No abstract

We observed 4 children with hemimegalencephaly, (3 boys, 1 girl aged 3-7 years). One child had a linear sebaceous nevus. All patients had a similar clinical, EEG, and neuroradiologic pattern. All patients had macrocrania, hemiparesis, hemianopsia, and psychomotor retardation of variable degree. All cases had an epileptic syndrome with onset during the first days or the first months of life. The seizures were consistently similar: partial motor seizures, generally hemiclonic and asymmetric brief tonic seizures, in series, involving predominantly one side of the body, contralateral to the cerebral damage. The EEG was initially characterized by a hemihypsarrhythmia and afterward, over the malformed hemisphere, by a rather high-frequency background activity associated with almost continuous transients of spikes, sharp waves, and spike and waves that progressively involved the contralateral hemisphere. Hemispherectomy was performed in two children with sudden and complete recovery from seizures and improvement in psychomotor development. Macroscopic examination was consistent with the computed tomography (CT) scan and magnetic resonance imaging (MRI) showing pachygyria and poorly represented white matter. Histologic examination showed the characteristic absence of cortical neuronal layering, and the presence of giant neurons, neuronal heterotopias, and gliosis. For hemispherectomy to be successful, we believe prolonged EEG monitoring is needed to verify that no seizures are originating in the normal hemisphere.

The clinical and electroencephalographic data of 31 children with cryptogenic West syndrome (WS), selected from a series of 103 WS patients, with a follow-up between 4 and 12 years, were studied retrospectively to verify whether this group included patients who fulfilled the criteria for an idiopathic etiology. The results identified a possible idiopathic etiology in 17 patients (55%), who had a family history of other forms of idiopathic epilepsy or febrile convulsions, or who developed, during the follow-up, an EEG genetic trait such as a photoconvulsive response or spike-and-wave discharges, or rolandic spikes. All 17 children had a favorable outcome and all had normal neuropsychological development. Four children (13%) fulfilled the criteria for a true cryptogenic etiology, a causative lesion being suspected, but never proved. At the end of the follow-up all four had seizures, or developmental delay or both, all signs that suggest an underlying cerebral lesion. The other 10 children, representing 32% of the cryptogenic cases, had a good prognosis, with early disappearance of spasms and hypsarrhythmia, and normal neurological development, but none had an EEG epileptic trait or family history of epilepsy or febrile convulsions; although they could have had an idiopathic WS, this was not proved. We conclude that among the children classified as having a cryptogenic WS, many--in our series at least 55%--fulfill the criteria for an idiopathic etiology.

No abstract