Abstracts

Rationale: Ictal electroencephalogram (EEG) of epileptic spasms (ES) is characterized by high-amplitude slow waves with superimposing fast activity. Slow waves are also observed in interictal epileptiform discharges (IIED) as post-spike slow waves. We analyzed the spatial distribution and correlation of fast (>10 Hz) and slow (<5 Hz) EEG components of ictal activities of ES and the IIED recorded by intracranial video EEG (IVEEG) in children with ES. We hypothesized that the mechanism of slow-wave generation in ES is distinct from that of slow-waves in IIED and that spatial distributions of fast and slow components in IIED correlate strongly while those in ES correlate weakly. Methods: We collected children with ES, who underwent IVEEG before resective surgery for epilepsy. We quantified the root-mean square (RMS) amplitude of fast and slow components of ictal activities of ES and IIED. We analyzed the spatial correlation between fast and slow components of ES and IIED using the RMS amplitude. Furthermore, we compared the concordance between spatial distributions of fast and slow components of ES and IIED. Results: Eight patients with ES were included in this study. The age at epilepsy onset ranged from two months to three years (median 5 months). Six patients had early onset of seizures under one year of age. Five out of eight patients had a history of infantile spasms. Four patients had additional seizure types: focal myoclonic seizures in one patient and focal seizures in three patients. The age at IVEEG ranged from three to 13 years (median 4.5 years). All patients were followed up for at least two years (2.2 to 5.7 years). Six underwent resective surgery and two declined it but agreed for brain biopsy. Three of six patients who underwent resective surgery became seizure free. Pathology reports indicated focal cortical dysplasia type I in three patients, hyaline protoplasmic astrocytopathy (filaminopathy) in two, tuberous sclerosis in one, cortical gliosis in one, and cortical gliosis and heterotopic neurons in one patient. Significant spatial correlation between fast and slow EEG components of ES was found in four out of eight children (50.0%), whereas that of IIED was found in all children (100%). There was larger concordance between spatial distributions of fast and slow components in IIED than ES (p = 0.0078). Conclusions: Spatial correlation between fast and slow EEG components was significantly different between IIED and ES. The mechanisms underlying the generation of slow EEG components may differ between ES and IIED. The slow EEG components of ES might indicate extensive epileptic network involving remote symptomatic zones for ES in either cortical or subcortical areas. The high spatial correlation between fast and slow components of IIED suggest an involvement of local inhibitory process within the epileptic cortex.