Age

Advanced age is one of the most commonly used factors in surgical risk stratification. Given the less invasiveness and avoidance of general anesthesia, CAS was initially considered a promising alternative to CEA, especially in elderly patients. Today, there is compelling information supporting CEA rather than CAS in the treatment of old patients, particularly octogenarians.^{20,31,110,111}

In a post hoc analysis of the SPACE trial, it was found that age (≥68 years) was significantly associated with the risk of stroke and death in the CAS group (P = .001) but not in the CEA group (P = .534).²⁰ Similarly, CREST showed that older age increases the risk for the combined endpoint of stroke/death or myocardial infarction (MI) for CAS (P = .02), with the efficacy of CAS and CEA approximately equal at the age of 70 years.³¹ CAS risk for the primary endpoint increased with age (P < .0001) by 1.77 times (95% confidence interval [CI], 1.38-2.28) per 10-year increment, whereas there was no evidence of increased risk for CEA (P = .27). The treatment-by-age interaction for CAS and CEA was not altered by symptomatic status (P = .96).³¹

Pooled analysis of the three European RCTs (EVA-3S, SPACE, and ICSS) from the Carotid Stenting Trialists’ Collaboration found that in patients 70 years or older, there was a twofold increase in the 120-day stroke or death risk with CAS over CEA: 12.0% versus 5.9% (risk ratio [RR], 2.04; 95% CI, 1.48-2.82; P = .0053). For patients younger than 70 years, the risk was comparable: 5.8% in CAS and 5.7% in CEA (RR, 1.00; 95% CI, 0.68 to 1.47).¹¹⁰

In the Cochrane meta-analysis of pooled RCTs, the odds ratio (OR) of CAS versus CEA for 30-day death or any stroke risk was 1.16 (95% CI, 0.80-1.67) in patients younger than 70 years and 2.20 (95% CI, 1.47-3.29) in patients 70 years or older.⁴²

The Society for Vascular Surgery Vascular Registry stratified outcomes of CAS and CEA by age and compared the composite outcome of death, stroke, or MI at 30 days among 1347 CEA and 861 CAS patients younger than 65 years and 4169 CEA and 2536 CAS patients 65 years and older. In patients 65 years and older, CEA had lower death (0.91% vs. 1.97%; P < .01), stroke (2.52% vs. 4.89%; P < .01), and composite death/stroke/MI (4.27% vs. 7.14%; P < .01) rates.⁸⁰ Significant difference in composite outcome for old patients persisted when the analysis was performed separately in symptomatic (5.27% vs. 9.52%; P < .01) and asymptomatic (3.31% vs. 5.27%; P < .01) subgroups.⁸⁰

Last, in-hospital data derived from a large number of carotid procedures (495,331) retrieved through the Nationwide Inpatient Sample (NIS) showed age 70 years and older to be an important predictor of postoperative stroke (OR, 1.7; 95% CI, 1.2-2.5; P < .0025) and cardiac complications (OR, 1.3; 95% CI, 1.0-1.6; P < .045) after CAS.¹¹¹

Although CAS is a less invasive approach, older patients frequently have extensive atherosclerotic diseases that translate into more frequent occurrence of multiple embolic sources, adverse conditions for endovascular approach (such as diffuse calcification or “shaggy” aortic arches), and impaired cerebrovascular reserve with higher sensitivity to minor cerebral emboli damage. Manipulations with wires, catheters, and sheaths in tortuous or diseased arteries proximal to the target lesion can contribute to the increased hazards of CAS in older patients.

Sex

Sex-specific differences in cardiovascular disease and CEA have been well described. However, whether such disparities apply to CAS remains uncertain as randomized and nonrandomized studies report conflicting results.^{33,48,112–115}

A secondary analysis of the CREST trial suggested that periprocedural risk of stroke/death appeared to be higher in women who had CAS than in those who had CEA: 5.5% versus 2.2% (P = .013).³³ The difference in rates was more evident in symptomatic women (7.5% vs. 2.7% in CAS vs. CEA, respectively; P = .030), whereas it was not confirmed in asymptomatic women (P = .28).³³

In the meta-analysis of the three European RCTs (Carotid Stenting Trialists’ Collaboration), an opposite trend was shown; a lower CAS to CEA risk ratio in women than in men (1.22 vs. 1.68) was assessed at 120 days.⁴⁸ A further meta-analysis including CREST and CAVATAS was con­sistent with the not real effect of sex on the risks of CAS or CEA.¹¹²

Outside RCTs, there is conflicting information in reporting the effect of sex on CAS outcome.^113–117 The potential and uncertain higher stroke risk in symptomatic women after CAS may be related to greater embolic potential in female carotid plaque.¹¹⁷ This hypothesis needs to be substantiated by more data.

Statewide risk-adjusted analysis of 18,320 women undergoing CEA and 2293 undergoing CAS found that the risk of in-hospital stroke/death was 1.7-fold higher in symptomatic and 3.4-fold higher in asymptomatic women receiving CAS compared with CEA.¹¹⁵

The ALKK-CAS registry showed that for the group of 1443 women, the total composite in-hospital endpoint of nonfatal stroke and death was comparable to that of men: 3.0% versus 3.4%. Despite the older age of women, all in-hospital outcomes of CAS in women were comparable to those of men in both symptomatic and asymptomatic patients.¹¹⁶

Currently, sex cannot be considered among high-risk criteria for CAS.

