Scalp recording of infraslow (<0.5 Hz) ictal activity is useful for localizing partial seizures (Vanhatalo et al., Neurology 2003a;60:1198-1104, Miller et al., Neuroimage. 2007;35:583-597). This study further characterizes these infraslow ictal shifts with invasive recordings. Invasive monitoring captured 82 seizures in 11 patients with a 64-channel direct-current amplifier coupled to arrays of subdural platinum electrodes with bandwidth of 0 to 100 Hz. Time of onset, location, amplitude, duration, and polarity of infraslow signals were determined. Infraslow ictal signals (800-10,000 μV), were seen in 10 patients, starting from 2 seconds before to 493 seconds after electrical ictal onset time on conventional recording. Seven patients had an infraslow ictal signal in at least one channel localizing ictal onset on conventional recordings. Nine patients had surgical resections, including five with infraslow localizations concordant with conventional EEG (five had Engel class IA outcome, 1 class IB, 1 class IIIA, and one with no follow-up). Seizure localization using infraslow ictal activity was concordant with conventional EEG for most patients and is useful for confirming localization. The high voltage of infraslow activity may explain why infraslow activity localizes seizures better than conventional EEG with scalp recordings.
From the *School of Medicine; and †Regional Epilepsy Center, Departments of Neurology and Neurosurgery, University of Washington, Seattle, Washington.
The authors of this manuscript do not have any conflicts of interest to disclose.
Wonsuk Kim’s work on this project was supported, in part, by the Medical Student Research Training Program (MSRTP) at the University of Washington.
Address correspondence and reprint requests to John W. Miller, M.D., Ph.D., Harborview Medical Center, Box 359745, 325 Ninth Ave., Seattle, WA 98104, U.S.A.; e-mail: firstname.lastname@example.org.