Influence of Interictal Epileptiform Activity on Language Networks in Children with Rolandic Epilepsy or DyslexiaDoctoral thesis
Date of Examination:2023-12-20
Date of issue:2024-01-05
Advisor:Prof. Dr. Niels K. Focke
Referee:Prof. Dr. Melanie Wilke
Referee:Prof. Dr. Susann Boretius
Referee:Prof. Dr. Knut Brockmann
Sponsor:Deutscher Akademischer Austauschdienst (DAAD)
Sponsor:Göttingen Graduate Center for Neurosciences, Biophysics, and Molecular Biosciences (GGNB)
Files in this item
This file will be freely accessible after 2024-12-19.
EnglishSelf-limited Epilepsy with Centrotemporal Spikes (SeLECTS) is the most common genetic epilepsy in children. It is characterized by seizures remission in puberty and highly frequent interictal epileptiform discharges (IEDs) during drowsiness and sleep. Developmental dyslexia (DD) is a learning disorder characterized by significant difficulties in reading and writing despite normal intellectual abilities. Both are considered neurodevelopmental disorders, sharing significant deficits in phonological awareness, reading comprehension, reading speed, non-words reading, writing, and working memory. According to previous publications, a percentage of the children with DD have IEDs in electroencephalography (EEG). Multiple EEG and neuroimaging studies have shown that IEDs are correlated to significant changes in functional connectivity (FC) in adults or children with epilepsy. At the same time, other researchers have proposed a causal link between IEDs and cognitive deficits in children with SeLECTS. However, it has also been argued that it is not clear that IEDs imply a higher risk of seizures or cognitive impairments in patients without seizures or epilepsy. Consequently, this doctoral project investigated the possibility that a relevant amount of IEDs can be found in children with DD, implying an additional change in their resting-state FC since previous studies have also indicated abnormal FC patterns in children with DD compared to controls. Likewise, FC changes concerning IEDs in children with SeLECTS were studied using EEG and functional resonance magnetic imaging (fMRI). This work is divided into two retrospective studies and a third prospective study. The first consisted of a retrospective search for EEGs of patients with DD to determine the existence of IEDs, and compare power and FC between DD children and controls. Clinical routine resting-state EEGs were collected, and power and FC were analyzed using source-reconstruction. In addition, neuropsychological assessment data was collected from patients with DD, and an association between cognitive performance and power or FC values was evaluated. When dividing the sample into children in 1st-to-4th and 5th-to-8th school years, the 1st-to-4th graders with DD showed a significant power decrease in the delta-theta range in left parieto-occipital regions and a significant FC decrease in the theta band in almost all regions of the left hemisphere and right parieto-occipital regions compared to matched controls. Meanwhile, 5th-to-8th graders with DD revealed widespread FC increase in the alpha band compared to matched controls. Moreover, the power values in the theta band of seven left parieto-occipital regions were significantly and positively correlated to their written spieling performance in children with DD. Similarly, the second study included a retrospective collection of clinical routine resting-state EEGs of patients with SeLECTS to investigate power and FC using source-reconstruction. Several comparisons were performed between selected EEG epochs of patients with SeLECTS and controls, but also between EEG epochs of SeLECTS patients with IEDs and without IEDs, from recent-onset SeLECTS against controls, from longer-duration (drug-receiving) SeLECTS against controls, and from recent-onset SeLECTS against longer-duration SeLECTS. Besides, associations between power or FC values of EEG epochs without IEDs and epilepsy-related variables were analyzed. All comparisons exhibited significant and widespread power increases in SeLECTS patients relative to controls. FC analyses of epochs without IEDs revealed focal and predominantly left-sided FC increases in the beta and decreases in the theta band relative to controls. Epochs with IEDs showed further FC increases in the delta band compared to epochs without IEDs, located in bilateral fronto-centrotemporal regions. Patients with recent-onset SeLECTS had a significant FC increase in the alpha band in right fronto-centro-temporal regions relative to controls. Patients with longer-duration SeLECTS showed a significant FC increase in the beta band in left centrotemporal regions, and FC decreases in the delta band in left temporo-parietal regions compared to controls. Lastly, FC values in the delta band in seven right central and frontal-opercular regions showed a significant and positive correlation with IEDs frequency in children with SeLECTS. A prospective and third study compared the activation of language-related and working memory-related brain regions between children with SeLECTS and controls, using task-based fMRI and resting-state fMRI. Additionally, it evaluated an association between the frequency of IEDs in resting-state EEG and performance in reading tests in patients with SeLECTS. Although no significant brain activation differences were observed, SeLECTS patients exhibited significant FC differences at rest and during tasks relative to controls. At rest, patients showed significant decreases between the right middle fronto-temporal regions and the bilateral frontal operculum. During a phonological task, SeLECTS children had significant increases between the right hippocampus and bilateral temporal regions. During a verbal working memory (VWM) task, SeLECTS patients revealed significant increases and decreases among multiple bilateral regions related to language and motor control. When comparing cognitive performance, children with SeLECTS showed poorer performance than controls but only a significant deficit in written spelling. Lastly, a significant correlation was found between IEDs frequency in drowsiness and sleep and performance in working memory subtests of WISC-V. In summary, this dissertation indicated the feasibility of source-reconstruction of clinical routine resting-state (low-density) EEG to investigate biomarkers of pediatric epilepsies and learning disorders. Additionally, as IEDs fulfilling internationally defined and recognized criteria were uncommon, no FC changes were demonstrated with IEDs in DD patients. Nevertheless, children with DD or SeLECTS showed different but abnormal power and FC patterns compared to controls. These EEG signatures are modulated by age in the DD sample and disease duration in the SeLECTS sample. In addition, children with SeLECTS show altered FC-fMRI patterns at rest, during phonological processing and VWM, suggesting compensatory neural mechanisms for adequate cognitive performance, yet relatively poorer than controls. Finally, this work suggests how neurophysiological variables might be related to the neurocognitive development variability in children with epilepsies or learning disorders.
Keywords: Self-limited epilepsy with centrotemporal spikes (SeLECTS); Dyslexia; Electroencephalography (EEG); Functional magnetic resonance imaging (fMRI); Language; Reading; Spelling; Verbal working memory