Endarterectomy or carotid artery stenting: the quest continues

Page 1

Scmt1fiG

0d

Review

Endarterectomy or carotid artery stenting: the quest continues

Michiel G. van der Vaart, M.D.a, Robbert Meerwaldt, M.D., Ph.D.b,Michel M.P.J. Reijnen, M.D., Ph.D.c, René A. Tio, M.D., Ph.D.d,

Clark J. Zeebregts, M.D., Ph.D.a,*aDepartment of Surgery, Division of Vascular Surgery, University Medical Center Groningen, 9700 RB Groningen, The Netherlands

bDepartment of Surgery, Isala Clinics, Zwolle, The NetherlandscDepartment of Surgery, Alysis Zorggroep, Lokatie Rijnstate, Arnhem, The Netherlands

dDepartment of Cardiology, University Medical Center Groningen, 9700 RB Groningen, The Netherlands

Manuscript received May 24, 2007; revised manuscript July 3, 2007

Abstract

Background: Carotid endarterectomy (CEA) is still considered the “gold-standard” of the treatment ofpatients with significant carotid stenosis and has proven its value during past decades. However, endo-vascular techniques have recently been evolving. Carotid artery stenting (CAS) is challenging CEA for thebest treatment in patients with carotid stenosis. This review presents the development of CAS accordingto early reports, results of recent randomized trials, and future perspectives regarding CAS.Methods: A literature search using the PubMed and Cochrane databases identified articles focusing on thekey issues of CEA and CAS.Results: Early nonrandomized reports of CAS showed variable results, and the Stenting and AngioplastyWith Protection in Patients at High Risk for Endarterectomy trial led to United States Food and DrugAdministration approval of CAS for the treatment of patients with symptomatic carotid stenosis. Incontrast, recent trials, such as the Stent-Protected Angioplasty Versus Carotid Endarterectomy trial and theEndarterectomy Versus Stenting in Patients with Symptomatic Severe Carotid Stenosis trial, (re)fuelled thedebate between CAS and CEA. In the Stent-Protected Angioplasty Versus Carotid Endarterectomy trial,the complication rate of ipsilateral stroke or death at 30 days was 6.8% for CAS versus 6.3% for CEA andshowed that CAS failed the noninferiority test. Analysis of the Endarterectomy Versus Stenting in PatientsWith Symptomatic Severe Carotid Stenosis trial showed a significant higher risk for death or any strokeat 30 days for endovascular treatment (9.6%) compared with CEA (3.9%). Other aspects–such as evolvingbest medical treatment, timely intervention, interventionalists’ experience, and analysis of plaque compo-sition–may have important influences on the future treatment of patients with carotid artery stenosis.Conclusions: CAS performed with or without embolic-protection devices can be an effective treatmentfor patients with carotid artery stenosis. However, presently there is no evidence that CAS provides betterresults in the prevention of stroke compared with CEA. © 2008 Excerpta Medica Inc. All rights reserved.

The American Journal of Surgery 195 (2008) 259–269

Keywords: Carotid endarterectomy; Embolic-protection device; Stenting; Stroke prevention

eppcdt

ocs

troke and stroke-related death are increasing causes ofoncern in the western world. Currently, stroke is the thirdost common cause of mortality [1,2]. A Swedish publica-

ion showed for the first time a stroke incidence of 213/00,000 persons annually [3]. This generates an enormousnancial burden to the western society, as shown by aerman cost analysis [4]. Direct medical costs for a first-

* Corresponding author. Tel.: �011-31-503613382; fax: �1-31-503611745.

sE-mail address: [email protected]

002-9610/08/$ – see front matter © 2008 Excerpta Medica Inc. All rights reservoi:10.1016/j.amjsurg.2007.07.022

vent, first-year survivor are �€18,517 (USD $25,016)/atient, and lifetime costs are �€43.129 (USD $58,257)/atient. This in turn accounts for 3% to 4% of total healthare costs in several European countries [4]. The estimatedirect and indirect cost generated so far by stroke in 2007 inhe United States is USD $62.7 billion [2].

Extracranial cerebral atherosclerosis causes 8% to 29%f all ischemic strokes [5]. Thrombotic emboli arising fromardiac origin are another more frequent cause of ischemictrokes [6–8]. The aim of treatment for patients with carotid

tenotic disease lies in decreasing the risk of disabling

ed.

