Clinical manifestations

Presentation and course

Localized bruits in the mid- and upper regions of the neck, below the angle of the jaw, correlate well with underlying occlusive disease at the carotid bifurcation. The bruit is typically systolic in timing but varies in intensity. Bruits usually appear when the diameter stenosis is 50% or greater (corresponding to a 75% reduction in cross-sectional area) and may actually disappear at a diameter stenosis of greater than 90%. As the stenosis progresses from 50% to 90%, the bruit becomes louder and higher pitched. A diastolic component is suggestive of more than 90% luminal narrowing with associated reduction in blood flow. A bruit over the carotid artery found in conjunction with a history of a hemispheric transient ischemic attack or retinal transient ischemic attack (amaurosis fugax) is likely to reflect a clinically significant, correctable carotid stenosis. Hemispheric symptoms include contralateral hemiparesis, contralateral paresthesia or hemisensory changes, visual field defects, and aphasia (dominant hemisphere) or neglect (nondominant hemisphere).

Prognosis and complications

Patients with asymptomatic carotid bruits are approximately three times more likely to have an ischemic stroke than an age- and sex-matched population without carotid bruits. Some investigators have suggested that a history of contralateral TIAs and renal dysfunction indicate increased stroke risk in patients with asymptomatic stenosis (33). Subsequent cerebral ischemic events were not significantly different in patients with localized versus diffuse bruits. Prognostic information from randomized trials is provided below.

A systematic review found that with improvements in medical therapy, there has been a decrease in the stroke rate with asymptomatic carotid stenosis over the past 25 years (01). In the past decade, the annual rate of stroke has fallen below 1.5% per year, in the range where it overlaps with the stroke rate in patients treated surgically.

Patients with asymptomatic carotid stenosis may be at increased risk for cognitive decline. A study followed 210 patients with greater than 70% stenosis with no previous history of stroke and 109 healthy control subjects (05). Patients with carotid stenosis, especially those with impaired hemodynamic response to breath holding, were at increased risk for cognitive worsening over a 3-year follow-up period (odds ratio 4.16).

Patients with asymptomatic stenosis need to live several years after carotid endarterectomy in order to benefit from the procedure and offset the upfront stroke risk from carotid endarterectomy. A study from Sweden showed that mortality among asymptomatic patients who undergo carotid endarterectomy is substantial (27). The 5-year survival in this series was 78.2% after five years and 45.5% after 10 years. Previous vascular surgery, cardiac disease, diabetes, and increasing age were predictors of decreased survival.

Using transcranial Doppler to assess for microemboli can be useful in risk stratification. Two studies found that the occurrence of active microemboli is fairly low in subjects with asymptomatic carotid stenosis (29; 43). In the Asymptomatic Carotid Emboli Study (ACES), 77 of 467 (16%) patients had evidence for microemboli at baseline. At two years, the risk of ipsilateral stroke was 3.6% in patients with microemboli and 0.7% in patients without microemboli (29). Patients with progression of stenosis on serial duplex ultrasound (DUS) studies by two or more stenosis categories have an increased risk of TIA or stroke (21).

Clinical vignette

A 49-year-old man was referred to the neurology clinic with an asymptomatic left carotid bruit. The patient had a history of hypertension, diabetes, hypercholesterolemia, and mild obesity. The patient denied previous symptoms such as monocular visual loss, right-sided weakness or numbness, or speech difficulty. He had a carotid ultrasound that revealed 50% to 79% left carotid stenosis, shown to be 70% on angiography. He was placed on aspirin, simvastatin, an angiotensin-converting enzyme inhibitor and continued on his regimen for diabetes. An exercise and weight loss regimen was encouraged. The patient was followed with serial ultrasound exams and the carotid stenosis remained stable over the next five years; he remains asymptomatic. The patient also remains free of ischemic heart disease.

Biological basis

Etiology and pathogenesis

Atherosclerosis causing a stenosis of more than 50% of the carotid diameter is the most common cause of this bruit. A bruit may also be caused by conditions that increase turbulent flow. This may be anatomical (eg, vessel kinking) or rheological (eg, lowered viscosity with anemia).

