Abstracts

Rationale: We previously described a model of infantile spasms based upon local neocortical infusion of tetrodotoxin (TTX) in infant rats. Many of these rats develop spasms which closely resemble those seen in the human disorder. The EEG findings are virtually identical to those seen in children, and during a spasm, there is typically a generalized high voltage slow wave, followed by a period of voltage attenuation commonly referred to as an electrodecremental response. As an initial step towards understanding the underlying mechanisms leading to seizure development, we have undertaken an analysis of the high frequency (HF) components associated with the ictal events. Methods: Infusion of TTX was started on postnatal day 11-12 and continued for 4 weeks. After the infusion period EEG monitoring sessions were conducted over the next several weeks using a digital sampling rate of 2048 Hz. HF EEG activity (30-500 Hz) associated with ictal events was quantified using 1) compressed spectral (FFT) arrays with a 100 msec sliding analysis window, and 2) narrow bandpass digital filtering. Analysis software was developed using the Matlab programming language. Results: Multiple seizures associated with spasms have been analyzed in 3 rats. In all cases the initial ictal event (corresponding to the onset of the slow-wave complex seen in conventional EEG) has been characterized by the simultaneous occurrence of multiple bands (typically 12-20) of discrete HF components ranging from 30 Hz to 500 Hz (and occasionally higher) with typical durations of 100-200 msecs. Narrow bandpass filtering at these specific frequencies reveals bursts of rhythmic activity during these times. During the remainder of the ictal event, which may last for several seconds, intermittent bursts of HF activity typically continue to occur, but with a gradual decrease of the highest frequency components until near the end of the seizure when higher frequencies may again increase in association with spike/sharp wave complexes. There is considerable variability in the pattern of HF components from seizure to seizure at a specific cortical site even in the same animal. There is also variability in the precise onset times of the various frequency components from time to time in different cortical locations. Conclusions: Very high frequency EEG components (30-500 Hz) accompany ictal events in the TTX model of infantile spasms and occur throughout the electrographic seizure, including periods of time when electrodecremental responses are observed in more conventional EEG recordings. The variability of the frequency pattern seen both from time to time at the same cortical site and between cortical sites, suggests that the ictal events may be triggered by projected, rather than primary cortical, mechanisms. An understanding of the neurobiological origins of this high frequency activity and it role in the pathophysiology of infantile spasms awaits further studies.