Population Pharmacokinetics and Dose Simulation of Levetiracetam in Patients with Epilepsy: Influence of Renal Function and Phenytoin Co‐Administration

Levetiracetam (LEV) exhibits considerable interindividual pharmacokinetic variability; however, data in Thai populations remain limited. This study aimed to develop a population pharmacokinetic model of LEV, identify covariates affecting clearance, and inform individualized dosing strategies. A total of 374 LEV plasma concentrations from 264 Thai patients with epilepsy were analyzed using nonlinear mixed-effects modeling. The median age was 65.3 years (range 18.09-97.71). LEV pharmacokinetics was characterized by a one-compartment model with first-order elimination and proportional residual variability. The typical apparent clearance and volume of distribution were estimated at 2.61 L/h and 56.3 L, respectively. Clearance was significantly influenced by creatinine clearance (CrCL), estimated using the Cockcroft-Gault equation with adjusted body weight, and by phenytoin co-administration. Phenytoin increased clearance by 43.1%, while reduced CrCL was associated with lower clearance. Monte Carlo simulations demonstrated that both trough concentrations and the probability of achieving therapeutic targets (12-46 mg/L) declined with increasing CrCL, especially in patients receiving phenytoin. Standard 12-h dosing was generally sufficient for patients with impaired renal function who were not receiving phenytoin. In contrast, subtherapeutic exposure was more frequent in those with preserved or augmented renal function, particularly when phenytoin was co-administered. These findings support the use of higher or more frequent dosing in these subgroups. The final model provides a framework for individualized LEV dosing. Prospective studies are needed to validate these findings and guide their implementation in clinical practice.