Summary:Purpose: To assess distribution of temporal lobe spikes across different states of sleep and wakefulness in simultaneous scalp and foramen ovale (Fo) recordings.Methods: The study included 12 patients with mesial temporal lobe epilepsy (MTLE). As part of their presurgical evaluation, patients underwent long-term video-EEG monitoring with combined scalp and foramen ovale electrodes (FoEs). In addition to traditional sleep scoring, waking was subdivided into eyes-opened and eyes-closed states, and rapid-eyemovement (REM) sleep was divided into phasic and tonic states. Spike counts were carried out visually for scalp and FoEs, and spiking rates were determined for each state. A ratio between FoE and scalp spiking rates also was calculated for each state.Results: Scalp spiking showed a significant increase during NREM3,4, whereas FoE spiking increased during NREM2. The scalp/FoE ratio significantly increased during NREM3,4. A significant difference in spiking rate also was found between phasic and tonic REM states as well as between waking with eyes opened and closed in FoE recordings.Conclusions: Our data provide evidence of a discrepancy in spike distribution across different states of sleep and waking monitored by scalp and FoE recordings. We suggest that these discrepancies may reflect differences in archicortical and neocortical spike synchronization. Key Words: Mesial temporal lobe epilepsy-Sleep-Interictal spike-Foramen ovaleHippocampus.Sleep-related modulation of epileptic activity is a well-known common feature of different epilepsy syndromes. Non-rapid-eye-movement (NREM) sleep has been shown by most authors to activate interictal epileptic activity in both partial (1) and primary generalized epilepsy syndromes (2), whereas REM and waking has a contrary effect. Considerable controversy surrounds the issue of whether light NREM or deep NREM sleep exerts a stronger facilitatory effect on spiking. There is accumulating evidence that this may depend on the type of epilepsy syndrome being studied. According to recent spectral analysis results, a range of idiopathic childhood epileptic syndromes (3-8) can be characterized by increased spiking in the sigma range related to spindle synchronization present mainly during NREM2. Adult partial epilepsies, conversely, are accompanied by increased spiking related to the delta frequency range (9,10). The same tendency in spike activation was reported earlier by several visually scored studies (11-13) finding highest levels of spike activation during deep NREM stages in most temporal lobe epilepsy (TLE) patients. Nevertheless, both visually scored and spectralanalysis studies revealed small groups of patients deviating from this trend, exhibiting maximal spiking rates during light NREM, REM, or waking.A second point of controversy concerns the spiking properties of mesiotemporal structures. There are reports describing maximal spike activation mainly during deep NREM stages (11,12), whereas other authors described increased spiking mainly during light NREM stages ...