Carotid Endarterectomy is Better than Carotid Artery Stenting for Asymptomatic Patients

I have no financial relationships to disclose relative to the issues of this debate; however, I do have a significant historical interest in carotid artery stenting. We wrote the 1st peer-reviewed article on carotid artery stenting in 1996.¹ At that time, we had an event rate of 10.6%, and now everybody around the world refers to Dr. Diethrich and his 10.6% event rate. I'm not particularly proud of it; we've managed to do better over the years. We also did the 1st televised case in the United States of carotid artery stenting; that was over 11 years ago.

Currently, about 75% of patients undergoing carotid artery stenting or carotid endarterectomy in the United States are asymptomatic. The same appears to be true in Europe, or at least in Italy, Germany, and France. Now ask yourself, “What is the primary purpose of any intervention in the carotid artery bifurcation?” It's the prevention of stroke. So how good are we at preventing stroke in primary asymptomatic patients?

From the outset, I would like to say that there is no level I evidence to support the contention that carotid artery stenting is better than endarterectomy or, for that matter, better than “best medical care.” What is the evidence, really? Most frequently cited (and I'm sure by my opponent) is the SAPPHIRE trial (Stenting and Angioplasty with Protection in Patients at High Risk for Endarterectomy), a randomized trial comparing endarterectomy and stenting in high-risk symptomatic and asymptomatic patients.

In this randomized trial, the 30-day death, stroke, and myocardial infarction rate in the stent group was 7.6%, and in the carotid endarterectomy group it was 11.2%. When those data were presented at the Society of Vascular Surgery, a very well-known surgeon from the United Kingdom, Sir Peter Bells, stood up and said, “Well, who exactly are those surgeons doing the carotid endarterectomy in this trial —orthopedic surgeons?”

A total of 740 patients were entered in the trial; 334 were randomized to the trial, and 406 were put in the registry. In other words, 55% of the patients who were entered and enrolled ended up in the registry, ostensibly because those patients were at “too high a risk for endarterectomy.” Since the purpose of the intervention is the prevention of stroke, I have to ask: If you were asymptomatic, would you accept the risk of either one of the 2 arms? I personally wouldn't.

It's also important to remember that what drove these outcomes was not necessarily neurologic events, but non-Q wave myocardial infarction (MI). The study had only 740 patients, so obviously it had virtually no power to detect true neurologic events. A non-Q wave MI was added to power the study, and that's why it is so distorted. A fairly famous vascular surgeon was quoted earlier this year as saying, “The evolution of SAPPHIRE is interesting but unfortunately flawed”; that is a common view among vascular surgeons today.

Since the time of the 1st carotid endarterectomy (done either by Dr. DeBakey down the street, or by Dr. Eastcott in London), the indication for endarterectomy (and now, for carotid artery stenting) has had its basis in the percentage of stenosis at the bifurcation. But are we in fact treating the wrong lesions? Are our selection criteria faulty? Do we need to change our thinking—particularly in asymptomatic patients?

We've been doing vein surgery for 100 years. When I learned vein surgery, we had to strip out the vein, make an incision in the groin, and locate and name all 5 tributaries. It was a big deal. What do we do today? We do this as an endovascular procedure. It's an outpatient procedure, and we never go into the groin, and never ligate the vessels. For over 100 years, we'd probably been doing the wrong thing. We didn't have the right idea, and we didn't get good results. So perhaps it is time for us to get away from this idea about percentage of stenosis as an indicator for endarterectomies—particularly in asymptomatic patients.

We already learned about “vulnerable plaque” from our coronary studies. Plaque characterization is extremely important in stratifying patients to decrease the possibility of embolization or neurologic events. Echolucent plaques are much more likely to result in an adverse neurologic event. In carotid lesions, are calcified plaques just waiting to embolize? Are we even looking at the morphology of the plaque in patients who undergo stenting? The answer to that is definitely no—we have no idea what these plaques look like. In the ACAS (Asymptomatic Carotid Atherosclerosis Study), there was no association between the stroke rate and the severity of the stenosis.

I'd like to make 2 critical points. First, the identification of a given degree of stenosis alone is no longer sufficient to identify patients at high risk for developing an acute cerebral event. Second, the risk of rupture is much more closely related to the morphologic characteristics of the plaque than to the degree of stenosis.

We have just had a study approved by the Food and Drug Administration (FDA) to look at morphologic characteristics, using intravascular ultrasound “virtual histology” to more precisely characterize the plaque and identify areas at risk of subsequent embolization and neurologic events.

I'm going to conclude with a provocative statement. I believe that, as of now, there has not been a study legitimately comparing the best medical care versus carotid artery stenting. There has been a study comparing stenting to endarterectomy, but it did not include modern angiotensin-converting enzyme inhibitors, 2nd-generation platelet inhibitors, current antihypertensive drugs, and other advanced pharmacotherapy. Comparing current invasive treatment options (such as carotid artery stenting and endarterectomy) with the “best medical care” available over a decade ago is absolutely absurd.

Fortunately, the Transatlantic Carotid Interventional Trial (TACIT) will be getting under way very soon. This is the 1st study that will look at endarterectomy, carotid stenting, and modern-day medical treatment in a 3-arm study. Is the risk of intervention greater than the risk of events with medical therapy alone? We really need to answer that question.

Two other trials are ongoing: ACT I (Carotid Angioplasty and Stenting Versus Endarterectomy in Asymptomatic Subjects With Significant Extracranial Carotid Occlusive Disease - Asymptomatic Carotid Stenosis, Stenting Versus Endarterectomy Trial) and CREST (Carotid Revascularization Endarterectomy Versus Stenting). Unfortunately, neither of these trials has a pure medical therapy arm, which, in today's advanced pharmaceutical arena, is essential.

So, what is the answer to the question of asymptomatic carotid artery stenting? An adequately powered randomized trial needs to be done. And until we have adequately powered randomized data, I seriously doubt that reimbursement for the broader application of carotid artery stenting in asymptomatic patients will be approved.

Ladies and gentlemen, for this debate, I rest my case. Until we have real data to the contrary, carotid stenting needs to be viewed as guilty until proved innocent, and it has certainly not been proved innocent.