Abstracts

Rationale: Intractable epilepsy may be treated successfully with surgery if the seizure focus can be localized. Ictal scalp EEG may not provide sufficient localization and intracranial recording of ictal EEG with depth and/or subdural grid electrodes may be required. These invasive studies may provide more accurate information when scalp EEG is inconclusive, discordant or shows diffuse bilateral ictal discharges. False localization of seizures on scalp EEG may occur due to volume conduction and impedance if a seizure starts from one hemisphere with rapid progression to the other side. This phenomenon that may cause false lateralization is known as “ping-pong” seizures, which describes the ictal EEG pattern that shift from one side to another while ceasing on the original side as seizure progresses (Fisher et al, J Epilepsy, 1995). We reviewed cases to determine the prevalence if this phenomenon among epilepsy surgery candidates at the Barrow Neurological Institute. Methods: We analyzed ictal EEG patterns in the intracranial EEG population retrospectively by reviewing patient’s medical record and ictal EEG information from their inpatient monitoring. The review was conducted on all patients who underwent invasive EEG monitoring between July 2006 and December 2008. The original charts and EEGs were reviewed by one author, and the findings confirmed by the second author. All patients had previous scalp EEG monitoring, which did not clearly localize the seizure onset necessitating the invasive monitoring. The protocol and collection of data were approved by the local Institutional Review Boards for this study. Results: Total of 103 patients underwent intracranial monitoring; 41 patients had depth electrodes, 46 had subdural grids or strips, and 16 had both depth electrodes and subdural grids. Of the 103 patients who underwent intracranial EEG monitoring at the BNI from 7/2006 through 12/2008, only 4 patients (3.9%) conclusively showed ping-pong seizure phenomenon on EEG, during which seizures clearly started on one side, then spread to the other side while normalizing the initial side. Patient 1 had a total of 9 events captured, of which 8 were ping-pong seizures. Patient 2 had 1 ping-pong seizure out of 13 events. Patient 3 had 2 ping-pong seizures out of 9 events and patient 4 had 5 ping-pong seizures out of 9 events. Conclusions: It is important to recognize that the presence of ping-pong seizures may cause false lateralization of seizure onset in epilepsy surgery candidates. The incidence of the ping-pong phenomenon in our patient population was significantly less than previously reported rate of 33% in 1994 even though the study was conducted at the same institution. Although the small sample size for this study precludes definitive conclusion, ping-pong phenomenon may not be as prevalent as initially estimated. Future study with a larger number of patients with different types of seizures and lateralization would also beneficial.