Carotid My In Symptomatic Patients

Carotid endarterectomy in symptomatic patients

INTRODUCTION — The proximal internal carotid artery and the carotid bifurcation are the locations most frequently affected by carotid atherosclerosis. Progression of atheromatous plaque at the carotid bifurcation results in luminal narrowing, often accompanied by ulceration. This process can lead to ischemic stroke or transient ischemic attack (TIA) from hemodynamic compromise, embolization, or thrombosis. Since the first carotid endarterectomy (CEA) was performed in 1954 [1], there has been a dramatic rise in the implementation of this procedure as a prophylaxis against stroke. As will be discussed below, CEA is indicated in some patients with clinically significant carotid atherosclerosis who have had evidence of nondisabling strokes and TIAs and in some who are asymptomatic. This review will discuss the major trials that have evaluated the efficacy of CEA in symptomatic carotid disease and the important clinical variables that impact the benefit and risk of CEA. Treatment with CEA for asymptomatic carotid stenosis is discussed separately. (See "Carotid endarterectomy in asymptomatic patients"). The preoperative evaluation, surgical technique, and complications of CEA are discussed separately. Other aspects of carotid occlusive disease, including risk factors, clinical manifestations, and diagnosis are also discussed separately. (See "Carotid endarterectomy: Preoperative evaluation; surgical technique; and complications", see "Secondary prevention of stroke: Risk factor reduction", see "Pathophysiology of symptoms from carotid atherosclerosis" and see "Evaluation of carotid artery stenosis"). EFFICACY — Two major trials, the North American Symptomatic Carotid Endarterectomy Trial (NASCET) and the European Carotid Surgery Trial (ECST), have demonstrated the efficacy of carotid endarterectomy (CEA) in selected patients who have carotid atherosclerosis and symptoms such as a transient ischemic attack (TIA) or a nondisabling stroke [2,3]. A third major trial, The Veterans Affairs Cooperative Trial 309 [4], was stopped prematurely because results from the NASCET were announced. Definition of symptomatic disease — Symptomatic carotid disease, as identified by the major clinical trials, is defined as focal ischemic symptoms that are referable to the appropriate carotid artery distribution, including one or more transient ischemic attacks characterized by focal neurologic dysfunction or transient monocular blindness, or one or more minor (nondisabling) ischemic strokes [2]. In addition, the definition is contingent on the occurrence of carotid symptoms within the previous four (NASCET) to six (ECST) months [2,3]. Thus, although there is no precise time limitation, remote carotid symptoms should not be considered as indicative of "symptomatic" carotid disease. Vertigo and syncope are not generally caused by carotid stenosis. Therefore, patients with these symptoms in isolation should be considered as asymptomatic with regard to carotid disease even if they are found to have carotid artery stenosis. Method of stenosis measurement — Comparison of the two major trials, NASCET and ECST, requires an understanding of how carotid artery stenosis was measured since these studies used different methodologies (show figure 1) [5]. •

NASCET measured the residual lumen diameter at the most stenotic portion of the vessel and compared this to the lumen diameter in the normal internal carotid artery distal to the stenosis [6].

•

ECST measured the lumen diameter at the most stenotic portion of the vessel, but compared this to the estimated probable original diameter at the site of maximum stenosis [3].

