Sensitivity of scalp EEG, cortical EEG, and somatosensory evoked responses during surgery for intracranial aneurysms

BACKGROUND We estimated the relative sensitivity and reliability of scalp EEG, cortical EEG and somatosensory evoked potentials (SSEPs) to detect significant changes during aneurysm surgery. METHODS Two observers independently reviewed data from 18 patients who were monitored with scalp EEG, cortical EEG, and SSEPs to determine which if any modality demonstrated significant changes during 25 different episodes of temporary intracranial vascular occlusion. RESULTS Kappa scores indicating the degree of agreement between the two observers were highest for the cortical strip EEG (kappa = 0.92) and the SSEPs (kappa = 0.82) and much greater than for the scalp EEG data (kappa = -0.07). The cortical strip recordings showed changes more often than either the scalp EEG or SSEP during temporary vascular occlusion. In no case did the scalp EEG change when neither the strip nor SSEP changed. In only 4% of events did the observers feel that the SSEP changed when the strip did not, but in 16% of cases, the strip changed without changes in any of the other modalities. CONCLUSIONS Recording of EEG from strip electrodes placed on the cortical surface detects changes more frequently than either scalp EEG or SSEPs during vascular occlusion. Different observers were more likely to agree on whether the cortical strip EEG changed during vascular occlusion than the other modalities. This argues for the possible advantages of recording from strip electrodes during aneurysm surgery. (C) 2002 by Elsevier Science Inc.