Abstracts

Rationale: Determining the localization and lateralization of seizure onset is an important step in the evaluation and treatment of patients with epilepsy. Recognizing clinical features of seizures that have localizing or lateralizing value can often provide clues for the precise location of seizure onset. Oral automatisms, for example, suggest the possibility of temporal lobe localization, although there has been no evidence in support oral automatisms having lateralizing value. Oral automatisms have many behavioral components, including lip, jaw, and tongue movement, swallowing, etc. Our hypothesis is that oral automatisms can be subcategorized into subtypes based upon differences in quality of the automatisms, and that certain oral automatism subtypes may correlate with lateralization of seizure onset.Methods: IRB approval was obtained prior to conducting this study. Subjects were selected from a database of resective epilepsy surgery candidate patients who were evaluated at the University of Michigan between January 1, 2011 through March 5, 2015. Subjects age 18 years and older with temporal lobe onset seizures and oral automatisms being a feature of the seizure semiology were included for review. Video was extracted from the video-EEG recordings, edited, and cropped to blind the reviewer to other lateralizing ictal features (e.g. unilateral tonic limb posturing). A speech-language pathologist reviewed each seizure and scored the presence of different subtype features (e.g. lip movement, jaw movement, swallowing, etc.). Oral automatism subtypes were compared for correlation to the known laterality of seizure onset using a Chi square test, calculated with Microsoft Excel.Results: A total of 52 temporal lobe onset seizures with oral automatisms visible on video recording were included for review. Chi square tests were performed to evaluate for correlation of oral automatism subtype to seizure lateralization (see Table 1). We found non-significant trends suggesting that oral automatisms that include prominent tongue movement (p = 0.1001) or jaw movement (0.1361) correlate with left temporal onset seizures.Conclusions: The trends observed in this study suggest that oral automatisms that include tongue or jaw movement may be associated with left temporal onset seizures. The lack of statistical significance for these results may be due to the sample size of this study. Thus, confirmation of the significance of these findings will require collection and analysis of additional data samples. If oral automatism subtypes that include prominent tongue or jaw movement do have lateralizing value in temporal lobe epilepsy, this information would add significantly to a growing knowledgebase about the lateralizing and localizing value of seizure semiology, which is critical for both the evaluation and treatment of patients with epilepsy.