Opening a narrowed carotid artery did not restore cognition

Published October 27, 2026 · Negative clinical result · Prespecified secondary outcomes of two randomized trials Primary source: CREST-2 cognition analysis, JAMA, PMID 42747844

A common rationale for treating severe asymptomatic carotid stenosis is that restoring blood flow might protect thinking. CREST-2 tested the cognitive part of that premise within two randomized comparisons: carotid stenting plus intensive medical management versus medical management alone, and carotid endarterectomy plus the same medical management versus medical management alone.

Among 2,366 U.S. patients, 2,165 completed baseline cognitive assessments. Composite cognitive z-score differences at four years were small and uncertain: 0.03 SD (95% CI −0.09 to 0.15) for stenting and 0.02 SD (−0.09 to 0.14) for endarterectomy. Neither comparison showed a significant benefit on the five individual cognitive tests. Follow-up testing averaged 2.24 years, despite the four-year modeled endpoint.

Strength and limitations. Large multicenter randomized parent trials, blinded central cognitive assessment, and prespecified analysis support the conclusion that clinically important cognitive benefit was not demonstrated. This was a secondary outcome, not a comprehensive test of every revascularization indication or stroke outcome. A smaller hemodynamic substudy also did not show cognitive advantage and shares the underlying CREST-2 population; it is not an independent trial. CREST-H substudy, PMID 42747834.

Clinical practice: Do not offer carotid revascularization solely in expectation of cognitive improvement on these data; assess stroke-prevention indications separately.