Intracranial EEG seizure onset-patterns correlate with high-frequency oscillations in patients with drug-resistant...

High-frequency oscillations (80-500Hz; HFOs) have been shown to be a specific biomarker of the seizure-onset zone. The relationship of HFOs with seizures having different intracranial electroencephalography (iEEG) morphological onsets, however, has shown significant relationships in experimental animals but has not been studied in humans. We investigated how interictal and ictal HFOs relate to different seizure-onset morphological patterns.

We analyzed the most representative seizure type of 37 patients with drug-resistant focal epilepsy who underwent iEEG for diagnostic evaluation. According to the morphology, 211 seizure-onset zone channels were classified in six patterns (low-voltage fast activity; sharp activity at ≤13Hz; low-frequency high-amplitude periodic spikes; burst of high-amplitude polyspikes; spike-and-wave activity; and delta brush). Interictal and ictal HFOs were compared between the six seizure-onset patterns.

Interictal ripple and fast ripple rates differed significantly across seizure-onset patterns (p<0.001). Significant differences were also found for ictal HFOs density across the different seizure-onset patterns (p<0.001). Sharp activity at ≤13Hz was associated with the lowest interictal HFO rate suggesting either that the mechanism that generates this type of EEG morphology do not generate HFOs or possibly that this pattern is more likely to be generated in a region of seizure spread. Regarding the difference in HFO density between pre-ictal baseline and seizure-onset section across the six patterns, burst of high-amplitude polyspikes and delta brushes had the highest densities of both ripples and fast ripples (p<0.001).

We demonstrated that distinct seizure-onset patterns correlate specific interictal and ictal HFO profiles confirming that seizures with different morphological patterns likely have different mechanisms of generation. This study emphazises that, in clinical practice, seizure-onset patterns should be distinguished and specified when analyzing HFOs, particularly if they are used in presurgical evaluation to better localize the seizure-onset zone.