The velocity profile of circulating blood is laminar and, therefore, motion is soundless. Murmurs are created when laminar flow is disturbed by alterations in biophysical characteristics such as velocity, viscosity, and vessel wall anatomy. Experimental work has shown that for any degree of stenosis there is a given blood flow rate that produces jet instability with associated turbulence that will produce a bruit (40). Increasing the degree of stenosis increases both the amount of turbulence and the likelihood of detecting a bruit. Carotid bruits have been described with a residual lumen diameter of 0.8 mm to 3.0 mm (normal mean carotid artery diameter 5.1 mm + 1.1 mm). It is, therefore, possible to hear a bruit with stenoses as low as 25% or even lower in high blood flow conditions. Theoretically, when the artery is occluded there should be no turbulence and, thus, no bruit. However, bruits are still audible in as many as 21% of patients with occluded internal carotid arteries (24). The most likely mechanism of bruit production with an occluded internal carotid artery is flow disturbance in a stenotic external carotid artery.

Epidemiology

The number of patients with asymptomatic carotid bruits is large; approximately 12.6% of the population older than 45 years of age have been noted to have some type of cervical murmur (41). Approximately one third of these murmurs are localized carotid bifurcation bruits, and about three fourths of the persons with localized bruits have had neither hemispheric nor retinal ischemic symptoms.

The number of patients with actual asymptomatic carotid stenosis is estimated at 4% to 6% in patients aged 65 and older. After a systematic review of the literature, the United States Preventive Services Task Force recommended that the general population should not be screened for asymptomatic carotid stenosis, primarily due to concerns that clinical trials of carotid endarterectomy may not be reproduced in clinical practice (48).

Prevention

All patients with carotid bruits should be evaluated for the presence of atherosclerotic risk factors, such as smoking, lipid disorders, hypertension, and diabetes mellitus. Atherosclerosis involving the cerebrovasculature is associated with the same risk factors as atherosclerosis elsewhere, particularly hypertension. A comprehensive approach to risk factor reduction including hypertension control, cessation of smoking, normalization of the lipid profile, attainment of ideal body weight, and regular exercise is appropriate in all patients with cerebrovascular disease.

The presence of carotid bifurcation disease is also a strong clinical marker for coronary atherosclerosis. The American Stroke Association has previously published a statement on coronary risk evaluation in patients with TIA or ischemic stroke (02). Due to the fact that 30% to 50% of patients with carotid atherosclerosis will have underlying coronary heart disease, the statement comments that clinicians should “consider coronary heart disease testing when the workup reveals carotid or other large-vessel atherosclerosis.” The evaluation and management of patients with combined carotid and coronary occlusive disease are controversial and must be individualized based on symptoms, anatomy, and hemodynamic considerations.

Differential diagnosis

Confusing conditions

Cervical murmurs can be classified by their location and acoustic characteristics (41). Venous cervical murmurs ("hums") are common in children but may occur in up to 27% of normal adults. Venous hums are usually heard at the base of the neck and their sound is low-pitched, roaring, usually continuous, maximal during diastole, loudest in the upright position, and decreased or abolished by the Valsalva maneuver. Arterial bruits in the neck can be divided into supraclavicular, vertebral, and carotid. Supraclavicular bruits are diffuse and can be heard along the length of the carotid artery. They are frequently bilateral and systolic in timing. These murmurs are commonly due to transmission of heart sounds (aortic stenosis and other aortic outflow murmurs, mitral insufficiency, coarctation of the aorta, and patent ductus arteriosus), structural changes affecting the vessels at the base of the neck (atherosclerosis, subclavian tortuosity, Takayasu disease, or conditions associated with thoracic outlet syndrome), or a hyperdynamic state (hemodialysis, fever, anemia, and hyperthyroidism). Vertebral bruits are heard from the supraclavicular fossa to the mastoid region, along the posterior border of the sternocleidomastoid muscle. The sound is usually systolic but may extend into diastole. Vertebral bruits are associated with stenoses of the origin of the vertebral artery, bilateral carotid artery occlusion, reversal of blood flow in the vertebral artery, and hemangioma of the brainstem. Localized bruits in the mid- and upper regions of the neck are, in most instances, due to atherosclerotic lesions in the proximal internal carotid artery. However, systolic bruits having identical acoustic characteristics may result from external carotid artery stenoses and may be heard with nonatherosclerotic lesions such as fibromuscular dysplasia or buckling and tortuosity of the carotid arteries.

