Figure 89-1. Stroke due to carotid bifurcation occlusive disease is usually caused by atheroemboli arising from the internal carotid artery, which provides the majority of blood flow to the cerebral hemisphere. With increasing degrees of stenosis in the carotid artery, flow becomes more turbulent, and the risk of atheroembolization escalates.
Figure 89-2. A: The carotid bifurcation is an area of low flow velocity and low shear stress. As the blood circulates through the carotid bifurcation, there is separation of flow into the low-resistance internal carotid artery and the high-resistance external carotid artery. B: The carotid atherosclerotic plaque typically forms in the outer wall opposite to the flow divider due in part to the effect of the low shear stress region, which also creates a transient reversal of flow during the cardiac cycle.
Figure 89-3. Duplex ultrasonography showing a carotid plaque with narrowing of the vessel lumen that can either embolize or cause low flow causing hypoperfusion. B-mode imaging shows a heterogeneous plaque with severe narrowing.
CLINICAL MANIFESTATIONS OF STROKE
Symptomatic carotid artery disease is classically thought of as a corresponding TIA, stroke, or amaurosis fugax. A TIA is a neurologic event that manifests stroke-like symptoms for <24 hours. These are considered a precursor of a more serious event and a large number will progress to a stroke. Symptoms correspond with a focal defect that can be related to carotid artery atherosclerosis. These symptoms typically include the anterior or middle cerebral artery circulations. Motor symptoms can include hemiparesis contralateral to the affected hemisphere. Sensory deficits can also occur in a similar fashion, such as numbness or paresthesia. Aphasia, dysphagia, or dysarthria can also occur. Amaurosis fugax is the temporary monocular blindness from cholesterol embolization to the retinal artery via the ophthalmic artery. Dizziness, syncope, vertigo, seizures, bowel or bladder incontinence, or migraines are not typically related to carotid disease and other causes must be sought. Once the symptoms progress past 24 hours, the TIA has become a full stroke. The full severity of a stroke can often take weeks to manifest as the penumbra either recovers or does not. Global ischemia is generally uncommon with carotid artery disease.
Fig. 89-4). Although labs differ in their ranges, patients are typically classified into: normal, 1% to 49%, 50% to 69%, 70% to 99% or occluded (Table 89-1). Additionally, the DUS can provide important information regarding the plaque morphology. The identification of high-risk echolucent plaques can aid the practitioner in estimating the risk of neurologic symptoms. The diagnostic accuracy of DUS does, however, depend on having a skilled sonographer. Magnetic resonance angiography (MRA) and computerized tomographic angiography (CTA) are becoming more popular for diagnosis with improvements in the technology. These imaging modalities can provide important information regarding the aortic arch, tortuosity, brachiocephalic disease, carotid bifurcation location, and intracerebral collateral circulation. They can, however, overestimate the amount of stenosis and often require correlation with DUS. Angiography remains the gold standard for diagnosis; however, the procedure itself carries approximately 1% risk of neurologic complications (Fig. 89-5). Transcranial Doppler (TCD) can provide information related to the significance of a stenosis and how it alters intracerebral hemodynamics.Carotid artery duplex ultrasonography (DUS) is the primary diagnostic tool for the evaluation of carotid artery disease. Although the role in screening is controversial, those with concerning symptoms should undergo DUS. The benefits of DUS include excellent sensitivity for the diagnosis of occlusive carotid artery disease, but also the avoidance of radiation and the rapid availability and ease of evaluation. The decision for carotid intervention is often taken solely on the information provided by the DUS. A complete examination includes evaluation of the common, external, and internal carotid arteries, as well as the vertebral arteries, and often the subclavian arteries. The peak systolic velocity (PSV), end-diastolic velocity (EDV), and the ratio of the common carotid artery to internal carotid artery (CCA/ICA) are used to determine the severity of stenosis (
Figure 89-4. A: Color power Doppler shows a severe stenosis at the origin of the internal carotid artery. B: The peak systolic velocity is 497 cm/s, correlating to a >70% stenosis.
