HFOs (80–450Hz) recorded with standard intracranial electrodes reflect epileptogenicity in children with focal cortical dysplasia

Introduction: High frequency oscillations (HFO) ranging between 80 and 500Hz recorded with intracranial electrodes have been linked to epileptogenic regions in patients with epilepsy. The removal of HFO-generating areas correlates with a good postsurgical outcome. Most studies have been performed with microelectrodes or custom made macroelectrodes. Data on children and different pathologies is limited. We here investigate HFOs in children with Focal Cortical Dysplasia (FCD) during a routine investigation with commercially available grid electrodes.

Methods: Patients under the age of 18 recorded in our centre between 2005 and 2009 and with pathologically confirmed FCD were included in this study. EEG was filtered at 450Hz and sampled at 1024Hz. HFOs (Ripples 80–200Hz, Fast Ripples 200–450Hz) were visually identified on bipolar montage in a five-minute segment of slow wave sleep using high-pass filters at 80Hz and 200Hz. Rates of HFOs in areas inside and outside the seizure onset zone (SOZ) were calculated for each channel and compared using ANOVA (p<0.05).

Results: Six children were included (mean age 13.6 years), four had FCD type 2a and two type 1b (Palmini classification). Rates of ripples (19.1/min) and fast ripples (2.5/min) were similar to those described before. They were significantly higher inside the SOZ (R: 29.1/min; FR: 4.7/min) than outside (R: 15.8/min, FR: 1.8/min, p<0.001). Two patterns of HFO occurrence were seen: in pattern 1, channels had a flat baseline without high frequency activity and intermittent very high voltage HFOs. Pattern 2 was defined as channels with a continuously oscillating baseline with intermittently occurring higher voltage HFOs. Of the SOZ channels 71.1% showed pattern 1 and 17.1% showed pattern 2. Of non-SOZ channels, 20.5% showed pattern 1 and 17.5% pattern 2. The remaining channels showed few or no HFO activity; nevertheless HFOs were generated in a large number of channels.

Conclusion: Children with FCDs showed a large extent of areas generating HFOs. Rates of HFOs could distinguish between channels inside and outside the SOZ. Therefore, HFOs recorded during only 5 minutes in a routine clinical setting in children may help to further delineate epileptogenic areas.

Two patterns of HFOs were distinguished, of which pattern 1 consisting of single high voltage HFOs may show a closer link to the SOZ. Whether these neocortical HFO patterns can be linked to FCD or its subtypes will have to be evaluated in larger studies.