Abstracts

Rationale: The existence of biorhythms and different states of vigilance of the human brain is well known for decades. In epilepsy, the circadian and multidien rhythms are known to modulate seizure risk and seizure recurrence. Invasive intracranial EEG (iEEG) studies in drug-resistant epilepsy patients revealed the dependence of interictal (seizure independent) markers of epilepsy, such as interictal spike discharges (IEDs) and high-frequency oscillations, on circadian rhythms and anatomical location. In this study, we analyzed changes in IEDs as a gold standard of epileptiform activity in contrast to relative entropy (REN) as an estimate of brain connectivity. Both features were analyzed in non-pathogenic tissue only. We hypothesized that epileptiform activity and electrophysiological biomarkers of connectivity have different rhythms, modulated by time of the day and anatomical location.

Methods: We analyzed 8 clinical stereotactic EEG recordings of patients with pharmacoresistant focal epilepsy who underwent monitoring as part of their evaluation for subsequent epilepsy surgery. The length of the recordings varied between 3-10 days with a sampling frequency of 5 kHz. The total time of the analyzed iEEG recordings was 65 days. All patients had excellent post-surgical outcome with Engel IA and at least one year of follow-up. The channels marked as seizure onset zone or located in the resected area of the brain were discarded from the analysis. IEDs were detected by an automated detector, and connectivity of the brain was estimated by the REN. IED counts and mean REN were computed for 30-minute bins across the whole recordings. The bin values were z-scored across subjects and anatomical structures. Only structures implanted in at least 6 patients were taken into account. The differences between day (6 AM - 10 PM) and night (10 PM - 6 AM) in separate anatomical structures were evaluated using the Mann-Whitney rank-sum test (alpha = 0.05) with Bonferroni correction to account for multiple comparisons.

Results: The anatomical structures fulfilling the criteria were the hippocampus and orbitofrontal, inferior frontal, middle temporal, and superior temporal gyri. REN showed a significant increase during the night in all analyzed anatomical structures. IED counts were increased during the night in the hippocampus and frontal lobe structures. Furthermore, IED counts were non-significantly increased during the day in the analyzed temporal lobe structures.

Conclusions: The analysis showed changes in biomarker values throughout the day in non-pathogenic tissue during clinical recordings. The IED counts, which are the most commonly used biomarker of epilepsy, showed unpredictable behavior across patients and anatomical structures, in contrast to REN which exhibited more stable behavior with well defined day-night phase, making it possibly a more reliable biomarker. The changes in biomarker values during the day should be considered in studies that analyze interictal biomarkers outside the pathological tissue.

Funding: Please list any funding that was received in support of this abstract.: Supported by Ministry of Education, Youth and Sports of the Czech Republic project no. LTAUSA18056 (programme INTER-EXCELLENCE).