Abstracts

Rationale: Hypersynchronous (HYP) and low voltage fast (LVF) activity are two different depth electrode-recorded electrographic seizure onset patterns in presurgical patients with suspected temporal lobe epilepsy. These patterns could arise from different mechanisms involving atrophic mesial temporal lobe (MTL) and possibly extra-MTL structures, yet the extent of extra-MTL structural abnormalities is not known. This study evaluated the location of cortical gray matter (GM) thickness reductions in patients with unilateral HYP and LVF ictal EEG onsets.Methods: A cortical pattern matching protocol was used to measure GM thickness from preoperative 1.5T MRI scans of 24 patients with unilateral MTLE and 24 matched controls (Table 1). Geometric models of the cortex using parametric surfaces were created and deformation maps were then built on the geometric models to spatially align cortical regions across subjects in order to generate group averaged cortical GM thickness maps.Results: Several features were similar between HYP and LVF onset seizures captured during depth EEG monitoring (Table1), but HYP onsets first appeared on fewer electrode contacts and spread to the contralateral (Cont) hemisphere after a longer delay than LVF onsets. Compared to controls, in patients with HYP onset up to 25% reduction in GM thickness appeared in ipsilateral (Ipsi) motor, somatosensory and visual cortices, temporopolar cortex and the inferior temporal gyrus. GM loss of a similar magnitude was observed across the same areas as well as the orbitofrontal areas in the Cont hemisphere. By contrast, LVF patients had a 10% reduction in GM thickness relative to controls in areas restricted to Ipsi motor and visual areas and inferior and middle temporal gyrus. Similar spatial patterns, but 15-20% GM thickness reduction, were found in Cont affected areas. Interaction of seizure onset pattern with years of epilepsy duration had a significant effect on GM thickness. HYP patients with a longer duration of epilepsy had thinner GM in Ipsi perirhinal cortex and anterior prefrontal areas, as well as Cont precuneus, posterior cingulate, motor and somatosensory cortices (Fig. 1). LVF patients with longer epilepsy duration had thinner GM chiefly in bilateral inferior prefrontal cortex and temporopolar areas. Finally, of patients who had resective surgery, 89% of those with HYP and 64% with LVF onsets were seizure-free.Conclusions: Results are consistent with the hypothesis that HYP ictal onsets are generated from a focal MTL network, and extensive areas of extra-MTL GM loss associated with HYP onsets detected here reflect uncontrolled seizure activity. By contrast, the findings suggest LVF onsets could arise from a regional uni- and possibly bi-lateral medial and lateral temporal lobe-inferior prefrontal network. Subtle structural alterations in all or some of these latter areas may be capable of supporting epileptic seizures. Functional studies may reveal epileptogenicity of extra-MTL GM loss associated with HYP and LVF ictal onsets, which could have important implications for resective surgery and postsurgical seizure freedom.