Abstracts

Rationale: Self-Limited Epilepsy with Centrotemporal Spikes (SeLECTS), the most common focal childhood epilepsy syndrome, is associated with sleep-potentiated interictal spike waves emerging from the motor cortex that can occur in a unilateral or bilaterally-independent distribution. Electroencephalography (EEG) connectivity studies have shown that focal spikes are associated with large and diffuse increases in brain connectivity and furthermore, that connectivity is increased even during spike-free periods of sleep when compared to children without epilepsy. Single pulses of transcranial magnetic stimulation (spTMS) paired with EEG can measure evoked connectivity from the stimulated cortex to other brain regions. Repetitive trains of TMS (rTMS) can modulate excitability for extended periods. The purpose of this study was to investigate whether low-frequency rTMS can reduce connectivity of the epileptogenic motor cortex in children with SeLECTS.

Methods: Eight children (eight to twelve years old, 75% male, 62.5% on antiseizure medications) underwent two rTMS sessions with a one week interval (Figure 1A). In one session, one thousand pulses of real rTMS were administered to the motor cortex of the hemisphere with the most spikes; on the other day, sham rTMS of similar duration was administered. Connectivity before and after rTMS was measured by administering 100 spTMS to the motor cortex and then calculating the weighted phase lag index (wPLI) of the beta frequency bands in the one second after each pulse. wPLI is a connectivity metric that is robust against volume conduction and can be measured in short time increments. We tested whether there was a change in connectivity between the stimulated motor region and 7 other regions of interest (Figure 1B) in either the real or sham condition using one sample t-test.

Results: Following real rTMS, there were significant decreases in connectivity between the stimulated motor region and multiple brain regions (Figure 1C). Specifically, real rTMS led to decreased connectivity from baseline at ipsi-frontal (-0.046, p=0.053 contra-frontal (-0.041, p=0.03), contra-central-motor (-0.056, p=0.039), ipsi-temporal (-0.062, p=0.019), contra-temporal (-0.042, p=0.028), ipsi-occipital (-0.052, p=0.049), contra-occipital (-0.046, p=0.054) regions. In contrast, sham rTMS did not produce significant changes in connectivity.  

Conclusions: Children with SeLECTS have global increases in brain connectivity both during spike waves and during sleep. Real but not sham rTMS to the motor cortex leads to a temporary and globally reduction in connectivity of this epileptogenic region. This suggests a mechanism by which rTMS could exert a therapeutic benefit in children with epilepsy.

Funding: XS receives support from the Stanford MCHRI (PTA: 1220552-169-DHPEU). FMB receives funding for her research efforts from the NINDS K23NS116110.