Project Summary/Abstract Epilepsy is the fourth most common neurological disease in the US. 20-30% of the patients with epilepsy (PWE) continue to experience seizures despite adequate drug treatment which puts them at risk for sudden unexpected death in epilepsy (SUDEP). SUDEP typically occurs in the postictal recovery phase and is characterized by impaired arousal followed by cardiorespiratory breakdown suggesting a shutdown of critical brainstem functions. However, respiratory disturbances severe enough to cause peri-ictal hypoxemia (PIH) and signs of impaired arousal such as widespread post-ictal EEG slowing (PIES), or postictal generalized EEG suppression (PGES) occur also during non-fatal seizures. The mechanisms of PIH, PIES and PGES and how they relate to SUDEP are not well understood. The overall goal of this project is therefore to use MRI to characterize structural and functional alterations of the brainstem systems controlling respiration and arousal in drug-resistant epilepsy and to investigate how they influence the occurrence of these non-fatal disturbances. This will not only allow for a better understanding of the mechanism underlying these common peri-ictal disturbances but also provide insights how they relate to SUDEP. Specifically, it is hypothesized that PIH is a consequence of structural and functional disturbances of the midbrain respiratory modulator complex but spares the respiratory pattern generator in the lower medulla. PIES are associated with abnormalities in the brainstem arousal network and PGES is hypothesized to be associated with severe abnormalities in the arousal as well as respiratory modulator network. Structural imaging and task-free fMRI will be acquired in 200 patients with drug-refractory, non-lesional epilepsy (PWE) undergoing long-term video-EEG telemetry and SpO2 monitoring in 3 epilepsy centers. The primary measure of interest is the occurrence (yes/no) of PIH and PIES/PGES. A new brainstem segmentation approach will be used to identify brainstem structures of interest. Dynamic functional connectivity analysis combining cluster analysis and graph theory will be used to identify different brain states in task free fMRI. 40 age and gender matched healthy controls (CON) will undergo the same imaging as PWE.