute kidney injury rates but did decrease time to therapeutic range and hospital length of stay in a sub-group analysis. ...