University of Twente Student Theses

Login

Observations on suppressed power following single pulse electrocortical stimulation : clinical relevance for the REC2Stim trial

Stoel, M.D. van der (2018) Observations on suppressed power following single pulse electrocortical stimulation : clinical relevance for the REC2Stim trial.

[img] PDF
9MB
Abstract:Purpose: Central lobe epilepsy (CLE) is a focal epilepsy arising from the pre- and/or postcentral gyrus, and is by nature often medically refractory with a high seizure frequency. Surgery is not an evident treatment due to the high risk of contralateral sensorimotor impairment, related to excision of eloquent cortex. Recently, a trial for a new therapy for CLE patients has been proposed: Rational Extra-Eloquent Closed-loop Cortical Stimulation (REC2Stim) using electrical stimulation within a local epileptogenic network to abort the build-up towards a seizure. The stimulation location is preferably connected to the seizure onset zone (SOZ), but outside eloquent cortex. Single pulse electrical stimulation (SPES) is often used to identify the SOZ based on delayed responses (DRs), but can also expose the underlying connections based on early responses (ERs). In time-frequency decompositions of SPES, suppression of power for frequencies <250 Hz is sometimes observed after stimulation. We hypothesized that power suppression after SPES might serve as a surrogate marker for suppression of epileptiform activity and could be a preferred site of cortical network stimulation. As ERs can identify cortical connections and were sometimes co-observed with the suppressions, we explored whether this suppressed power (SP) was associated with the occurrence of ERs and therefore, a direct connection between cortex underneath the stimulus pair and response electrode. Furthermore, we explored whether timing of the stimulation and thus the phase of the background electrocorticography (ECoG) signal determines the occurrence of an SP. Method: Refractory epilepsy patients who underwent intracranial subdural electrocorticography monitoring at the UMC Utrecht were retrospectively analysed. SPES (10 pulses: 0.2 Hz, 1 ms, 4-8 mA) was routinely administered on adjacent electrode pairs. A machine learning algorithm (support vector machine (SVM)) was developed to detect SP based on features ‘area’ and ‘duration’ (chapter 2). ERs and SP were detected and visually checked in a total of 34600 responses across ten subjects. Six subjects had (one of) their grids implanted in the central lobe. The other four subjects had grids implanted elsewhere. We determined the number of response electrodes in which both ER and SP, either ER or SP, and neither ER nor SP were evoked by the same stimulus pair. A chi-squared test was used to determine whether SP was associated with the occurrence of ERs (chapter 3). The relevance of the phase of the ECoG signal at the moment of stimulation for the occurrence of power suppression was researched with the inter trial coherence (ITC) and instantaneous phase between stimuli with a suppression (chapter 4). Results: Ten subjects were included (four females, median age 15, range 9-41 years). In all subjects, the number of response electrodes with SP was smaller than those with ERs. All subjects considered, 8% of the responses contained both ER and SP, 16% had an ER without SP, 3% had an SP without an ER and 73% had neither ER nor SP. In each subject, significantly more electrodes with both ER and SP were found than would be expected based on chance (p<.001). This was also found when combining subjects (p<.001). For each subject a stimulation pair outside functional area, causing suppression in (part of) the SOZ, could be found. For one subject (female, 31 years), the ITC was determined for individual stimuli. No significant difference in ITC between the stimuli with suppression and the stimuli without suppression was found. A difference in the distribution between up and down phase was seen. Conclusion: The occurrence of SP was strongly associated with the occurrence of ERs. However, response electrodes with SP are no perfect subset of response electrodes to whom an ER is evoked. Further research is needed to investigate whether cortical stimulation, suppressing power after SPES, is also effective in reducing epileptiform activity. In one subject we did not identify significance in phase coherence right before stimulation. Studying more subjects would be appropriate to reach a definite conclusion about the influence of the phase on the occurrence of power suppression after SPES.
Item Type:Essay (Master)
Faculty:TNW: Science and Technology
Subject:44 medicine, 50 technical science in general
Programme:Technical Medicine MSc (60033)
Link to this item:http://purl.utwente.nl/essays/76608
Export this item as:BibTeX
EndNote
HTML Citation
Reference Manager

 

Repository Staff Only: item control page