The maximum stenosis is generally in the carotid bulb, a wider portion of the artery than the distal segment. As a result, a given stenosis would be more severe using the ECST method compared with that of NASCET. ECST methodology also requires an assumption of the true lumen, which increases the risk of interobserver variability. Despite these differences, the results of different methods have a nearly linear relationship to each other and provide data of similar prognostic value. A 50 percent stenosis with the NASCET method is equivalent to a 65 percent stenosis for the ECST method, while a 70 percent stenosis with the NASCET method is equivalent to an 82 percent stenosis for the ECST method. (See "Evaluation of carotid artery stenosis"). NASCET trial — The North American Symptomatic Carotid Endarterectomy Trial (NASCET) was initiated in the mid-1980s to investigate the efficacy of CEA in symptomatic patients [6]. This randomized, prospective, multi-center trial enrolled 659 patients who had had a hemispheric or retinal TIA or a nondisabling stroke within the 120 days before entry and had stenosis of 70 to 99 percent in the symptomatic (ipsilateral) carotid artery. The study was prematurely terminated by the National Institutes of Health because of the clear evidence of benefit from surgery in this selected group of patients. At the time of study termination, patients had been followed for a mean of 18 months. Although the risk of stroke and death was higher at 30 days in the patients treated with CEA (5.8 versus 3.3 percent with medical therapy), longer follow-up revealed the following significant benefits with CEA [2]: •

A lower cumulative risk at two years of any ipsilateral stroke (9 versus 26 percent, p<0.001) (show figure 2)

•

A lower risk of major or fatal ipsilateral stroke (2.5 versus 13.1 percent, p<0.001) (show figure 3)

The trialists concluded that CEA was highly beneficial for patients with recent TIAs or nondisabling strokes with ipsilateral stenosis of 70 to 99 percent [2]. In a subsequent report, The NASCET trial demonstrated that CEA also had a moderate degree of benefit for patients with 50 to 69 percent symptomatic ipsilateral stenosis compared with those receiving medical therapy [7]. However, the overall statistical significance favoring surgery was marginal; the five-year rate of any ipsilateral stroke for endarterectomy versus medical treatment was 15.7 versus 22.2 percent (p = 0.045). Patients with stenosis of less than 50 percent did not benefit from surgery. Follow-up of the NASCET cohort continued for an additional seven years. The benefits of CEA in patients with 70 to 99 percent stenosis continued to be evident after this prolonged period [8]. Subset analysis of patients ages 75 and older found that elderly patients with 50 to 99 percent stenosis benefited more from CEA than younger patients [9]. Others have reported similar findings [10]. These findings suggest that CEA should not be withheld from appropriately selected, fit patients over the age of 75. Predictors of medical complications with CEA were a history of myocardial infarction or angina and hypertension [11]. Patients with a recent history of myocardial infarction, unstable angina pectoris, or recent congestive heart failure were excluded from NASCET, perhaps in part explaining the low perioperative medical complication rate. The NASCET surgical results study examined rates of perioperative stroke and death at 30 days, stroke severity at 90 days, variables that influenced perioperative risk, and the durability of CEA [12]. Five baseline variables were predictive of increased surgical risk: hemispheric (versus retinal) TIA as the qualifying event; left sided procedure; contralateral carotid occlusion; ipsilateral ischemic lesion on CT scan; and irregular or ulcerated ipsilateral plaque. ECST trial — The European Carotid Surgery Trial (ECST) was a multi-center, prospective trial that randomly assigned 2518 patients with a nondisabling ischemic stroke, TIA, or retinal infarct due to a stenotic lesion in the ipsilateral carotid artery to medical therapy with aspirin or to surgery [3]. The first report included 374 patients with a mild stenosis (0 to 29 percent) and 778 patients with severe stenosis (70 to 99 percent). After a three-year follow-up, the following findings were noted in the interim report:

•

Patients with mild stenosis had little risk of ipsilateral ischemic stroke; possible benefits of CEA were small and were outweighed by the early risks (show figure 4).

•

At 30 days, the incidence of stroke or death was 7.5 percent in the patients with a severe stenosis who underwent CEA; the risk of these complications was not related to the severity of the stenosis (show figure 5).

•

At three years, patients treated with CEA had significant reductions in the incidence of ipsilateral ischemic stroke (2.8 versus 16.8 percent with aspirin alone, p<0.0001) and in the total risk of surgical death, surgical stroke, ipsilateral ischemic stroke, or any other stroke (12.3 versus 21.9 percent). The risk varied with age and sex, with benefit being less likely in women and over a narrower range of carotid stenosis in younger patients (show figure 6) [13].

