Carotid Endarterectomy: Who Needs It? Who Doesn't?

Although it is a potentially valuable prevention measure for ischemic brain infarction, carotid endarterectomy does carry serious risks. Careful attention to the guidelines discussed here will help you determine which patients will benefit most from this surgical procedure.

By Seemant Chaturvedi, MD

Dr. Chaturvedi is associate professor of neurology and associate director of the stroke program at Wayne State University in Detroit, Michigan.

Cerebrovascular disease is the third leading cause of death and a leading cause of disability among adults in the United States. Currently in this country, approximately 700,000 new strokes occur each year, and there are about 4 million stroke survivors. Those numbers are certain to grow as the population ages. Roughly one third of persons who have had a stroke require assistance with daily activities.

Carotid endarterectomy (CE) is a potentially valuable measure to prevent the most common type of stroke, ischemic brain infarction, which accounts for 80% of all strokes. However, because CE also carries serious risks, the clinician must clearly understand the guidelines when determining which patients will benefit most from this surgical procedure.

This article will review the key clinical points that will help you select appropriate candidates for CE, and it will discuss the strengths and weaknesses associated with various carotid artery imaging techniques.

back to top

Underlying Pathologic Condition

The underlying pathologic condition for which CE is performed is proximal internal carotid artery (ICA) atherosclerosis. The proximal ICA is the site at which atherosclerotic disease most often develops in patients with cerebrovascular disease. The second most common site in the carotid system is located distally, in the so-called carotid siphon.

Risk factors for proximal ICA disease include the usual suspects-hypertension, cigarette smoking, diabetes mellitus, and hypercholesterolemia. Of course, age itself predisposes one to atherosclerotic disease, as does a history of stroke.

Carotid territory strokes typically occur in response to the formation of artery-to-artery emboli or to hemodynamic insufficiency. In the first instance, a small plaque from the proximal ICA breaks off and occludes a distal small or medium-sized vessel, such as the middle cerebral artery (MCA) or one of its branches. In the second instance, progressive carotid stenosis decreases regional cerebral blood flow to the point where transient or permanent cerebral ischemia results.

back to top

Four Clinical Presentations

Carotid artery disease typically appears in one of four manifestations: as a large, disabling stroke; as a small, nondisabling stroke; as a transient ischemic attack (TIA) only; or with no symptoms at all (asymptomatic carotid stenosis). The risk/benefit ratio for CE varies significantly with each of these four different possibilities. Therefore, it is important to recognize what is and is not true carotid symptomatology.

The major symptoms of carotid disease are transient visual loss in one eye, contralateral weakness or numbness, dysphasia in the dominant hemisphere, and sensory neglect in the nondominant hemisphere (see table, below). These symptoms are caused by either retinal or hemispheric ischemia. It should be noted that vascular-related binocular visual loss would not develop from blockage of only one carotid artery. Therefore, carotid stenosis in a patient with binocular visual loss should be considered asymptomatic.

Similarly, patients with carotid disease who have insufficient blood flow to the vertebrobasilar system in the back of the brain should be regarded as asymptomatic in terms of their carotid disease. Classic symptoms of vertebrobasilar disease are vertigo, facial numbness, dysarthria, diplopia, and ataxic gait (see table, above). Finally, there are many patients with carotid disease who have nonspecific or vague symptoms, such as lightheadedness, giddiness, or isolated headache. These patients should not be considered to have symptomatic carotid disease.

If a patient presents with symptoms of a potential carotid TIA, such as weakness on one side of the body and speech difficulty that have lasted for 30 minutes, other diagnostic possibilities must be considered. The most relevant conditions would be a focal seizure; a migrainous process, especially in young women; and a structural lesion, including brain tumors or demyelinating plaques, which occasionally produce acute symptoms.

