High-Frequency Oscillations in Self-Limited Epilepsy With Centrotemporal Spikes: Potential Predictors of Attention...

The aim of this study was to examine the possible effect of HFOs detected in children with SeLECTS, who have rolandic spikes with or without ADHD, in predicting cognitive comorbidities with the fully automatic ripple detector program we developed.

A total of 40 patients diagnosed with SeLECTS with at least a 1-year follow-up were included in this study. The patients were divided into two groups: those diagnosed with SeLECTS only and those diagnosed with SeLECTS+ADHD. For ripple detection, EEG data recorded for at least 10 min during non-REM stage 2 sleep with a sampling frequency of 2000 Hz was analysed in the MATLAB environment. After the data in each channel was filtered at 80-200 Hz, ripple detection was made with the fully automatically developed ripple detector program.

At least one ripple was detected in 29 of 40 patients (72.5%). The total number of spikes in both groups had a mean of 1435.8 ± 1626.9 (5-6183). The number of spikes in the rolandic region was found to be statistically significantly higher in the SeLECTS+ADHD group (p = 0.042). The total number of ripples in both groups was the mean: 9.5 ± 26.5 (0-158). The highest ripples count was detected in a patient in the SeLECTS+ADHD group; 158 ripples were counted, and the ripple distribution was found to be 33 ripples in the centrotemporal region and 125 ripples in the frontal region. The ripple of number (p = 0.009) and ripple ratio in the 'Fz-Cz' electrode were found to be statistically significantly higher in the SeLECTS + ADHD group (p = 0.009, p = 0.019, respectively).

Our study showed that the presence of interictal scalp HFOs has the effect of predicting neurocognitive comorbidities. We think that ripple analysis with the can be used as a potential biomarker to predict neurocognitive comorbidities.