EEG power changes are more sensitive than spectral edge frequency variation for detection of cerebral ischemia during carotid artery surgery: a prospective assessment of processed EEG monitoring.

To investigate the basis of ambiguous reports of the validity and utility of processed electroencephalogram (EEG) detection of cerebral ischemia, 19 patients monitored during surgical procedures requiring clamping of the carotid artery were studied. The EEG was recorded and observed for detection of alteration of spectral edge frequency versus EEG power. Electrodes were positioned at the P3-C3' locations over the left hemisphere and P4-C4' areas over the right hemisphere (10-20 system of electrode placement). Maximum sensitivity was used for recordings of the processed EEG. Twelve of 19 patients had bilateral carotid vascular stenosis. Nine of 19 patients studied with EEG monitoring had EEG changes suggestive of cerebral ischemia during interruption of carotid artery blood flow by surgical manipulation, defined as a decline in EEG power of greater than 40% or a decline in spectral edge frequency of at least 3 Hz. Eight of these episodes occurred at the time of carotid vascular clamp placement. These changes were confirmed by the raw EEG. Whereas power band monitoring detected 9 episodes of suspected ischemia, alteration of spectral edge frequency was sufficient to detect only 2 of these episodes. One patient sustained a right hemispheric stroke detected intraoperatively by a 47% decline in EEG power; however, these changes were unaccompanied by intraoperative alteration of spectral edge frequency. It is concluded that monitoring of EEG power with processed EEG devices is a more sensitive indicator of cerebral ischemia than monitoring only the spectral edge frequency of the EEG.