Abstracts

Rationale: There is abundant information of the origin of sleep architecture (spindles, vertex waves, K-complex and slow waves) from animal data; however there is less data available in humans, and even minimal to no data of similar correlations in children. The purpose of this study is to correlate sleep architecture on scalp EEG (Cz electrode) with different areas of the cortex during intracranial monitoring via strips and grids placed for phase 2 monitoring of intractable epilepsy. Extensive areas of frontal, parietal and temporal regions including mesial regions were sampled. Methods: We analyzed data from four children, ages of 4 to 21 years, 3 males and 1 female, correlating sleep architecture on Cz electrode with areas of the cortex on implanted electrodes showing maximum amplitude. Results: We found that spindles were visualized maximally in the central peri-sulcal region at times arising a few milli-seconds earlier, but also seen in the frontal regions, including the mesial areas; they were not seen in electrodes representing the hippocampus and mesial temporal areas. Interestingly, we found slow waves of sleep in a more extensive distribution across the cortex, more pronounced over the frontal areas. The spindle component of the K-complex was visualized maximally over the central sulcus region, while the slow wave component was seen persistently with higher amplitude in the contacts anteriorly in the frontal cortex. In-spite of postulated acoustic origin, K complexes were not visualized in the temporal cortex. Vertex waves were seen maximally and persistently over the frontal-central regions. Conclusions: Preliminary data from our study shows that there is a good cortical representation of sleep architecture correlating with traditional scalp electrodes. Additional information exploring abnormalities or differences in scalp/depth correlation not only in non-REM but also in REM sleep need to be done.