TREATMENT OF CAROTID ARTERY OCCLUSIVE DISEASE
Patients with carotid artery occlusive disease are placed into two main categories: asymptomatic or symptomatic. A recent (<6 months) TIA, stroke, and amaurosis fugax are considered symptoms of occlusive carotid disease. These patients have the highest risk of recurrence of symptoms or possible ipsilateral stroke. In patients who have a TIA, 15% will go on to a CVA, 10% within 90 days.7 Several studies have examined the risk reduction of stroke with medical and surgical management.
Patients with symptomatic lesions are at the highest risk of ipsilateral stroke, and the degree of stenosis corresponds with the risk. Validated in multiple trials – most prominently in the North American Symptomatic Carotid Endarterectomy Trial (NASCET) study – the degree of stenosis was correlated with Stroke risk. A stenosis <50% was unlikely to cause neurologic manifestations. In patients with >50% stenosis, those treated with medical therapy were more likely to progress to ipsilateral stroke than those that underwent surgical management. This benefit to surgical intervention was highest in the 70% to 99% stenosis range. In the NASCET study, symptomatic patients were assigned to medical therapy or endarterectomy with symptomatic disease and a stenosis >50%. After 2 years, there was a significant reduction in ipsilateral stroke in those who underwent a CEA ([50% to 69% stenosis: 22.2% in medical patients and 16.7% in surgical patients over 5 years] and [70% to 99% stenosis: 26% in medical patients and 9% in surgical patients over a 2-year period]).14 The European Carotid Surgery Trial (ECST) showed similar findings for severe stenosis, however, no benefit was found to surgery over medical management with mild stenosis.15 Medical therapy has advanced since the NASCET trial with the introduction of improved antiplatelet medications such as clopidogrel, and HMG-CoA reductase inhibitors. In the NASCET trial, aspirin was the medical therapy that surgery was compared against. Aspirin is an important medication in secondary stroke prevention. It can be used alone or in combination with dipyridamole or clopidogrel. Most neurologists recommend aspirin and clopidogrel in symptomatic patients for prevention of stroke until surgical intervention is performed. Statins lower stroke risk by 30% by plaque stabilization and are important for the reduction of stroke.16
Figure 89-5. A carotid angiogram with digital subtraction angiography reveals a patch repair after prior endarterectomy.
Timing of intervention after stroke continues to be debated. Traditionally, surgeons waited >6 weeks prior to intervention but there is a chance of recurrent stroke which is highest in the short term. Performing the surgery too soon can also have devastating consequences such as hemorrhagic conversion of an ischemic lesion. In patients that have returned to baseline from their symptoms, the current thought is to undergo surgery somewhere between around 2 days and 2 weeks after the event.17 If the patient doesn’t return to normal or has a dense lesion on imaging, it is best to wait a few weeks. Crescendo TIA and evolving stroke are different entities and although the risk of stroke with intervention is slightly higher, urgent revascularization is warranted to prevent a large stroke.
18 One of the first and most important trials regarding asymptomatic patients was the Asymptomatic Carotid Atherosclerosis Study (ACAS) trial. In ACS, asymptomatic patients with >60% stenosis were randomly assigned to either medical management or CEA. After 5 years, best medical therapy was found to be inferior to surgery in regard to ipsilateral CVA (11% for medical management vs. 5.1% for CEA over 5 years). Therefore with endarterectomy, there was a risk reduction in stroke of 53%.19 In Europe, the Asymptomatic Carotid Surgery Trial (ACST) showed similar results, with a benefit in patients undergoing surgery over medical management in patients with ≥60% stenosis.20 The biggest benefit comes from severe stenosis and it is generally accepted that surgical intervention be pursued for stenosis ≥80%, with 60% to 79% remaining controversial. Antiplatelet therapy is also recommended in the form of aspirin for asymptomatic stenosis.21 There does not appear to be a benefit to the addition of clopidogrel.22 Best medical management with risk factor modification is mandatory for all patients with severe carotid stenosis, whether symptomatic or asymptomatic. Currently, several clinical trials are comparing carotid intervention versus best medical management for the treatment of asymptomatic carotid stenosis.Treatment for patients with symptomatic disease is well established in the literature, however, the treatment of asymptomatic carotid disease is not so clear-cut. Detection of a carotid stenosis is typically based on a DUS finding or evidence of a carotid bruit. The rate of neurologic symptoms with an asymptomatic bruit was estimated to be 4% per year.
