Abstracts

Rationale: Surgical resection is the most effective treatment for focal drug-resistant epilepsy (DRE) but is not feasible if the seizure focus is in eloquent cortex. Chronic subthreshold cortical stimulation (SCSC) is a novel brain stimulation technique that lowers seizures probability and offers an alternative therapy (Lundstrom et al., JAMA Neurol., 2016; Lundstrom, Worrell, Stead, & Van Gompel, Exp Rev Neurotherapeut., 2017). There are not well-developed methods for choosing appropriate stimulation parameters for brain stimulation techniques. Here, we summarize the results of patients treated with SCSC and describe several intrinsic excitability measures (IEMs) that could potentially be used to estimate seizure probability during SCSC.  Methods: Intracranial electroencephalography (iEEG) data from seven patients with DRE were analyzed respectively upon admission (EpiAdm) and after the initiation of cortical stimulation (EpiStim). Patients were implanted with subdural grid electrodes. Six 15-minute blocks of EEG data for each of 16 electrodes were reviewed per patient. Spikes were detected via a previously validated method for automated detection (Barkmeier et al., Clinical Neurophysiology, 2012). Slow oscillation phase and synchrony were measured for oscillations in the 0.1-2 Hz frequency band. The seizure onset zone (SOZ) was defined as the contacts involved at immediate seizure onset. Results: SCSC lowers seizure probability and focally decreases the rate of interictal spikes at the SOZ (Lundstrom et al., 2016). Average spike rate was decreased, spike phase in relation to underlying slow oscillations (0.1-2 Hz) was later, and a global synchrony measure was decreased for EpiStim compared to EpiAdm. Further, spike rate and phase were altered in the SOZ compared to the non-SOZ. Used in conjunction these measures predicted contacts that were more likely to be in the SOZ. Conclusions: SCSC is a brain stimulation technique that is an alternative treatment option for focal DRE. To assist in parameter adjustments, subjective assessments of interictal spike rate are used. Quantitative assessments of spike rate as well as other excitability measures such as spike phase in relation to slow oscillations and synchrony are potentially useful to guide parameter adjustments. An efficient means for estimating seizure threshold would improve the efficacy of brain stimulation techniques such as SCSC. Funding: Data collection was supported by grant NIH UH2-NS095495 (GW).