Model-Informed Precision Dosing of Vancomycin in Vietnamese Children: Innovative Midpoint Concentration Monitoring...

Bayesian area under the curve (AUC)-guided vancomycin dosing, preferably with 2 samples (2S), is recommended for pediatric patients. However, the timing of sampling may not always be convenient in the clinical setting. In low- to middle-income countries, the implementation of model-informed precision dosing (MIPD) using the Bayesian approach remains limited because of high software costs. Using MIPD with Bayesian analysis, the authors compared the accuracy and precision of midpoint and guideline-recommended 2S strategies and evaluated the clinical utility of an open-source software package and compared it with PrecisePK.

This retrospective cohort study was conducted at 2 pediatric hospitals with a combined 2000 beds from April 2022 to April 2023. The authors enrolled patients aged from 3 months to 16 years who received at least 2 doses of vancomycin and had at least 2 measured drug concentrations: Cmax (1-2 hours after the end of infusion), Cmid (midpoint of the dosing interval), and Ctrough (30 minutes-1 hour before the next dose). The accuracy and precision of pharmacokinetic parameters were calculated using and , respectively.

Eighty children with 226 vancomycin concentrations were included. The median age was 1.6 (interquartile range 0.88-3.01) years, the average weight was 10.0 (8.0-12.0) kg, the baseline serum creatinine concentration was 0.43 (0.38-0.53) mg/dL, and the empirical vancomycin dose was 60 (57-61) mg/kg/d. Most subjects received empirical vancomycin for pneumonia (75%) and bacteremia (26%), and 75% were admitted to the intensive care unit. Compared with the 2S monitoring state, the accuracy and precision of the midpoint for the Bayesian-derived AUC 24 values were 5.02% and 6.45%, respectively. Compared with PrecisePK, the AUC 24 estimation by Shiny exhibited an accuracy and precision of -5.58% and 6.09%, respectively.

Midpoint concentration offers a convenient sampling approach to accurately monitor vancomycin therapy in hospitalized pediatric patients. Shiny serves as an alternative Bayesian MIPD tool for dosing and monitoring vancomycin therapy in children in Vietnam.