Predictive Value of the First Ictal Recording in Determining Localization of the Epileptogenic Region by Scalp/Sphenoidal EEG

Abstract

The number of seizures recorded during a noninvasive evaluation for epilepsy surgery varies across centers. We retrospectively studied the accuracy of the first recorded seizure in predicting the final localization of the epileptogenic region. Sixty-six consecutive patients undergoing continuous EEG monitoring with scalp and sphenoidal electrodes were studied. The first recorded seizure was determined to be either well localized or nonlocalized and was compared with the ultimate localization after multiple seizures were recorded. The first seizure was well localized in 28 and nonlocalized in 38 patients. In the localized group, the first seizure correctly predicted the final localization of the EEG monitoring study in 26 patients, whereas 2 patients had bilateral independent temporal seizures. In the nonlocalized group, ictal onset remained nonlocalized in 34 patients despite recording of multiple (median of five) seizures. This gave a sensitivity of 87% and specificity of 94% for the first recorded seizure to predict the final results of noninvasive EEG monitoring. We conclude that the first recorded seizure is highly predictive of the final results of prolonged noninvasive ictal EEG recordings but it cannot exclude the possibility of multiple epileptogenic foci.