Carotid Stenting, Endarterectomy Equally Effective at Preventing Severe Strokes in Symptomatic Patients

Download this article's Factoid (PDF & PPT for Gold Subscribers)

Long-term data from the largest trial of symptomatic patients randomized to carotid artery stenting (CAS) or carotid endarterectomy (CEA) show similar outcomes for fatal or disabling stroke, according to findings published online October 14, 2014, ahead of print in the Lancet.

Investigators for the International Carotid Stenting Study (ICSS), from which 30-day follow-up data was previously released, say that while CAS and CEA appear equally effective at preventing severe strokes that lead to death or disability, the increase in minor, nondisabling strokes with stenting should be considered and weighed against the risk of surgical complications such as procedural MI in the choice of treatment.

Researchers led by Martin M. Brown, MD, of the National Hospital for Neurology and Neurosurgery (London, England), randomly assigned 1,713 patients with symptomatic carotid stenosis to CAS (n = 855) or CEA (n = 858) at 50 centers around the world from May 2001 to October 2008. Baseline characteristics were similar between groups. Follow-up lasted a median of 4.2 years (IQR 3.0-5.2; maximum 10 years).

Five-year cumulative rates of fatal or disabling stroke—the primary endpoint—were similar in the CAS and CEA groups in the intention-to-treat analysis. However, the risks of any stroke and the combination of procedural stroke, procedural death, or ipsilateral stroke were both higher with CAS (P < .001 and P < .01, respectively). All-cause mortality was similar between groups (table 1).

Table 1. Outcomes at 5 Years



(n = 853)


(n = 857)

HR (95% CI)

Fatal or Disabling Stroke



1.06 (0.72-1.57)

Any Stroke



1.71 (1.28-2.30)

Procedural Stroke, Procedural Death, or Ipsilateral Stroke



1.72 (1.24-2.39)

All-Cause Death



1.17 (0.92-1.48)


 The 5-year cumulative risk of any stroke was also higher with CAS compared with CEA in the per-protocol population (8.9% vs 5.8%; P = 0.04). 

According to the study authors, the difference in the rate of any stroke was attributable mainly to an excess of nondisabling strokes (73 in the CAS group vs 27 in the CEA group). Exploratory subgroup analyses demonstrated that no baseline characteristics significantly altered the risk of the primary or secondary endpoints.

Additionally, there was no difference between the treatment groups in functional outcome as measured by modified Rankin scale scores at the end of follow-up or at either 1 or 5 years after randomization.

Training, Experience Addressed

Numerous other investigators have labeled ICSS “ethically questionable,” due to the minimal endovascular experience required of participating operators. ‘Experienced centers’ were required to have operators with at least 50 total stenting procedures, of which 10 had to be CAS cases. ‘Supervised centers’—where operators did not meet those requirements—could perform tutor-assisted procedures.

In the new report, Dr. Brown and colleagues say that the 2 investigators from separate centers whose cases were linked to an excess of adverse events faced consequences. The poor results led to both operators being barred from treating any further patients in the trial and the centers being suspended from further participation.

On the basis of the interim safety analysis showing no significant difference in risk between CAS and CEA in supervised (n = 199) and experienced centers (n = 1,511), the ICSS investigators chose to combine the results from all centers in the long-term analysis. The 5-year data again show no interaction related to the level of experience and the treatment effect of CAS vs CEA.

‘Uncertain’ Future for Carotid Stenting?

In a commentary accompanying the paper, Marco Roffi, MD, of University Hospital (Geneva, Switzerland), who has been a vocal critic of the safety and ethics of not only ICSS but EVA-3S and SPACE, adds that no difference was noted in patient outcomes after stratification by lifetime stenting experience of the operators at the time of the procedure (tertiles of 0-16 cases, 17-37 cases, and more than 37 cases).

“Although from a methodological point of view the conclusion is sound, it needs to be underlined that two-thirds of the patients in these trials were treated by operators with lifetime experience of 37 carotid stenting interventions or fewer at the time of the procedure,” Dr. Roffi writes. With no further studies randomly allocating symptomatic patients to CAS or CEA in sight, he concludes that “the future of carotid stenting is uncertain.”

But Dr. Brown and colleagues say the ICSS primary analysis demonstrates that “stenting is as effective as endarterectomy in preventing fatal or disabling stroke in patients with symptomatic carotid stenosis up to 10 years after treatment.” They add that the increased risk of minor nondisabling strokes associated with CAS “must be weighed against the increased risk of procedural myocardial infarction, cranial nerve palsy, and access-site hematoma associated with endarterectomy.”

Tying Up Loose Ends

In a telephone interview with TCTMD, William A. Gray, MD, of Columbia University Medical Center (New York, NY), said the ICSS report “recapitulates much of what we saw in CREST... in that there were a few more minor strokes in the stent group, but at 6 months there were no differences between groups, and major stroke and all-cause death were not different between stenting and surgery even early on.”

Overall, while the populations were different, the results should be considered “confirmatory” of the CREST findings since they are “more alike than different,” he said.

Dr. Gray noted that one limitation of the study not acknowledged by the researchers is that MI was not routinely measured as an endpoint, which he said is expected in the United States for device trials. “I think their ascertainment of MI was probably lower than what it otherwise would have been had they been measuring all along,” he said. “In CREST about 2 to 2.5% of all patients undergoing endarterectomy experienced MI compared with only about .5% in ICSS. It would have been interesting to see if that same kind of differential was seen.”

Additionally, Dr. Gray said the low use of embolic protection in ICSS (approximately 70%) “is not a small issue and should be noted, because it’s hard to justify that as ‘modern stenting’ technique.”

As to the “future of carotid stenting,” he said reimbursement issues—which are unrelated to the data—currently are the major concern for clinicians and patients in the United States.



1. Bonati LH, Dobson J, Featherstone RL, et al. Long-term outcomes after stenting versus endarterectomy for treatment of symptomatic carotid stenosis: the International Carotid Stenting Study (ICSS) randomized trial. Lancet. 2014;Epub ahead of print.

2. Roffi M. The struggle of carotid artery stenting [comment]. Lancet. 2014;Epub ahead of print.


  • The trial was funded by grants from the Medical Research Council and Stroke Association (both United Kingdom), Sanofi-Synthélabo, and the European Union.
  • Drs. Brown, Roffi, and Gray report no relevant conflicts of interest.

Related Stories:


We Recommend