Grant: Project Grant

More than 54,000 Canadians have epilepsy with seizures that often occur daily and are poorly controlled by medication. In these cases, surgical removal of seizure generating tissue (which contains the critical seizure onset zone) is their best chance of seizure freedom and relief from psychiatric and cognitive problems associated with epilepsy. Currently, the clinical gold standard for identifying the seizure onset zone is EEG recordings during seizures. Unfortunately, the surgical success rate is <50%, mostly due to imprecise identification and removal of the seizure onset zone. We propose here a strategy to correct this. Mounting evidence indicates that high frequency oscillations (HFOs), detected between seizures using electrodes implanted into the brain (intracranial electrodes), better localizes the seizure onset zone than seizure recordings. However, intracranial EEG only samples a few brain areas. By combining intracranial EEG with functional MRI, which can measure responses across the entire brain, we have observed HFO-associated brain activity both at the conventionally measured seizure onset zone and in nearby regions. We propose that these additional regions are critical parts of the full seizure onset zone. As a result, we predict better surgical outcomes when they are also resected. We also postulate that the degree to which the seizure onset zone is connected to the rest of the brain (measured by functional MRI) predicts clinical outcome: fewer connections, better outcome. We will study 120 patients with uncontrolled epilepsy. We will determine if a better surgical cure rate is associated with i) the extent of surgical removal of the seizure onset zone as defined by intracranial EEG-functional MRI, and ii) fewer pre-operative functional connections of the seizure onset zone. Our study will offer an unprecedented ability to accurately identify the seizure onset zone and improve surgical and neurocognitive outcomes in persons with epilepsy.