Abstracts

Rationale: High frequency oscillations (HFOs) in the range of 40 to 250 Hz (gamma and ripples) have recently been described on the scalp EEG of patients affected by focal and generalized epilepsy. In focal epilepsy, HFOs were shown to be more specific to the seizure-onset zone (SOZ) than spikes, suggesting they can be used as interictal EEG marker of the epileptogenic areas. Fast ripples (FRs) are more specific and accurate than ripples to identify the SOZ in intracerebral EEG. To our knowledge, no studies have recorded FRs on the scalp EEG of epileptic patients. The aim of this study was to evaluate if FRs can be detected on the scalp EEG of adult patients with focal epilepsy and to assess their relationship with the SOZ. Methods: Patients who had at least 10 unambiguous spikes on the subdermal EEG during the first consecutive 30 minutes of N3 sleep were included. Recordings were selected from 55 patients with intractable focal epilepsy who underwent combined subdermal - intracranial EEG (500Hz filter and 2000Hz sampling) from January 2010 to May 2014 at the Montreal Neurological Institute for presurgical evaluation. Subdermal electrodes were placed at positions Fz, Cz, Pz, F3, F4, C3, C4, P3, P4. Sleep was manually scored in the subdermal EEG according to AASM 2.0. FRs were visually marked, using a bipolar montage, in the first 30 minutes of N3 sleep. FRs were defined as events containing at least 4 consecutive sinusoidal oscillations and amplitude clearly greater than the background and with a frequency > 250 Hz. We excluded the events that have morphological features resembling artifacts and the oscillations associated with artifacts identifiable in the original unfiltered record (also reviewed in expanded time scale). After marking all the events, the EEGs were re-analyzed by the same reviewer together with two HFO experts and only events for which there was consensus were retained. The SOZ was defined using depth recordings during the presurgical investigation, independently of this study. Events recorded in subdermal electrodes were categorized as related to the SOZ if localized in the same brain lobe, and not related to the SOZ, if detected in other lobes. Results: Ten patients fulfilled our criteria. FRs, of generally low amplitude, were found in 60% (6/10) of patients. The mean rate of FRs was 0.19/minute. Considering the 6 patients with FRs, 2 exhibited events related to the SOZ (patients 5 and 7) and 3 showed no relationship between FRs and SOZ (patients 3, 6 and10). In the remaining patient (patient 8) the SOZ could not be identified (Table 1). Conclusions: FRs can be detected with surface EEG using subdermal electrodes in patients with focal epilepsy. FRs were not present in all patients analyzed, and when present, occurred at a very low rate. The distinction between FRs and artifacts is challenging. The relationship between surface FRs and SOZ remains unclear. Further studies, with a higher spatial sampling on the surface EEG, are necessary to clarify this issue.