The risk of disabling or fatal stroke was also reduced in patients treated with CEA. However, the incidence of these complications in the aspirin group seemed to diminish after the first year; thus, a delay of surgery by just a few months after clinical presentation could have negated the overall difference. A subsequent final report from ECST, based upon an ultimate total of 3024 patients followed for a mean of six years, noted two major findings [13]: •

CEA was beneficial for symptomatic carotid stenosis of 80 to 99 percent (show figure 7).

•

The risk of a major ischemic stroke ipsilateral to the unoperated carotid artery increased with the severity of the stenosis, particularly above 80 percent, but only for two to three years after randomization (show figure 7).

Overall, the ECST confirmed the results of the NASCET trial, demonstrating a benefit with CEA in symptomatic patients with severe ipsilateral carotid stenosis, although age and sex were important considerations in a decision about surgery. The reduced risk of recurrent stroke associated with CEA was durable during at least 10 years of follow-up [14]. Even in the areas where NASCET and ECST appeared to disagree (eg, in patients with less than 80 percent stenosis), a reanalysis of the data suggests that if the same measurement criteria were used, these differences would disappear [15]. (See "Method of stenosis measurement" above). Pooled analysis — The efficacy of CEA in patients with symptomatic carotid disease has been calculated from a pooled analysis of the major trials that included data from ECST, NASCET, and the Veterans Affairs trial 309 [5]. Prerandomization carotid angiograms from ECST were reassessed by the NASCET method, and outcomes were standardized to achieve comparability among the trials. The following observations were reported: •

CEA was beneficial for patients with greater than 70 percent symptomatic stenosis (but not near occlusion). The number needed to treat (NNT) to prevent one stroke over five years for this group was 6.3, with an absolute risk reduction (ARR) of 16 percent.

•

No significant benefit of CEA with near occlusion of the internal carotid artery (ICA) was observed. (Near occlusion is defined by the angiographic appearance of a collapsed ICA distal to the stenosis, accompanied by faster filling in the external carotid artery, and preferential filling of the intracranial circulation from collateral vessels). The ARR over two years with CEA for near occlusion was 5.6 percent, but the lack of benefit became apparent over five years, when the ARR was negative (-1.7 percent).

•

CEA was beneficial for patients with 50 to 69 percent symptomatic stenosis. The NNT to prevent one stroke over five years in this group was 22, with an ARR of 4.6 percent.

•

CEA was not beneficial for symptomatic carotid stenosis of 30 to 49 percent, and CEA was harmful for symptomatic patients with less than 30 percent stenosis.

FACTORS INFLUENCING BENEFIT AND RISK — A large number of factors may influence the risk/benefit ratio with carotid endarterectomy (CEA) and thus the decision to treat individual patients with carotid occlusive disease. Risk factors for morbidity and mortality associated with CEA should be identified in order to avoid surgery in those who may face unacceptably high surgical risk [16]. These risk factors are discussed separately. (See "Carotid endarterectomy: Preoperative evaluation; surgical technique; and complications", section on Preoperative evaluation).

In addition, it is important to exclude patients who are unlikely to benefit from CEA. These include patients who have severe comorbidity due to other surgical or medical illness, patients with an ipsilateral stroke associated with persistent disabling neurologic deficits, and patients with total occlusion of the internal carotid artery. A number of additional factors appear to impact benefit and risk in symptomatic patients with carotid stenosis. These include the timing of CEA after the ischemic event, the perioperative complication rate, patient sex, and the presence of contralateral carotid stenosis or occlusion. This section will discuss these factors. A separate issue is that low-dose aspirin (81 to 325 mg/day) treatment is recommended for all patients who are having CEA. Aspirin should be started prior to surgery and continued for at least three months after surgery. This topic is discussed separately. (See "Carotid endarterectomy: Preoperative evaluation; surgical technique; and complications", section on Aspirin). Timing of surgery — No prospective, randomized studies have evaluated the timing of CEA after a recent stroke or TIA as a primary outcome measure, and the optimal timing of CEA in symptomatic patients has been the subject of considerable debate [17]. Mild stroke or TIA — A pooled analysis of the NASCET and the ECST trials, representing the largest experience to date, found that CEA within two weeks of a nondisabling stroke or TIA significantly improved outcomes compared with later surgery (show figure 8) [10]. The following observations were reported. •

