Abstracts

Rationale:
Mesial temporal lobe epilepsy (MTLE) is often intractable and is one of the most common forms of pharmacologically resistant epilepsy. Scalp EEG is limited at detecting interictal and ictal abnormalities within deeper brain regions. Intracranial recordings from foramen ovale (FO) electrodes can reveal abnormalities not seen from the scalp and offer an opportunity to understand the underlying epileptic interictal and ictal networks. We recently discovered highly reproducible interictal spike networks from neocortical epilepsy patients with subdural grid electrodes using directed direct transfer function (dDTF).  Here, we applied this approach to patients with FO and scalp EEG to develop a better understanding of the interictal spike networks in temporal lobe epilepsy to improve surgical and non-surgical treatment approaches.
Method:
We collected 6 ten-minute segments, 3 awake and 3 asleep, of simultaneous interictal FO and scalp EEG from 10 consecutive epilepsy patients suspected for having MTLE.  Interictal spikes were marked using an established algorithm, time-segments were isolated where spikes appeared synchronous, and dDTF was used to measure interictal spike propagation network. FO spike networks were aligned with lesion location, scalp EEG, and seizure onset patterns.
Results:
Each patient had an entirely unique spike propagation pattern that was highly reproducible between awake and asleep time segments and occurred at frequency bands ranging from delta to gamma oscillations. Interictal spike propagations traveled both anterior and posterior along each mesial temporal lobe as well as between temporal lobes.  While for lesional patients most spike propagations remained ipsilateral, non-lesional patients had more interhemispheric temporal lobe spike propagations that occurred mostly between posterior temporal lobe regions.  Very few (≤ 5%) mesial temporal spikes were associated with scalp detected spikes, but were not time-locked to suggest a directionality of spread. Interictal spike networks surrounded the margins of tumor and sclerotic brain regions that matched seizure onset zones.
Conclusion:
Interictal spike propagation networks are ‘hard-wired’ like what we have described for neocortical spike networks.  Mesial temporal spikes travel both within and between mesial temporal lobe structures and less than 5% are detected by scalp EEG.  Spike networks are spatially organized around lesions where they are mostly unilateral and closely linked to seizure onset, perhaps explaining why lesional patients have better surgical outcomes than non-lesional patients.   Expanding our understanding of interictal spike networks with larger scale studies could provide more reliable identification of seizure onset zones that could lead to improved surgical and non-surgical outcomes for temporal lobe epilepsy.
Funding:
:This work was supported by NIH NS109515 and NS083527 (JAL) and Biswajit Maharathi received Pre-doctoral Education for Clinical and Translational Scientists (PECTS) Fellowship (NIH UL1TR002003).