Abstracts

The aims of the present study are (1) to identify the regions of altered glucose metabolism in juvenile myoclonic epilepsy (JME) patients compared with normal controls, and (2) to correlate generalized spike or polyspike-wave (GSW) with its possible metabolic concomitants using a combined PET/EEG method. We studied 19 patients with unequivocal JME with ¹⁸F-fluoro-2-deoxy-[italic]D[/italic]-glucose (FDG)-PET and simultaneous EEG recording. Neither clinical nor electrographic seizure was recorded during the EEG monitoring. The duration of visually identified GSW, either single or repetitive, was measured from the spike which interrupts the background activity to the end of slow wave. Total cumulative time was measured by addition of each GSW duration. Firstly, using statistical parametric mapping (SPM), significant increases and decreases in regional glucose metabolism in the patient group were estimated by comparing their normalized PET images with those of 19 control subjects (uncorrected [italic]P[/italic] [lt] 0.001, voxel [gt] 50). Secondly, in order to identify brain regions where glucose metabolism significantly correlated with spike-wave duration in patients with JME, general linear model with the cumulative GSW time as a covariate was tested for each voxel. This analysis was performed using the covariate analysis using SPM2 (uncorrected [italic]P[/italic] [lt] 0.001, voxel [gt] 50). Group comparison: When comparing 19 JME patients with 19 controls at uncorrected [italic]P[/italic] [lt] 0.001, hypermetabolic regions were found in the right thalamus (Talairach coordinates, 8/ -7/ 13; [italic]Z[/italic] score, 3.37; voxel count, 90) and left thalamus (Talairach coordinates, -6/ -5/ 11; [italic]Z[/italic] score, 3.59, voxel count, 155) in the patients group. No significant hypometabolic regions across the whole brain were shown at the same threshold.
Covariate analysis: When we correlated glucose metabolism of JME patients with cumulative GSW time in a 40-min EEG recording, significant positive correlations ([italic]P[/italic] [lt] 0.001) were found in bilateral thalami, cerebellar hemispheres, and midbrain. No significant regions of negative correlations were found at the same threshold. With a combined PET/EEG method, we observed that interictal thalamic metabolism is increased in patients with JME and that increased interictal spike frequency correlated with an increase in thalamic glucose metabolism. This is, to our knowledge, the first study to document a correlation between the interictal epileptiform discharges and thalamic metabolic activity in JME. This study provides evidence for the thalamus having an important role in the generation of interictal spike-wave discharges in JME.