Background: Several studies have reported that trough levels may not be optimal for monitoring vancomycin therapy, because of overexposure and nephrotoxicity risks. Therefore, we developed a population pharmacokinetic model to optimize vancomycin dosing and monitoring in pediatrics.
Methods: Data were retrospectively collected on 76 pediatric patients 1-12 years of age, admitted to general pediatric wards or intensive care units at King Saud University Medical City, Riyadh, Saudi Arabia. The predictability of 3 methods for calculating the area under the curve (AUC) at steady state was assessed for optimum vancomycin therapy monitoring. The 3 methods were simple linear regression, Bayesian approach and the 2-sample pharmacokinetic equation method. We also used Monet Carlo simulations to evaluate the dosing of vancomycin.
Results: A 1-compartment model adequately described the data. A strong correlation occurred between the observed and predicted AUC from 0 to 24 hours (AUC0-24h) calculated using the Bayesian approach with a trough sample only or pharmacokinetic equations based on 2 measured samples (R = 0.93 and 0.92, respectively). For the simple linear regression method with a trough sample only, the predicted AUC0-24h at steady state with vancomycin trough levels of 10, 15 and 20 µg/mL were 413, 548 and 714 µg·hour/mL, respectively. The target AUC0-24h above 400 was achieved in 46% and 95% of individuals with trough values of 7-11 and 11-15 µg/mL, respectively. Monte Carlo simulations showed that 60-80 mg/kg/d doses are needed to optimize vancomycin therapy.
Conclusions: In conclusion, targeting vancomycin trough levels above 15 µg/mL in pediatrics would overshoot the target AUC0-24h above 400 and expose them to unnecessary adverse events.