In the subgroup of patients with 70 percent or greater carotid stenosis, CEA was associated with a 30.2 percent reduction in absolute risk of stroke in patients randomized within two weeks of their last event, compared with a 17.6, 11.4, and 8.9 percent absolute reduction in those randomized 2 to 4, 4 to 12, and more than 12 weeks from their last event.

•

For patients with 50 to 69 percent stenosis, clinically important benefits from CEA were noted only in patients randomized within two weeks of their last event.

Further analysis of the pooled NASCET and ECST data showed the following additional observations [18]: •

The decline in benefit of CEA over time was more rapid in women than in men.

•

Surgical benefit in women was confined to those who had CEA within two weeks after their last event, irrespective of the degree of stenosis.

•

CEA within two weeks of a nondisabling hemispheric stroke was not associated with an increased operative risk.

Moderate to severe or progressing stroke — The benefit of CEA for patients with moderate to severe ischemic stroke has not been evaluated in randomized clinical trials, as patients who have disabling stroke were not eligible for NASCET or ECST. Given that there have been no prospective trials addressing the question of timing of surgery, it is difficult to generalize observations from the NASCET and ECST trials to all patients undergoing CEA. In particular, patients who have a large infarction with brain swelling, hemorrhagic brain infarction, or progressing stroke have long been thought to have high perioperative risk with early CEA [19-22]. On the other hand, delay in CEA may expose the patient to an increased risk of recurrent stroke. A number of retrospective studies have evaluated the timing of CEA after ischemic stroke, and a 1997 review of these trials made the following observations [17]: •

Patients undergoing CEA who have fixed neurologic deficits after a stroke probably represent a heterogeneous group whose risks vary according to clinical and radiological features.

•

Several reports documented satisfactory outcomes in patients undergoing surgery within six weeks of cerebral infarction.

•

Intracerebral hemorrhage after CEA is most likely due to postoperative hyperperfusion and can possibly be reduced by meticulous attention to blood pressure perioperatively. (See

"Carotid endarterectomy: Preoperative evaluation; surgical technique; and complications", section on Hyperperfusion syndrome). An evidenced-based review of CEA published by the American Academy of Neurology (AAN) in 2005 identified six retrospective cohort studies comparing the timing of CEA (early versus late) in patients after stroke [23]. None of the studies found any differences in outcomes of perioperative morbidity or longer-term follow-up. All six studies were limited by small size. Acute stroke — The 2005 AAN evidenced-based review also identified four observational studies that evaluated emergent CEA for progressing stroke [23]. The studies were limited by conflicting findings, small size, and methodological flaws. Subsequent studies of CEA for acute stroke suffer from the same limitations [24,25]. The AAN review concluded that no recommendation could be made regarding the value of emergent CEA in patients with progressing neurologic deficits. Perioperative complications — Combined morbidity and mortality that exceeds 6 percent for patients with symptomatic stenosis could eliminate the benefit gained from CEA [26-28]. Relatively small increases in major surgical and angiographic complication rates may tip the balance against surgery, particularly for patients with moderate symptomatic carotid stenosis (50 to 69 percent by NASCET criteria) [26]. However, the utility of these data are limited in practice, since it may be difficult to obtain the appropriate perioperative complication rates for individual surgeons or centers. A review of Medicare recipients undergoing CEA found a higher mortality rate in hospitals that did not participate in these carotid surgery trials, particularly in those centers performing fewer procedures (2.9 versus 1.4 percent in NASCET and ACAS trial hospitals) [29]. (See "NASCET trial" above and see "Carotid endarterectomy in asymptomatic patients", section on ACAS trial). Low patient volume (less than three CEAs performed every two years) and a greater number of years since licensure of the surgeon are associated with worse outcomes following CEA [30]. Gender — The benefit of CEA may be greater for men than for women. Clinical trials including NASCET and ECST did not have enough statistical power to determine the effect of surgery in specific patient subgroups. (See "NASCET trial" above and see "ECST trial" above). However, an analysis of pooled patient data from ECST and NASCET [10], and a subsequent analysis of data from NASCET and the Aspirin Carotid Endarterectomy (ACE) trial [31], provided the following observations regarding the benefit of CEA for symptomatic patients by gender: •