Timing in Symptomatic Patients

In recent years, there is evidence recommending treatment with “early” CEA in symptomatic patients to prevent the risk of recurrent stroke, which is four to five times higher in the early period after the onset of symptoms.¹¹⁸ Whereas for patients undergoing CEA there is increased benefit from early surgery performed within the first 7 days of the symptomatic event, the same data are conflicting when they are applied to CAS.^119–122 The Carotid Stenting Trialists’ Collaboration meta-analysis of the three European RCTs observed that patients undergoing CAS within 14 days of their most recent symptom were almost three times more likely to have a procedural stroke or death (8.6% vs. 3.2%; OR, 2.7; 95% CI, 1.36-5.45; P = .30) than patients undergoing CEA.⁴⁸ Updated data from the same group recently showed that the stroke risk in CAS compared with CEA appeared to be greatest in patients treated within 7 days of symptoms. Patients treated with CAS in this period had 9.4% risk of periprocedural stroke/death compared with 2.8% for CEA, leading to more than threefold increased risk of CAS versus CEA adjusted for age, sex, and type of event (hazard ratio [HR], 3.4; 95% CI, 1.01-11.8).¹¹⁹

Analysis of 482 symptomatic patients included in the CAPTURE registry showed that an interval of 0 to 13 days from symptoms (stroke or transient ischemic attack) to CAS was an independent predictor of adverse 30-day outcome (OR, 2.52; 95% CI, 1.33-4.78; P = .0047).⁶³ Topakian et al¹²⁰ analyzed the effect of timing in 77 CAS patients and identified old age and treatment within 2 weeks as the only predictors of increased risk for 30-day complications; stroke and death rate was 26% versus 1.9% in those treated later.

Conversely, in a cohort of 320 patients from two centers described by Gröschel et al,¹²¹ the 30-day complication rate after CAS was 8.45%, but early stenting (<2 weeks) was not associated with an increased risk of stroke.

Lin et al,¹²² using a single-center prospective registry with 224 CAS procedures in symptomatic patients, found that the risk of periprocedural stroke was 3.45% in the early (<4 weeks) and 5.95% in the late (>4 weeks) intervention groups (P = .5). The 30-day stroke/death and MI rates were also similar (6.03% vs. 8.33%).

It is intuitive that endovascular manipulation of a carotid lesion recently symptomatic would have increased embolic risk based on knowledge of carotid plaque pathology as reviewed in Chapter 97. Nevertheless, management of patients by CAS in the early period after the onset of a transient ischemic attack or minor stroke is a challenge to be solved by larger prospective studies focused on timing.

Asymptomatic Carotid Stenosis

Data from CREST and SAPPHIRE suggest that in properly selected asymptomatic patients, CAS, performed by experienced interventionalists, may be equivalent to CEA. Nevertheless, these findings must be interpreted in light of recent studies and guidelines suggesting that for many asymptomatic patients, especially those with short life expectancy or with recurrent stenosis more than true carotid plaque (like many of those included in SAPPHIRE), best medical therapy may be the optimal treatment modality.^123–128 Currently, there are insufficient data for CAS to be recommended as a primary therapy for asymptomatic patients with severe carotid stenosis. RCTs are ongoing.^38–41

Combined Carotid and Coronary Artery Disease

Because the incidence of concomitant carotid and coronary atherosclerosis is significant and presence of carotid disease has been associated with an increase in the incidence of stroke in the perioperative period for cardiac surgery, management of concomitant carotid and coronary disease has been the subject of considerable debate. The use of CAS would be a logical alternative to CEA in this setting because of its less invasiveness. However, data are not sufficiently robust to support definitive recommendations.^123–129 A review of the NIS found that the risk of perioperative stroke was 62% greater in patients undergoing CEA plus coronary artery bypass grafting (CABG) than in those undergoing staged CAS before CABG (OR, 1.62; 95% CI, 1.1-2.5; P = .02). Nevertheless, rates of in-hospital mortality were similar with the two approaches (5.2% vs. 5.4%).¹³⁰

A meta-analysis of published studies of CAS plus CABG by Naylor et al¹³¹ showed a stroke/death rate at 30 days of 9.1%. Although CAS was suggested as a valuable alternative in combined carotid and coronary operations, the stroke/death risks remained questionable because most carotid stenoses treated were asymptomatic.¹³¹ Whether the low rate of complications with CAS plus CABG shown in other small studies^132–134 reflects case selection bias or an intrinsic safety advantage remains uncertain and should be clarified in properly designed prospective studies.

One of the most challenging issues to be solved is that catheter-based carotid interventions require treatment with a potent platelet inhibitor such as clopidogrel and dual antiplatelet therapy, which can increase the risk of major bleeding associated with cardiac surgery. Conversely, delay of antiplatelet therapy raises the risk of stent thrombosis and stroke.

Aortic Arch

Factors involving the aortic arch that may increase technical complexity during CAS include the presence of extensive aortic wall irregularities with multiple atheromas (shaggy aorta) and severe aortic calcification (eggshell aorta). A patient with a shaggy aorta may have an absolute contraindication to CAS from the high risk of catastrophic distal embolism not only to the brain but, rarely, to the viscera and lower extremities. An eggshell aorta is associated too with the risk for intimal disruption and embolism because of lack of compliance while the guide wires, catheters, and sheaths are being directed. Last, the stiffness of the vessels may decrease the torqueability of wires and catheters and cause resistance to progression of the sheath or stent delivery system to the target lesion.

Arch morphology can be variable and becomes more elongated and tortuous with advancing age. The upper inner aspect of the arch becomes a fulcrum; with increasing tortuosity, the ascending aorta and the transverse arch can elongate and push the aortic valve and the origins of the innominate artery and left CCA downward. As the arch becomes more tortuous, the origins of the major branches become more difficult to negotiate by remote endoluminal access.

