Carotid endarterectomy (CEA) requires an understanding of anatomy from the aortic arch to the intracranial vessels. The relationship of multiple vital structures within a small anatomical space is crucial to proper assessment and treatment.

The anatomy begins with the aortic arch, from which the great vessels originate. In approximately 90% of individuals the aortic arch gives rise to three separate branches: the right brachiocephalic (innominate), the left common carotid artery (CCA), and the left subclavian artery. The two most common variants are a common origin of the left common and subclavian arteries (the so-called bovine arch) and a direct origin of the left vertebral artery from the aorta. The location of the origin of the great vessels within the aortic arch has recently become of keen interest because of the need to cannulate these vessels for carotid stent procedures.

Carotid stenosis is often first diagnosed as an asymptomatic finding. This is often because a duplex scan or other imaging studies such as magnetic resonance arteriography (MRA) or computerized tomographic arteriography (CTA) is performed for vague neurologic symptoms that usually are not directly attributable to carotid artery disease or for evaluation of a cervical bruit detected on physical examination. Numerous studies have demonstrated a poor correlation between a cervical bruit and the severity of carotid stenosis, although the presence of a bruit is usually indicative of the presence of carotid stenosis and warrants further investigation with a duplex scan. Additionally, a carotid duplex scan might be performed as a “screening” test in patients with other symptoms of or risk factors for atherosclerosis. Symptomatic presentation can take several forms, which usually reflects atheroembolic disease originating from plaque in the carotid artery, or rarely more global ischemic changes from overall diminished cerebral blood flow.

Transient ischemic attacks (TIAs) or an ischemic stroke may be the presenting symptom. Clinically, an ischemic stroke is determined to have occurred if the neurologic deficit is present for over 24 hours. Complete resolution of a neurologic deficit in less than 24 hours is considered to be a TIA, regardless of the severity of the transient deficit. From a practical point of view, most TIAs last only several minutes. Some patients with a clinical presentation of TIA may have evidence of acute or chronic cerebral infarction on neuroimaging by computerized tomographic (CT) scan or magnetic resonance imaging (MRI) of the brain.

When an ischemic stroke or a TIA occurs, an assessment must be performed that will identify the location of the lesion in the brain and the mechanism of the deficit. This includes a thorough history and physical examination, including a detailed neurologic examination. In the setting of an acute stroke, decisions about initiating urgent thrombolysis will require consideration.

Embolic sources of strokes include not only the ICA, but also the heart, the aortic arch, and the great vessels. A history of atrial fibrillation or intracardiac thrombus may point to a cardiac source, and plaque in the aortic arch and brachiocephalic or CCAs may provide evidence of these anatomic areas as sources of atherosclerotic emboli. The distribution of the deficit can be determined clinically and radiologically. This can often be broken down into patterns associated with specific large intracranial vessels, or more punctuate lesions associated with smaller vessels within the brain cortex. Large, named intracranial vessels include the MCA, the ACA, the posterior cerebral artery (PCA), the basilar artery, and vertebral arteries. The MCA, ACA, or PCA distributions can cause weakness in the face, extremities, and/or aphasia. The motor deficits typically occur in the side of the body opposite the hemisphere of the neurologic deficit. Classic cerebral hemispheric TIAs related to disease in the right carotid artery include left upper and/or lower extremity weakness, numbness, or paralysis. Classic hemispheric TIAs related to disease in the left carotid artery include aphasia or other speech difficulties in addition to right upper and/or lower extremity weakness, numbness, or paralysis. Atheroemboli to the ophthalmic artery, a branch of the ICA as it travels intracranially, can present with visual loss or with the classic symptom of amaurosis fugax. Amaurosis fugax is temporary monocular blindness ipsilateral to the diseased carotid artery, and is often described by the patient as a window shade coming down over the visual field and subsequently resolving. In addition to clinical amaurosis fugax, Hollenhorst plaques, which represent evidence of atheroemboli to branches of the retinal arteries, can sometimes be seen on ophthalmologic examination.

The severity of neurologic deficits can range from severe hemiplegia to minor fine motor deficits or cognitive and behavioral disturbances. Symptoms of ataxia, dizziness, vertigo, diplopia, or circumoral numbness may be caused by vertebral basilar TIAs. Syncope, light-headedness, or seizures are rarely caused by carotid or vertebral disease.

Imaging studies in combination with a history and physical examination consistent with symptomatic carotid artery disease will confirm the diagnosis. The initial test of choice is a duplex ultrasound examination, which combines B mode ultrasound with Doppler velocities. Visualization of plaque using the B mode along with increased flow velocities and disturbances of flow patterns in the Doppler examination are evidence of stenosis. The duplex scan may be supplemented with CT angiography, MR angiography, or digital subtraction angiography for additional anatomic data, including the status of the vessels proximal to the carotid artery (aortic arch and great vessels) and/or the intracranial circulation. However, if the symptoms are typical symptoms in the carotid distribution, many clinicians will perform intervention on the basis of the duplex scan alone.

