Abstracts

There are few data on the incidence of neuroimaging abnormalities in children who present with a first unprovoked seizure. In addition the relationship of mesial temporal sclerosis and prior seizure history in children who have only had a few seizures is not well understood. We performed MRIs on a cohort of children followed from the time of their first unprovoked seizure.
In a prospective study, 407 children with a first unprovoked seizure were identified and followed for a mean of 12.8 years. We performed MRIs with thin cuts of the temporal lobes 10 or more years after the inital seizure. To date 173 scans have been obtained and read. Mean age at first seizure was 7.2 years and mean age at time of MRI was 19.6 years. All imaging studies were reviewed by two neuroradiologists blinded to the clinical details and outcomes. Interrater reliability was assessed.
A total of 56 (33%) of MRIs were abnormal, including 7 scans with incidental abnormalities such as Chiari I. If we limit the analysis to the 159 children with a normal neurological examination, then 45 (28%) of MRIs were abnormal, including 7 with incidental abnormalities. The most common abnormalities were hippocampal atrophy (n=28), focal atrophy (n=22), and disorders of cerebral migration (n=5). MRIs were somewhat more likely to be abnormal in children with partial seizures (27 of 67, 40%) than in children with generalized seizures (29 of 106, 27%) (p=0.210) and in children with recurrent seizures (32 of 82, 39%) than in those who who only experienced a single seizure (24 of 91, 26%) (p=0.210). However, the differences were not statistically significant.
Hippocampal atrophy with or without abnormal T2 signal was present on 28 (16%) scans including 7 who met radiologic criteria for mesial temporal sclerosis (atrophy and abnormal T2 signal). Of these 28 cases, hippocampal atrophy was an isolated abnormality in 18 cases including 3 with radiologic mesial temporal sclerosis and was associated with other abnormalities such as heterotopias or focal atrophy of the temporal lobe in 10 cases. The proportion of children with status epilepticus, or recurrent seizures was not higher in the hippocampal atrophy group (n=28) nor was the mean seizure duration. Of the 18 children with isolated hippocampal atrophy, only 4 ever had an episode of status epilepticus. Ten (56%) only had a single seizure which was not status epilepticus.
MRI abnormalites are common even in those who have only had isolated seizures. The most common abnoramlities are hippocampal and may be missed on routine imaging. There is a substantial incidence of hippocampal atrophy and even of mesial temporal sclerosis which does not appear to be related to the occurrence of either prolonged or frequent seizures in this population. Further studies correlating MRI and seizure characteristics and outcomes are in progress.
[Supported by: NIH grant NS 26151 from NINDS]