Abstracts

Rationale: Pathological HFOs are still difficult to differentiate from physiological HFOs. Recently, we reported that ripples (80-200Hz) exhibiting less suppressive effect of rapid eye movement sleep may provide a specific marker of epileptogenicity (Sakuraba et al., 2015). Here, we investigated the effect of sleep stages on distribution of interictal fast ripples (FRs) (200-500Hz) and correlation to epileptogenic area.Methods: The subjects comprised 9 patients (average age, 30.0 years; range, 14–58 years; 7 males) with drug-resistant epilepsy who underwent extraoperative intracranial electroencephalography (EEG) monitoring by a combination of depth electrodes (median, 4 contacts per patient; range, 0–11 contacts) and subdural electrodes (median, 48 contacts per patient; range, 36–52 contacts). All patients underwent surgical resection and seven achieved freedom from seizures postoperatively. Intracranial EEG signals were sampled and recorded at 2000 Hz simultaneously with scalp EEG and electromyography for sleep staging. The recorded signals were filtered between 200 and 500 Hz, and interictal FRs were automatically detected on 5-min EEG samples derived from different sleep stages. FRs were defined by events above three times the standard deviation of baseline activities and containing at least four consecutive oscillations. The occurrence rate of FRs was compared between REM sleep and NREM sleep. High-rate FR electrode was defined as electrode presenting FRs at higher than 50% of the maximum rate and at least once every minute. The relationship between the area of surgical resection and location of the high-rate FR electrodes were compared between REM and NREM sleep stages.Results: In total, 18,842 and 1,470 FRs were identified during NREM and REM sleep, respectively, from 351 and 167 electrodes of a total of 442 intracranial electrodes across all patients. The occurrence rate of FRs was significantly lower during REM sleep (mean, 0.7/min; range, 0.0–60.6/min) than during NREM sleep (mean, 8.5/min; range, 0.0–297.8/min) (p < 0.0001, Wilcoxon test). In seven patients with postoperative seizure freedom, high-rate FR were observed in 15.9% (20/126) of the electrodes inside resection and 1.9% (4/216) of the electrodes outside the resection during NREM sleep, and 9.5% (12/216) of the electrodes inside resection and none (0/216) of the electrodes outside resection during REM sleep, respectively.Conclusions: Sleep stages influence the occurrence of FRs. Generally, the occurrence rate of FRs was lower during REM sleep than during NREM sleep. High-rate FRs during REM sleep can serve as more specific marker of the epileptogenic zone than those during NREM sleep.