CAROTID ENDARTERECTOMY VERSUS CAROTID STENTING
The choice for intervention in symptomatic (>50% stenosis) and asymptomatic (>80% stenosis) patients is currently widely accepted. The next decision is to which treatment modality should be undertaken. CEA traces back to the 1950s and is considered the gold standard for the surgical treatment of occlusive carotid artery stenosis. Since the Food and Drug Administration (FDA) approved CAS in 2004, there was a proliferation of the technology and use for the treatment of symptomatic and asymptomatic carotid disease. High-risk patients from an anatomic or medical reason, as well as elderly patients were thought to benefit from CAS over CEA. The Stenting and Angioplasty with Protection in Patients at High Risk for Endarterectomy (SAPPHIRE) trial examined CAS versus CEA for high-risk patients (medical comorbidities and anatomical features), and CAS was associated with a lower risk of major events at 1 year, with a greater advantage in asymptomatic patients.23 The Endarterectomy Versus Angioplasty in Patients with Symptomatic Severe Carotid Stenosis Trial (EVA-3S) trial was a European trial for symptomatic patients with >60% stenosis that was stopped prematurely due to higher 30-day stroke risk with CAS. However, embolic protection device was optional and its use was associated with similar results to CEA.24 The Stent-Supported Percutaneous Angioplasty of the Carotid Artery versus Endarterectomy trial (SPACE) trial examined patients with ≥70% symptomatic stenosis failing to prove noninferiority of CAS to CEA.38 Most recently, the Carotid Revascularization Endarterectomy versus Stenting Trial (CREST) examined CAS versus CEA in standard-risk patients for both symptomatic and asymptomatic patients. Overall, there was no difference between CAS and CEA long-term, but this was likely due to a higher rate of MI with CEA, and a higher rate of periprocedural CVA with CAS. Despite the similar rate of complications overall, stroke-related complications portended a worse long-term survival.25 Currently, the Centers for Medicare and Medicaid Services (CMS) limits reimbursement for CAS to high-risk symptomatic patients with carotid stenosis >70%.
25,26 Symptomatic patients undergoing CAS appear to have a higher risk of stroke or death if performed in <7 days after the symptomatic event.27 There are no sufficient data to support CAS for asymptomatic patients outside high risk for CEA by anatomical factors (stoma, previous neck radiation, contralateral vocal cord paralysis, or high lesion) or severe medical comorbidities (coronary or pulmonary disease) (Table 89-2). Furthermore, patients with significant anatomical factors that put them at risk for CAS should undergo CEA (Table 89-3). Currently, ongoing clinical trials seek to determine more understanding of the patients that will benefit best from each therapy as well as the assessment of new stent designs, such as mesh-covered stents.Despite some uncertainty regarding CAS versus CEA, there are some subpopulations that do better than others with a particular treatment option. Contrary to initial belief regarding advanced age as a benefit for carotid stenting, age ≥68 was associated with a significantly higher risk of stroke and death for carotid stenting.
