Structural changes in patients with primary generalized tonic and clonic seizures.

According to the International Classification of Epilepsies, idiopathic generalized epilepsy (IGE) is characterized by absence of cerebral abnormalities. However, this is contradicted by findings of migrational disturbances, cerebral distortions, and neuroreceptor and metabolite changes in IGE.

In the current study, we investigated whether IGE is associated with structural changes and whether the location of these changes could be relevant for the behavioral expression of seizures. Two MR-based measurements were employed: voxel-based morphometry (VBM) of the entire brain and MR volumetry of selective regions of interest.

Discussion

Seizure semiology is crucial in studies of IGE, and we wanted to include patients with a single seizure type. The diagnosis IGE with pure GTCSs seems adequate considering the age at seizure onset, the objective seizure description, lack of lateralized abnormalities on MRI, and presence of bilaterally synchronous epileptiform EEG activity.

The two methods used are complementary and may show different results in regions with well-defined structural landmarks but a poor gray–white matter contrast. This can explain that in our patients, the VBM was normal in caudate and putamen despite the reduced volumes. Notably, when reducing the search space, patients showed reduced gray matter in putamen.

The observed changes were confined to the frontotemporoparietal neocortex, the thalamus, basal ganglia and cerebellum. This distribution is congruent with the reported increases in the blood oxygenation level–dependent signal during interictal epileptiform activity in IGE and with elevations of cerebral blood flow during generalized seizures following electroconvulsive therapy. The respective networks could be directly engaged in the major manifestations of GTCSs, considering that impairment of consciousness is associated with altered frontotemporoparietal and thalamic signaling and that thalamus mediates motor functions via connections from the ventral anterior and lateral nuclei to the motor cortex, basal ganglia, and cerebellum. One possibility is that the changes in our patients resulted from seizures. However, no quantitative relationship was found with seizure load (duration and frequency), possibly because of the narrow range. Alternatively, they could reflect microscopic heterotopias. Theoretically, the changes could be related to valproate (a general suppression of blood flow has been reported with valproate), but also this seems improbable, as the changes were regional.

The proposed mechanisms are not mutually exclusive, and subtle developmental malformations could, for example, increase the susceptibility to seizure activity, resulting in excitotoxic lesions.
The current data allow no cause–effect conclusions. Nevertheless, by use of two independent MR methods, they add to several previous notions that morphologic changes indeed exist in human IGE. Furthermore, the finding of regional changes in GTCSs provides new arguments to the discussion as to whether generalized seizures are truly generalized and calls for a reevaluation of the current diagnostic criteria for IGE.

We conducted MR volumetry and voxel-based morphometry in 19 patients who had idiopathic generalized epilepsy (IGE) with tonic clonic seizures as the only seizure type and 52 control subjects. Patients exhibited elevated frontal lobe fraction of CSF and reduced fraction of gray matter in the frontal, parietal, temporal cortex, thalamus, and cerebellum. The thalamus and cerebellum also showed reduced volumes, as did the caudate and putamen. Contrary to current diagnostic criteria, regional tissue abnormalities may be present in IGE.