LOCALIZATION OF ICTAL ONSET WITH DC SOURCE LOCALIZATION
Abstract number :
2.154
Submission category :
Year :
2004
Submission ID :
4676
Source :
www.aesnet.org
Presentation date :
12/2/2004 12:00:00 AM
Published date :
Dec 1, 2004, 06:00 AM
Authors :
1Wonsuk Kim, 1John W. Miller, 1Mark D. Holmes, and 2,3Sampsa Vanhatalo
Ictal discharges contain frequency components that are substantially below conventional recording bandwidth ([lt]0.5Hz). We have previously shown that such slow activity may, on occasion, lateralize seizure origination better than the conventional EEG. In this study we expand on these observations by performing long-term (24-72 hour) direct current EEG recording with 27 channels, followed by analysis of the ictal recordings by principal component analysis. A total of 65 seizures in 15 patients with focal epilepsy were recorded using a commercial system (NuAmps, Compumedics Neuroscan). A full array of Ag/AgCl electrodes was applied using the 10:10 system. A special connector was used so that conventional long-term video EEG monitoring could continue simultaneously with the DC recording. Afterwards, ictal events were exported to an EEG analysis software package (BESA 5.0, Megis Software). Segments for the 60-second interval preceding AC electrical seizure onset were decomposed into principal components, and current sources were modelled to correspond to the dominant components. This analysis was applied for sequental short intervals, studying the evolution of current sources. Ictal onset was determined to be when and where a current source first became dominant, being stationary for at least 5 seconds. Our results show that, for sources with high signal-to-noise ratio, and not obscured by slow artifact, clear localization consistent with other clinical data can be determined. In such patients, localization of the dominant DC component (PCA weighting [gt] 90%) can be discerned up to 20 seconds before apparent AC onset, even when lateralization via traditional EEG techniques was inconclusive. When available, subsequent invasive monitoring confirms the DC localization. DC-EEG recordings can give useful additional information to noninvasively localize the epileptogenic focus. The value of this method is increased by mathematical analysis that can reveal highly localized changes not apparent on direct visual inspection.
References: Vanhatalo S, Holmes MD, Tallgren P, Voipio J, Kaila K, Miller JW. Very slow EEG responses lateralize temporal lobe seizures: An evaluation of non-invasive DC-EEG. Neurology, 2003; 60:1098-1102. (Supported by The Regional Epilepsy Center)