Abstracts

Rationale: Recent studies suggest that a non-negligeable proportion of drug-resistant epilepsy have an epileptogenic zone that involves the insula. We hypothesized that proton magnetic resonance spectroscopy (¹H-MRS) could detect insular metabolic abnormalities in patients with insular epileptogenicity.Methods: MRI and voxel-based ¹H-MRS were performed on 8 patients with suspected insular epileptogenicity who subsequently underwent insular and peri-insular subdural and depth electrode recordings and 16 age-matched controls. Voxels were tailored to the anterior and posterior insular regions to measure peak areas for N-acetylaspartate (NAA) and creatinine (Cr). Comparisons of the NAA/Cr ratios between the normal and epileptogenic insula, and between patients and controls were made using a one-way ANOVA. Asymmetry indices (AI) were also carried out to evaluate potential intra-individual metabolic variations. Results: Insular interictal spikes were recorded in all 8 patients (100%) (in addition to other areas of spiking activity). Insular seizures, however, were recorded only in 2 of 8 patients (25%). MRI revealed hippocampal asymmetry in 3 of 8 (38%) patients but none had an insular abnormality. When compared to controls, both ipsi and contralateral insular NAA/Cr ratios were significantly higher in patients with insular epileptiform activity (2.19 0.10 and 2.22 0.17 respectively versus 2.00 0.15, P=0.003). No difference was found between patients with insular seizures and patients with insular spikes only. No difference in the mean NAA/Cr ratio was observed between the ipsilateral and contralateral insula. Mean AI in patients with insular epileptiform activity were not different from controls (P=0.2). Lateralization of the AI to the epileptogenic insular cortex was found in only one of 8 (13%) patients. Conclusions: ¹H-MRS reveals paradoxical bilateral insular increase in NAA/Cr ratio in patients with insular epileptiform activity. The significance of these counterintuitive findings needs to be determined.