The shape and curvature of the aortic arch can be categorized into three types, depending on the position of the innominate artery as defined by two horizontal lines drawn across the highest point of the outer and inner curvatures of the arch. In a type I aortic arch, the great vessels arise above or in the same horizontal plane as the outer curvature of the arch. In a type II aortic arch, the origin of the innominate artery lies between the horizontal planes of the outer and inner curvatures of the aortic arch. In a type III arch, the innominate artery lies below the horizontal plane of the inner curvature of the arch (Fig. 101-1). The more inferior the origin of the supra-aortic vessels (type II or III arch), the greater the difficulty in gaining access to the carotid arteries and catheter guidance and exchange. A type III aortic arch may lead to prolonged catheter manipulation with a possible risk for aortic plaque embolization.

In addition to these arch configurations, some congenital variations in the origin of the great vessels are relatively common. The so-called bovine arch, present in up to 27% of patients,¹³⁷ may take two different anatomic patterns. In the more frequent type 2 (8% to 10% of cases), the innominate artery and left CCA share a common origin (Fig. 101-2); in the other type, the common carotid branch takes off from the innominate artery. A “pure” bovine arch, extremely rare, occurs when there is a common brachiocephalic arterial trunk originating from the arch that gives rise to both subclavian arteries (right and left) and a single bicarotid trunk. The bicarotid trunk then branches into two separate trunks, left and right CCA.¹³⁷

Carotid Tortuosity and Calcification

Heavy circumferential calcification, especially if it is associated with tortuosity along the carotid vessels, increases the difficulty of accessing the lesion. Severe distal ICA kinking or coiling may prevent positioning of the distal EPD with a landing zone sufficient for stent deployment and may predispose to vascular spasm at the end of the procedure. When the angulation is located at the distal end of carotid plaque, stent deployment may change the conformation of the vessel and cause flow-limiting angulation. On occasion, extreme tortuosity may preclude a patient from undergoing intervention.

Finally, extensive calcification in the area of the stenosis may make stent delivery more difficult and may lead to insufficient expansion after deployment and recoiling of the dilated stenosis after deflation of the balloon.

Plaque Morphology

Quantitative and qualitative carotid plaque analysis, primarily achieved with duplex ultrasound, has been suggested as an important parameter for assessing the risk of stroke during CAS.^138–142 Details of carotid duplex ultrasound are provided in Chapter 98. There appears to be a higher potential for embolism during CAS in the presence of hypoechoic or echolucent soft plaque.^139–142 However, the definition of plaque echogenicity is operator dependent and difficult to classify in terms of different grades of severity. Novel computerized ultrasound technology, such as the gray scale median score, has been proposed¹⁴¹; however, its validation and reproducibility are not universally accepted for routine use in clinical practice.¹⁴³ In general, the reliability of ultrasound evaluation and correlation with pathologic specimens remain under study, and the clinical applicability of these observations awaits solid confirmation.

Contralateral Carotid Occlusion

There are conflicting data in the literature supporting the increased risks of stroke for patients with contralateral carotid occlusion.^144–147 The ALKK-CAS Registry showed overall major stroke and death rates in patients undergoing CAS comparable to those of patients without contralateral carotid occlusion.¹⁴⁴ Similar information was provided by the Illinois experience, which found only 2.6% periprocedural stroke in CAS patients with contralateral carotid occlusion.¹⁴⁵ Brewster et al,¹⁴⁷ in a small prospective study, found two transient ischemic attacks and no perioperative stroke in the group of CAS patients with contralateral carotid occlusion, and 30-day and midterm outcomes were comparable to those of patients without contralateral carotid occlusion.

Currently, the observed outcomes do not support use of contralateral carotid occlusion as a selection criterion for CAS over CEA in the absence of other indications.

Adverse Neck Conditions

CAS is preferred to CEA in patients with local factors that increase complexity of CEA, such as tracheostomy, prior contralateral nerve palsy, lesions that extend proximally near to the clavicle or distally to the C2 vertebral body, and situations in which local tissues are scarred and fibrotic from prior ipsilateral surgery or external beam radiotherapy.^123–127 A meta-analysis of 27 articles including 361 CAS and 172 CEA procedures performed for carotid stenosis after radiation therapy showed perioperative risks for any cerebrovascular events of 3.9% with CAS and 3.5% with CEA. However, risk of cranial nerve injury was 0% after CAS and 9.2% after CEA, whereas CAS was followed by higher risk of restenosis over time.¹⁴⁸ Another systematic review including 211 surgical procedures and 510 CAS procedures for post-radiotherapy carotid stenosis showed 10.3% cranial nerve palsy occurrence (0.6% permanent) in the surgical group. There were no statistically significant differences in 30-day rates of stroke/death between surgery and CAS in both symptomatic (OR, 0.52; 95% CI, 0.14-1.98; P = .38) and asymptomatic (OR, 0.55; 95% CI, 0.06-5.42; P = .99) patients.¹⁴⁹

A complete outline for decision making based on clinical and anatomic criteria to select CAS, CEA, or medical management in patients with carotid stenosis is described in Chapter 99.

Adequate Fluid Administration

Adequate administration of fluids is the first fundamental step in reducing the risk for contrast-induced nephropathy and intraprocedural hypotension in preparing patients for CAS.

