Lennox-Gastaut syndrome: a secondary network epilepsy
AffiliationAustin Academic Centre
Document TypePhD thesis
Access StatusThis item is embargoed and will be available on 2020-07-11.
© 2018 Dr. Aaron Warren
Lennox-Gastaut syndrome (LGS) is a severe epilepsy that usually begins in childhood. Diverse aetiologies can cause LGS, including structural and genetic abnormalities. However, the electroclinical phenotype that emerges from these aetiologies is remarkably consistent across patients and includes tonic seizures and generalised interictal epileptiform discharges on scalp electroencephalography (EEG). Frequent epileptic activity is thought to contribute to cognitive impairment in LGS (‘epileptic encephalopathy’). The mechanisms by which diverse aetiologies lead to LGS, and by which epileptic activity causes impaired cognition, are poorly understood. This thesis aimed to determine the brain regions underlying epileptic activity in LGS, and to assess how LGS may alter functional organisation of brain networks that support cognition. Simultaneous EEG with functional magnetic resonance imaging (EEG-fMRI) was performed in 38 patients, including 11 children with recent-onset LGS and 27 older patients with longstanding LGS. To limit motion during scanning, younger patients were anaesthetised with low-dose isoflurane and remifentanil. Data from the older patients, who tolerated scanning without anaesthesia, were compared to resting-state fMRI in 29 age-matched healthy controls. The primary hypothesis was that LGS reflects abnormal expression of a shared network that is ‘secondary’ to the specific aetiology of LGS. Five studies were performed. Study 1 examined technical feasibility of anaesthetised fMRI in children with LGS. Resting-state networks previously described in non-anaesthetised subjects were observable in the anaesthetised patients with LGS, demonstrating that isoflurane-remifentanil anaesthesia can extend the utility of fMRI to young and intellectually disabled patients while retaining interpretability of fMRI. Study 2 explored brain areas underlying interictal epileptiform discharges in LGS using EEG-fMRI. In both anaesthetised and non-anaesthetised patient groups, discharges recruited widespread areas of frontal, parietal, and temporal cortex, as well as the thalamus and brainstem. Activation was similar across individual patients with structural, genetic, and unknown aetiologies of LGS. Further analysis using dynamic causal modelling suggested a cortically driven process underlying discharges. Study 3 explored network mechanisms of impaired cognition in LGS by comparing resting-state functional connectivity between patients and healthy controls, focusing on corticocortical interactions. Patients showed abnormal integration and segregation of major cognition-related networks, including the executive-control and default-mode networks. Altered connectivity persisted during periods without epileptic activity on patients’ in-scanner EEG, potentially contributing to pervasively impaired cognition in LGS. Study 4 tested thalamocortical circuits and found abnormally enhanced connectivity in LGS, maximally involving mediodorsal and ventrolateral nuclei. Finally, study 5 examined neural changes accompanying successful treatment of LGS by re-scanning one patient who experienced seizure remission following resection of a cortical lesion. After surgery, the patient showed reversal of abnormal network patterns seen in non-operated patients, with improved network integration and segregation. This thesis provides a new way of conceptualising LGS as a ‘secondary network epilepsy’, where the syndrome’s characteristic epileptological and cognitive features reflect abnormal expression of a shared brain network, rather than the specific initiating aetiology. The epileptic process in LGS is cortically driven, affects specific thalamic subregions, and may be reversible with early surgical intervention.
Keywordsepilepsy; Lennox-Gastaut syndrome; EEG; fMRI; EEG-fMRI; functional connectivity; intellectual disability
- Click on "Export Reference in RIS Format" and choose "open with... Endnote".
- Click on "Export Reference in RIS Format". Login to Refworks, go to References => Import References