Page 2

sefvmertphacaatts5pmdA(aa[l[mc

baosictdpred

M

dCfr

H

wbiIccc

tEA(tIscdtomrswhgfdeciTatspa

s(p45osaabppttWeitft

eSpfpDsvqn

260 M.G. van der Vaart et al. / The American Journal of Surgery 195 (2008) 259–269

troke or stroke-related death as consequences of thrombo-mbolism. Different medical-treatment strategies evolvedrom studies initially aimed at treating patients with cardio-ascular disease. For a long time, treatment consisted of 2ain modalities: medication and/or open surgery (carotid

ndarterectomy [CEA]) [9–12]. For most patients with ca-otid stenosis, surgical endarterectomy, rather than medicalreatment, became the treatment of choice for stroke pro-hylaxis, with proven efficacy. Symptomatic patients whoave carotid stenoses between 50% and 99% and perioper-tive rates of stroke and/or death �6% are best treated by aombination of best medical treatment (BMT) and surgeryccording to guidelines of the American Heart Associationnd results of large randomized trials [12–16]. The Asymp-omatic Carotid Atherosclerosis Study (ACAS) concludedhat also asymptomatic patients with a carotid artery steno-is �60% are good candidates for CEA, with a reduced-year ipsilateral stroke risk. Especially for this category ofatients, CEA should be performed with low morbidity andortality rates in order to achieve a considerable risk re-

uction warranting the risks of the operation [17]. Themerican Heart Association therefore recommended that

grade A) CEA be performed in asymptomatic patients withcarotid stenosis of 60% to 99% if perioperative risk rates

re �3% and if the patient has a life expectancy �5 years1,12]. Results from these studies are discussed in detailater in this review. Results, as shown by Mullenix et al18], show that CEA is a safe, effective, and durable treat-ent even when not performed in “high-volume” CEA

enters [18].Despite the proven efficacy of CEA, great interest has

een generated in carotid angioplasty and stenting (CAS) asn alternative to surgical therapy. The assets of CAS seembvious in patients with hostile necks because of previousurgery and/or radiotherapy [19]. Moreover, CAS is lessnvasive compared with CEA and has decreased risk forranial nerve damage as well as the ability to treat lesionshat are beyond the reach of CEA [20]. During the lastecade, several trials and series have been published com-aring CAS with CEA [21–29]. The aim of this article is toeview the literature concerning the results of CAS and tolucidate on its current status. In addition, future options areiscussed.

ethodsA literature search using the PubMed and Cochrane

atabases identified articles focusing on the key issues ofEA and CAS. Manual cross-referencing was also per-

ormed, and relevant references from selected papers wereeviewed.

istoryThe first successful extracranial CEA (ICEA) procedure

as performed in 1953. However, it took almost 20 yearsefore the results for therapy of patients with (symptomatic)nternal carotid artery (ICA) stenosis were reported [30,31].n the 1980s, CEA was the most-performed vascular pro-edure. It was not until the last two decades of the 20thentury that results from large randomized controlled trials

onsidering BMT versus surgery were published supporting o

his previously largely unfounded practice [13–15,32]. Theuropean Carotid Surgery Trial (ECST) and the Northmerican Symptomatic Carotid Endarterectomy Trial

NASCET) are the 2 most-referred trials on the subject ofhe treatment of patients with symptomatic carotid stenosis.nclusion criteria consisted of patients who had had a tran-ient ischemic attack or nondisabling stroke in the internalarotid flow tract �6 months before enrollment. Despiteifferences in carotid stenosis analysis, both trials came tohe same conclusions [15]. Findings in the NASCET dem-nstrated a decreased 2-year stroke risk from 26% in theedical group to 9% in the CEA group, yielding an absolute

isk reduction of 17% (for patients with �70% carotidtenosis). Perioperative risk rates for stroke and/or deathere 5.8% in the surgical arm. Patients in this study whoad undergone surgical correction of high-grade stenosisained a durable benefit lasting �8 years [33]. It was furtheround that the efficacy of CEA increased with increasingegree of stenosis, previous stroke presentation, and pres-nce of ulceration. Furthermore, the presence of diabetes,oronary heart disease, or hypertension increased stroke riskn the medically treated group but not in the CEA group.he ECST showed a decrease in the 3-year risk of strokend/or death from 26.5% in the medical group to 14.9% inhe CEA group. Interestingly, the early (30-day) rates oftroke and/or death were higher in women (10.6%), inatients with systolic blood pressure �180 mm Hg (12.3%),nd in patients with peripheral vascular disease (12.3%).

From the results of the pooled data (6,092 patients), aignificant 16% absolute risk reduction during 5 yearsnumbers needed to treat 6.3) was shown for symptomaticatients with a stenosis �70% (without near occlusion). A.6% absolute risk benefit was shown for patients with a0% to 69% stenosis (numbers needed to treat 22). Overallperative risk of stroke and/or death within 30 days afterurgery was 7.1%. For patients with near occlusion, thebsolute risk reduction was 5.6% during 2 years (P � .19),nd �1.7% during 5 years (P � .9). Others have debated theenefit of CEA in patients with near occlusion. For exam-le, Fox et al [34] showed no apparent benefit of CEA inatients with near occlusion. The absolute risk reduction inhe near-occlusion group was 4.2% compared with 17.8% inhose with severe stenoses but without near occlusion [34].

ith near occlusion, there is relative protection againstmboli because arterial diameter is decreased, and this mayn part explain the relatively low long-term stroke risk inhese patients. Numerous posthoc analyses of subgroupsrom the NASCET and the ECST have been published, buthey are beyond the scope of this review article.

