Diagnostic difficulties and treatment implications.

Abstract

Differentiation between types of epileptic seizures has been aided in recent years by the introduction of intensive neurodiagnostic techniques and the development of increasingly detailed classification systems. Paradoxically, these developments have not simplified the task of matching the appropriate antiepileptic drug to a particular seizure type. It is reasonable to assume that anticonvulsant drugs will have different effects on different types of seizures, but faulty, circular reasoning can enter the picture if one also assumes that responses of seizures to different drugs signify different seizure types. There are several examples of differential diagnoses that can fall prey to this problem, including the diagnosis between partial seizures with secondary generalization and generalized tonic-clonic seizures, and the diagnosis between complex partial seizures and absence seizures with automatisms, among others. Considerations of etiology in future classification systems can further complicate the problem: should one then choose an anticonvulsant drug on the basis of individual seizure type or on the basis of the type of epilepsy? Ramifications of this issue extend even to the drug approval process. Official sanction is not given for use of a drug for a seizure type not included in the original efficacy studies, even if later scientific evidence shows that seizure type to be related to a type that is included. New trials must be undertaken. These problems arise from how we choose to classify seizures.