The utility of the electroencephalogram (EEG) to evaluate clinical situations in epilepsy is reviewed from the Department of Clinical Neurophysiology, King’s College Hospital, London, UK. Based on an analysis of articles available in Medline, Cochrane, and the Internet, and personal experience, the utility of the EEG is not satisfactorily addressed. A referral for an EEG should contain a clear statement of the clinical problem and the reasons for obtaining an EEG. Lack of communication between the clinician and the neurophysiologist may result in abuse of the EEG and minimize its usefulness. The diagnosis of epilepsy is made essentially on clinical evidence, and a referral for a routine EEG for this purpose is usually an abuse. In patients with epilepsy, the first EEG will confirm the diagnosis in 80%. Epileptiform activity in the EEG considerably enhances the likelihood of epilepsy. It may also answer a frequent question "Is it epilepsy?" Video EEG telemetry or ambulatory EEG may be required to distinguish epilepsy vs nonepileptic attack disorder, or absence vs daydreaming. The EEG may be used to distinguish a seizure type or syndrome and its associated clinical prognosis; it can rarely determine etiology. It may provide the first localizing evidence for a lesion, and its utility in presurgical evaluation of epilepsy has not been entirely displaced by imaging techniques. In children, the EEG can help to determine when it is safe or appropriate to discontinue antiepileptic drugs. Interpretation of the EEG requires detailed information regarding the clinical situation, and close liaison between the referring clinician and the EEG department. 
COMMENT. The authors conclude that the EEG has many uses in the evaluation of epilepsy, but attention to detail is essential in the referral request. Our own Epilepsy Center at Children’s Memorial Hospital, Chicago, directed by Drs Nordli, Stack, and Kelly, requires a detailed referral form to be completed, when requesting an EEG. If this practice was more generally adopted, the utility of the EEG in the management of epilepsy would be increased.
EEG in evaluation of the first nonfebrile seizure is addressed in a recent report of the quality standards subcommittee of the American Academy of Neurology, Child Neurology Society, and American Epilepsy Society . (Reprints: QSS, American Academy of Neurology, 1080 Montreal Ave, St Paul, MN 55116). Based on a review of available evidence in the literature, routine EEG was recommended as part of the diagnostic evaluation of the first nonfebrile seizure in children. The authors disagreed with the conclusion of Gilbert and Buncher  that the EEG should not be routinely performed after a first seizure because it does not alter treatment. In contrast, the committee concludes that the EEG helps in determination of seizure type, the diagnosis of epilepsy syndromes, and risk of recurrence; it provides information on long-term prognosis; it influences the decision to perform neuroimaging studies; and it may affect further management. The optimal timing of the EEG is not clear. It is most likely to show abnormalities when obtained within 24 hours of the seizure, although postictal slowing may be transient and must be interpreted with caution.