Neuroimaging abnormalities in children with an apparent first unprovoked seizure

Objective: To determine the incidence and type of neuroimaging abnormalitie s: in children presenting with a first seizure. Methods: In a prospective o bservational study, 411 children with a first afebrile seizure were seen be tween 1983 and 1992. Imaging studies were performed in 218 (53%). For this analysis we examined the most sensitive neuroimaging study performed which included 159 computed tomography scans and 59 magnetic resonance imagings ( MRI). Results: Four children were found to have lesions requiring intervent ion (blain tumor in two, neurocysticercosis in two). The remaining 407 were enrolled in a follow-up study of children with a first unprovoked seizure. After a mean follow-up of > 10 years, none have developed clinical evidenc e of a tumor. In these 411 children, 45 (21%) of 218 imaging studies were a bnormal. The most common abnormalities were focal encephalomalacia (n = 16) and cerebral dysgenesis (n = 11). Although children with partial seizures were more likely to be imaged (64%) than children with generalized seizures (43%) (P < 0.001), the fraction of abnormal imaging studies was similar in both groups. Six children with a normal neurological examination who were initially classified as cryptogenic were subsequently found to have errors of cerebral migration on MRI. The incidence of lesions requiring acute inte rvention in children presenting with a first seizure is low. A significant proportion will have neuroimaging abnormalities particularly on MRI. Conclu sions: Neuroimaging should be considered in any child with a first seizure who does not have an idiopathic form of epilepsy. (C) 2001 Elsevier Science B.V. All rights reserved.