Carotid Stenosis Stenting and Cognition: CREST-H Results

A randomized trial led by Columbia University and published in JAMA found that opening severe carotid blockages with stenting or surgery did not improve cognition compared with medication alone. The study, called CREST-H, enrolled 291 patients with asymptomatic carotid stenosis of 70% or greater. Cognition improved slightly in both groups after one year, with no added benefit from revascularization.

What happened

CREST-H was a part of a larger trial, CREST-2, designed to find the best way to prevent strokes in people with severe carotid artery blockages. According to the report, CREST-2 found that medication plus stenting, but not blockage-removing surgery, decreased strokes twofold compared with medication alone. Study leader Randolph Marshall, MD, said these results appear to have changed practice already, with more of these patients receiving stents.

CREST-H asked a different question: does restoring blood flow also improve thinking? Cognition is known to decline when the brain's main vessels are narrowed. Severe carotid blockages are also associated with cognitive impairment even without stroke, particularly when cerebral blood flow is reduced. The hypothesis that revascularization could help had support from animal studies and case series, but it had never been tested in a randomized clinical trial.

The trial assessed cerebral blood flow and cognition at enrollment and again one year later. It compared patients who had a procedure to increase blood flow with patients treated with medication only. Both groups improved slightly, and the difference between them was not significant. The authors concluded that reversing cognitive decline by revascularization in these patients cannot be expected.

Why it matters

Carotid revascularization has a clear job: preventing stroke. This trial shows that cognitive improvement should not be added to the list of expected benefits for patients with asymptomatic stenosis and hemodynamic impairment. The JAMA title refers to "hemodynamic impairment," which fits with the study's focus on cerebral blood flow.

The finding also shows the gap between a biologically sensible idea and a proven one. Hypoperfusion plausibly impairs cognition, and animal data and case series pointed toward benefit. A randomized trial with a medication-only comparison did not confirm it. This is a useful reminder when you weigh evidence levels on exam questions.

Marshall's takeaway was about prevention: because revascularization cannot be expected to reverse cognitive decline, preventing carotid stenosis in the first place deserves more emphasis.

What to know for the USMLE

• Pathogenesis: Carotid stenosis is usually caused by atherosclerosis at the carotid bifurcation, affecting the proximal internal carotid artery (ICA). Risk factors include hypertension, smoking, diabetes, and hyperlipidemia.
• Clinical presentation: Many patients are asymptomatic. Symptomatic disease presents as amaurosis fugax (transient monocular vision loss from retinal artery embolization), TIA, or ischemic stroke in the MCA territory. A cervical bruit may be heard but is neither sensitive nor specific.
• Diagnosis: Carotid duplex ultrasonography is the usual first-line test. CT angiography or MR angiography is used to confirm and plan treatment.
• Revascularization options: Carotid endarterectomy (CEA) removes the plaque surgically. Carotid artery stenting (CAS) is endovascular. Classic teaching favors CEA for most surgical candidates with symptomatic stenosis of 70–99% (NASCET). CAS is considered for high surgical risk, a hostile neck (prior radiation or surgery), or surgically inaccessible lesions.
• Procedure-specific complications: CEA can cause cranial nerve injury (hypoglossal, marginal mandibular branch of the facial nerve), MI, and cerebral hyperperfusion syndrome. CAS carries a risk of periprocedural embolic stroke.
• Medical therapy for everyone: Antiplatelet therapy, high-intensity statin, blood pressure control, glycemic control, and smoking cessation. In CREST-2, medication alone was the comparison arm.
• Vascular cognitive impairment (vascular dementia): Classically it has a stepwise course with executive dysfunction and slowed processing. Mechanisms include chronic hypoperfusion, microemboli, small-vessel disease, and white matter injury. It is a differential diagnosis against Alzheimer disease, which has a gradual decline with early memory loss.
• Test-taking angle: If a stem asks about the goal of revascularization in carotid stenosis, the best answer is stroke prevention, not reversal of cognitive decline. The distinction between asymptomatic and symptomatic disease still matters for deciding who gets a procedure at all.

Limitations

• Source detail is limited. This summary relies on a news report, not the full paper. Effect sizes, specific cognitive tests, and confidence intervals are not given, so we do not report them.
• Population. CREST-H enrolled patients with asymptomatic stenosis of 70% or greater. The results do not automatically apply to symptomatic stenosis, lower-grade stenosis, or other patient groups.
• Follow-up. Cognition was reassessed at one year. The report does not say whether longer follow-up would show a different pattern.
• Both groups improved slightly. The source does not explain why. Repeated cognitive testing can raise scores through familiarity with the tests, but the report does not say whether the authors considered this.
• Procedure comparison. The report describes "stenting or surgery" as a combined procedure group. It does not describe results for stenting versus endarterectomy separately, so do not infer differences between them.
• Single trial. One randomized trial is strong evidence but not the final word. Replication and longer follow-up would add confidence.
• Not individual advice. This article is educational. Treatment decisions belong to patients and their clinicians.

FAQ

Did stenting or endarterectomy improve cognition in CREST-H?

No. In 291 patients with asymptomatic carotid stenosis of at least 70%, cognitive scores improved slightly in both groups after one year. The group that had a procedure did not differ from the medication-only group.

Does this mean carotid revascularization is not useful?

No. CREST-H tested cognition, not stroke prevention. According to the report, the parent trial, CREST-2, found that medication plus stenting decreased strokes twofold compared with medication alone, while surgery did not. The new result only suggests that cognitive improvement should not be expected as a benefit.

How could this appear on USMLE Step 1 or Step 2 CK?

Expect questions on the pathophysiology of carotid atherosclerosis and on indications for CEA versus CAS. Also expect questions on the features of vascular cognitive impairment, and on the role of medical therapy (antiplatelet, statin, risk-factor control). Knowing the stated goal of the procedure, stroke prevention, helps you avoid distractors.

Key takeaways

• In CREST-H, a randomized trial of 291 patients with asymptomatic carotid stenosis of 70% or greater, revascularization did not improve cognition at one year compared with medication alone.
• The purpose of carotid revascularization remains stroke prevention. CREST-2, per the report, showed fewer strokes with medication plus stenting.
• For the USMLE, link carotid atherosclerosis to its presentations, diagnostic workup, CEA versus CAS indications and complications, and the stepwise pattern of vascular cognitive impairment.