The risk of stroke ipsilateral to a symptomatic carotid stenosis is significantly lower for medically treated women than men.

•

The perioperative risk of death from CEA is significantly higher in women than in men.

•

CEA is clearly beneficial for women with 70 to 99 percent symptomatic carotid stenosis, and the five-year absolute risk reduction in stroke is similar for men and women (17.3 and 15.1 percent, respectively). The number of patients needed to treat (NNT) with CEA to prevent one ipsilateral stroke in five years is also similar for men and women (6 and 7, respectively).

•

CEA is not beneficial for most women with 50 to 69 percent symptomatic carotid stenosis, in contrast to men. However, a proportion (29 percent) of women in the 50 to 69 percent group with the highest risk factor profile, as determined by a modified version of the Stroke Prognosis Instrument-II (SPI-II) [32], may have benefit from CEA.

A further analysis of ECST suggested that the generally smaller size of women may contribute to their higher risk of perioperative stroke and death, perhaps because of limited access to the surgical field in smaller people and/or difficulties in operating on smaller vessels [33]. A subsequent report confirmed that women have smaller carotid arteries than men, even after adjusting for body and neck size, age and blood pressure [34]. Retinal versus hemispheric ischemia — In patients with symptomatic carotid disease, transient retinal ischemia (ie, transient monocular blindness [TMB]; also called amaurosis fugax) portends a lower risk of ipsilateral carotid stroke than hemispheric TIA. In a subset analysis from NASCET, 198 medically treated patients with TMB had a three-year risk of ipsilateral stroke that was approximately one-half that of 417 medically treated patients with hemispheric TIA [35]. Six factors were associated with a higher risk of stroke in patients with TMB [35]: •

Age ≥75 years

•

Male sex

•

History of hemispheric TIA or stroke

•

History of intermittent claudication

•

Carotid artery stenosis of 80 to 94 percent

•

Absence of collateral circulation

The three-year risk of stroke with medical treatment for patients with zero to one, two, and three or more risk factors was 1.8, 12.3, and 24.2 percent, respectively. The corresponding three-year absolute risk reduction (ARR) for stroke associated with CEA was -2.2, 4.9, and 14.3 percent, respectively. The corresponding NNT to prevent one stroke over three years did not apply to the lowest risk group with a negative ARR, while the NNT was 20 in the group with two risk factors and seven in the group with three or more risk factors. Contralateral carotid stenosis or occlusion — CEA is likely to be beneficial for patients who have symptomatic ipsilateral severe carotid stenosis and coexisting severe contralateral carotid stenosis or occlusion, although patients with a severe contralateral carotid artery stenosis may be at higher perioperative risk. These points were illustrated in a report from the NASCET cohort that analyzed medical versus surgical therapy in 659 patients with a recent ischemic event referable to a 70 to 99 percent stenosis of an ipsilateral carotid artery [36]. The contralateral carotid artery had a 70 to 99 percent stenosis in 8.6 percent and was totally occluded in 7 percent. After a two-year follow-up, the following observations were reported: •

Medically treated patients with an occluded contralateral carotid were twice as likely to have an ipsilateral stroke compared with those with severe or mild to moderate disease (hazard ratio 2.36 and 2.65, respectively).

