Investigation of Human Epileptic Networks by fMRI

Investigation of Human Epileptic Networks by fMRI Based Effective Connectivity: a New Approach to Identify the Neuronal Drivers of the Pathological Activity in Surgically Remediable Epilepsies

Background: In the Emilia-Romagna Region approximately 19.000 people are affected by epilepsy. About 25% of epileptic patients are drug-resistant (DRE) and some of them are eligible for resective surgery of the epileptogenic zone (EZ). The precise EZ localization is crucial for a good surgical outcome. Intracranial EEG (icEEG) recordings remain the gold-standard to localise the EZ. New neuroimaging techniques, like simultaneous recording of functional MRI and EEG (EEG-fMRI), with advanced methodological approaches as effective connectivity analysis (i.e. Dynamic Causal Modelling-DCM) might improve the EZ localization. Objectives: (1) To develop a non-invasive protocol for the investigation of the epileptic network in patients with surgically remediable epilepsies; (2) To shed light on the patho-physiological mechanisms of drug resistance in DRE; (3) To provide a validation of effective connectivity applied to fMRI data in epilepsy. Methods: Two Research Units (RU1, RU2) will identify and characterize a cohort of DRE patients eligible for resective surgery. RU1 will be in charge to perform the EEG/fMRI coregistration and data analysis. RU2 will be responsible for the surgical resection of epileptic foci. IcEEG recordings will be performed at the Claudio Munari Epilepsy Center, Ospedale Niguarda, Milano. RU1&RU2 will evaluate the fMRI data results and compare with icEEG findings or expert's surgical decision. The principle measures of outcome are: (a) percentage of concordance of fMRI results with icEEG findings or electro-clinical features in term epileptic network identification; (b) percentage of concordance between DCM findings and EZ/IZ localization; (c) percentage of concordance of DCM findings with icEEG or electro-clinical features regarding the causal hierarchy within the epileptic network.

Background: About 25% of epileptic patients are drug-resistant (DRE) and some of them are eligible for resective surgery of the epileptogenic zone (EZ). The precise EZ localization is crucial for a good surgical outcome. Intracranial EEG (icEEG) recordings remain the gold-standard to localise the EZ. New neuroimaging techniques, like simultaneous recording of functional MRI and EEG (EEG-fMRI), with advanced methodological approaches as effective connectivity analysis (i.e. Dynamic Causal Modelling-DCM) might improve the EZ localization. This innovative tool will have the advantage to be non-invasive and safe with significant decrease of injuries, hospitalization, with a resulting favourable cost/benefit ratio. Objectives: (1) to provide a validation of effective connectivity applied to fMRI data in epilepsy. (2)To develop a non-invasive protocol for the investigation of the epileptic network in patients with surgically remediable epilepsies; (3) To shed light on the patho-physiological mechanisms of drug resistance in DRE. Methods: Two Research Units (RU1, RU2) will identify and characterize a cohort of DRE patients eligible for resective surgery. RU1 will perform patients' recruitment, presurgical evaluation and EEG/fMRI coregistration and data analysis. RU2 will perform patients' recruitment, presurgical evaluation and surgical resection of epileptic foci. IcEEG recordings will be performed at the Claudio Munari Epilepsy Center, Ospedale Niguarda, Milano . RU1&RU2 will evaluate the fMRI data results and compare with icEEG findings or expert's surgical decision. The principle measures of outcome are: (a) percentage of concordance of fMRI results with icEEG findings or electro-clinical features in term epileptic network identification; (b) percentage of concordance between DCM findings and EZ/IZ localization; (c) percentage of concordance of DCM findings with icEEG or electro-clinical features regarding the causal hierarchy within the epileptic network.

The presented project will include only one arm constituted by patients affected by pharmacoresistant epilepsies elegible for respective surgery. Patients will be identified by RU1 and RU2. The definition of drug-resistant epilepsy requires: (a) the failure of at least two first-line AEDs; (b) the occurrence of an average of one seizure per month for > 18 months; (c) no more than 3-month seizure free hiatus during those 18 months (Berg et al., 2006).

All the patients recruited will undergo to the following interventions: 1. Presurgical assesment protocol of the recruited patients (RU1 and RU2) that will includes: (a) detailed medical and epilepsy patients' history; (b) neurologic evaluation; (c) structural 3 Tesla MRI; (d) prolonged video-EEG monitoring including one-overnight EEG recording; (e) neuropsychological evaluation including quality of life evaluation. 2. EEG-fMRI coregistration: All subjects will undergo to a Video-EEG/fMRI study. This protocol will include a 32-channels EEG recorded simultaneously to fMRI data acquisition (3T scanner, Philips). This phase will be articulated in two steps: (i) acquisition of the EEG-fMRI data; (ii) fMRI data analysis (conventional and DCM analysis). RU1 will be charged for both steps. 3. Surgical resection of the epileptic foci and/or intracranial electrode recording (icEEG). The DRE patients in whom an intracranial recording will be considered unnecessary, will be operated at the RU2.

The non-invasive EEG-fMRI (conventional analysis) study will reveal the epileptic network in more than 80% of the DRE patients recruited.

DCM based on fMRI will identified the causal hierarchy within the epileptogenic network in more than 80% of the patients studied. Particularly the following outcome measures will be considered: (a) localization of epileptogenic zone (EZ); (b) localization of irritative zone (IZ).

Identification of possible mechanisms of drug-resistance in refractory epilepsies The outcome measure system relies on: (a) clinical data collection recorded in an electronic dedicated case report form (CRF), (b) the qualitative results of EEG-fMRI recordings. Particularly the level of concordance between fMRI maps and icEEG/ expert's surgical decision will be assessed by looking the distance (in cm) between the area of maximum BOLD changes (Global Maxima) and the defined EZ, as already validated by others; (c) surgical outcome (at 3-6-9-12 months after surgery) in those DRE patients operated as measure of the DCM success.

Inclusion Criteria: - Adults patients (≥18yrs) diagnosed with DRE (drug-resistant epilepsy), candidate for epilepsy surgery who will undergo directly to the resection of the epileptic focus or to icEEG recordings for a better EZ definition Exclusion Criteria: Patients with idiopathic generalized epilepsies; Patients with focal epilepsy responders to AED; Patients with refractory focal epilepsy but contraindicated to perform a MRI; Patients who refute to have the EEG-fMRI; Patients whose cognitive status is too impaired to complete the necessary study forms.