AimsVigabatrin is an antiepileptic drug used to treat some forms of severe epilepsy in children. The main adverse effect is ocular toxicity, which is related to the cumulative dose. The aim of the study is to identify an acceptable exposure range, both through the development of a population pharmacokinetic model of vigabatrin in children enabling us to calculate patient exposure and through the study of therapeutic response.MethodsWe performed a retrospective study including children with epilepsy followed at Necker‐Enfants Malades hospital who had a vigabatrin assay between January 2019 and January 2022. The population pharmacokinetic study was performed on Monolix2021 using a nonlinear mixed‐effects modelling approach. Children treated for epileptic spasms were classified into responder and nonresponder groups according to whether the spasms resolved, in order to identify an effective plasma exposure range.ResultsWe included 79 patients and analysed 159 samples. The median age was 4.2 years (range 0.3–18). A 2‐compartment model with allometry and creatinine clearance on clearance best fit our data. Exposure analysis was performed on 61 patients with epileptic spasms. Of the 22 patients who responded (36%), 95% had an AUC0–24 between 264 and 549 mg.h.L−1.ConclusionsThe population pharmacokinetic model allowed us to identify bodyweight and creatinine clearance as the 2 main factors explaining the observed interindividual variability of vigabatrin. An acceptable exposure range was defined in this study. A target concentration intervention approach using this pharmacokinetic model could be used to avoid overexposure in responder patients.