Key points in a patient's history include the length and frequency of the episodes, the presence of a headache or any confusion or alteration in consciousness during the episodes, and associated symptoms. If symptoms such as tingling, pain, or sparkling lights are present, they would suggest an excitatory brain process such as an epileptiform event or a migraine. Carotid TIAs typically last 2 to 30 minutes. If altered consciousness has occurred, seizure should be considered as the cause, because unilateral carotid disease typically will not cause drowsiness or syncope.

back to top

Diagnositic Testing

The initial screening test that is most often used in evaluating patients for possible carotid stenosis is duplex ultrasound evaluation. The results of this test are usually given as a range of stenosis; for example, 1% to 15%, 16% to 49%, 50% to 79%, or 80% to 99%.

Duplex ultrasound examination does have certain limitations. It can falsely indicate a patent vessel as totally occluded. The accuracy of the test result depends greatly on the skill of the sonographer. In addition, because the quality of ultrasound laboratories can vary, an accredited laboratory should always be used.

Magnetic resonance angiography is another noninvasive test that can be used to assess both the extracranial and intracranial cerebral vasculature. However, compared with conventional cerebral angiography, this test tends to overestimate the degree of stenosis in the neck. It can also be inaccurate in cases of flow turbulence or vessel tortuosity.

The major clinical trials of CE that have been conducted to date have relied on conventional angiography to determine the degree of stenosis. Although this procedure is invasive and potentially hazardous-it carries a 1% risk of causing iatrogenic stroke-I still favor angiography, performed by experienced radiologists, to guide me in making surgical decisions. Some clinicians have argued that such decisions can be based solely on the results of noninvasive imaging methods. From an evidence-based point of view, however, clinical decision-making is best supported by angiographic images.

For example, what would your decision be in the case of a patient with evidence on duplex ultrasound of stenosis in the range of 50% to 79% and who has had a recent episode of transient monocular visual loss? Clinical trials indicate that patients with retinal TIA benefit from CE only if angiography reveals stenosis greater than 70%. The range of 50% to 79%, as determined by duplex ultrasound evaluation, is too broad to be useful in this case.

back to top

Results of Nascet Trial

What would you do for a patient with definite carotid symptoms and at least 50% carotid stenosis on the side that would produce those symptoms? How should you proceed? A review of the results of the major clinical trials of CE for symptomatic patients will help you make this determination.

The North American Symptomatic Carotid Endarterectomy Trial (NASCET) enrolled patients who had carotid territory TIA or minor stroke and randomly assigned them to receive "best medical therapy" or "best medical therapy plus CE" (New England Journal of Medicine, vol. 325, p. 445, 1991). Patients were eligible if they had stenosis ranging from 30% to 99% on conventional angiography and if they had symptoms within the previous six months. Included among the several exclusion criteria were a known source of cardioembolism, such as atrial fibrillation, and a life expectancy of less than five years.

The most important finding of NASCET was that patients derived significant benefit from CE if they had recent symptoms and stenosis greater than 70%. In this group, the absolute risk of ipsilateral stroke at two years was reduced by 17%, corresponding to a number needed to treat of six patients. However, only certain patients with stenosis ranging from 50% to 69%, such as men and patients who suffered hemispheric events, benefited from the procedure, and the benefit was much more modest-at five years, the reduction of absolute risk was 6.5%. Women and patients with retinal ischemia in this stenosis range did not benefit. No benefit was noted for patients with less than 50% stenosis. Results of the European Carotid Surgery Trial also found that not all patients benefit equally from CE (The Lancet, vol. 354, p. 2165, 1999).

The question as to why women with stenosis in the 50% to 69% range do not benefit from CE is controversial. Some early studies found that perioperative complication rates from CE were higher among women, but NASCET did not confirm this finding. In NASCET, women receiving medical therapy alone appeared to have fewer strokes during long-term follow-up evaluation.

back to top

Nature of the Symptoms

The nature of a patient's clinical symptoms is also an important factor to consider when determining his or her eligibility for CE. Patients with symptoms of a hemispheric stroke or TIA, such as contralateral weakness or speech difficulty, have more to gain from CE than patients with symptoms of a retinal ischemic event, such as transient monocular visual loss (see table, below). In patients in NASCET who had severe stenosis (70% to 99%), the two-year stroke risk in the medical therapy group was 44% for those who had suffered a hemispheric event and only 17% for those who had a retinal event.