Indications for extracranial cerebral revascularization have been well defined for both symptomatic and asymptomatic patients in large prospective, randomized trials. The North American Symptomatic Carotid Endarterectomy Trial (NASCET) established that symptomatic patients with greater than 70% diameter reduction of the ICA have a significant reduction in the incidence of stroke with CEA when compared with medical management alone. Symptoms consist of ipsilateral hemispheric TIA or stroke. Similarly, the Asymptomatic Carotid Atherosclerosis Study (ACAS) demonstrated improved stroke prevention in asymptomatic patients with greater than 60% stenosis treated with CEA versus those treated medically. These data were obtained from patients undergoing conventional angiography, with a caliper measurement of the stenotic portion of the ICA and similar measurement of the normal size of that same ICA more distally. Correlation of duplex scan results with conventional angiography or MR or CT angiography should be known for the laboratory performing the examination. These correlations often conform to a range of stenosis. Most practitioners use a threshold of 50% or greater for any stenosis as an indication for treatment of a symptomatic lesion. For asymptomatic lesions, some adhere to the 60% stenosis cut point established by the ACAS study, and others are more conservative, reserving intervention for patients with an 80% stenosis or greater. In general, nonspecific symptoms such as syncope, headache, lightheadedness, or dizziness are not considered to be indications for carotid artery revascularization.

The comorbidities and general health of the individual also enter into the clinical decision. Factors such as severe heart or pulmonary diseases that may increase the risk of the procedure must be taken into account. Similarly, in as much as the risk of stroke in many individuals is calculated over years rather than months, the age and life expectancy based on the comorbidities may significantly impact on the benefit of a prophylactic operation.

Lastly, indications for surgery or angioplasty are based on the assumption that the risk for stroke and death from a procedure is <3% for asymptomatic patients and <6% for symptomatic patients. If the surgical team treating the patients cannot achieve these results, the benefit of the prophylactic intervention cannot be justified based on available data.

Once a decision has been made that a patient has an appropriate indication for CEA, preoperative planning includes an evaluation of the patient’s overall medical condition and fitness to undergo surgery, and planning of the technical aspects of the procedure. With regard to the preoperative medical assessment, this would generally consist of a history and physical examination, electrocardiography (EKG), chest X-ray, and routine laboratory tests. As many patients with carotid bifurcation disease will have some degree of concomitant coronary artery disease, some clinicians perform routine preoperative assessment for coronary disease, including some form of stress testing. However, the majority of patients having CEA can undergo surgery without an extensive evaluation for coronary disease. Stress tests or other extensive cardiac testing is appropriate for those individuals with severe angina, congestive heart failure, or other serious cardiac condition. Antiplatelet therapy consisting of either aspirin therapy or clopidogrel is generally started as soon as the diagnosis of significant carotid disease is made, and this is continued up to and including the day of surgery, and through the postoperative phase. This is based on clear data that have documented a reduction in thromboembolic events after carotid surgery for patients on antiplatelet therapy.

With regard to the planning of the procedure itself, the surgeon must consider the method of anesthesia to be used, the method of cerebral monitoring to be used, the likelihood of the need for placement of a shunt during the procedure, and whether or not the patient has any specific anatomical issues that might require modification of the standard CEA procedure.

CEA can be performed under either general or local/regional anesthesia, and the method of anesthetic is intimately tied to the way in which cerebral function is monitored during the period of cessation of ICA blood flow during the procedure. When performing the procedure under local/regional anesthesia, which is the authors’ preference, the patient remains awake, and direct monitoring of the patient’s neurologic function is performed to assess for cerebral ischemia. However, in order to perform the operation successfully under local/regional anesthesia, one must have a cooperative patient who is able to adequately communicate. Although we have found that local/regional anesthesia for CEA is feasible in at least 85% of patients who undergo CEA, issues such as claustrophobia, a language barrier, or neurologic deficit from a prior stroke can make this technique less realistic in certain cases. When performed under general anesthesia, the surgeon must employ some other method of determining whether cerebral ischemia occurs during the period of carotid clamping. The most commonly utilized techniques are intraoperative EEG (electroencephalographic) monitoring, and measurement of an ICA “back pressure” when the carotid artery is clamped. This measurement reflects the adequacy of collateral blood flow to the cerebral hemisphere from collateral pathways such as the vertebral–basilar system and contralateral carotid distribution. Other less commonly practiced forms of intraoperative monitoring include transcranial Doppler (TCD) and somatosensory evoked potentials (SSEP).

With any of the above methods, when a patient is felt to have inadequate hemispheric perfusion during the period of carotid artery clamping, an intraarterial “shunt” is inserted; the shunt maintains perfusion to the cerebral hemisphere while the endarterectomy is being performed. A variety of shunts are available; the most commonly used are plastic tubes that are secured to the common and ICAs using specially designed shunt clamps, vessel loops placed around the corresponding artery, or balloons located on the shunt device itself. Under local/regional anesthesia, approximately 10% of patients undergoing CEA will require the placement of a shunt; this proportion increases to approximately 30% when total occlusion of the contralateral carotid artery exists. Many surgeons will place a shunt prophylactically in specific clinical situations, even if the patient does not demonstrate any signs of intraoperative cerebral ischemia; the
most common of these situations would be a recent stroke, or contralateral carotid artery occlusion. Some surgeons use a shunt in all CEA cases performed under general anesthesia; however, because shunt placement can make the performance of the operation more cumbersome, and can rarely be associated with complications or vessel trauma, most surgeons do not employ this approach.

Finally, preoperative evaluation must include review of the patient’s imaging studies to delineate any anatomical issues that might represent a technical challenge during surgery, or compromise the results of the surgery. These would include concomitant great vessel occlusive disease, concomitant intracranial occlusive disease, severe tortuosity or calcification of the artery, and a high carotid bifurcation. The location of the carotid bifurcation is variable; when it is located at a level above the second cervical vertebrae, special maneuvers may be required to achieve an adequate dissection in this area.