Anticoagulation

As in most vascular procedures, intraprocedural anticoagulation is necessary with CAS. Adequate anticoagulation is usually achieved with unfractionated heparin (70 to 100 units/kg) administered after arterial access is gained and before manipulation of catheters in the aortic arch. An activated clotting time not longer than 250 to 300 seconds is recommended to avoid the risk for intracerebral hemorrhage from reperfusion. Heparin is rarely reversed at the end of the procedure. Given that the half-life of heparin is about 90 minutes, the procedure is usually finished before the anticoagulant effect of the initial dose has completely worn off. Bivalirudin may be used as an alternative.¹²⁴

Atropine

Hypotension associated with marked bradycardia is a common feature after balloon dilatation, especially in elderly patients with heavily calcified stenoses.^150–152 Intravenous atropine (0.4 to 1 mg given intravenously) may be administrated before stent deployment and balloon inflation to suppress the hemodynamic response to stretching of carotid baroreceptors. In the case of a severe hemodynamic response, aggressive volume expansion and an additional dose of atropine or use of vasopressors, including intravenous phenylephrine (1 to 10 µg/kg/min) and dopamine (5 to 15 µg/kg/min) infusion, may be necessary.¹²⁴ Moderate hypotension may last 24 to 48 hours before the carotid sinus adapts to the radial force of the self-expanding stents, and pharmacologic support is occasionally required. The hemodynamic response may be exaggerated in patients with baseline bradycardia or those taking beta blockers or digoxin, whereas patients with a denervated carotid sinus because of previous CEA or with pacemakers are at lower risk. Intraprocedural transcutaneous temporary cardiac pacing is rarely indicated for CAS.^153,154

Vasodilators and Hypotensive Drugs

Accurate pressure control during and after the procedure is essential in decreasing the risk of complications. Hypertension occasionally develops immediately before, during, or rarely after CAS, and maintenance of systolic blood pressure below 180 mm Hg is advised to minimize the risk of intra­cranial hemorrhage.¹²⁴ More commonly, hypotension may develop after stent deployment and persist a few days as described before.

Spasm may appear after EPD retrieval in the distal ICA. Most cases of spasm resolve spontaneously in a few minutes. However, when spasm is severe and persistent, administration of vasodilators may be required. Commonly, nitroglycerin (100 µg) is administered directly into the ICA through the introducer sheath or guiding catheter (500 µg diluted in 10 mL and 2 mL). Additional doses may be administered every 3 to 5 minutes. The patient must be observed for possible exaggerated hypotension.

Antiplatelet Drugs

Rapid thrombus formation inside the stent and consequent potential embolization immediately after CAS provide the rationale for double antiplatelet therapy.^{123–125,127–129,155} Aspirin, the most studied antiplatelet agent, causes irreversible inhibition of platelet cyclooxygenase by decreasing production of thromboxane A₂.^155–158 However, there is interpatient variability in the response of platelets to aspirin as well as occasional resistance to the drug.¹⁵⁹ Thienopyridine derivatives, such as ticlopidine and clopidogrel, or the newer ticagrelor and prasugrel, which act as adenosine diphosphate receptor antagonists through a separate pathway from inhibition of cyclooxygenase, have been considered a valid alternative or adjunctive treatment to aspirin.^160–162

Clinical evidence supporting the use of double antiplatelet treatment to prevent vascular events is mainly derived from studies involving coronary interventions; there are few data in CAS patients.^163–170 The benefit is less clear when dual treatment is applied to patients who had ischemic stroke or transient ischemic attack.^171,172 The MATCH trial (Management of Athero Thrombosis with Clopidogrel in High-risk patients with recent transient ischemic attack or ischemic stroke) found only a nonsignificant 6.4% reduction in the relative risk for primary endpoints (stroke, MI, vascular death, rehospitalization) with a dual antiplatelet regimen.¹⁷² Similar results were obtained in the CHARISMA trial (Clopidogrel for High Atherothrombotic Risk and Ischemic Stabilization, Management and Avoidance).^173,174 The benefit of combination therapy became marginally significant in high-risk subgroups, such as patients with symptomatic vascular disease; conversely, the related risk of bleeding obviated the benefits of treatment in low-risk subgroups.^173,174

Despite uncertain information about the effectiveness of dual antiplatelet treatment in preventing cerebrovascular ischemic events during CAS, there is consensus that patients undergoing CAS should receive a regimen similar to that of those undergoing coronary stenting, as recommended by guidelines from the Society for Vascular Surgery, American College of Cardiology, American Heart Association, European Society of Cardiology, and European Society for Vascular Surgery.^{123–125,127} The standard regimen includes aspirin, 75 to 325 mg/daily, and clopidogrel, 75 mg/daily, both starting at least 4 days before the procedure. Alternatively, a loading single dose of clopidogrel (300 to 600 mg) at least 4 to 6 hours before the procedure is advocated. For patients intolerant to clopidogrel, ticlopidine (250 mg twice daily) may be used.¹²³

There is limited evidence to support the benefit of dual antiplatelet therapy with aspirin and dipyridamole after CAS.^128,175,176

Today, most centers administer dual antiplatelet therapy for 4 weeks after carotid stenting as suggested by studies documenting the occurrence of adverse events during the first 24 days and up to 30 days after the procedure and taking into account that stent endothelialization lasts 28 to 96 days.^124,127,177 Longer dual antiplatelet treatment is suggested in patients at high risk for restenosis or major cardiovascular events.^{127,163,165,168} A low dose of aspirin (75-100 mg/daily) may be considered effective for long-term treatment in standard-risk patients.