•

Among surgically treated patients, the perioperative risk of stroke and death was higher in those with a totally occluded or mild to moderately stenotic contralateral vessel (4 and 5 percent, respectively) compared with those without contralateral disease.

•

Despite this increased risk, patients who had CEA on a recently symptomatic and severely stenosed ipsilateral carotid still had a significantly better outcome than patients treated medically.

Advances in medical therapy — At the time when most of the major CEA trials were underway (generally the late 1980s to mid 1990s), the best medical therapy for carotid disease was generally considered to be antiplatelet treatment with aspirin. Since the completion of these trials, newer medical regimens have emerged that reduce the risk of stroke. These therapies include aggressive regimens of statins, antiplatelet agents, and antihypertensive agents. As an example, statin treatment can delay progression and induce regression of atherosclerotic lesions both in native coronary vessels and in coronary bypass grafts. Statin treatment can also reduce the incidence of first and recurrent stroke. (See "Clinical trials of cholesterol lowering in patients with coronary heart disease or coronary risk equivalents", section on Improvement in atherosclerosis and see "Secondary prevention of stroke: Risk factor reduction", section on Dyslipidemia). It is unknown if these newer medical therapies would change the relative risk reduction afforded by CEA for carotid disease, but they might increase the number of patients needed to treat to prevent one stroke. This may be of particular significance for patients with asymptomatic carotid disease in whom the NNT is already high compared with symptomatic patients. (See "Carotid endarterectomy in asymptomatic patients", section on Efficacy and section on Advances in medical therapy). Evaluation of carotid artery stenosis — The benefit of CEA for patients with symptomatic carotid disease was established by clinical trials designed in the 1980s that employed conventional contrast angiography to determine the degree of carotid stenosis. (See "Efficacy" above). Conventional contrast angiography is still considered the gold standard for the evaluation of internal carotid artery stenosis. However, noninvasive tests such as carotid duplex ultrasound (CDUS), magnetic resonance angiography (MRA), and computed tomographic angiography (CTA) are often preferred in current clinical practice, mainly because contrast angiography is an invasive procedure associated with a risk of stroke.

The available clinical data suggest that the noninvasive tests (CDUS, MRA, and CTA) are sufficiently accurate and reliable that they can be used to select symptomatic patients with carotid stenosis who can benefit from CEA. This issue is discussed separately. (See "Evaluation of carotid artery stenosis" and see "Carotid endarterectomy: Preoperative evaluation; surgical technique; and complications", section on Preoperative evaluation). Risk modeling — Since any benefit of CEA is dependent on the absolute risk of adverse outcome with or without treatment, it is useful to systematically consider the risks and benefits of CEA based on individual patient characteristics whenever possible [37]. A risk model for patients with recently symptomatic carotid stenosis has been derived from the ECST trial data and validated by showing that the predicted stroke risk for patients assigned to medical treatment in the NASCET trial was close to the observed risk [37]. (See "ECST trial" above and see "NASCET trial" above). The calculation of baseline ipsilateral stroke risk for recently symptomatic patients is primarily dependent on the following factors [37]: •

Patient age

•

Patient sex

•

Type of presenting symptomatic event (eg, ocular TIA, hemispheric TIA, minor stroke, or major stroke)

•

Degree of carotid stenosis

•

Carotid plaque morphology (smooth versus ulcerated or irregular)