Diagnostic workup

Although the clinical implications of neck auscultation remain uncertain, invasive diagnostic tests are not indicated in every patient with a carotid bruit. Noninvasive carotid evaluation by duplex scanning facilitates accurate correlation of cervical bruits with extracranial arterial anatomy. Combining ultrasound imaging and Doppler assessment of blood flow, the duplex scan provides relevant information regarding the carotid bifurcation. Doppler techniques measure the frequency of the sound reflected from red blood cells. The faster the speed of the red cells, the higher the frequency of the reflected sound. An increasing stenosis produces a smaller lumen leading to a higher flow velocity. The ability of duplex imaging to distinguish between severe stenoses and mild stenoses is excellent, with a reported specificity of 84% and sensitivity of 99% (36). Some centers recommend surgery based on duplex scanning alone. In some centers, however, preoperative, confirmatory contrast arteriography is recommended. Contrast angiography, however, remains the gold standard for the diagnosis of carotid artery stenosis. Either conventional or digital subtraction techniques provide excellent images of the head and neck and the intracranial circulation. Computed tomography of the brain and MRI are used to identify prior infarcts as well as to exclude other intracranial lesions. Magnetic resonance angiography may provide an appropriate alternative to conventional contrast angiography in some patients, and some centers perform carotid surgery on the basis of a combined battery of duplex ultrasound and MRA, reserving angiography for discordant cases. Contrast-enhanced MRA may improve the accuracy of conventional MRA imaging and can be used in a noninvasive battery of carotid stenosis tests (47). CT angiography is also increasingly being performed in some centers. An overview of the pros and cons of different imaging modalities has been reviewed (37).

The hemodynamic significance of a carotid bruit (sufficiency of distal flow preservation or collateralization) can be evaluated by ocular pneumoplethysmography (systolic pressure reduction in the ophthalmic artery reflects a hemodynamically significant lesion) or transcranial Doppler. Transcranial Doppler also allows the assessment of microemboli in patients with extracranial carotid stenosis. A study of 319 patients who had greater than 60% asymptomatic carotid stenosis and transcranial Doppler exams with subsequent clinical follow-up indicated that patients with microemboli were at significantly higher risk for stroke at one year (15.6%) compared to patients without microemboli (1%) (42). Ninety percent of these patients did not have microemboli at baseline, suggesting that the majority of patients are at low risk.

With newer imaging modalities, there is the potential to investigate “vascular inflammation.” A report described MRI with a contrast agent, an ultrasmall superparamagnetic iron oxide (USPIO), in the asymptomatic carotid artery contralateral to a symptomatic stenosis (45). In a preliminary study with 20 patients, 57% of regions analyzed on the asymptomatic side had inflammatory activity compared to 71% on the symptomatic side (p=0.007). Further studies are needed to see whether this is a useful marker for risk stratification.

Management

The detection of an asymptomatic carotid bruit invariably engenders considerable debate and uncertainty about proper evaluation and management. Which noninvasive tests should be performed? Should brain imaging be performed to search for asymptomatic infarcts? When is carotid endarterectomy or carotid stenting indicated for asymptomatic stenosis?

All patients with cervical bruits, whether or not they are symptomatic, should be evaluated by noninvasive Doppler ultrasound to determine if the bruit reflects a significant internal carotid artery stenosis. Some centers recommend CT scans or MRI of the brain on all patients to either delineate old strokes in asymptomatic patients or exclude other intracranial lesions in symptomatic patients. However, routine CT scans on all patients with carotid stenoses may not be cost effective.

Asymptomatic bruits. If, based on a thorough history, the patient is truly asymptomatic and Doppler studies demonstrate a stenosis of less than 80%, then intensive medical therapy with antiplatelet therapy such as aspirin, high potency statins, and risk factor modification is appropriate for most patients. Anticoagulation with warfarin is not recommended. Patients with greater than 60% stenosis should be treated for hyperlipidemia, and the LDL level should be targeted for less than 70 mg/dl. Statin treatment will be required in the majority of patients. In carotid endarterectomy specimens, use of statins has been associated with reduced macrophage infiltration and matrix metalloproteinase 9 area, suggesting that statins can be useful in stabilizing carotid plaques (28). Smoking cessation should be strongly encouraged. Although the study was done in symptomatic patients, the Treat Stroke to Target study, which randomly assigned patients to a moderate or intensive LDL target, found benefit with a lower LDL target (03). Therefore, it appears reasonable to extrapolate an LDL of less than 70 mg/dl as a reasonable target for patients with asymptomatic carotid stenosis. Less than 55 mg/dl can be considered in patients judged to have higher vascular risk, such as carotid stenosis with diabetes.