The NASCET study also compared patient outcomes in terms of lacunar versus nonlacunar infarcts. Lacunar infarcts are strokes that result from the occlusion of small, penetrating blood vessels deep in the brain. Clinically, these infarcts can be identified by the restricted nature of the patient's symptoms-only weakness may be present, for example-and the absence of cortical symptoms such as dysphasia.

In patients who have a lacunar infarct and coexisting moderate-to-severe carotid stenosis on the same side, it is still not clear whether the carotid narrowing is causative or incidental. In NASCET, patients with lacunar infarcts were eligible for the study if they had moderate to severe carotid stenosis on the same side. Although the study did not clear up the controversy regarding the relevance of carotid stenosis to lacunar infarcts, it did find that patients with such infarcts do not benefit as much from CE as do patients who have nonlacunar infarcts. Among patients with lacunar infarcts, the reduction of relative risk associated with CE was 35%; in those with nonlacunar infarcts, it was 61%.

back to top

Important Prognostic Information

Cerebral angiography can provide important prognostic information that may help you determine whether a patient requires CE or only medical therapy. This imaging technique can reveal morphologic characteristics of the plaque, for example, which is a subject of long-standing interest.

To what degree do plaque characteristics, in addition to the degree of stenosis, influence future stroke risk? The NASCET results revealed that in patients with severe stenosis (70% to 99%), the risk of stroke associated with an ulcerated carotid lesion, as demonstrated on angiography, is two times higher than that associated with a plaque with a smooth contour.

The adequacy of collateral circulation can also be evaluated on angiography. The results of NASCET revealed that in patients with severe carotid stenosis who were undergoing medical therapy, the two-year risk of hemispheric stroke was 27.8% in those patients without collateral flow versus 11.3% in those with collateral flow. This latter group also had fewer perioperative strokes than did the group without collateral flow, although the difference was not significant (1.1% versus 4.9%, respectively). On balance, the absolute benefit from CE was greater for patients without collateral flow.

It is important to remember that subgroup analyses from NASCET or any other clinical trial should be viewed cautiously. However, some of the points mentioned above, such as the greater stroke risk with hemispheric events, have been confirmed in the European Carotid Surgery Trial.

Finally, in your determination of a patient's eligibility for CE, you should consider his or her age and coexisting conditions. Several studies have shown that perioperative mortality rates are higher among elderly patients receiving CE. In a study of more then 100,000 Medicare beneficiaries in the United States, the perioperative death rate among patients undergoing CE was 1.9% at average-volume hospitals and 3.6% among patients older than 84 years of age. That 1.9% death rate at the average-volume hospital is almost twice as high as the 1.1% perioperative mortality rate in the NASCET study, a discrepancy that raises the question of whether clinical trial results can be translated into routine daily practice. Obviously, in deciding whether to recommend CE to a patient, the intended surgeon's performance history and experience are critical.

back to top

Evaluating Asymptomatic Patients

As mentioned earlier, there are four clinical presentations of carotid stenosis. The first is a large, disabling stroke, such as that seen in a patient with complete inability to speak and hemiplegia. In such a case, there would be little to gain in performing CE, as the purpose of the procedure is to prevent a stroke. The next two presentations, minor strokes and TIAs, were addressed by the NASCET study.

The final presentation is demonstrated by the patient who has carotid stenosis but either no symptoms of a stroke syndrome or only nonspecific symptoms such as vertigo, binocular visual loss, and the others previously noted. The most important point to keep in mind is that the subsequent risk of stroke in asymptomatic patients is much lower than it is for symptomatic patients. Studies have demonstrated that for patients who have asymptomatic stenosis ranging from 60% to 99%, the risk for future ipsilateral strokes is 2.0% to 2.5% per year. In contrast, the risk for symptomatic patients is 10% to 15%. Assuming the perioperative stroke and death rate associated with CE at your institution is approximately 6%, it follows that the risk/benefit ratio for asymptomatic patients is much less favorable.

