Point/Counterpoint

William A. Gray, M.D. and Anthony J. Comerota, M.D.

Stents are best in high-risk patients. | Stents aren't proven to reduce stroke risk.

While some surgeons will argue that there is no such thing as a high-surgical risk patient, I think the data show that some patients are clearly at significant risk for undergoing surgery. Carotid stenting is the best treatment option in these high-risk patients, assuming no anatomical exclusions.

At particularly high risk with endarterectomy for extracranial carotid artery disease are those patients with a failed prior surgery, prior radiation, age over 75 years, congestive heart failure, coexisting coronary disease requiring bypass, renal failure, or contralateral carotid occlusion.

Studies consistently show that the risk of 30-day stroke and death in patients with some of these risk factors is doubled (up to 10% or more), compared with low-risk surgical patients.

For example, in landmark trials done at expert institutions by expert operators, such as the North American Symptomatic Carotid Endarterectomy Trial (NASCET), the risk of stroke and death at 30 days in patients with contralateral carotid occlusion was 14%--well above the 6% limit for symptomatic patients as set by an American Heart Association consensus panel.

Furthermore, some retrospective data suggest the in-hospital stroke and death rates are tripled in patients with renal insufficiency, coronary disease, and chronic obstructive pulmonary disease, compared with lower-risk patients.

It is also important to consider adverse outcomes other than stroke and death in high-risk patients.

In NASCET, there was a 9% combined incidence of infection and bleeding, and an 8.5% risk of cranial nerve palsy. Most--but not all--of these problems resolve, but some can be life threatening or altering.

Also, there was a nearly 10% 30-day risk of medical complications, most of which prolonged hospitalization. These kinds of complications are important to patients and cannot be disregarded.

Restenosis rates with surgery vs. stenting also have to be considered. Data from the Asymptomatic Carotid Atherosclerosis Study (ACAS) trial suggest that with endarterectomy, the restenosis rates are high at 5%-10%.

This was confirmed in the randomized Stenting and Angioplasty with Protection in Patients at High Risk for Endarterectomy (SAPPHIRE) trial, where the reintervention rate in the first year was 4% following carotid endarterectomy. In contradistinction, the reintervention rate following carotid artery stenting (CAS) was less than 1% and averages 1% per annum in 3-year data recently reported from the pivotal CAS trials.

Whereas carotid endarterectomy is a very effective tool in patients in selected at-risk populations, there is a major difference in outcomes between high- and low-risk surgical patients.

Neither randomized, controlled, nor multicenter data supporting its use in the high-risk surgical population exist, nor are there data comparing carotid endarterectomy with medical therapy.

We have a great deal of data, however, suggesting that stenting is effective in this high-risk population.

In more than 3,000 patients treated in at least seven multicenter controlled trials with at least 3 years of follow-up in the United States--that's more than the total number of patients in landmark trials establishing endarterectomy as a treatment for revascularization--the average major stroke and death rate was approximately 3%.

That's darn good, compared with an operation that would give you at least a 10% stroke and death rate, according to the available data.

And when one looks at direct comparisons made between surgery and stenting in the SAPPHIRE trial, there are significant differences found in favor of stenting, no difference found in stroke prevention, and effective stroke prevention (less than 1% stroke rate per year) out to 3 years with stenting.

Finally, when we look at how the data transfer to the real world, we see that outcomes with surgery are even poorer outside of trials--both in the trial hospitals and in other hospitals--with stroke and death rates that are two to three times those seen in the trials.

However, stenting has proved the ability to deliver this therapy outside the hospital setting without any loss of safety, as demonstrated in the CAPTURE (Carotid Acculink/Accunet Post Approval Trial to Uncover Rare Events) and CASES (Carotid Artery Stenting Education System) postmarket approval studies.

The data in favor of stenting are very compelling in the high-risk population. Comparable surgical data do not, never have, and never will exist.

Our job is not to do a procedure; it is to help the patient, so let's consider what we're doing to help the patient.

The real question is whether any procedure we do reduces the risk of stroke. In fact, there are no published data showing that carotid angioplasty and stenting do reduce the risk of stroke.

In fact, the currently available objective data show an unacceptably high morbidity with stenting, and the procedure remains unproven in the types of high-risk patients that Dr. Gray mentioned.

As for whether stenting or endarterectomy is best in these patients, it is important to consider the specific types of risks we're talking about.

If the risks that are being referred to include those of renal failure, high carotid lesions, prior radiation, contralateral occlusion, advanced age, and so on, then consider that we often safely operate on patients on dialysis, and high carotid lesions can safely be accessed with subluxation of the mandible--which provides good exposure and allows the operation to proceed as usual.

As for radiation arteritis, I'm not keen on operating on these patients, but data presented at the recent Vascular Society meeting in Philadelphia suggest that there's not much difference in outcomes over 3 years with stenting vs. surgery.

Other data presented at that meeting suggest that the recurrent restenosis rates that are seen in patients undergoing angioplasty and stenting are high, at 19%.

Treating those recurrent cases of stenosis with angioplasty and stenting does not appear to be better than treatment with carotid endarterectomy.

An 11-year follow-up in 66 asymptomatic patients with 50%-79% recurrent stenosis following carotid endarterectomy showed that only 6% progressed over time, and of 3 patients with 80% stenosis, 2 underwent reoperation after becoming symptomatic and 1 remained asymptomatic.

A separate report on the operative approach to recurrent carotid stenosis showed that the operative stroke and death rate as well as the 5- and 10-year stroke-free survival rates were very acceptable, and that the 1.5% of patients who had cranial nerve injury generally showed resolution of the condition within several days to a week.

The data suggest a strong trend favoring surgical intervention for recurrent restenosis and stroke free survival.

Evidently, most of the lesions found in high-risk patients are not better managed with angioplasty and stenting.

This is true in the older population as well.

Whereas aging patients are indeed at increased surgical risk, their exclusion from large trials of endarterectomy and stenting--which was pointed out as an indication of their risk--was not really due to any surgical risk factors but due to the fact that those patients weren't expected to live long enough to provide adequate follow-up data for a quality randomized trial.

Available data suggest stenting in patients over the age of 80 years is associated with an unacceptably high rate of stroke and death that is much higher than with carotid endarterectomy.

In patients enrolled in the ARCHeR (Acculink for Revascularization of Carotids in High Risk Patients) trial with atherosclerotic lesions--those lesions we're most concerned with--stenting was found to be associated with a 9.5% 30-day stroke and death rate. This is unacceptable.

In patients with renal disease on dialysis, stroke or death following carotid angioplasty and stenting was 29%. Angioplasty and stenting is certainly not doing these patients any favors.

In the SAPPHIRE trial, the stroke and death rates observed with angioplasty and stenting were found to be much better. But in this study the endarterectomy patients were found to actually do slightly better (5.8% vs. 4.6%).

In medically treated historical control patients over the same period, the risk was 0.9%, according to data from NASCET.

I would argue that any procedure offered to a patient should reduce the risk of stroke. There are no data showing that angioplasty and stenting does this.

If a procedure-related stroke and death rate is not less than that seen with best medical care, the procedure should not be offered.

Since it does not appear that carotid angioplasty and stenting is clearly better than best medical care, it should not be offered to high-risk surgical patients unless they are treated as part of a randomized trial.