Abstracts

Rationale: Seizure freedom is the ultimate goal of epilepsy surgery. In neocortical epilepsy only 25-40% of non-lesional cases and 65-70% of lesional cases achieve seizure freedom. The likelihood of favorable outcome is thought to increase by presence of a resectable cortical lesion, recognizing that there is an epileptogenic zone that may extend beyond the lesion. Variability in patterns of seizure propagation in a given patient may indicate more extensive underlying network involvement, with some networks involved in some seizures and not in the others. Patterns of seizure propagation were analyzed and compared with post surgical outcomes in a small group of patients.Methods: The patients studied here were initially selected for analyzing patterns of seizure termination and consists of 14 neocortical patients with 151 seizures with both synchronous and asynchronous termination. Analyses were extended to examine the propagation patterns of these partial seizures. Visual analysis of the seizures was supplemented with measures of signal complexity changes derived from time-frequency analysis and subsequent calculations of a related complexity measure, the Gabor atom density (GAD) index (Jouny et. al. 2003). Multichannel propagation maps were assembled from the single channel analyses. Patients were divided into 2 groups with uniform propagation patterns or variable propagation patterns. Seizures with uniform propagation patterns all had clear signal complexity changes involving the same electrodes for each seizure. Variable propagation patterns occurred when different seizures in a given patient involved different electrodes in the grid. The seizure surgery outcomes of these two groups were compared.Results: Of 14 neocortical patients 8 had uniform propagation pattern and 6 had variable propagation pattern. The 8 patients with uniform propagation pattern had total of 96 seizures. The number of seizures per patients ranged from 5 to 17 with an average of 12 seizures per patient. The 6 patients with variable propagation pattern had total of 55 seizures. The number of seizures per patients ranged from 5 to 18 with average of 9.1 seizures per patient. The pattern of seizure propagation varied among seizures within a given patient with at least one or more of the seizures being different than others. At one year post surgery 6 of the 8 patients with uniform propagation pattern were seizure free (Engel 1) while 2 of the 8 patients had Engel 3 and 4 outcomes. From the 6 patients with variable propagation pattern only one patient achieved seizure freedom at one year (Engel 1), one patient had Engel 3 outcome and 4 patients had Engel 4 outcomes. Conclusions: Seizure dynamics at the focus are often quite stereotyped. Propagation patterns, although more variable may also be similar in certain patients. Uniform propagation patterns may suggest better surgical outcomes. Further correlation with imaging and pathology in additional patients may yield additional insights. Supported by NIH grant NS 48222