Regarding the patients scheduled for combined carotid and coronary surgery, the American College of Cardiology and American Heart Association guidelines recommend stopping clopidogrel therapy at least 5 days before CABG.^{123–125,127,178} Recent data suggest that clopidogrel may be safely continued through the perioperative period without excessive bleeding risk, and the last Society for Vascular Surgery carotid guidelines suggest that it is reasonable to individualize perioperative management of clopidogrel therapy.¹²³

Antithrombotic Drugs

Few studies have compared antiplatelet with anticoagulant therapy after CAS. In a small RCT that included 47 patients undergoing CAS, the 30-day neurologic complications rate was 0% with clopidogrel and aspirin versus 25% (P = .02) with aspirin plus heparin. The unacceptable level of complications in the anticoagulant group resulted in early termination of the study.¹⁷⁹

Warfarin is recommended for primary and secondary prevention of stroke in patients with atrial fibrillation.^180,181 On the basis of these trials in patients with carotid stenosis and a concurrent risk for cardioembolic stroke, it is reasonable to maintain the antithrombotic therapy without adjunctive antiplatelet drugs to avoid the increased hemorrhagic risk. Warfarin may be converted to intravenous heparin before the procedure.

Statins

Statins (3-hydroxy-3-methylglutaryl coenzyme A reductase inhibitors) have been found to be highly effective in preventing stroke in both male and female patients with cardio­vascular disease.¹⁸² Review and large multicenter studies analyzing the effect of statins in patients undergoing vascular surgery showed significant improvement in postoperative outcomes.^183–185 A number of studies have specifically focused on outcomes of carotid surgery, raising the question of whether similar effects would affect the outcomes of CAS, but few studies have addressed this issue.^185–189 The supposed benefit in patients undergoing carotid revascularization appears to be largely independent of the cholesterol-lowering effect of the drug; it is due to so-called pleiotropic effects because statins have been shown to stabilize atherosclerotic plaque and show anti-inflammatory, antithrombogenic, antiproliferative, and antileukocyte adhesion properties.

In a single-center experience, the incidence of cardiovascular events after CAS was 4% in statin users versus 15% in nonusers (P < .05).¹⁸⁸ Another large series showed 1.5% versus 4% 30-day stroke/death (P < .009) for users and nonusers.¹⁸⁹ On the basis of available data, it is reasonable to prescribe statins early before the procedure but also to monitor liver function and creatine kinase levels thereafter to detect potential side effects of the drug.^124,127

Access

Remote Access to the Target Lesion

Stable sheath access in the proximal CCA is a critical step. Vessels that originate below the apex of the aortic arch are more difficult to cannulate because wires and catheter insertion and exchange become increasingly difficult with a type II or III aortic arch.

In procedures with remote access, two main techniques to advance the catheter into the CCA are commonly used. Once it is in place, the guiding catheter or long sheath side port is intermittently irrigated or attached to a slow continuous infusion of saline to avoid stagnation of blood in the sheath and then connected to a blood pressure monitoring system.

Arteriography

When a stable working platform has been reached in the CCA, an arteriogram through the guiding catheter or sheath should be obtained. Anteroposterior or lateral projection is needed to obtain minimal overlap of the ICA and ECA, optimal visualization of the target lesion, and complete intracranial carotid vessel distribution. When differentiation between the two carotid branches is not easily obtained with two projections, an adjunctive oblique projection is performed following the best view obtained from the preprocedural three-dimensional computed tomographic angiography reconstruction. Rotational angiography (the image intensifier rotates around the patient and acquires a series of images that are then reconstructed through software algorithms into a three-dimensional image) is occasionally used, when it is available, in selected cases of difficult visualization of the carotid lesion (Fig. 101-3).

Stent System Delivery

Categories of Embolic Protection Devices

Today there are a large variety of EPD models with different mechanisms that extend its applicability to disparate anatomic conditions. Three conceptually different methods, each with advantages and drawbacks, have been developed.

Selection of Embolic Protection Device Type

The selection of filter device versus proximal or flow reversal systems is guided by the patient’s anatomy but often related to operator preference. It is hoped that technologic improvements including lower profile of the device, better efficacy in recruiting debris, and reduced need of manipulation for the proximal occlusion balloon systems can improve the safety of the procedure.*

A number of studies, mostly industry sponsored, have been conducted to show the safety of different EPDs.^203,213,214 Randomized (e.g., PROFI trial)²¹⁴ and mainly not randomized studies, including small samples of patients, on direct comparison between proximal occlusion and distal filter, seem to show lower rates of periprocedural stroke^213,214 or reduced rates of microembolic lesions with the proximal occlusion systems,^{198,201,202,215} but this awaits confirmation in larger population studies.

Despite considerable improvements, no EPD model has been shown to be completely effective, and the problem of cerebral emboli during CAS in different steps of the procedure (cannulation, placement of the EPD, balloon dilatation, and stenting) cannot be considered completely solved. The operator’s skill is crucial, and each interventionalist should be familiar with at least two different classes of EPD so that the technique can be tailored to the individual case.¹⁷⁸

Closed Cell versus Open Cell Stents

An important technologic feature of stents for the carotid artery has been the development of two main configurations: closed cell, in which all stent struts are interconnected; and open cell, in which not all struts are interrelated. Open cell stents are more conformable than closed cell stents and preferred in tortuous anatomy (Figs. 101-5 and 101-6).

Stent selection is often based on the physician’s preference. Although it has been suggested that closed cell stents with a smaller free cell area and a greater percentage of wall coverage offer better scaffolding, reducing embolization risk compared with the open cell structure, there is no consensus on that point.⁷²

In a multicenter experience of 3179 CAS procedures, Bosiers et al⁷² reported a significantly lower 1.3% adverse event rate in patients with closed cell stents and 3.4% in those with open cell stents (P < .0001), mainly for symptomatic patients. At the opposite, in another multicentric study of 1648 patients, Schillinger et al⁷³ reported a similar 30-day stroke and death rate for closed cell and open cell stents (3.1% vs. 2.4%; P = 0.38). A small randomized study of open versus closed cell stents by Timaran et al²¹⁶ failed to show increased stroke/death rate or new cerebral lesion by magnetic resonance angiography (53% vs. 47%) or cerebral embolic Doppler signals (264 vs. 339) in procedures performed with the two different types of stents. Histologic analysis of material retrieved within the distal filter suggested that the use of open cell stents was associated with a larger mean particle size but had no impact on procedural outcomes or total number of embolic particles compared with closed cell.²¹⁷ More data are needed for final conclusions to be drawn about open versus closed cell stent selection or other characteristics that may ultimately determine optimal stent performance.

