Uncovering Patterns of Seizure Variability Within Individual Patients with Focal Epilepsy

Abstract

One of the primary challenges in treating epilepsy is that it is a dynamic disorder, with fluctuations in pathological brain activity and symptoms. Interestingly, seizures them selves can manifest in different ways, resulting in a range of spatiotemporal seizure evo lutions within the same patient. Although such diversity may affect patient treatments and outcomes, the extent and characteristics of within-patient seizure variability are un known. Here I investigated within-patient seizure variability by analysing seizure network evolutions in intracranial EEG (iEEG) recordings from patients with focal epilepsy. I first developed an approach for objectively comparing seizures and used my resulting “seizure dissimilarity” measure to characterise the extent and features of seizure variability. I then investigated how seizures changed over time in short-term recordings, revealing that variability in seizure evolutions could be explained by fluctuations over circadian and/or slower timescales. I next examined the relationship between seizure evolutions and seizure durations and found that these features could vary independently in the same patient: seizures with the same evolution could have dramatically different durations due to tempo ral “elasticity,” and seizures with similar durations could have distinct evolutions. Finally, using chronic iEEG recordings, I explored how the occurrence and duration of different seizure networks changed over multiple timescales that were revealed by fluctuations in interictal spike rate. I found that all patients had seizure network feature(s) that were associated with specific spike rate fluctuations, suggesting that seizure evolutions were modulated over these timescales. My work provides the first extensive characterisation of variability in within-patient seizure evolutions as well as a framework for quantita tively comparing seizures. Moreover, my results suggest that various modulatory factors, operating over different timescales, influence seizure evolutions. Understanding the mech anisms that shape seizure features within the same brain could provide new opportunities for controlling the full repertoire of seizures in each patient.