Effect of treatment and additional disease on pharmacokinetic of valproic acid in children with epilepsy.

BACKGROUND Most of bayesian pharmacokinetic studies and the influence of clinical variables have been carried out in adults. PURPOSE The aim was to estimate population-based pharmacokinetic of valproic acid (VPA) and to determine the effect of treatment and additional disease on its performance in children with epilepsy. MATERIAL AND METHODS For the study steady-state serum concentrations of VPA were determined from 108 epileptic patients (44 females and 64 males) who were receiving the anticonvulsant as main drug of treatment with age range since 1 to 16 years (median 4y, 6m) and weight since 5.2 to 50 kg (median 17.5 kg). All patients had their renal, hepatic and nutritional functions normal. One compartment model using interactive two-stage Bayesian approach was employed in the analysis. RESULTS; Population estimates of CL/F and V/F for VPA were 0.022 +/- 0.013 L/h and 0.217 +/- 0.134 L/kg, respectively. These estimates were significantly affected by weight, age, carbamazepine (CBZ) and gastroesophageal reflux (GER). The final regression models were: CL/F (L/h) = 0.0696 + 0.0031 (Age) + 0.0075 (Weight); and V/F (L) = 0.674 + 0.0308 (Age) + 0.0756 (Weight). Prediction of VPA serum concentration in other validation group revealed an important improvement in the predictive performance of VPA concentrations in comparison with the basic model that did not include any co-variables. CONCLUSIONS Based on a population model of children with epilepsy, the pharmacokinetic of VPA could be altered by weight, age and the administration of CBZ and additional GER to epilepsy.