Carotid Artery Stenting: The New Treatment Standard for Asymptomatic Carotid Stenosis?
January 1, 2026
By Alan Z. Segal, MD
Synopsis: CREST-2 demonstrates that, in asymptomatic carotid stenosis, carotid artery stenting modestly reduced four-year stroke risk compared with intensive medical management, whereas carotid endarterectomy did not.
Source: Brott TG, Howard G, Lal BK, et al; CREST-2 Investigators. Medical management and revascularization for asymptomatic carotid stenosis. N Engl J Med. 2025; Nov 21. doi: 10.1056/NEJMoa2508800. [Online ahead of print].
In 1991, the landmark North American Symptomatic Carotid Endarterectomy Trial (NASCET) established that surgical treatment of carotid stenosis with carotid endarterectomy (CEA) in patients with a prior history of transient ischemic attack (TIA) or stroke provided a robust benefit in secondary stroke prevention.1 The two-year risk of ipsilateral stroke was 9% with CEA compared to 26% with medical therapy. NASCET was important not only because of its outcomes, but also because of its groundbreaking methodology.
The NASCET method of measuring carotid stenosis (using the narrower normal distal carotid as the denominator rather than the dilated bulb as in prior studies) assured that there would not be overestimation of stenosis. This enriched the treatment population. Patients with lesions of > 70% severity using older methods were not included in NASCET, which measured the same lesions at only about 40% stenosis. NASCET furthermore was unique in that it assured that surgery was not handicapped by subpar surgeons (since it was restricted to those with < 6% complication rates).
The profound benefit seen in NASCET, enrolling symptomatic patients, stood in contrast to the Asymptomatic Carotid Atherosclerosis Trial (ACAS), which also was positive, but far less dramatic.2 There was a 5% stroke rate for CEA compared to 11% in the medical arm. Furthermore, ACAS required five years of follow-up to prove a 6% improvement compared to NASCET, which showed a much higher 17% benefit after only two years. ACAS hinged on surgeons having even lower complication rates (< 3%) than in NASCET. Furthermore, while NASCET subsequently was shown to be positive even in patients with moderate stenosis of 50% to 69%, ACAS floated in a relative no-man’s land (including patients with > 60% stenosis, likely diluting any treatment effect).
With the emergence of endovascular stroke therapies, a new question emerged, pitting CEA up against the relatively less-invasive procedure of carotid artery stenting (CAS). Twenty years after NASCET, in 2010, the highly anticipated Carotid Revascularization Endarterectomy vs. Stent Trial (CREST) attempted to settle this dispute.3 Instead, it showed relative equipoise. CREST suggested that CEA was superior in older patients, because of a high risk of periprocedural stroke in CAS and found the reverse in younger subjects (aged < 70 years), who were better candidates for stenting. CREST incorporated both symptomatic and asymptomatic patients, leaving the question of whether any intervention at all was justified for asymptomatic patients unanswered.
After another long hiatus (15 years), the publication of the CREST-2 trial in November 2025 represents the next paradigm shift in the management of carotid stenosis. CREST-2 studied approximately 2,500 patients, divided into four groups (each with approximately 600 patients), split into two parallel trials. CREST-2 found that the four-year stroke risk with CAS was 2.8%, significantly lower than with medical therapy alone (6%). In contrast, the stroke risk for CEA was 3.7% compared to 5.3% with medical therapy, a nonsignificant difference. In the CAS group, the total number of strokes was 15. This compared to 28 patients getting medical therapy. Eight of the strokes in the CAS group were peri-procedural, occurring within the first 44 days of follow-up.
In the CEA group, the total number of strokes was 19. This compared to 26 patients getting medical therapy. Nine of the strokes in the CEA group (again, about half) were peri-procedural. Despite any favorable results in CREST-2 for CAS or CEA, operating on an asymptomatic patient and immediately precipitating a stroke continues to have significant practical implications.
Patients in the medical group of CREST-2 had aggressive treatment of stroke risk factors — particularly systolic blood pressure (SBP) and low-density lipoprotein (LDL) cholesterol. The CREST-2 protocol of intensive medical management (IMM) included active coaching for weight loss, smoking cessation, and diabetic care, as well as frequent telephonic follow-ups — a program known as INTERVENT. Despite this optimization, SBP < 130 mmHg and LDL cholesterol < 70 mg/dL were achieved in only 61% and 67% of subjects, respectively.
Overall, the patients in CREST-2 had stenosis in the lower end of the inclusion range. About 30% of patients, distributed relatively equally among the four treatment groups, had a demonstrated peak systolic velocity on ultrasound of > 389 cm/sec. This parameter corresponds to a stenosis of ≥ 80%. Therefore, the majority of enrolled patients (70%) had carotid stenosis in the 70% to 80% range.