Asymptomatic patients with stenoses between 60% and 79% should be re-evaluated every 6 to 12 months with noninvasive studies to determine if there has been progression of stenosis. If, during follow-up, the patient becomes symptomatic or the lesion progresses, then further evaluation is warranted. Plaque morphology, as it becomes better defined and classified, may significantly influence clinical decision making in the future. Asymptomatic patients with high-grade (greater than 80%) stenoses by Doppler ultrasound who are good operative risks with excellent 5-year life expectancies should receive further evaluation with either MRA or contrast angiography.

The Asymptomatic Carotid Atherosclerosis Study evaluated patients with 60% to 99% asymptomatic carotid stenosis and found a modest reduction in ipsilateral stroke with carotid endarterectomy (CEA) (04). At five years, the projected rate of ipsilateral stroke was 11.0% in the medically treated patients and 5.1% in the surgery group. The annual absolute risk reduction of 1.2% per year is small, and some vascular neurologists have argued that it is not clinically meaningful. In addition, concerns have been raised about whether the low perioperative stroke rate in the Asymptomatic Carotid Atherosclerosis Study can be reproduced routinely in the community. Current recommendations are that for asymptomatic patients, the perioperative stroke rate should be less than 3% (07), yet the stroke rate for asymptomatic patients was 4.6% in the Aspirin and Carotid Endarterectomy trial, which featured experienced surgeons at stroke centers. In addition, in up to 45% of patients with asymptomatic carotid stenosis, the future cause of an ipsilateral stroke may not be related to the carotid stenosis, with the cause being either cardioembolism or small vessel occlusive disease (25). If a practitioner is confident that the perioperative stroke rate is less than 3% at their institution and if the patient is otherwise healthy and with a 5-year life expectancy, then there may be a small benefit for performing carotid endarterectomy in patients with 60% to 99% stenosis. This position was articulated in the American Academy of Neurology carotid endarterectomy guidelines (10). However, whether CEA is still of benefit compared to modern intensive medical therapy is unclear (see below).

The Asymptomatic Carotid Surgery Trial (ACST) results have been published (MRC ACST Collaborative Group 2004). This was a multicenter international study in which patients with more than 60% stenosis were randomized to immediate carotid endarterectomy or deferred carotid endarterectomy. Patients in the deferred carotid endarterectomy group were treated medically. Approximately 90% of patients were on antiplatelet therapy, 70% were on antihypertensive agents, and 40% were on lipid lowering medication. Enrolled were 3120 patients who were followed for a mean period of 3.3 years. The 5-year projected rates of total ischemic stroke were 11.8% in the deferred carotid endarterectomy group and 6.4% in the immediate carotid endarterectomy group. About half of strokes in both groups were disabling. The benefit was smaller in women compared to men. Patients younger than 75 years of age demonstrated benefit, but it was unclear if patients older than 75 years of age derived benefit because their death rate was higher during the follow-up period. The overall perioperative morbidity and mortality rate was 3.1% in the ACST.

The ACST investigators have also reported 10-year results (19). At 10 years, there was still a benefit for carotid endarterectomy, but it was lower in absolute terms. The immediate carotid endarterectomy group had a 10-year stroke risk of 13.4% compared to 17.9% in the deferred carotid endarterectomy group (absolute gain 4.6%). Patients aged 75 years and older once again did not show clear benefit. Stroke risk in both the carotid endarterectomy and medical therapy group were substantially lower in the patients who received lipid-lowering treatment. Patients with remote symptoms (greater than 6 months previously) and patients with radiologic infarcts had a higher 10-year stroke risk, with a 5.8% increase over 10 years (44). An analysis from the Oxford Vascular Study also reports that patients with 80% to 99% stenosis have a higher 5-year stroke risk compared to patients with less than 80% stenosis (22).

