Investigation of Electrical Focus and Generator Mechanism for Photoparoxysmal Discharges with a Dipole Tracing Method

Abstract

Purpose: Photoparoxysmal discharges induced by photic stimuli are generally observed on the scalp EEG in cases of photosensitive epilepsy. There have been many reports about the clinical and electroencephalographic aspects of photosensitive epilepsy, but the mechanisms for the photoparoxysmal discharges, especially in the regions of the electric source generator and its expansion, have not yet been determined. The dipole tracing method is a computer‐aided noninvasive technique for estimating the location of epileptic discharges from a scalp EEG. In this study, we examined the location of the electrical generator for the photoparoxysmal discharges with scalp‐skull‐brain dipole tracing (SSB‐DT). Methods: We studied eight patients with epilepsy who had photoparoxysmal discharges. We divided these cases into groups A and B (four cases each), composed of patients with and without visually induced seizures, respectively. In this method, the generators corresponding to the paroxysmal discharges were estimated as equivalent current dipoles (ECDs) by one‐ and two‐dipole analyses. The locations of the ECDs were estimated by iterative calculations. Algorithms minimizing the squared difference between the electrical potentials recorded from the scalp EEG and those calculated theoretically from the voluntary dipoles were used. In the SSB model, the scalp shell was reconstructed from the helmet measurements, and the shapes of the skull and brain were three‐dimensionally reconstructed from the computed tomography (CT) images. A dipolarity > 98% was adopted to indicate the accuracy of the estimation. We recorded the 21–channel monopolar scalp EEG. To stimulate the photoparoxysmal discharges, we used intermittent light stimulating with a red filter at flash rates of 20 Hz. Each of the spikes was sampled and analyzed at 10 points around the peaks of ≤10 spikes in each patient by using the SSB‐DT method. The ECDs were then superimposed on the MRI for each patient to identify the anatomic region more exactly. Results: The findings from the one‐dipole method showed focused locations, and the dipolarity was 98% or more in all cases, whereas those from the two‐dipole method showed scattered locations, so that the analyzed signals were considered to be generated from a single source. Visually induced seizures appeared in case 1 from group A during light stimulation. The ictal EEG at first revealed an occipital‐dominant initial focal spike and the subsequent diffuse spike‐and‐wave complex bursts, dominant at the left hemisphere. The result of the superimposed ECDs, of the initial spike located at the visual cortex in the right occipital lobe, and those of the diffuse spikes, corresponded to the small area adjacent to the lateral geniculate nucleus on the MRI. The ECDs of the diffuse spikes in three cases from group A included case 1 located at the lateral geniculate nucleus. In contrast, the ECDs of the diffuse spikes in all four cases from group B corresponded to the corpus callosum. Three cases (one case in group A, case 1; and two in group B, cases 5 and 6) had the initial focal spikes, and their ECDs were all located in the visual cortex. Conclusions: This study suggests that the pathways of epileptic discharges from the epileptic focus are different between patients with and without visually induced seizures. It is supposed that neural activity of the lateral geniculate nucleus might be responsible for the generator mechanism of photoparoxysmal discharges, which evokes photosensitive epilepsy.