Studying the mechanisms of absence seizure generation using EEG with functional MRI

© 2012 Dr. Patrick William CarneyPublications included in thesis:Mullen, S. A., Crompton, D. E., Carney, P. W., Helbig, I. & Berkovic, S. F. (2009). A neurologist’s guide to genome-wide association studies. Neurology, 72(6), 558-565. DOI: 10.1212/01.wnl.0000341942.29513.bdCarney, P. W., Masterton, R. A. J., Harvey, A. S., Scheffer, I. E., Berkovic, S. F. & Jackson, G. D. (2010). The core network in absence epilepsy: differences in cortical and thalamic BOLD response. Neurology, 75(10), 904-911. DOI: 10.1212/WNL.0b013e3181f11c06Masterton, R. A., Carney, P. W. & Jackson, G. D. (2012). Cortical and thalamic resting-state functional connectivity is altered in childhood absence epilepsy. Epilepsy Research, 99(3), 327-334. DOI: 10.1016/j.eplepsyres.2011.12.014Carney, P. W., Masterton, R. A. J., Flanagan, D., Berkovic, S. F. & Jackson, G. D. (2012). The frontal lobe in absence epilepsy: EEG-fMRI findings. Neurology, 78(15), 1157-1165. DOI: 10.1212/WNL.0b013e31824f801dCarney, P. W., Harvey, A. S., Berkovic, S. F., Jackson, G. D. & Scheffer, I. E. (2013). Siblings with refractory occipital epilepsy showing localized network activity on EEG-fMRI. Epilepsia, 54(2), e28-e32. DOI: 10.1111/epi.12076Absence seizures are brief epileptic events characterised by loss of awareness, often with subtle motor features such as blinking. Although seizures are brief, with a rapid onset and offset, they may be very frequent and are no longer considered benign given the potential impact on attention, learning and memory. Absence seizures are accompanied by bursts of high amplitude, rhythmic, usually 3 Hz, spike and wave activity arising from an otherwise normal electroencephalography (EEG) background. A number of pathophysiological models have been developed to explain the mechanism of absence seizure generation. These models rely heavily on observations from animal studies of induced and spontaneous absence like seizures and have shown clearly that oscillations within thalamo-cortical networks are central to seizure generation. Animal studies have also implicated a role for other brain regions outside the thalamo-cortical network, as well as identifying possible focal cortical generators of seizures. Studying the structural and functional relationships between brain regions in humans is hampered by the need for non-invasive techniques. EEG with functional magnetic resonance imaging (EEG-fMRI) is one such technique that provides an opportunity to explore the cortical and sub-cortical interactions between brain structures involved in absence seizure generation. EEG-fMRI uses the blood oxygen level dependent (BOLD) signal as an endogenous contrast agent which provides an indirect measure of neuronal activity and hence can demonstrate regions of increased or decreased activity during absence seizures. In this thesis, event-related analysis of absence seizures as well as time course analysis and functional connectivity was used to explore the mechanisms of absence seizure generation. A large and diverse cohort of subjects with new onset, familial and refractory epilepsy with absence seizures was recruited. Using EEG-fMRI, we identified a core network of structures involved in absence seizure generation. The core network comprises the thalamus, midline and lateral parietal cortex (the default mode network [DMN]), caudate nuclei and the reticular structures of the pons. The only component displaying an increase in BOLD signal relative to the resting state is the thalamus. The other structures show relative decreases in BOLD signal and importantly the only cortical change, in the group analysis of new onset epilepsy, is reduced BOLD signal in the DMN. Time course analysis demonstrates that the earliest change in BOLD signal occurs in the DMN, prior to the onset of epileptiform events. This region also shows altered functional connectivity in patients with absence seizures. Hence it appears that this region could be central to our understanding of how absence seizures are generated. Furthermore, we have identified that there could be variable BOLD changes in the dorso-lateral pre-frontal cortex during absence seizures and have speculated that this change could have both phenotypic and mechanistic implications for absence epilepsy. We also studied the contribution of genetic and phenotypic factors on BOLD signal change and found that phenotypic and genotypic similarity led to greater correlation in BOLD signal change. We have also shown a more extensive absence network in subjects with a refractory absence phenotype and other seizure types. The results of this thesis provide validation of the pathophysiological models of absence seizure generation derived from animal models and propose the contribution of other important structures to seizure generation