Abstracts

Rationale: Recently, the utility of “scalp” High Frequency Oscillations (HFOs) has been reported and applied to various fields in epilepsy1). We also reported the usefulness of scalp HFOs as a predictor of prognosis in epilepsy2). Then, we focused on the relation between the extension of scalp HFOs and the control of epilepsy..Methods: Five young patients (mean age: 10.4 years; range: 9-12 years) with focal epilepsy underwent yearly scalp EEG with 1,000 Hz sampling frequency for 2-4 years. HFOs coincident with focal spike were confirmed macroscopically and analyzed by time frequency analysis (10-200 Hz; +/-500ms of the spike; reference period on first 200ms) in “all” electrodes of 10-20 international system. The number of electrodes which showed significantly (p<0.05) increased amplitude over 80 Hz was counted in each EEG. The counting was done without knowing patients’ clinical information and then compared with epilepsy control.Results: A total of 12 EEGs were analyzed and the number of electrodes with HFOs over 80 Hz decreased with the age in all patients. The extension of such electrodes was broader, including contralateral hemisphere of the focal spike, at first EEG but settled into electrodes around the spike. In general, the number decreased from teens to zero and the attrition rate was accorded with control of seizures. For example, a 12-year-old patient, who developed first nocturnal seizure at 4 years old with Rolandic discharges, repeated seizures monthly and required four antiepileptic drugs but his last seizure at 9 years old was between 17 electrodes (Figure 1) at 8 years old and 0 electrode (Figure 2) at 10 years old.Conclusions: The number of HFOs may reflect epileptogenicity at the time of recording EEG and could be the indicator for treatment of epilepsy. Kobayashi et al. reported1) that spike-related HFOs in idiopathic partial epilepsy of childhood contain more information about epileptogenicity than the spikes themselves. In this study, the number of electrodes and their distribution includes far more information than spike-related HFOs in one electrode. The number and distribution of electrodes with significant HFOs is a candidate for “new biomarker” of epilepsy treatment besides conventional EEG. Whereas the treatment of Rolandic epilepsy tends to be excessive due to continuance of Rolandic discharges, the number of significant HFOs could suggest the timing of treatment discontinuation. Longer follow-up and comparison with “intracranial” EEG data are needed for further analysis, including the strategy of proper epilepsy treatments and the veracity of “scalp” HFOs. References 1. Kobayashi K, et al. High-frequency oscillations in idiopathic partial epilepsy of childhood. Epilepsia 2011 ;52: 1812-1819. 2. Nagasawa T, et al. Predictive indicator for prognosis of childhood epilepsy with Rolandic discharges using HFOs. Journal of the Japan Epilepsy Society 2014 ;32: 411.