One of the accepted factors of antiseizure medication resistance is the action of
P-glycoprotein (P-gp), limiting the access of drugs to the nervous system. But if we ask which antiseizure
medications are substrates of P-gp and which are not, the available bibliography will not
allow us to obtain a clear answer. In this review, we focus on clarifying this response. The reviewed
studies have been conducted both in cell lines and in mice that have been administered a P-gp inhibitor,
artificially induced with drug-resistant epilepsy, or had a P-gp gene knockout. A limited number
of studies have been conducted in dogs, primates, brain sections of known epilepsies, or human volunteers,
including pharmacokinetic studies in healthy volunteers and symptomatic response to treatment.
Notably, in human cases, allele variation studies check if having one allele or another of P-gp
varies the pharmacokinetics in question. As we see, the approach to P-gp and antiseizure medication
can be done using very different methods, which undoubtedly complicates the interpretation of the
findings. We cannot be categorical in our results, but we can mention probabilities. Regarding the
weighting of studies, we will consider those conducted in humans as more important, followed by
animal studies, and we will give less weight to studies showing contradictory results compared to the
general bibliographic base. Based on the published bibliography, we propose that, among the anticrisis
medications, the following are likely substrates of P-glycoprotein: Phenytoin, Phenobarbital,
Oxcarbazepine, Lamotrigine, Topiramate, and Lacosamide (less evidence). The following are probably
not substrates: Brivaracetam, Zonisamide, Valproic acid, Perampanel, Gabapentin, and Vigabatrin.
We have not obtained enough information about: Carbamazepine, Eslicarbazepine, Levetiracetam,
Tiagabine, Felbamate, Pregabalin, Rufinamide, Ezogabine, and Retigabine.