Carotid endarterectomy (CEA) is the elective surgical procedure to prevent stroke due to stenosis of the carotid bifurcation. During a period of 17 years the Authors performed 215 operations on the carotid arteries of 168 patients. The average age was 64.6 and the male/female ratio was 3/1. Patients were symptomatic in 75.8% of cases and asymptomatic in the remaining 24.2%. Preoperative investigations consisted of echo-Duplex scanning, arteriography, cerebral CT or MRI. Indications for surgery were: stenosis wider than 70% in 173 cases, ulcerated or "high-risk" stenosis in symptomatic patients in 37 cases, and carotid malformation in 5 cases. The intraoperative use of shunt (12% of the operations) was selective, depending from the results of our monitoring system: stump pressure and transcranial Doppler (TCD) of the middle cerebral artery (MCA) ipsilateral to the procedure. The global major stroke/mortality rate was 3.3% (7/215), the minor morbidity was 8.8% (19/215). Mortality rate was 0.5% (1/215). The major stroke/mortality rate for symptomatic patients was 4.2% and for asymptomatic patients was 0%. The average follow up was 58 months (range 1-192) for 200/215 patients, with 15/215 patients (7%) lost. The postoperative incidence of stroke after 4 years was 8.5% (17/200), with an annual mortality rate of 1.6% (min. after 2 months, max. 118, average 55 months). CEA is a safe procedure to prevent cerebral infarctions, but it still carries an operative risk. A better monitoring would allow to understand the mechanisms of clamp-induced ischaemia and prevent it, therefore decreasing the operative risks and extending the surgical indications to a higher ratio of asymptomatic subjects. TCD is becoming essential for our goal: it is useful in deciding to insert an intraoperative shunt, check the carotid flow, recognize embolic events, and also during the initial phase of carotid preparation.