There was a flurry of late strokes in the CEA group. The Kaplan-Meier curves comparing CAS and CEA were overlapping during the entire trial until the final six to eight months of the four-year follow-up period, when the stroke rate in CEA bumped from the low 2% range to the final 3.7%. The stroke rate in CAS was static at 2.8% during this time interval. Without this increase of approximately four to five stroke events in the CEA group, this therapy also would have proved significantly beneficial.
Surgeons and interventionalists in CREST-2 underwent rigorous vetting prior to enrollment. CAS operators were required to submit their most recent 25 consecutive cases and required a self-reported lifetime experience of at least 100 CAS procedures. Surgeons submitted their most recent 50 cases. Both CAS and CEA operators required a < 3% complication rate (the established standard in asymptomatic carotid trials).
Approximately two-thirds of patients in the medical management group had non-disabling strokes with a higher proportion of disabling strokes occurring in the interventional arms. As noted in the editorial comment, the major difference between the IMM and CAS groups was driven by 20 nondisabling strokes in the IMM cohort (patients with a favorable clinical status who then could go on to a procedure for symptomatic disease). With these patients with nondisabling stroke eliminated, there was no difference between CAS and IMM.
Commentary
Taken at face value, the results of the CREST-2 study are clear — CAS is superior to medical therapy, but CEA is not. Therefore, any patient with asymptomatic carotid stenosis ≥ 70% should undergo CAS as opposed to CEA or IMM.
However, from a different perspective, regardless of management choices, the stroke risk from asymptomatic disease in CREST-2 was quite low — in the 3% to 5% range over four years. Stated another way, more than 19/20 patients did not have a stroke in CREST-2, a proportion closer to 19.5/20 when nondisabling strokes are excluded. Furthermore, with an absolute difference of 3.2% between CAS and IMM, the number needed to treat is 31 to prevent one stroke.
The low overall stroke rate in CREST-2 likely was driven by a component of selection bias. Patients with asymptomatic stenosis in the 90% to 99% range likely were not referred for the trial. This likely applied to many patients in the 80% to 89% range as well. However, it is impressive that even with low event rates in a relatively healthier population (having 70% to 79% stenosis), CREST-2 was able to show a statistically significant benefit for CAS.
Crucial to any clinical trial performed under highly regulated conditions is the question of whether the results are generalizable to real-life patients in non-tertiary settings. The editorial notes that despite the best efforts of the investigators, patients were not optimally managed in the IMM group (with blood pressure and LDL goals only achieved in about two-thirds of patients). Their implication is that any difference between intervention and IMM would have been reduced or eliminated if risk factor management was more effective. Among other advances, they highlight the treatment of cholesterol with PCSK-9 inhibitors rather than statins. However, a more realistic viewpoint is that medical management in real-life likely will be worse than in the study, widening (not narrowing) the gap between intervention and IMM. Interestingly, high-volume community-based hospitals have demonstrated CEA complication rates comparable to tertiary settings, but CAS in the community (as opposed to academic centers) does produce more complications.
As noted earlier, there was a “flurry” of strokes in the CEA group at the very end of the CREST-2 follow-up period. These are important because at that late juncture there is a reasonable chance that these strokes had nothing to do with the carotid at all but rather occurred as the result of other stroke mechanisms for which these patients are at risk. These strokes might have been caused by occult atrial fibrillation or small vessel ischemic disease rather than carotid stenosis in the 70% to 79% range.
Given the financial realities of our capitalism-driven medical system, will CREST-2 open the floodgates of patients undergoing CAS for asymptomatic carotid stenosis? There are many stakeholders who perform CAS, including vascular surgeons, neurosurgeons, cardiologists, and interventional radiologists. Given this competitive landscape, the fear is that the more circumspect approach outlined here will only be a neglected afterthought.
Alan Z. Segal, MD, is Associate Professor of Neurology, Weill Cornell Medicine.
References
1. North American Symptomatic Carotid Endarterectomy Trial Collaborators; Barnett HJM, Taylor DW, Haynes RB, et al. Beneficial effect of carotid endarterectomy in symptomatic patients with high-grade carotid stenosis. N Engl J Med. 1991;325(7):445-453
2. [No authors listed]. Endarterectomy for asymptomatic carotid artery stenosis. Executive Committee for the Asymptomatic Carotid Atherosclerosis Study. JAMA. 1995;273;(18):1421-1428.
3. Brott TG, Hobson RW 2nd, Howard G, et al. Stenting versus endarterectomy for treatment of carotid-artery stenosis. N Engl J Med. 2010;363(1):11-23.