Clamping ischaemia is responsible for the 20-30% of the neurological complications during carotid surgery. It is impossible at present to determine preoperatively for certainty the patients who are at risk and the ones who are not at risk for clamping ischemia. Intraoperative monitoring is mandatory in all the cases to point out cerebral ischemia and to prevent neurological deficit with an adequate brain protection. Methods used to predict brain ischaemia include local anesthesia, EEG, SEP, stump pressure measurement, transcranial doppler, regional cerebral blood flow measurement and evaluation of venous blood gases from the internal jugular vein. Local anesthesia is a safe and simple method of assessing cerebral ischemia during carotid clamping but has some limits particularly in case of brain ischaemia for the technical difficulties to install an indwelling shunt in an awake and often troubled patient. Moreover the only possibility of cerebral protection under local anesthesia is an increase in systemic blood pressure with a resulting growth in cardiac morbidity and mortality. For these reasons operation should be performed under general anesthesia to assure a better farmacogical brain protection when preoperative evaluation demonstrates a reduction of cerebral vasoreactivity, with a consequent high risk of clamping ischemia. At present EEG is one of the most used technique of cerebral monitoring under general anesthesia with a sensibility and specificity in the diagnosis of brain ischaemia that is about 90%. The best cerebral protection is obtained with a selective shunting. Burst suppression, with high dose thiopentone, has recently demonstrated its reliability alone or in association with an indwelling shunt in patients at high risk for clamping ischaemia. Of course a systematic cerebral protection together with an accurate control of the blood pressure is necessary in all the patients. This behaviour allowed us to obtain good results in the last 500 patients with a morbidity and mortality respectively of 1% and 1.6% with no significant differences between patients who tolerated carotid clamping and those with clamping ischaemia.