Abstracts

Rationale: Patients with partial epilepsy may show diffuse bilateral EEG discharges believed to be propagation from their focus. In some, this is the only scalp EEG manifestation, making it difficult to distinguish it from a generalized seizure disorder. Previous studies described the EEG-fMRI characteristics of generalized spike and wave discharges (GSWD) in idiopathic generalized epilepsy (IGE). Here we describe the EEG-fMRI response to diffuse bilateral discharges in focal epilepsy patients. Methods: Fourteen patients with focal epilepsy and diffuse bilateral discharges underwent an EEG-fMRI study. Some had focal interictal discharges as well, which were analyzed separately. EEG discharges were marked and convolved with four hemodynamic response functions (peaks at 3-9s). Significant responses were defined with %t%>3.1 and spatial extent of 4 voxels (p=0.05, corrected). Visual analysis and comparison with EEG and clinical data was performed. Results: Five patients had diffuse bilateral EEG events without side predominance: four showed lateralized EEG-fMRI results. The fifth had symmetric fMRI activation and deactivation that did not match the known EEG-fMRI pattern of IGE. Nine patients had lateralized EEG predominance: seven of them had matching EEG-fMRI activation asymmetry, and two others had matching deactivation asymmetry (matching was defined as higher %t% in the corresponding side or region of EEG maximum). All 14 patients had some to most of the components of the default mode network in deactivation maps: precuneus, posterior cingulate gyrus, medial prefrontal lobe, and parts of associative parietal cortices bilaterally. However, this pattern was not complete and showed asymmetry in eight patients. Activation patterns varied among patients, although activation of frontal, insular, and perisylvian regions, similar to responses to GSWD, was frequently seen. A thalamic response was less reliable than what was reported in IGE: five patients had no response, six showed activation stronger in anterior thalamus, two had a deactivation in posterior thalamus, and one showed both thalamic activation and deactivation. Caudate response was uncommon: only two patients showed a caudate deactivation -a common finding in IGE. Conclusions: Our results show EEG-fMRI result differences between diffuse bilateral discharges and the known patterns of IGE. Cortical activation seems the most reliable finding, showing lateralization or focal accentuation. Cortical deactivation was less lateralized and consistently overlapping parts of the default mode network. Thalami showed activation, deactivation, or both in some of the patients. An ipsilateral stronger thalamic response was the most likely. These findings may be due to a stronger BOLD response in the primary focus of the discharges, followed by a weaker response related to propagation. In conclusion, EEG-fMRI can be considered as a method to distinguish diffuse bilateral discharges in focal epilepsy from a genuine generalized process, especially in the absence of a more localizing feature.