Comparing paediatric intravenous phenytoin doses using physiologically based pharmacokinetic (PBPK) modelling software

Purpose: To use a physiologically based pharmacokinetic (PBPK) modelling system to predict the serum levels achieved by two different intravenous loading doses of phenytoin. Methods: A phenytoin pharmacokinetic model was used in the SimcypTM population-based ADME simulator, simulating 100 children age 2-10 years receiving intravenous phenytoin (18 and 20 mg/kg). Visual checks were used to evaluate the predictive performance of the candidate model. Results: Loading with doses of 18 mg/kg, blood levels were sub-therapeutic in 22/100 (concentration at 2 h post infusion (C2h) 20 mg/mL). Loading with 20 mg/kg, the percentages were 15, 59, and 26, respectively. Increasing from 18 to 20 mg/kg increased the mean C2h from 16.0 to 17.9 μg/mL, and the mean AUC from 145 to 162 μg/mL/h. A C2h > 30 μg/mL was predicted in 4% and 8% of children in the 18 and 20 mg/kg doses, with 3% predicted to have a C2h > 40 μg/mL following either dose. For maintenance doses, a 1st dose of 2.5 or 5 mg/kg (intravenous) given at 12 h (after either 18 or 20 mg/kg loading) gives the highest percentages of 10-20 μg/mL serum concentrations. For sub-therapeutic concentrations following intravenous loading (20 mg/kg), a 1st maintenance dose (intravenous) of 10 mg/kg will achieve therapeutic concentrations in 93%. Conclusion: Use of PBPK modelling suggests that children receiving the 20 mg/kg intravenous loading dose are at slightly increased risk of supra-therapeutic blood levels. Ideally, therapeutic drug monitoring is required to monitor serum concentrations, although the dose regime suggested by the BNFc appear appropriate