A total of 231 patients with atheromatous ulcerations of the carotid artery and obstructions due to critically stenotic plaques or kinks underwent reconstructive surgery mainly of the internal carotid artery, under general anesthesia, normocarbia, and without increasing the systemic blood pressure. Continuous electroencephalography and phot pulses from the cutaneous territories of the internal and the external carotid arteries were used to monitor 250 carotid operations. An intraoperative test, clamping of the common carotid artery and its two branches for 3 min, was performed on each patient. An internal shunt was routinely used in 18 patients. In the remaining 232 operations, an internal shunt was selectively used in only 13 of the 15 patients in whom there was focal showing on the EEG and/or a reduction in the amplitude of the ipsilateral supraorbital pulses to less than 25% of their preclamping values. In the group in which no shunt was used, the duration of the carotid clamping varied from 6 to 59 min, with a mean of 26 min. Seven patients died, but death was related to brain ischemia during surgery in only two. There were no ischemic neurological complications, except in the patients who died. This results in a mortality and morbidity rate of 0.8%. In those patients in whom a shunt was used, CCPPG recordings clearly indicated obstruction of the shunt during surgery. Thrombotic occlusions of the distal end of the shunt in one case and of the repaired vessel in another two were diagnosed. Electroencephalographic and CCPPG monitoring provided useful information of the state of cerebral function, the level of anesthesia, and the occurrence of cardiac dysrhythmia or hypotension and its effect on brain circulation. It is believed that this monitoring technique greatly contributed to the decreased mortality and morbidity in our series.