In a combined analysis of Asymptomatic Carotid Atherosclerosis Study and ACST, it was noted that there was no clear benefit for carotid endarterectomy in asymptomatic women (39). One of the potential reasons for this finding is that women seem to do better with medical therapy. In a study evaluating endarterectomy specimens, women had more features of stable plaques such as low macrophage staining and strong smooth muscle staining (20).

Carotid artery stenting (CAS) is increasingly being considered as an alternative to carotid endarterectomy and in fact, most carotid stenting is being performed in asymptomatic patients. The SAPPHIRE trial found that patients at high medical risk for endarterectomy who were subsequently randomized to stenting or endarterectomy had a lower risk of stroke, myocardial infarction, or death with stenting (12.2% at 1 year) compared to surgery (20.1%) (49). However, the study did not include a control group who were treated medically, and it remains unclear as to whether stenting is superior to intensive medical therapy in patients considered high risk for endarterectomy. In the 3-year results of the SAPPHIRE study, there was not a clear difference between carotid endarterectomy and carotid artery stenosis (17). The relatively high death rate in both groups raises questions about whether either procedure is necessary in “high risk for carotid endarterectomy” asymptomatic patients.

The Carotid Revascularization Endarterectomy versus Stenting Trial (CREST) was reported in 2010 (08). This study enrolled conventional risk patients and subjects were assigned to either carotid endarterectomy or carotid artery stenting. The study originally included symptomatic patients only, but asymptomatic patients with more than 60% stenosis on angiography or more than 70% stenosis on ultrasound were included in 2004. The study was not powered to detect a difference in the 1176 asymptomatic patients, however. The primary endpoint of stroke, myocardial infarction, or death within 30 days or ipsilateral stroke after 30 days did not differ between the two groups (carotid artery stenting 3.5%, carotid endarterectomy 3.6%) (08). At four years, there was a trend favoring carotid endarterectomy if periprocedure myocardial infarction events are not included (carotid artery stenting 4.5%, carotid endarterectomy 2.7%, p=0.07).

Carotid artery stenting has also been compared to carotid endarterectomy in the Asymptomatic Carotid Trial 1 (ACT-1) (38). This randomized trial enrolled patients younger than 80 years of age, and the results were comparable with carotid endarterectomy and carotid artery stenting. The 30-day rate of stroke, myocardial infarction, or death was 3.8% with carotid artery stenting and 3.4% with carotid endarterectomy.

Another study that has reported comparative outcomes between carotid endarterectomy and carotid stenting is the Asymptomatic Carotid Surgery Trial 2 (ACST 2) (18). The study randomized patients who, in the view of the treating physician, were suitable for treatment with either procedure. Both procedures had an approximately 1% rate of disabling periprocedural stroke. Overall, periprocedural stroke or death was noted in 3.5% of stenting patients and in 2.6% of endarterectomy patients. There was no difference in the 5-year outcomes of the procedures. ACST 2 provides a similar message to other studies in that periprocedural strokes are slightly lower with carotid endarterectomy, but long-term outcomes are similar.

In community practice, however, carotid endarterectomy appears to have better outcomes than carotid stenting. In an analysis of over 20,000 patients treated at 186 hospitals in the United States, the in-hospital stroke and death rate was 4.0% with carotid stenting and 1.5% with carotid endarterectomy (12). This difference remained statistically significant (p< 0.001) after adjustment for baseline characteristics.

A systematic analysis found that results of carotid endarterectomy are improving. Registry and clinical trial showed an average 6% reduction in major complications during the period 1991 to 2010 (32). The authors proposed that the new benchmarks for an asymptomatic patient should be stroke/death less than 1.2% and a periprocedural mortality rate of less than 0.4%.

Evidence is accumulating that improvements in medical therapy have led to a reduction in stroke risk for patients with asymptomatic carotid stenosis. In the Oxford Vascular Study, 101 patients with greater than 50% carotid stenosis were treated medically (30). During a mean of three years follow-up, only one minor stroke and five transient ischemic attacks occurred. The annual stroke risk was 0.34%. A study from the Netherlands followed 293 patients with asymptomatic stenosis of 50 to 99% for a mean period of 6.2 years. The any territory risk of stroke was 0.4% in patients with 50 to 99% stenosis and 0.5% in patients with 70 to 99% stenosis (13). This study, along with the transcranial Doppler studies mentioned above, suggests that carotid revascularization may not be necessary in the majority of patients with asymptomatic carotid stenosis. An analysis of data from the U.S. VA Hospital system also found that the benefit of endarterectomy has been reduced in recent years and that intensive medical therapy is a reasonable first option (26).