Table 89-2 Criteria for High-Risk Carotid Endarterectomy
There are many decisions a surgeon has to consider for a CEA. First, there is the choice of anesthetic: local, regional, or general. Depending on the choice of anesthetic, there is a choice of cerebral monitoring. If the patient has the procedure under local anesthesia, their ability to respond to commands dictates the adequacy of collateral flow during clamping. If the patient is under general anesthesia, there is the choice of electroencephalogram (EEG), TCD, stump pressure measurement, or somatosensory evoked potentials (SSEPs) that guide the use of a shunt for cerebral protection. EEG is the most widely used technique of cerebral monitoring. It requires the addition of 8 to 16 leads and monitors brain activity. The use of a shunt is based on a positive test, which is a drop of fast background activity of 50%. EEG, however, is likely overly sensitive, overestimating the number of people that require a shunt.28,29 TCD works in a similar manner, evaluating the middle cerebral artery (MCA) for changes in flow with clamping. Measuring a stump pressure requires clamping the common and external carotid artery while maintaining the internal carotid artery open and using a transducer to check the back pressure. A minimum mean pressure of 40 mm Hg has been most widely adopted for the threshold for shunting. Stump pressure measurement does, however, have a small risk of embolization. Others opt for routine shunting, allowing a consistent endarterectomy technique without anxiety. Shunting completely relieves intracerebral ischemia, but requires a safe technique with shunt placement above the superior aspect of the plaque. Routine shunting has been met with excellent results.30,31
Table 89-3 Contraindications to Carotid Artery Stenting
Correct patient positioning is paramount for a successful endarterectomy, especially for a high lesion. The patient’s neck is slightly hyperextended and turned away, with a roll under the shoulder to open the neck (Fig. 89-6). A longitudinal incision is made along the anterior border of the sternocleidomastoid muscle (SCM). The use of ultrasound intraoperatively can help guide the surgeon to perform a limited incision over the bifurcation. The platysma is divided. Care must be taken to avoid the anterior jugular vein and carefully divide any venous tributaries. By retracting the SCM laterally the internal jugular vein is identified in the carotid sheath. The facial vein is an important landmark for identification of the bifurcation and is divided, however, frequently the hypoglossal nerve can traverse at this site and care must be taken to avoid it. Often the superior belly of the omohyoid muscle is divided for exposure. Inferior and medial to the internal jugular vein is the common carotid artery which is carefully exposed. Care must be taken to avoid injuring the vagus nerve that is typically posterior to the artery, but can travel anteriorly in some cases. After the common carotid artery is dissected out, the external carotid and superior thyroid arteries are controlled. Often, dissection at the carotid bifurcation can cause reactive bradycardia because of stimulation of the carotid body. This can be improved with administration of 1% lidocaine directly into the carotid body. Care must be taken in dissecting out the internal carotid artery to avoid excessive manipulation to prevent embolization and avoid the more common location of the hypoglossal nerve that traverses the proximal artery. For high lesions, extension of the incision posteriorly in a periauricular fashion and division of the posterior belly of the digastric muscle can aid in distal exposure. If a lesion is felt to be very high, preoperative nasotracheal intubation and mandibular subluxation with temporary fixation can assist in exposure of the distal ICA.
Figure 89-6. The patient’s neck is slightly hyperextended and turned to the contralateral side. An incision is made along the anterior border of the sternocleidomastoid centered over the carotid bifurcation.
Once all vessels are controlled, intravenous heparin is administered (100 units/kg) prior to clamping (Fig. 89-7). As discussed previously, the choice of cerebral monitoring or routine shunting dictates how the surgeon proceeds. Typically, the internal carotid artery is clamped first to avoid embolization after ensuring the clamp site is in a normal segment of artery distal to the palpable plaque. The external and common carotid arteries are then clamped. A longitudinal arteriotomy is created from the distal common carotid artery into the internal carotid artery past the plaque. Often the ansa cervicalis limits exposure as it traverses the carotid artery, which requires its ligation once it is indeed identified as the ansa cervicalis. If necessary, a shunt can be placed from the common carotid artery to the internal carotid artery to maintain antegrade flow (Fig. 89-8). An endarterectomy can be performed to remove the plaque. Typically, a layer is teased out in the media, and the entire plaque is removed after either cutting the plaque from the proximal and distal artery or carefully pulling free trying to avoid a flap. The plaque extending into the external carotid artery is most easily dealt with by way of an eversion technique. The entire endarterectomized surface must be examined to remove flaps or debris. This is of most importance at the distal aspect in the internal carotid artery. If needed, tacking sutures with 7-0 prolene can be used to keep down a concerning endpoint to avoid a flap once flow resumes (Fig. 89-9).
