Cerebral vasoreactivity does not predict cerebral ischaemia during carotid endarterectomy.

Aim. Assessment of cerebrovasoreactivity (CVR), obtained by transcranial Doppler (TCD) and the acetazolamide test to predict cases requiring selective carotid shunting on the basis of neurologic monitoring. Methods. A consecutive series of 87 carotid endarterectomy (CEA) cases was studied. Before surgery CVR was evaluated by measuring the mean velocity of the middle cerebral artery (mvMCA) using TCD at the basal condition and at 30 min after intravenous administration of acetazolamide (1 g). Carotid shunting was performed using neurologic monitoring under local anesthesia. Receiver operating characteristic (ROC) curve was calculated for sensitivity and specificity for various CVR cut offs. Results. The ROC curve demonstrated that there was no single CVR cut off with both sensitivity and specificity above 80%. Conclusion. The present study, which employed neurologic monitoring as the method of comparison, did not prove that CVR, as calculated by TCD and the acetazolamide test, is a valid preoperative test for predicting cerebral ischaemia caused by carotid clamping.