Abstracts

Rationale: Successful surgical treatment of intractable focal epilepsy largely depends on the precise identification and removal of the epileptogenic zone (EZ). Stereo-electro-encephalography (SEEG) is one of the invasive methods that aims to delineate EZ. Fast activity is considered as one of the most important biomarkers of the EZ in SEEG recording (de Curtis et al., 2009), but it is also often seen in the regions of propagation. Recently, several studies aimed to find more specific biomarkers of the EZ. Our previous study showed that a specific time-frequency pattern in the epileptogenic zone with a combination of (i) sharp transients or spikes; preceding (ii) multiband fast activity concurrent; with (iii) suppression of lower frequencies is a good marker of the EZ (Grinenko et al., 2018). On the other hand, some studies with epilepsy patients suggested that ictal slow shift (infraslow or direct current shift) appeared in epileptic foci during an ictal period with invasive recordings (Ikeda et al., 1996, Thompson et al., 2016). Based on these facts, slow shifts can potentially complement fast activity in EZ detection. The aim of this study is to reveal the characteristic of slow shifts in the EZ in SEEG recordings and to clarify the relationship of slow shift and fact activities. Methods: Seven patients with intractable focal epilepsy who underwent SEEG evaluation at the Epilepsy Centre of Cleveland Clinic in 2015 with good seizure outcome and had seizure onset initiated with low voltage fast activity which sustained 3 seconds or longer were included. To evaluate the combination of fast activities and slow shifts in the EZ, we selected only the contacts localized in the EZ as it was determined based on our pervious study. These contacts expressed 'fingerprint pattern' characterized by multiband fast activity associated with suppression of low frequencies and preceded by spikes. Visually we analyzed the occurrence ratio of the ictal slow shifts in the EZ, the time of onset of ictal slow shift in relation to the onset of fast activity. Results: We analyzed five patients, because two patients had no contacts with 'fingerprint pattern'. The ictal slow shifts were observed in all patients, in total, in 97.1% of the contacts with the fingerprint pattern. In Patient 1, 4 and 5, the onset time of ictal slow shifts preceded the onset of fast activity and in Patient 2 and 3, it observed simultaneously with fast activity. The complexity of ictal slow shifts differed across patients, and across the lobes of contacts in Patient 4. Conclusions: Ictal slow shift at the seizure onset consistently observed in the EZ and co-occur with the fingerprint pattern. It can precede fast activity or can appear simultaneously with fast activity. The complexity of ictal slow shifts could differ across patients and lobes. Funding: No funding