Approximately 20 to 30% of patients with newly diagnosed epilepsy do not have their seizures controlled with currently available antiepileptic drugs. The clinical need for new antiepileptic drugs is therefore clear.In recent years, as our understanding of the molecular basis of epilepsy has unfolded, several novel candidate antiepileptic drugs have become available for clinical evaluation. The major emphasis has been on the development of more potent and effective antiepileptic drugs, and also drugs with fewer adverse effects than existing therapies. This has resulted in 7 new drugs being licenced around the world in the last 5 years (felbamate, gabapentin, lamotrigine, oxcarbazepine, piracetam, vigabatrin and zonisamide). In addition, 7 other promising drugs are in various stages of development [eterobarb, fosphenytoin, levetiracetam (ubc L059), remacemide, stiripentol, tiagabine and topiramate].Numerous advantages over existing antiepileptic drugs can be identified for some of these new drugs. A mechanism of action has been determined for lamotrigine, tiagabine and vigabatrin. This may prove particularly useful therapeutically since it allows a more rational treatment strategy. Eterobarb, fosphenytoin, oxcarbazepine and remacemide are prodrugs. This is a particular advantage for fosphenytoin, which is metabolised to phenytoin. Gabapentin, piracetam and topiramate are not metabolised and vigabatrin is minimally metabolised. These drugs do not exhibit significant binding to blood proteins. Therefore, these drugs are not susceptible to significant pharmacokinetic drug interactions. Oxcarbazepine also exhibits minimal drug interactions. This is in contrast to felbamate, lamotrigine and stiripentol, drugs with which pharmacokinetic interactions can be clinically problematic.All drugs, with the exception of piracetam, are effective treatments for partial or secondarily generalised seizures. Piracetam and zonisamide are effective in myoclonus, and felbamate has been licenced for use in children with Lennox-Gastaut syndrome. All 14 drugs have the potential to induce adverse effects, mostly CNS-related. Whilst treatment recommendations can be made for some of the drugs, these cannot be considered definitive since they are based largely on data from controlled clinical studies in highly selected patients. Further treatment recommendations for different seizure types and epilepsy syndromes will inevitably develop as clinical experience with the drugs increases.