Atypical benign partial epilepsy of childhood (ABPE = Pseudo-Lennox syndrome) shows semiologic parallels to Lennox-Gastaut syndrome, however--besides the lack of tonic seizures--it has an entirely different etiology and prognosis. Recently Hahn et al [17] investigated the long-term evolution of 43 cases with ABPE. Symptomatology, EEG findings, and course were found to overlap with Rolandic epilepsy, Landau-Kleffner syndrome and ESES. The incidence of seizures in relatives was determined in the whole series investigated by Hahn et al [17]. Five of 56 siblings suffered from seizures (3 Rolandic seizures; one febrile convulsions; one unclassified). Three fathers reported grand mal. In 29 families of the series of Hahn et al EEG recordings were performed: 22 brothers, 19 sisters and 16 pairs of parents. In 29% of the siblings a sharp wave focus was demonstrable. The rate rose to 40% when only siblings investigated at the age of maximum expression (3 to 10 years) were considered. Sharp wave foci were mostly multifocal and indistinguishable from those observed in siblings of children with Rolandic epilepsy. Photoparoxysmal response and generalized spikes and waves during rest and hyperventilation were also found to be significantly elevated (26% and 13% respectively). We conclude that ABPE is a subgroup of idiopathic partial epilepsy of childhood (representing a less benign part of a spectrum) that has to be ranked in a continuum with Rolandic epilepsy. The different clinical phenotype might be caused by a higher expressivity of the identical genetic trait, possibly facilitated by other genetic or acquired factors. Genetic heterogeneity represents another possibility.

Severe myoclonic epilepsy in infancy (SMEI), severe idiopathic generalized epilepsy of infancy (SIGEI) with generalized tonic clonic seizures (GTCS), and myoclonic astatic epilepsy (MAE) may show semiological overlaps. In GEFS+ families, all three phenotypes were found associated with mutations in the SCN1A gene. We analyzed the SCN1A gene in 20 patients with non-familial myoclonic astatic epilepsy -- including 12 probands of the original cohort used by Doose et al. in 1970 to delineate MAE. In addition, 18 patients with sporadic SIGEI -- mostly without myoclonic-astatic seizures -- were analyzed. Novel SCN1A mutations were found in 3 individuals. A frame shift resulting in an early premature stop codon in a now 35-year-old woman with a borderline phenotype of MAE and SIGEI (L433fsX449) was identified. A splice site variant (IVS18 + 5 G --> C) and a missense mutation in the conserved pore region (40736 C --> A; R946 S) were detected each in a child with SIGEI. We conclude that, independent of precise syndromic delineation, myoclonic-astatic seizures are not predictive of SCN1A mutations in sporadic myoclonic epilepsies of infancy and early childhood.

No abstract