Due to the advances in medical therapy, some have advocated multi-center clinical trials to compare aggressive medical therapy (AMT) alone versus AMT plus carotid revascularization (09). The CREST 2 study is in progress to compare AMT alone versus either AMT plus carotid endarterectomy or AMT plus carotid stenting (11). AMT consists of aspirin, high dose statin therapy with a LDL goal of less than 70 mg/dl, blood pressure reduction targeted to national guidelines, and lifestyle modification (23). This North American study will have two parallel trials comparing AMT alone versus AMT plus carotid endarterectomy and AMT alone versus AMT plus carotid stenting. As of December 2023, over 2300 patients have been enrolled, with an ultimate goal of 2480 patients. The CREST 2 trialists have published useful algorithms for systematic treatment of lipids and blood pressure (46).

An asymptomatic carotid stenosis that has been reported is the SPACE 2 study (35). The study was halted prematurely due to slow enrollment and loss of funding. The study enrolled patients between ages 50 to 85 years and with extracranial internal carotid stenosis of at least 70%. Patients were assigned to either CEA + best medical therapy (BMT), CAS+ BMT, or BMT alone. The primary endpoint was stroke or death within 30 days plus ipsilateral stroke beyond 30 days. The endpoint rates were 2.5% with CEA + BMT, 3.1% with BMT alone, and 4.4% with CAS + BMT. The authors concluded that neither CEA or CAS were shown to be superior to BMT alone but that the results should be viewed with caution. It is notable that the annual stroke risk with BMT alone decreased from 2.2% in the ACAS trial to 0.6% per year in SPACE 2.

Symptomatic bruits. Patients with hemispheric transient ischemic attacks or amaurosis fugax associated with lesions of less than 50% by ultrasound should be treated with aspirin and risk factor modification. If the patient has had recent symptoms (in the previous 6 months) and if there is severe (greater than 70%) stenosis, then carotid endarterectomy will be of benefit in reducing the stroke rate if the patient is an otherwise stable surgical candidate and the surgery can be performed with a less than 6% stroke rate (34). For patients with 50% to 69% stenosis, carotid endarterectomy is likely to be of a greater benefit in patients with hemispheric as opposed to retinal symptoms, patients with ulcerated lesions as opposed to smooth lesions, and men compared to women (06; 10).

Special considerations

Pregnancy

Little information is available. Carotid bruits, like cardiac outflow murmurs, may be more frequent in pregnant women. This may be related to an increased cardiac output (increased flow) or decreased blood viscosity (decreased hematocrit) associated with pregnancy.

Anesthesia

All patients during carotid endarterectomy should at least have continuous monitoring of blood pressure (by intraarterial catheter) and ECG. The most commonly used regional techniques include superficial cervical plexus block, local infiltration, and deep interscalene block. The advocates of regional techniques claim a better ability to assess the patient's neurologic status, a lower incidence of perioperative myocardial infarction (50), a decreased incidence of blood pressure fluctuations in the perioperative period, and a hospital cost savings related to a decrease in intensive care unit requirements (15). The evaluation of neurologic status under regional anesthesia allows for a rapid and accurate decision regarding the need for intraluminal shunting during carotid endarterectomy. General anesthesia with EEG monitoring, on the other hand, has the advantage of reducing the cerebral metabolic demand of the brain while increasing cerebral blood flow and provides for good airway control, reduced patient anxiety, and a quiet surgical field. A trial of 3526 patients in which patients were randomly assigned to treatment with general or local anesthesia found no difference in the rate of stroke, death, or myocardial infarction within 30 days of surgery (16).

Regardless of the anesthesia technique chosen, two objectives must be met in all patients undergoing carotid endarterectomy: (1) to provide a safe approach in patients with multisystem vascular disease, and (2) to allow for reliable, continuous neurologic monitoring.