Selection of Patients for Carotid Endarterectomy versus Carotid Artery Stenting

Carotid Stenting Versus Endarterectomy: Major Trial Data

The highest-quality data accrued on the comparison of CAS with CEA is based on a small number of well-conducted randomized trials that have been completed during the last few years. However, selecting the ideal patient population who may be best treated with CAS remains difficult. Data from four trials conducted with perhaps the best methodology and lack of bias deserve discussion.

The CREST study, a randomized trial designed to compare the efficacy of stenting in patients with average surgical risk, was completed in 2008. It included 2502 patients, 53% of whom were symptomatic, with a statistical power to detect small differences between CAS and CEA. The trial’s primary endpoint was the combined incidence of stroke, myocardial infarction (MI), and death from any cause. Secondary endpoints included incidence of restenosis, post-procedure morbidity, quality of life, cost effectiveness, and outcomes among different subgroups considered at high risk. To compare the best possible stenting with the best quality surgery, the CREST trial included a strict accreditation phase designed to eliminate the learning curve associated with the interventional technique for CAS and to select surgeons with a documented low complication rate with CEA.

At a median of 2.5 years of follow-up, the incidence of the composite endpoint of death, MI, and stroke was not different between CAS and CEA (7.2% vs. 6.8%; hazard ratio [HR], 1.11; p = .51), with no difference in the treatment effect based on symptomatic status or gender. The lack of difference in the composite primary endpoint between the two groups was mainly produced by the higher incidence of MI in the endarterectomy group (1.1% vs. 2.3%; p = .03). On the other hand, the incidence of any stroke within 30 days after the procedure was significantly higher in the CAS group (4.1% vs. 2.3%; p = .01), but there was no difference in the incidence of major stroke between the groups (0.9% for CAS vs. 0.7% for CEA).

An important finding of the CREST trial was that the impact on quality of life at 1 year of intervention was greatly reduced in patients with major or minor stroke but not in those with MI. This may be caused by the use of biomarkers in the diagnosis of myocardial ischemia that included many myocardial events of questionable clinical significance, inflating the incidence of MI. The difference in incidence of MI between CREST and other trials is particularly striking when compared with the International Carotid Stenting Study (ICSS) published in Lancet in 2010, in which there was no difference in the incidence in MI rates (0.35% for CAS vs. 0.46% for CEA), but with a greater fatality for MI on the CAS group.

Consistent with previous trials, age appeared to influence treatment outcomes. In CREST, there was a slight difference in outcomes at 4 years of follow-up, with patients older than 70 years faring better with CEA, and patients younger than 70 years faring better with CAS.

The EVA-3S, SPACE, and ICSS trials were conducted in France, Germany, and Great Britain, respectively, and their results were published between 2006 and 2010. These three trials included only patients with symptomatic carotid artery disease. Proponents of CAS criticized the requirements of these three trials for qualification of physicians to conduct carotid intervention. They argued that the experience of those performing carotid stenting in these trials was insufficient, and that the higher neurologic complication rates of CAS compared with CEA reflected in the trials was a result of interventional inexperience. Although the small numbers of CAS procedures required for trial participation was in our opinion a genuine concern, there was no difference in neurologic complication rates between the experienced and inexperienced interventionalists participating in those trials, and therefore inexperience did not appear to be the reason for the difference in results.

In the EVA-3S trial, at 4 years of follow-up, the cumulative probability of procedural stroke or death and nonprocedural stroke was higher in the CAS group (11.1 vs. 6.2; p = .03). However, this difference was derived mainly from the 30-day periprocedural complication rates, and after that period the risk of ipsilateral stroke was equivalently low for both procedures.

The SPACE trial, at 2 years of follow-up, found no significant difference in the incidence of the composite endpoint of ipsilateral stroke and periprocedural stroke or death between CAS and CEA (9.5% vs. 8.8%; p = .31). During follow-up, the incidence of greater than 70% recurrent stenosis, as assessed by ultrasound, was significantly higher with CAS (10.7% vs. 4.6%; p = .0007), but there was no modification in velocity criteria used for the assessment of intrastent stenosis. Therefore, it is likely that the stent restenosis rate was overestimated.

Finally, the ICSS trial early results published in 2011 revealed a higher incidence of the combined endpoint of stroke, death, and MI at 4 months from randomization in the CAS group compared with CEA (8.5% vs. 5.2; p = .006). Interestingly, and in contradiction to the CREST trial, the ICSS trial found no difference in the incidence of MI between the two treatment groups, but there were three deaths from MI in the CAS group, whereas no cardiac-related deaths occurred in patients treated with CEA.

A preplanned meta-analysis of the early results (120 days from inclusion in the trial) of these three trials including a total of 3433 patients found that the incidence of any stroke or death in patients undergoing CAS was significantly higher than that for CEA (8.9% vs. 5.8%; p = .0006). This meta-analysis suggested that the difference in early outcomes is derived from the twofold higher rate of complications for CAS that occurred in patients 70 years of age or older, compared with those undergoing CEA in the same age group (12% vs. 5.9%; p = .0053). The meta-analysis found no difference in the risk of stroke or death between CAS and CEA in patients younger than 70 years.