Brain activation patterns of versive, hypermotor, and bilateral asymmetric tonic seizures.

Purpose: Patients who have seizure onset from different brain regions can produce seizures that appear clinically indistinguishable from one another. These clinically stereotypic manifestations reflect epileptic activation of specific networks. Several studies have shown that ictal perfusion single photon emission computed tomography (SPECT) can reveal propagated ictal activity. We hypothesize that the pattern of hyperperfusion may reflect neuronal networks that generated specific ictal symptomatology.

Methods: All patients were identified who were injected with (99m)Tc-hexamethyl-propylene-amine-oxime (HMPAO) during versive seizures (n = 5), bilateral asymmetric tonic seizures (BATS; n = 5), and hypermotor seizures (n = 7) in the presurgical epilepsy evaluation between 2001 and 2005. The SPECT ictal–interictal difference image pairs of each subgroup were compared with image pairs of 14 controls using statistical parametric mapping (SPM 2) to identify regions of significant hyperperfusion. Hyperperfused regions with corrected cluster-level significance p < 0.05 were considered significant.

Results: We have identified a distinct ictal perfusion pattern in each subgroup. In versive seizure subgroup, prominent hyperperfusion was present in the frontal eye field opposite to the direction of head version. In addition, there was associated caudate and crossed cerebellar hyperperfusion. The BATS subgroup showed pronounced hyperperfusion supplementary sensorimotor area ipsilateral to the epileptogenic region, bilateral basal ganglia, and contralateral cerebellar hemisphere. The hypermotor seizure subgroup demonstrated two clusters of significant hyperperfusion: one involving bilateral frontomesial regions, cingulate gyri, and caudate nuclei, and another involving ipsilateral anteromesial temporal structures, frontoorbital region, insula, and basal ganglia.

Discussion: We have identified distinct hyperperfusion patterns for specific ictal symptomatology. Our findings provide further insight into understanding the anatomic basis of seizure semiology.