Objective: A population analysis of the kinetics of valproic acid (VPA) in children with epilepsy was performed in order to characterize the covariates which influence VPA clearance (CL).
Methods: A total of 770 steady-state serum concentration samples was analysed. These were collected during VPA therapy from 255 children, aged 0.1-14 years and weighting 4-74 kg. Age, total body weight (TBW), VPA daily dose, sex and comedication with carbamazepine (CBZ) were considered as covariates. Population analysis was made with NONMEM program, assuming a one-compartment model, fixing the VPA absorption rate, bioavailability and distribution volume at values found in the literature. The results of the population pharmacokinetics analysis were validated in a group of 45 epileptic patients.
Results: The final regression model for VPA clearance, that included TBW (kg), daily dose (mg/kg) and CBZ comedication as covariates with a significant influence on this parameter, was as follows: CL (L/h) = 0.012 TBW0.715 DOSE0.306(1.359 CBZ). The coefficient of variation for interpatient variability in CL was 21.4% and the residual variability estimated was 23.9% for a concentration of 65 mg/l. In order to estimate the predictive performance of the selected final model, predictions of the VPA serum concentrations were calculated and compared with VPA measured concentrations in the validation group. This assessment revealed an important improvement in the predictive performance of VPA concentrations in comparison with the basic model that did not include any covariates (root squared mean error: 19.50 vs. 39.73 mg/l).
Conclusion: A population pharmacokinetic model is proposed to estimate the individual CL for paediatric patients receiving VPA in terms of patient's dose, weight and concomitant CBZ, in order to establish a priori dosage regimens.