Spatial distribution of intracranially recorded spikes in medial and lateral temporal epilepsies *Irina I. Goncharova, *Hitten P. Zaveri, *yRobert B. Duckrow, *yzEdward J. Novotny, and *ySusan S. Spencer *Departments of Neurology, yNeurosurgery, and zPediatrics, Yale University School of Medicine, New Haven, Connecticut, U.S.A. SUMMARY Purpose: Although seizures and interictal spikes are not always colocalized, there may be valuable localizing information in the spatial distribution of spikes. To test this hypothesis, we studied the spa- tial distribution of intracranially recorded interictal spikes in patients with medial temporal (MT) and lateral temporal (LT) neocortical seizure onset. Methods: A total of 21 patients (MT n = 12, LT n = 9) who completed intracranial monitoring were selected for this study. Two 4-h intracranial electroencephalography (icEEG) epochs were analyzed, one during wake and one during sleep, both at least 6 h removed from seizures. Spikes detected automatically in medial temporal struc- tures [hippocampal formation (H) and entorhinal cortex (EC)], and in five cortical areas (occipital, frontal, parietal, lateral temporal, and inferior temporal) were tabulated. Results: Interictal spikes occurred broadly over medial temporal structures and cortical areas in MT and LT patients. The patients differed with a greater number of spikes in medial temporal structures in the MT group (p < 0.05 for H and p < 0.001 for EC) and a greater number of spikes in parietal (p < 0.01) and frontal (p < 0.001) areas in the LT group. There were sleep-related increases in spike rates in inferomedial temporal structures in both groups. The two groups could be separated with a classifier based on medial temporal and parietal and frontal spikes (p < 0.0001). Discussion: MT and LT patients have different spa- tial distributions of interictal spikes and can be dis- tinguished by the relative spike rates in medial temporal and extratemporal areas during sleep and wake. KEY WORDS: Intracranial EEG, Interictal spikes, Temporal lobe epilepsy, Sleep, Wake. Interictal spikes recorded on electroencephalography (EEG) are known to be highly correlated with the presence of epilepsy (Carren ˇo & Lüders, 2001). However, many epilepsy centers do not consider their localization in eval- uation for surgical resection. Interictal spikes recorded in the individual patient have considerable variability in their spatial and temporal characteristics. Furthermore, the spa- tial relationship between the location of interictal spikes and the seizure onset zone, and the temporal relationship between the occurrence of spikes and the occurrence of seizures, are poorly understood. Previous studies of spikes conducted with scalp EEG focused on idiosyncratically selected individual spike occurrences and attempted to discover their localization or origin with methods such as source localization (see Tao et al., 2007 for a review). More recently, spike-triggered functional magnetic reso- nance imaging (fMRI) has been used to determine the areas involved in spike generation by measurement of fMRI activation synchronized to occurrence of spikes (Krakow et al., 2001; Al-Asmi et al., 2003). Intracranial EEG offers an opportunity to understand spikes recorded closer to the source of their generation. Studies of spikes recorded on intracranial EEG, however, have been scarce. Challenges to the study of spikes with intracranial EEG are both the limited spatial sampling provided by intracranial electrodes, and the considerable number of spikes that can be observed with intracranial electrodes. If salient and robust characteristics of spikes that relate to the seizure onset location are uncovered with the study of intracranial EEG, these findings could present Accepted June 5, 2009; Early View publication August 8, 2009. Address correspondence to Dr. Irina I. Goncharova, Yale University School of Medicine, PO Box 208018, New Haven, CT 06520-8018, U.S.A. E-mail: irina.goncharova@yale.edu Wiley Periodicals, Inc. ª 2009 International League Against Epilepsy Epilepsia, 50(12):2575–2585, 2009 doi: 10.1111/j.1528-1167.2009.02258.x FULL-LENGTH ORIGINAL RESEARCH 2575