Noninvasive Biomarkers for Assessing the Excitatory/Inhibitory Imbalance in Children with Epilepsy

Abstract

Epileptic seizures involve a cortical excitation–inhibition imbalance driven by dysfunctional interneurons that contribute to gamma oscillation generation. While impaired gamma oscillations are commonly reported in epilepsy, the dynamics of broadband and narrowband gamma as well as beta oscillations remain underexplored. These oscillations may serve as noninvasive electrophysiological markers associated with altered cortical dynamics, potentially reflecting underlying excitation–inhibition imbalance in epilepsy. Here, we recorded high-density electroencephalography and magnetoencephalography data to investigate visual stimuli-elicited cortical oscillations in 48 neurotypical (20 females) and 49 children (26 females) with epilepsy. We found that epilepsy is characterized by reduced amplitude and prolonged latency of evoked cortical response compared with controls after visual stimulation, with alterations in N1 peaks and M100, M150, and M250 components ( p  < 0.05). Additionally, source imaging revealed disrupted oscillatory features in epilepsy patients, including suppressed power, reduced amplitude, and increased latency in evoked and induced beta and gamma oscillations from the visual cortex ( p  < 0.05). These alterations were consistent across diverse epilepsy subtypes, including focal, generalized, and nonlesional epilepsy cases. Utilizing these differences, we developed a novel classification model that differentiates individuals with epilepsy from controls with high accuracy, offering potential clinical utility in epilepsy diagnosis. Our results suggest disrupted beta and gamma oscillations may be associated with impaired inhibitory mechanisms and altered cortical dynamics, potentially indicative of an excitation–inhibition imbalance. Our findings highlight the potential of noninvasive electrophysiological biomarkers to capture cortical dynamics possibly influenced by excitation–inhibition imbalance in epilepsy, supporting their use in early diagnosis and disease monitoring.