Large Medicare Study Compares Carotid Stenting, Endarterectomy

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

Carotid artery stenting (CAS) yields in-hospital and 1-year stroke rates similar to those of carotid endarterectomy (CEA), according to a study of over 10,000 Medicare patients. And while mortality and myocardial infarction (MI) rates at 1 year are increased with CAS, that may be due to the patients’ higher baseline risk, according to results reported online May 26, 2011, ahead of print in Stroke.

Researchers led by Fen Wei Wang, MD, PhD, of Creighton University (Omaha, NE), looked at 10,958 Medicare patients identified from a 5% nationwide random sample who received CAS (n = 1,323) or CEA (n = 9,635) between 2004 and 2006. The majority of the patients in the study (87.5%) had asymptomatic disease. CAS procedures increased from 8.2% of carotid revascularizations in 2004 to 16.8% in 2006. CAS patients were older with a higher risk profile, including history of CAD or MI as well as more heart failure, renal failure, and peripheral vascular disease.

In-hospital outcomes were similar between the 2 groups (table 1).

Table 1. In-Hospital Death and Stroke


(n = 1,323)

(n = 9,635)

P Value









At 1 year, although stroke rates remained similar, mortality and MIs were increased with CAS, resulting in a higher combined rate of stroke/MI/death in patients receiving stenting (table 2).

Table 2. One-Year Outcomes


(n = 737)

(n = 6,724)

P Value




< 0.001








< 0.001




< 0.001

In addition, there was a higher rate of stroke at 1 year in symptomatic patients who received CAS compared with CEA (18.9% vs. 10.3%; P = 0.016). This was not true in asymptomatic patients (3.4% with CAS vs. 3.2% with CEA; P = 0.79).

On multivariable analysis, CAS was not an independent predictor of stroke, but the procedure was associated with increased risk for mortality (adjusted HR 1.32; 95% CI 1.02-1.71), MI (adjusted HR 1.54; 95% CI 1.06-2.25), and combined endpoints (adjusted HR 1.34; 95% CI 1.10-1.63).

Analyze and Reanalyze

To further adjust for potential confounders, the researchers performed a propensity analysis to account for those variables that were statistically unbalanced at baseline between CAS and CEA patients. The results were the same as with the traditional multivariable analysis. Dr. Wang and colleagues then performed an additional analysis, a sensitivity study to estimate the potential effect of unmeasured confounders on the study outcomes with regard to mortality and MI.

The researchers used different estimates of an unmeasured confounding factor that ranged in prevalence from 40% to 60% in the CAS group with an effect size ranging from small (10%) to large (50%). They found that an unknown risk factor with either a very small effect on mortality and somewhat large difference in prevalence or with a large effect and a small difference in prevalence could render the difference in mortality between CAS and CEA no longer significant. For MI, this would hold true if there were both a larger effect size and a large difference in prevalence for an unmeasured confounder.

On a final interaction analysis, Dr. Wang and colleagues found no significant interactions between age, gender, asymptomatic vs. symptomatic disease presentation, or treatment choice (CAS vs. CEA) on any of the 1-year outcomes studied.

Comorbidities a Likely Culprit

“CAS [and CEA] patients had similar in-hospital postoperative stroke and in-hospital mortality rates,” the researchers conclude. “By 1 year, CAS patients had a similar stroke risk but an increased mortality rate compared with CEA, likely because of increased baseline mortality risk.”

They note that while differences in the 2 procedures may play a role in the mortality difference at 1 year, other factors argue against that. For instance, “if stent implantation per se were associated with increased mortality, then the expectation would be that late stent thrombosis and fatal stroke would be the putative mortality mechanism,” they write. “This possibility is unlikely insofar as nonfatal stroke was similar in the 2 groups. Further, the incidence of CAS late stent thrombosis reported in the literature is quite low. More likely is the possibility that CAS patients were at higher baseline risk for mortality.”

According to William A. Gray, MD, of Columbia University Medical Center (New York, NY), the study “is an analysis that needed to be done in terms of not only identifying the differences in baseline characteristics between stenting and surgery in that era, but also what effect those baseline differences have on longer-term mortality and clinical outcomes.”

In a telephone interview with TCTMD, Dr. Gray added that the study supports “a reasonable premise,” which is that “short-term outcomes are not materially different between the 2 procedures given that there’s clearly a difference in the stent population in terms of comorbidities. It’s remarkable, actually, that there’s no difference.”

And at 1 year, “we do see higher mortality rates,” he said. “Again, not surprising given that these are higher-risk patients who have significant comorbidities.”

Medical Therapy a Better Choice for Some Patients?

Dr. Gray added that the study must be taken in context. For example, the time frame of 2004 to 2006 represents the beginning of the postmarket period for CAS, whereas results assessed today would be expected to be improved. In addition, “we know that stenting does very well in patients who are younger,” he said. “So in a Medicare population of patients aged 66 years or older, we’ve excluded patients who probably benefit the most from stenting as opposed to surgery.”

Dr. Gray also commented on the fact that the majority of patients in the study were asymptomatic, noting that perhaps in some cases no revascularization at all is the more appropriate choice. “If you’re asymptomatic and you’re in this group, it’s reasonable to ask what else we should be doing,” he said. “I think that’s a fair question.”

Dr. Wang and colleagues agree, noting that “state-of-the-art intensive medical treatment without revascularization may be an appropriate therapeutic option in some patients . . . but this hypothesis requires formal testing.”

For instance, such an approach might be appropriate for a patient with significant comorbidities, Dr. Gray noted. “You would counsel them and say, ‘Look, if you have a 90% lesion and you’re asymptomatic, but you’re diabetic and in renal failure, the risk of revascularization might not be justified based on your longer-term outcomes,’” he said. “The patient and the doctor have to make that decision together because it takes at least 3 years for the benefit of revascularization to play out. We know that from the surgical groups, and it’s probably true for stenting, too.”


Wang FW, Esterbrooks D, Kuo Y-F, et al. Outcomes after carotid artery stenting and endarterectomy in the Medicare population. Stroke. 2011;Epub ahead of print.



Related Stories:

Jason R. Kahn, the former News Editor of TCTMD, worked at CRF for 11 years until his death in 2014…

Read Full Bio
  • Drs. Wang and Gray report no relevant conflicts of interest.