EVA-3S: Stenting again inferior to endarterectomy in patients with symptomatic carotid stenosis

Susan Jeffrey

October 18, 2006

Boston, MA - Results of a randomized trial show that carotid stenting, compared with carotid endarterectomy, failed to meet noninferiority criteria in patients with symptomatic stenosis of 60% or more[1]. At both one and six months after the procedure, death and stroke rates were lower with endarterectomy.

These results, from the Endarterectomy versus Angioplasty in Patients with Symptomatic Severe Carotid Stenosis (EVA-3S) trial, are published in the October 19, 2006 issue of the New England Journal of Medicine.

The EVA-3S report appears one week after the publication of the Stent-Supported Percutaneous Angioplasty of the Carotid Artery versus Endarterectomy (SPACE) trial[2], in the October 7, 2006 issue of the Lancet, which also showed a slightly higher rate of ipsilateral ischemic stroke and death at 30 days in the patients undergoing carotid stenting than in those undergoing endarterectomy.

"The next step is to precisely identify the factors that are associated with an increased risk of stroke or death after carotid stenting," EVA-3S principal investigator Dr Jean-Louis Mas (Hôpital Sainte-Anne, Paris, France) said. These factors include the learning curve associated with carotid stenting for operators, the technique (including devices or adjuvant therapies), and factors related to the patient, such as the type of stenosis, the patient's age, and whether the plaque is unstable, he added.

"We will look a these factors in the EVA-3S database, but a large number of patients will be needed to perform meaningful subgroup analyses and to identify prognostic factors," Mas added. A meta-analysis on individual data from three European trials—EVA-3S, SPACE, and the ongoing ICSS trial—will be performed as soon as the data from ICSS are available, he noted.


EVA-3S was a multicenter, randomized, noninferiority trial comparing carotid stenting with endarterectomy in patients with symptomatic carotid stenosis of at least 60%. The trial was stopped after the enrollment of 527 patients, short of the target of 872 patients, for "reasons of both safety and futility," the authors write.

The 30-day incidence of any stroke or death, the primary end point of the trial, was significantly higher with stenting than with endarterectomy, with a relative risk of 2.5 for stenting vs the surgical procedure. "The absolute risk increase was 5.7%, suggesting that one additional stroke or death resulted when 17 patients underwent stenting rather than endarterectomy," they write, which is well short of the 2% noninferiority boundary.

The 30-day incidence of disabling stroke and death was also higher with stenting than with endarterectomy.

Carotid stenting vs endarterectomy in EVA-3S: Any stroke or death and any disabling stroke or death at 30 days

End point Endarterectomy (%) Carotid stenting (%) Relative risk (95% CI)
Any stroke or death 3.9 9.6 2.5 (1.2-5.1)
Disabling stroke or death 1.5 3.4 2.2 (0.7-7.2)

A greater proportion of strokes occurred on the day of the procedure in the stenting group than in the endarterectomy group, they note: 17 of 24 with stenting vs three of nine with carotid endarterectomy.

At six months, the incidence of any stroke or death continued to be lower with surgery (6.1% vs 11.7% after carotid stenting; p=0.02).

There were more major local complications after stenting, they note, and more systemic complications, mainly pulmonary complications, after surgery, but the differences were not significant. Cranial-nerve injury was more common after endarterectomy.

Not the final word

In an editorial accompanying the paper[3], Dr Anthony J Furlan (Clinic Foundation, OH) suggests that the jury is still out on the safety and efficacy of carotid stenting. Currently, the only FDA-approved indication for carotid stenting is in patients with high-grade symptomatic stenosis who are at high operative risk, based on results from the SAPPHIRE trial.

The benefits of carotid endarterectomy were established in the large NASCET and ACAS trials, in symptomatic and asymptomatic patients, respectively. However, endarterectomy had been performed for 30 years before those trials were done, whereas carotid stenting is still much earlier in its evolution, Furlan said. "What we need is a NASCET-like trial, but none of the studies of carotid stenting to date have been definitive," he said. "They're all flawed in one way or other."

For example, he pointed out, both SPACE and EVA-3S were stopped early, leaving them with smaller-than-anticipated sample sizes and therefore underpowered to provide definitive conclusions. In EVA-3S, different stents and cerebral-protection devices were used by the various participating centers. The eligibility criteria for participating interventionalists were different between trials, as were the antiplatelet regimens. "When these studies were designed, we didn't pay quite as much attention to these issues, but it might be important in terms of in-stent thrombosis."

"So you learn from each of these studies," Furlan concluded. "I think the (ongoing) CREST study is our last best chance of getting a definitive answer. The sample size is much larger [at about 2500 patients], it's a more standardized protocol where they're only using one stent and one embolization-protection device and vetting the interventionalists. I think it is more rigorous than in some of the other studies."

Enrollment has continued to be slow in CREST, as it is in many device trials, he acknowledged. However, he said, "it is hard to imagine launching another CREST-like study [because of] the difficulties they've had in that study with patient enrollment, so I would urge physicians that if they have patients like this, that they enroll them in CREST, so that we can . . . get a definitive answer."

The study was supported by a grant from the Programme Hospitalier de Recherche Clinique of the French Ministry of Health, Assistance Publique-Hôpitaux de Paris. Mas reported no conflict of interest related to this article. Disclosures for coauthors appear in the paper.


Comments on Medscape are moderated and should be professional in tone and on topic. You must declare any conflicts of interest related to your comments and responses. Please see our Commenting Guide for further information. We reserve the right to remove posts at our sole discretion.