At this point, closure of the arteriotomy is accomplished. Primary repair is not typically recommended due to improved short- and long-term outcomes with patch closure, and CMS uses patch closure as a quality metric during endarterectomy.32 The choices of patch include synthetic material, such as Dacron or polytetrafluoroethylene, a biologic material, such as bovine pericardium, or an autogenous vein (Fig. 89-10). Prior to closure, the shunt is removed and careful sequential flushing of each of the vessels is performed to remove air of debris. Additionally, after closure, the internal carotid artery clamp remains on for a few seconds to allow flushing through the external carotid artery to ensure any remaining debris of thrombus does not proceed into the cerebral circulation. The heparin is typically reversed with protamine sulfate. Once hemostasis is accomplished, closure is performed by approximating the sternocleidomastoid back, closing the platysma and skin. A full neurologic examination is performed prior to extubation or exit from the operating room if performed under local anesthesia.
An eversion endarterectomy is another technique for CEA and has the benefit of not requiring prosthetic material for a patch. In the eversion technique, exposure of the vessels is similar to the standard CEA. However, with the eversion technique, the internal carotid artery is transected at the bulb. The edges of the vessel are peeled back in a plane removing the plaque as it proceeds distally. A downside of the eversion technique is the inability to easily use distal tacking sutures. Closure is primarily accomplished at the bifurcation and doesn’t compromise the internal carotid artery lumen.
Figure 89-7. A: During carotid endarterectomy, vascular clamps are applied to the common carotid, external carotid, and internal carotid arteries. Carotid plaque is elevated from the carotid lumen. B: Carotid plaque is removed and the arteriotomy is closed either primarily or with a patch angioplasty.
Avoidance of a perioperative stroke is the primary objective during and after the endarterectomy. To minimize that risk, typically caused by a technical error such as a flap, thromboembolism, or carotid artery thrombosis, completion studies should be considered. At the completion of arterial closure, continuous wave Doppler analysis can be used to check the patency of the vessels by identifying normal Doppler flow or an area of stenosis. This is unlikely to pick up a small intimal flap or subtle stenosis. DUS is an excellent modality for examination of blood flow waveforms and B-mode imaging to identify small flaps or intimal defects. Angiography is considered the “gold standard” with contrast either injected directly into the common carotid artery or via a transfemoral approach. However, these studies have shown mixed results in the identification of defects, or can be overly sensitive and pick up defects that might not cause problems while repair increases the morbidity and mortality.33,34
Figure 89-8. A temporary carotid shunt is inserted from the common carotid artery (long arrow) to the internal carotid artery (short arrow) during carotid endarterectomy to provide continuous antegrade cerebral blood flow.
Figure 89-9. The distal transition line (left side of the figure) in the internal carotid artery where the plaque had been removed must be examined carefully and should be smooth. Tacking sutures (arrows) are placed when an intimal flap remains in this transition to ensure no obstruction to flow.
Figure 89-10. A: An autologous or synthetic patch can be used to close the carotid arteriotomy incision, which maintains the luminal patency. B: A completion closure of carotid endarterectomy incision using a synthetic patch.