•

Time since last symptomatic event

These factors can be applied to estimate the absolute risk of ipsilateral stroke for individual patients with symptomatic carotid disease who are candidates for CEA (show figure 9). As an example from the model, a 70-year-old woman who presents with an ocular TIA, a 70 to 99 percent ipsilateral carotid stenosis, an ulcerated irregular plaque, and time greater than 12 weeks since the last symptomatic event has a predicted absolute five-year stroke risk of 15 to 20 percent. However, if the same patient presents with a cerebral (hemispheric) TIA and time less than two weeks since the last symptomatic event, the predicted absolute five-year stroke risk rises 35 to 40 percent. While the model does not account for the risk and benefit of surgery, it may be useful for assessment of baseline stroke risk and patient selection for CEA. A version of the ECST stroke risk model calculator is available online (www.stroke.ox.ac.uk). SUMMARY AND RECOMMENDATIONS — Our recommendations regarding carotid endarterectomy (CEA) for symptomatic carotid stenosis are based on our clinical experience, data from clinical studies, an evidenced-based review of CEA published by the American Academy of Neurology in 2005 [23], guidelines for the prevention of stroke published by the American Heart Association/American Stroke Association in 2006 [38], and guidelines published in 2008 by the National Institute for Health and Clinical Excellence in the United Kingdom [39]. •

We recommend CEA for patients with recently symptomatic carotid stenosis of 70 to 99 percent who have a life expectancy of at least five years, provided that the perioperative risk of stroke and death for the surgeon or center is less than 6 percent (Grade 1A). The number needed to treat (NNT) to prevent one stroke over five years for this group is 6.3. (See "Efficacy" above, see "Pooled analysis" above and see "Perioperative complications" above).

•

CEA is beneficial for patients with 50 to 69 percent symptomatic stenosis, and the NNT over five years in this group is 22. However, women with 50 to 69 percent symptomatic carotid stenosis have not shown clear benefit. We suggest CEA for men with recently symptomatic carotid stenosis of 50 to 69 percent who have a life expectancy of at least five years, provided that the perioperative risk of stroke and death for the surgeon or center is less than 6 percent (Grade 2A). We suggest medical management rather than CEA for women with recently symptomatic carotid stenosis of 50 to 69 percent (Grade 2B). (See "Efficacy" above, see "Perioperative complications" above and see "Gender" above).

•

CEA is not beneficial for symptomatic carotid stenosis of 30 to 49 percent, and CEA is harmful for symptomatic patients with less than 30 percent stenosis. We recommend medical management rather than CEA for patients with symptomatic carotid stenosis that is less than 50 percent (Grade 1A). (See "Efficacy" above and see "Pooled analysis" above).

•

CEA is likely to be of greatest benefit if performed within two weeks of the last symptomatic event in patients with nondisabling stroke or transient ischemic attack (TIA) who have 70 percent or greater carotid stenosis. However, CEA may not be beneficial if performed two weeks or longer after the last event for patients with 50 to 69 percent stenosis. In addition, CEA may not be beneficial if performed beyond two weeks after the last event for women irrespective of the degree of stenosis. We recommend that CEA be performed without delay, preferably within two weeks of the last symptomatic event, for patients with carotid stenosis and a nondisabling stroke or TIA (Grade 1B). (See "Timing of surgery" above).

•

A number of additional factors may influence the benefit and risk of CEA for patients with symptomatic carotid stenosis. CEA is not associated with significant benefit for patients with near occlusion of the symptomatic ipsilateral internal carotid artery. Older patients may have a greater benefit than younger patients. Patients with hemispheric TIA appear to have greater benefit than patients with transient retinal ischemia. Coexisting severe contralateral carotid stenosis or occlusion may increase perioperative risk but does not cancel out benefit. (See "Pooled analysis" above, see "Factors influencing benefit and risk" above, see "Risk modeling" above, see "NASCET trial" above, see "Retinal versus hemispheric ischemia" above and see "Contralateral carotid stenosis or occlusion" above).

•

The benefit of CEA for patients with moderate to severe ischemic stroke has not been evaluated in randomized clinical trials, and patients with an ipsilateral stroke who have persistent disabling neurologic deficits are unlikely to benefit from CEA. The risks of CEA for such patients probably vary according to clinical and radiologic features. (See "Factors influencing benefit and risk" above and See "Timing of surgery" above).

•

We recommend aspirin (81 to 325 mg/day) treatment starting before surgery for all patients having CEA and continuing for at least three months after surgery (Grade 1A). (See "Carotid endarterectomy: Preoperative evaluation; surgical technique; and complications", section on Aspirin).