While both ephedrine and phenylephrine are currently used to treat hypotension occurring during carotid endarterectomy (CEA) under general anaesthesia, phenylephrine may have deleterious effects on the cerebral watershed, due to its exclusively vasoconstrictive action. In this controlled, double‐blind randomised trial, we compared the effects of ephedrine and phenylephrine administered in a standardised algorithm to treat the first hypotensive event occurring since induction of anaesthesia until carotid cross‐clamping. The algorithm consisted of 1‐to‐3 boluses of 6 mg of ephedrine or 50 μg of phenylephrine, after a goal‐directed fluid therapy. In case of failure, the treatment switched to the other study drug. Cerebral tissue oxygen saturation (SctO2) was monitored by near infrared spectroscopy (NIRS), and the primary outcome was the restoring effect of SctO2 (ipsilateral to surgery) to baseline values. Secondary postoperative outcomes were: contralateral SctO2, neurological outcomes, and plasma S100B protein measured at discharge from post‐anaesthesia care unit. Ephedrine treatment provided a higher rate of restoration of ipsilateral SctO2 than phenylephrine (93.2% vs. 85.1%, P=0.034); this was also noted for contralateral SctO2 (93.5% vs. 90.7%, P=0.026). The gain in SctO2 on the lowest value during hypotension was also higher under ephedrine than phenylephrine (6.4% vs 4.3% ipsilateral, 5.1% vs. 4% contralateral), but not significantly so. Clinical outcomes were unaffected by the treatment, but S100B protein plasma concentration was higher in the phenylephrine group. To conclude, this pilot trial, focusing on intermediate outcomes, suggests that ephedrine should be preferred to phenylephrine to treat hypotension during CEA. This article is protected by copyright. All rights reserved.