BACKGROUND AND PURPOSE: Intraoperative monitoring of brain function may influence the outcome of carotid endarterectomy (CEA). METHODS: We performed transcranial Doppler (TCD) monitoring of middle cerebral artery blood flow velocities (VMCAs) and eight-channel electroencephalographic (EEG) recording simultaneously in 82 patients undergoing CEA. Thiopental narcosis limited EEG interpretation in 11 patients, thus allowing direct comparison of both methods in 71 patients. RESULTS: There was a significant correlation between VMCA decrease and the frequency of EEG changes after carotid clamping (P < .001). Eight patients (11%) showed a VMCA decrease exceeding 60%, accompanied by EEG changes in 7 patients. Altogether, 16 patients (22%) showed severe or moderate EEG changes. Stenosis or occlusion of the contralateral carotid artery led to an increase of abnormal findings with both monitoring methods, which was, however, significant only for TCD (P < .05). Four patients (4.8%) suffered intraoperative transient ischemic attacks. In 3 of these patients, there were no abnormal findings with either of the methods. The events were thus unpredictable and probably of embolic origin. The fourth patient showed VMCA decrease to 0 and severe EEG changes. Nine patients had severe or moderate EEG changes without significant VMCA decrease and without complications. EEG monitoring alone in these would have led to unnecessary use of a shunt with the increased risk of embolism. CONCLUSION: EEG and TCD monitoring are complementary techniques. Their results showed a good overall correlation but with marked differences in the individual patient. TCD monitoring alone was sensitive enough to prevent ischemic intraoperative complications. EEG findings are of limited value when barbiturates are used.
BACKGROUND AND PURPOSE: Intraoperative monitoring of brain function may influence the outcome of carotid endarterectomy (CEA). METHODS: We performed transcranial Doppler (TCD) monitoring of middle cerebral artery blood flow velocities (VMCAs) and eight-channel electroencephalographic (EEG) recording simultaneously in 82 patients undergoing CEA. Thiopentalnarcosis limited EEG interpretation in 11 patients, thus allowing direct comparison of both methods in 71 patients. RESULTS: There was a significant correlation between VMCA decrease and the frequency of EEG changes after carotid clamping (P < .001). Eight patients (11%) showed a VMCA decrease exceeding 60%, accompanied by EEG changes in 7 patients. Altogether, 16 patients (22%) showed severe or moderate EEG changes. Stenosis or occlusion of the contralateral carotid artery led to an increase of abnormal findings with both monitoring methods, which was, however, significant only for TCD (P < .05). Four patients (4.8%) suffered intraoperative transient ischemic attacks. In 3 of these patients, there were no abnormal findings with either of the methods. The events were thus unpredictable and probably of embolic origin. The fourth patient showed VMCA decrease to 0 and severe EEG changes. Nine patients had severe or moderate EEG changes without significant VMCA decrease and without complications. EEG monitoring alone in these would have led to unnecessary use of a shunt with the increased risk of embolism. CONCLUSION: EEG and TCD monitoring are complementary techniques. Their results showed a good overall correlation but with marked differences in the individual patient. TCD monitoring alone was sensitive enough to prevent ischemic intraoperative complications. EEG findings are of limited value when barbiturates are used.