Design: Prespecified secondary analysis of two randomised trials (CREST-2)
Population: 2,165 US patients with ≥70% asymptomatic carotid stenosis, mean age 70
Primary outcome: Adjusted composite cognitive z score at 4 years
Effect: Stenting +0.03 SD (95% CI −0.09 to 0.15); endarterectomy +0.02 SD (−0.09 to 0.14)

CREST-2 comprised two parallel trials in patients with at least 70% asymptomatic carotid stenosis: stenting plus intensive medical management versus medical management alone, and endarterectomy plus intensive medical management versus medical management alone. This prespecified secondary analysis tested cognition annually by telephone, blinded to treatment, in 2,165 patients (mean age 70).

At four years there was no difference in the composite cognitive score in either trial (stenting: difference 0.03 SD, 95% CI −0.09 to 0.15; endarterectomy: 0.02 SD, −0.09 to 0.14). No individual test differed.

Asymptomatic stenosis is associated with poorer cognition, and restoring flow has been offered as a reason to intervene. These results do not support that. Whether to revascularise asymptomatic stenosis rests on stroke prevention, not memory.

Do not recommend carotid revascularisation to improve memory or thinking.
Base the decision on stroke risk and the main CREST-2 results.
Intensive medical management — blood pressure, lipids, antiplatelet, smoking — is the common foundation in both arms.
Evaluate cognitive complaints in a patient with carotid stenosis as you would in any other older adult.

Why it matters

It removes cognitive protection from the list of arguments for intervening in asymptomatic carotid disease.

The statistics, in plain English

The confidence intervals for both trials stay far below 0.5 SD, the smallest difference the investigators judged clinically meaningful, so a worthwhile cognitive benefit is effectively ruled out.