Nevertheless, because a substantial number of patients undergoing CE in the United States are asymptomatic, the clinical trial data in this population should be reviewed. The Veterans Affairs Asymptomatic Study included 444 male patients and found no benefit in performing CE to prevent stroke in asymptomatic men (New England Journal of Medicine, vol. 328, p. 221, 1993).

The Asymptomatic Carotid Atherosclerosis Study (ACAS) randomly assigned 1662 patients who had asymptomatic carotid stenosis measuring at least 60% to undergo medical treatment or CE. At five years, the risk of stroke was 11.0% in the medical treatment group and 5.1% in the surgery group (JAMA, vol. 273, p. 1421, 1995). This absolute risk reduction of 5.9% over five years translates into only a 1.2% annual reduction.

In other words, patients who have asymptomatic carotid stenosis measuring 80% to 99% have a 98% chance of being stroke free one year after undergoing medical therapy. If CE is chosen, the patient will have a 99% chance of being stroke free. Many authorities have used these findings to argue that CE does not provide clinically meaningful stroke prevention in asymptomatic patients, because those persons already have such a low risk of stroke.

In addition, the ACAS results were based on excellent surgical performance; the perioperative death rate was 0.1% and the stroke and death rate was 2.6% (with intention to treat). In fact, a major conclusion of the study was that CE can be effective for asymptomatic patients only if the surgical stroke and death rate is kept below 3%. Whether that goal is realistic at most North American hospitals is questionable. In the recently published Aspirin and Carotid Endarterectomy (ACE) trial, in which most of the centers had very experienced surgical teams, the stroke and death rate among asymptomatic patients was 4.6% (The Lancet, vol. 353, p. 2179, 1999). The old saying that statistical significance does not always equate with clinical significance is worth remembering when evaluating patients who have asymptomatic carotid disease.

However, such patients do require aggressive medical management. They should be given antihypertensive agents, lipid- and homocysteine-lowering agents, and antiplatelet medications, when indicated. A smoking cessation program, if applicable, is also appropriate.

back to top

Ongoing Study

One ongoing study is addressing the asymptomatic carotid stenosis issue-the Asymptomatic Carotid Surgery Trial. The design of this study is similar to that of the earlier CE trials, in which patients are randomly assigned to undergo either surgery or medical treatment. One major difference is the large sample size (3200 patients), which will allow greater exploration for subgroup analyses (European Journal of Vascular Surgery, vol. 8, p. 703, 1994). Another is the study's evaluation of the relevance of ultrasound-based plaque morphology-specifically, the proportion of the plaque that is echolucent-in determining future stroke risk.

One subgroup analysis that will be of great clinical interest concerns the appraisal of the difference in benefit that men and women receive from CE. In the ACAS study, asymptomatic women benefited less from the procedure than men did. If the new trial confirms this finding, the result will be noteworthy.

Suggested Reading Barnett HJM, et al.: Benefit of carotid endarterectomy in patients with symptomatic moderate or severe stenosis. N Engl J Med 339:1415, 1998. Barnett HJM, et al.: Do the facts and figures warrant a 10-fold increase in the performance of carotid endarterectomy on asymptomatic patients? Neurology 46:603, 1996. Henderson RD, et al.: Angiographically defined collateral circulation and risk of stroke in patients with severe carotid stenosis. Stroke 31:128, 2000. Inzitari D, et al.: The causes and risk of stroke in patients with asymptomatic internal carotid artery stenosis. N Engl J Med 342:1693, 2000. Inzitari D, et al.: Risk factors and outcome of patients with carotid artery stenosis presenting with lacunar stroke. Neurology 54:660, 2000. Silvestrini M, et al.: Impaired cerebral vasoreactivity and risk of stroke in patients with asymptomatic carotid stenosis. JAMA 283:2122, 2000. Streifler JE, et al. The risk of stroke in patients with first-ever retinal vs. hemispheric transient ischemic attacks and high-grade carotid stenosis. Arch Neurol 52:246, 1995. Wennberg DE, et al.: Variation in carotid endarterectomy mortality in the Medicare population. JAMA 279:1278, 1998.

Printable version of this article

E-Mail this Article