The majority of patients tolerate the endarterectomy well and are discharged home after 24 hours. Complications can range from mild to life threatening. Postoperative blood pressure control often requires pharmacologic assistance due to carotid sinus manipulation during the procedure, but rarely persists after 24 hours. An acute ipsilateral stroke is the most dreaded complication after a CEA, especially if performed for asymptomatic disease. Cerebral ischemia can occur due to hypoperfusion or embolization during the procedure. Hypoperfusion is related to the collateral flow through the contralateral carotid artery and posterior circulation, and often relies on the status of the circle of Willis. Embolization can occur from shunt placement, the clamp site, thrombus formed during the procedure, or inadequate flushing prior to arteriotomy closure. Less frequently, the carotid artery can occlude causing a stroke. This can be related to an intimal flap, platelet adhesion from the patch, dissection, or anastomotic stricture. The importance of neurologic evaluation prior to leaving the operative suite is the quick identification of a problem, and usually requires immediate reexploration. Alternatively, if there was a normal postoperative duplex, angiography can be performed for the identification of the etiology of the stroke. The perioperative risk after CEA differs depending on the indication for the intervention. In symptomatic patients, the 30-day risk of stroke is 3.2% to 5.4%, MI 1.8% to 2.6%, and death around 1.1%.14,24,25,38 For asymptomatic patients, the 30-day risk of stroke is 1.4% to 5.4%, MI is 2%, and death is 1.1%.14,24,25,38 In CREST, the 30-day stroke and death rate after CEA was 3.2% in symptomatic and 1.2% in asymptomatic patients. Additionally, when identifying symptomatic patients treated by vascular surgeons, the 30-day risk of stroke and death was 1.3%, which increased to 3.9% when combined with MI. Overall with endarterectomy, CREST showed a periprocedural risk of death at 0.3%, CVA of 2.3%, and MI of 2.3%.25
Cardiac complications are common after CEA, and are a major cause of perioperative death.35,36 This risk is due to the high number of patients with concomitant coronary artery disease, upward of 50% of patients. Postoperative bleeding can cause a significant hematoma that can cause respiratory compromise. An expanding hematoma should be evacuated emergently and the bleeding source identified. Cranial nerve injuries are a common complication, however, the incidence varies greatly with a range between 1% and 30%, more common experience showing an incidence of 0.5% to 4.7%.25,37 Cranial nerve injuries are shown in Table 89-4. Cerebral hyperperfusion is an uncommon complication after CEA. The cause of hyperperfusion is disordered autoregulation of arteriolar cerebral blood flow after return of flow after a tight stenosis. Symptoms start with a headache and progress to seizures and intracerebral hemorrhage with a high mortality rate. Severe hypertension is a risk factor and requires urgent lowering of the blood pressure and anticonvulsant therapy. Contralateral severe internal carotid artery stenosis is another important risk factor for the development of cerebral hyperperfusion syndrome. Restenosis can occur after CEA. If occurring within 2 years, it is likely related to intimal hyperplasia and can improve. After 2 years, it is likely caused by atherosclerosis. Patch infection and pseudoaneurysm formation are also other complications following CEA.
CAS requires careful patient identification and examination of adequate preoperative imaging for procedural success. Prior to intervention, axial imaging is highly recommended. This is most generally an MRA or CTA with thin cuts. This should include imaging from the aortic arch all the way to beyond the circle of Willis. Examination of the aortic arch will identify those with anatomy that would make brachiocephalic vessel cannulation and the tracking of wires and catheters difficult. With excessive manipulation, or not identifying arch pathology, a stroke could be possible. The angle of the aortic arch, brachiocephalic ostial disease, tortuosity of brachiocephalic vessels, arch thrombus, or calcification must all be identified for safe carotid stenting. Furthermore, there are specific lesion characteristics that can make stenting more difficult or risky, such as, hypoechoic plaque, ulceration, lesion length, circumferential calcification, or thrombus. A careful physical examination must be performed to identify femoral and iliac artery anatomy that would make arterial access difficult. Although the transfemoral approach is the most common, there are a variety of other approaches that are possible, including transcervical or transradial.
Table 89-4 Cranial Nerve Injuries Associated with Carotid Endarterectomy