Abstracts

Rationale: Despite routine use of eletrocorticography (ECOG) to identify abnormal electrical activity associated with ictal onset, the electrocortical distribution of sleep architecture has not been sytematically described in humans. ECOG provides a means to study the cortical appearance of well described scalp potentials characteristic of sleep as recorded by electroencephalography (EEG). An understanding of the normal topographical appearance of such electrical potentials should facilitate the accurate identification of abnormal cortical electrical activity relating to pathology.Methods: A 60 year old female with refractory epilepsy of suspected right temporal lobe origin was admitted for surgical placement of subdural electrodes prior to epilepsy surgery. 96 grid and strip electrodes were placed, including a 32-contact lateral temporal grid, a 16-contact lateral frontal grid, a 16-contact occipital grid, a 4-contact orbitofrontal strip, a 6-contact temporal polar strip, a 6-contact anterior subtemporal strip, a 6-contact medial subtemporal strip and an 8-contact posterior subtemporal strip (Figure 1). CZ, C3, C4, O1, O2, A1 and A2 scalp electrodes, outer canthus electrodes and submental electromyography electrodes were added to identify stages I, II, III, IV, and rapid eye movement (REM) sleep. We created a referential montage with each channel referenced to electrode A1 and examined the EEG and ECOG recorded during sleep. We identified 18 vertex waves, 25 K complexes and 61 sleep spindles in the EEG record and studied the coincident ECOG record to determine if similar waveforms were present or absent in the channel corresponding to each subdural recording electrode.Results: We identified vertex sharp waves in the superior anterior region of the frontal grid and to a lesser extent in the posterior inferior corner of the frontal grid and along the superior margin of the temporal grid. We identified sleep spindles throughout the frontal grid, in the anterior superior corner of the temporal grid, the posterior superior region of the occipital grid, the mesial half of the posterior subtemporal strip and the orbital frontal strip. We identified K complexes in the anterior superior region of the frontal grid, the superior posterior corner of the temporal grid and the anterior superior portion of the occipital grid (Figure 2). Interictal epileptiform spikes were confined to the temporal polar strip, the anterior and middle subtemporal strips and the inferior half of the lateral temporal grid.Conclusions: ECOG as part of the standard presurgical evaluation of epilepsy patients provides a means of coincident characterization of cortical potentials during sleep. The cortical distribution of ECOG potentials bearing similar morphology to coincident scalp recorded vertex sharp waves, K complexes and sleep spindles appears to display regional specificity and a definable topography. Further investigation of the topography of normal cortical electrical potentials should enrich our knowledge of neurophysiology and improve our ability to characterize the location and extent of pathological cortical rhythms.