Spasm and Thrombosis

Some ischemic complications may be related to the use of an EPD. Spasm at the site of filter deployment is common but usually minor and does not require intervention or antispasmodic drug use (e.g., nitrate 200 µg). On occasion, severe spasm, dissection, or debris-full filter may be severe enough to cause flow interruption. Arrest of the flow should be managed by slow closure and withdrawal of the filter and eventually spasmolytic drug use (Fig. 101-7). Flow-limiting dissection would best be managed with further stenting.¹²⁷

Acute stent thrombosis is a rare but potentially disastrous complication after CAS. It can be caused by stent misplacement, severe recoiling due to annular calcification, or massive plaque protrusion and requires immediate stent removal with endarterectomy.²¹⁸

Embolization

Distal brain embolization from atherosclerotic plaque or catheter-generated thrombus is the most common complication reported after CAS. The incidence varies and in some cases may be related to operator experience and patient selection. Advanced age, difficult arch, and long or multiple lesions have been implicated as predictors of embolic stroke.²¹⁹ When neurologic symptoms occur and intracranial arterial occlusion is visible on the angiogram, a number of neurorescue maneuvers have been suggested for retrieval of distal emboli, including catheter-directed chemical thrombolysis with urokinase or recombinant tissue plasminogen activator, thrombus maceration, aspiration thrombectomy, snare removal, and glycoprotein IIb/IIIa receptor inhibitor (e.g., ReoPro) administration, all with variable results. Presently, prevention remains the best option to avoid disastrous consequences of distal cerebral embolization.^23,24,73

In general, consequences of cerebral embolization can vary from silent cerebral infarction to transient ischemic attack and stroke of varying severity. Studies with TCD or diffusion-weighted magnetic resonance imaging (MRI) showed increased incidence of cerebral infarcts in patients undergoing CAS than in CEA.²²⁰ In the ICSS trial, about three times more patients in the CAS group than in the CEA group had new ischemic lesions on the diffusion-weighted MRI postprocedural scan (50% vs. 17%; adjusted OR, 5.21; 95% CI, 2.78-9.79; P < .0001).²²¹ Echolucent lipid-rich carotid plaques are more likely to be associated with increased embolic risks,^222,223 whereas recurrent carotid stenosis from intimal hyperplasia and fibrous lesions are associated with a decreased occurrence. Postprocedural ischemic brain lesions detected by diffusion-weighted MRI can similarly occur in patients with symptomatic or asymptomatic carotid stenosis and are found in approximately half of asymptomatic patients after CAS.^220,224 The proportion of patients with new ipsilateral diffusion-weighted MRI lesions might be significantly lower after protected versus nonprotected CAS, but data are not robust to confirm this hypothesis.^202,215,224

Although most studies with TCD and diffusion-weighted MRI showed varying and at times high rates of intraprocedural embolization with different plaque compositions and size, most of these signals were not associated with acute clinical consequences. However, in general, when symptoms of focal neurologic injury develop during CAS and no immediate cause is found, it is best to complete the procedure, to retrieve the catheters, and to reassess the patient.

Neurocognitive Complications

Although most of the identified new cerebral infarcts after CAS are subclinical, there are concerns that these lesions might be associated with subtle long-term neurologic changes and deterioration of cognition.²²⁵ The ICSS trial in a subgroup of patients undergoing neuropsychological examination showed that differences between CAS and CEA on cognition at 6 months after revascularization were small and not statistically relevant, despite new ischemic lesions found twice as often after CAS than after CEA (RR, 2.1; 95% CI, 0.0 to 4.4; P = .04).²²⁶ Wasser et al²²⁷ prospectively analyzed neuropsychological outcomes with a multiple domains test battery and diffusion-weighted MRI lesions after 24 CAS and 31 CEA procedures. New diffusion-weighted lesions were detected in 71% of CAS patients and 4% of CEA patients; however, none of the cognitive domains decreased significantly at 72 hours, and the overall cognitive performance was not significantly different between patients with and without new diffusion-weighted MRI lesions.²²⁷ A number of studies and literature reviews seem to support similar negative findings, whereas some studies reported an overall improvement in cognitive function after CAS. Neurocognitive effects after silent cerebral embolization from CAS remain largely uncertain.^227–230 The most common finding in the literature is that new brain subclinical lesions after CAS seem not to affect cognitive performance of the patient.

Intracranial Hemorrhage

Intracranial hemorrhage is a rare (<1%) and devastating complication after carotid revascularization that has been particularly detrimental after CAS.^231,232 Large NIS data including 215,012 CEA and 13,884 CAS procedures showed that patients undergoing CAS had sixfold increased risk of intracranial hemorrhage (OR, 6.07; 95% CI, 4.7-7.8; P < .0001).²³¹ Reasons for these CAS versus CEA disparities are unclear and may be related to the dual antiplatelet regimen used for CAS or baroreceptor deregulation after stent deployment in the carotid body, leading to cerebral hyperperfusion.²³³ However, similar higher risks of intracranial hemorrhage for CAS are not supported by findings from RCTs comparing CAS and CEA. In a subgroup analysis, the SPACE trial investigators found ipsilateral intracerebral bleeding after one CAS procedure (0.17%) and five CEA procedures (0.86%), resulting in a nonsignificant lower trend in favor of CAS (OR, 0.19; 95% CI, 0.004 to 1.74).¹⁹ The CREST trial found that intracranial hemorrhages were severe and devastating (two thirds fatal) but not more common in the CAS arm (4 in 1262 CAS vs. 3 in 1240 CEA).²³⁴ Similarly, in the EVA-3S trial, three intracranial hemorrhages occurred in the CAS (n = 261) and three in the CEA (n = 259) arm.¹⁷ In these RCTs, intracranial hemorrhage occurred days after intervention, raising doubts about the benefit of the underlying dual antiplatelet therapy and suggesting the imperative need of blood pressure control days after carotid revascularization.

Patients with symptomatic carotid stenosis, hypertension, and preocclusive internal carotid stenosis, especially in association with contralateral severe carotid stenosis or occlusion, are more likely to develop intracranial hemorrhage after CAS.^235,236

Hypotension

Hypotension can occur in 19% to 51% of CAS patients immediately after stent deployment and after dilatation; it is usually transient and rarely symptomatic.^{150,151,239–241} However, it may last longer than 24 hours in 3% to 4% of the cases. There is currently no consensus as to which patients require vasopressor agents.¹²⁷ Bradycardia is also common after CAS, with a variable incidence (2.3%-37%) in cases of prophylactic atropine administration and more frequent occurrence (23%-62%) without the use of atropine before stent deployment.*

In CREST, 53 cases of severe hypotension (systolic blood pressure ≤80 mm Hg or pressors administrated ≥24 hours) and 41 cases of bradycardia occurred in the CAS arm. Rates were lower after CEA.²⁸ The EVA-3S similarly showed 4.2% hypotension or bradycardia requiring prolonged monitoring or treatment in the CAS group versus 0% in the CEA group.¹⁷

Increased age, symptomatic lesions, diabetes, and presence of calcification have been found to be significant predictors of bradycardia.^150,151,239 Nitinol stents might be associated more with hypotension and bradycardia, but data are limited.^242,243

The clinical relevance of such hemodynamic changes on cardiovascular mortality after CAS remains debatable.^{150,151,153,239–244} Preventive measures include fluid infusion in the preoperative period, intraprocedural atropine administration, and infusion of pressor amines. Prophylactic placement of a temporary pacemaker is indicated only in the presence of preexisting relevant rhythm disorders. Atropine is routinely administered prophylactically in many centers, but its selective use has been debated because of its potential side effects, including tachycardia, which increases cardiac oxygen demand.¹²⁷

Sapphire

The Stenting and Angioplasty with Protection in Patients at High Risk for Endarterectomy (SAPPHIRE, sponsored by Cordis, Johnson & Johnson Company, Warren, NJ) was designed to test the noninferiority of CAS versus CEA.^15,16 The primary endpoint of the trial was the cumulative incidence of death, stroke, or MI within 30 days after the procedure or death or ipsilateral stroke between 31 days and 1 year. The study enrolled patients with 50% or greater symptomatic carotid stenosis or 80% or greater asymptomatic stenosis with conditions that potentially increased the risk of surgery. An ancillary registry included patients ineligible for randomization. All patients undergoing CAS were treated with a self-expandable nitinol stent (Smart or Precise, Cordis) with EPD (Angioguard or Angioguard XP, Cordis). The randomized trial was ended earlier than planned because of slow enrollment; 334 patients were randomly assigned to CEA (n = 167) or CAS (n = 167), and an additional 406 patients had been included in an ancillary registry of the study. The 1-year cumulative primary endpoint rate was lower in the CAS than in the CEA group (12.2% vs. 20.1%; P = .004). The difference was more evident in the asymptomatic (9.9% in CAS vs. 21.5% in CEA) than in the symptomatic group of patients (16.8% in CAS vs. 16.5% in CEA) and became less evident over time (at 3 years, 24.6% for CAS vs. 26.9% for CEA; P = .71).¹⁶

EVA-3S

The Endarterectomy Versus Angioplasty in Patients with Symptomatic Severe Carotid Stenosis (EVA-3S) trial is a multicenter noninferiority European RCT that compared CAS with CEA in 527 patients with greater than 60% symptomatic carotid stenosis.^17,18 Various models of carotid stents and EPDs were used, but EPD use was not routine in the first phase of the trial. The primary endpoint was any stroke or death within 30 days after the procedure.¹⁷ The trial was stopped prematurely because of a significant 2.5-fold higher risk for 30-day stroke and death in the CAS group (9.6% vs. 3.9% in CAS vs. CEA). However, it was noted that in the CAS group, the 30-day incidence of stroke or death was lower among patients who received cerebral protection (7.9%) than among those treated without (25%; P = .03). At 4 years, the combined rate of 30-day stroke or death was higher with CAS than with CEA (11.1% vs. 6.2%; HR, 1.97; 95% CI, 1.06-3.67; P = .03).¹⁸ The 4-year difference was largely accounted for by the higher periprocedural risk of CAS. The conclusion of this trial was that the safety of carotid stenting needed to be improved before it could be used as an alternative to CEA in patients with symptomatic carotid stenosis.

SPACE

The Stent-Supported Percutaneous Angioplasty of the Carotid Artery versus Endarterectomy (SPACE) trial randomized 1196 European patients with symptomatic carotid stenosis (≥70% according to European Carotid Surgery Trial criteria) to CAS (n = 607) or CEA (n = 589) to test the noninferiority of CAS.^19–21 The primary endpoint was ipsilateral stroke or death between randomization and 30 days after treatment. Different models of carotid stents and EPDs were used, but an EPD was used in only 27% of the procedures. The trial was stopped before reaching the planned sample for reasons of futility and costs.^19,20 The rate of death or ipsilateral stroke was 6.9% with CAS and 6.5% with CEA (90% CI, −1.89 to 2.91; P = .09), but the difference in upper CI limits was greater than the predefined threshold to establish noninferiority of CAS versus CEA. No significant differences were detected in procedures with and without the use of an EPD. At 2 years, the cumulative rate of any periprocedural stroke or death plus ipsilateral strokes did not differ between CAS and CEA (9.5% vs. 8.8%; HR, 1.10; 95% CI, 0.75-1.61; P = .6).²¹ The authors concluded that their trial failed to prove noninferiority of CAS compared with CEA for the higher periprocedural complication rate. Consequently, the widespread use of CAS for treatment of carotid artery stenosis had yet to be considered justified.

ICSS

The International Carotid Stenting Study is a multicenter international trial that enrolled 1713 patients with symptomatic carotid stenosis of greater than 50% (North American Symptomatic Carotid Endarterectomy Trial criteria).²² The trial established a minimum of 50 stenting procedures with at least 10 in the carotid artery (or with proctoring) as credentialing requirements for operators to join the study. Use of an EPD, with different device models, was not mandatory (used in 72%). The primary outcome measure, the 120-day rate of stroke, death, or MI, was double in the CAS group (8.5% vs. 5.2%; HR, 1.69; 95% CI, 0.16-2.45; P = .006). MI rates were comparable. The majority of adverse events recorded at 120 days occurred during the initial 30-day postprocedural period; cumulative incidence of stroke/death and MI at 30 days was 7.4% versus 4.0% (P = .003) in CAS and CEA, respectively. Thus, while waiting for completion of long-term data, the authors concluded that CEA should remain the treatment of choice for patients with symptomatic carotid stenosis suitable for surgery.²²

CREST

The Carotid Revascularization Endarterectomy versus Stenting Trial (CREST) is a U.S. trial cosponsored by the National Institutes of Health and an industry partner.^28–37 To date, it is the most recent and largest RCT designed to compare the efficacy of CAS and CEA in standard-risk patients with statistical power to identify significant differences between the two treatments.²⁸ The trial included 2502 patients with symptomatic or asymptomatic carotid stenosis of 70% or greater (based on ultrasound criteria) or 50% or greater (based on angiographic North American Symptomatic Carotid Endarterectomy Trial criteria). CAS was performed with the RX Acculink stent (Carotid Stent System, Abbott Vascular, Abbott Park, Ill), and a distal EPD (RX Accunet Embolic Protection System, Abbott Vascular) was used in 96.1%. The primary endpoint was the composite of stroke, death, or MI during the periprocedural period or any ipsilateral stroke within 4 years after randomization. The 477 surgeons and 224 interventionalists who were allowed to perform procedures within the trial each met a set of standards for training and experience. In CREST, there was no significant difference in the estimated 4-year rate of the primary endpoint between the CAS (n = 1262) and the CEA (n = 1240) groups (7.2% vs. 6.8%, respectively; HR, 1.11; P = .51; 95% CI, 0.81-1.51). There was also no differential treatment effect at 4 years according to symptomatic status (8.0% vs. 6.4% in symptomatic patients and 4.5% vs. 2.7% in asymptomatic patients after CAS and CEA, respectively). Nevertheless, there were important differences detected among treatment groups at 30 days; risk of periprocedural stroke/death was significantly higher in CAS than in CEA (4.4% vs. 2.3%; P = .005), and that of MI was significantly higher in CEA (1.1% vs. 2.3%; P = .003). Periprocedural risk of stroke/death was higher after CAS for symptomatic patients (6.0% vs. 3.2%; P = .02), whereas it did not achieve statistical significance in the asymptomatic group (2.5% vs. 1.4%; P = .15). Thus, CREST confirmed higher periprocedural stroke hazards from CAS in symptomatic patients but raised concerns about the cardiac risks after CEA.

Although it has been argued that MI in the setting of carotid revascularization is not as important as stroke in terms of overall health, there are also data to support that MI as well as cranial nerve lesions may indeed be relevant complications in carotid procedures. A subanalysis of CREST showed that perioperative MI complications were independently associated with a nearly threefold increased future mortality. Mortality at 4 years was significantly higher for both patients who developed MI (HR, 3.40; 95% CI, 1.67-6.92) and those with cardiac biomarker increase only (HR, 3.57; 95% CI, 1.46 to 8.68) after the carotid procedure. The increased risk remained unchanged after adjustment for baseline risk factors.³⁰ However, perioperative strokes have an important short- and long-term impact on carotid revascularization. In CREST, quality of life analyses among survivors at 1 year indicated that periprocedural stroke had a greater adverse effect on health status domains (physical and mental) than did MI.²⁸ Furthermore, CREST supported that stroke complications implied a poorer long-term survival, with almost three times more risk of death at 4 years; the estimated 4-year mortality after periprocedural stroke (including minor) was 21.1% versus 11.6% for patients without stroke (HR, 2.78; 95% CI, 1.63-4.76).³⁷ The CREST authors suggested that the relative increase in long-term mortality risk in patients with perioperative stroke was similar in magnitude to that of patients who had a perioperative MI.³⁷

• The equivalence or noninferiority of CAS related to CEA has not been definitively proved.

• CAS outcomes are strictly related to the interventionalist’s level of experience, patient selection, and technology improvements and require well-trained centers.