Other studies, such as the Asymptomatic Carotid Ath-rosclerosis Study (ACAS) and the Asymptomatic Carotidurgery Trial (ACST), were designed to investigate whetheratients with asymptomatic stenotic lesions were eligibleor CEA [35,36]. In ACST, a total of 3,120 asymptomaticatients with stenotic lesions �60%, as seen on duplexoppler ultrasound, were included. The 5-years risk for

troke (minor and major) in surgical patients was 6.4%ersus 11.8% for patients who deferred surgery. Conse-uently, a significant absolute risk reduction of 5.4% wasoted, although a subgroup analysis showed clear benefits

nly for patients �75 years old. Fifty percent of people

Page 3

oFmrdhsrmtctcipsnanc

Aatp

EE

snpaowarespoeu

saassf

samoai4wwicpith5nv

C

ppoacorComctmb

TE

A

RTBDWYVWHT

261M.G. van der Vaart et al. / The American Journal of Surgery 195 (2008) 259–269

lder in age died of unrelated causes �5 years of follow-up.urthermore, the efficacy of CEA in women compared withen was also one-third less based on higher perioperative

isks in women. Overall operative risk of stroke and/oreath within 30 days after surgery was 3.1%. The ACASad similar results: There was risk reduction of 5.9% inurgical patients with a stenosis �60%. The 2.3% operativeisk of stroke and/or death in this trial was low. Approxi-ately 50% of the strokes in the CEA arm were related to

he surgical procedure, whereas the others were related toontrast arteriography. The ASA and Carotid Endarterec-omy randomised controlled trial published in 2003, whichompared periop erative complications with CEA depend-ng on different acetylsalicylic acid dosages, disclosed aerioperative complication risk of 4.6% [37]. Notable, re-ults from the ACAS trial show that 20 CEAs would beeeded to prevent 1 stroke in 5 years of follow-up. Twonalyses performed afterwards showed that despite the highumbers needed to treat, CEA in asymptomatic patients isost-effective [38,39].

Guidelines for performing CEA were distilled by themerican Heart Association from these data. In symptom-

tic patients, the risk of stroke and/or death resulting fromreatment by CEA should be �6% and for asymptomaticatients should be �3% [40].

ndovascular treatment for carotid stenosisarly reports

As a result of the widespread use of angioplasty andtenting in the treatment of patients who have arterial ste-osis in the context of coronary artery disease, treatment ofatients with peripheral stenotic arterial vascular diseaselso evolved. With the advancing techniques of percutane-us transluminal angioplasty (PTA), treatment of patientsith carotid stenosis also became feasible. The first balloon

ngioplasty for carotid stenosis was performed in 1979, andeports in the 1980s included balloon occlusion to decreasembolic complications [41–43]. Meanwhile, carotid arterytenting has been presented in an increasing variety ofublications as a viable alternative to CEA in the treatmentf patients with extracranial carotid stenosis [43–50]. Sev-ral arguments have been brought forward to advocate its

able 1arly carotid stent series data

uthor Year n Symptomaticstenosis (%)

oubin et al [50] 2001 528 52heron [43] 1996 69 NSergeron et al [49] 1999 99 58iethrich et al [46] 1996 110 38aigand [126] 1998 50 28adav et al [29,47] 1997 107 64ozzi [127] 1997 22 45holey et al [44,57] 1997 108 56enry et al [51] 1998 163 65eitelbaum et al [52] 1998 22 68

NS � not specified.

se. The minimally invasive nature of the procedure made it w

uitable to treat patients with severe concomitant cardiacnd/or pulmonary disease. Other advantages include easyccess in patients with hostile necks because of previousurgery and/or radiotherapy. In addition, patients whosetenoses extended onto the base of the skull were accessibleor treatment.

Risk of embolic stroke limited early enthusiasm. Initialtrategies focused on neurologic rescue by fibrinolyticgents or techniques to remove embolic debris. Later treat-ent shifted from rescue to protection. Most of the results

f carotid PTA proved promising, with rates of strokend/or death ranging between 0% and 7.9%, but most stud-es were rather small and nonrandomized (Table 1) [43,4,46,47,49–52,126,127]. Inclusion and exclusion criteriaere diverse, and PTAs were randomly carried out with orithout stenting. PTA alone had its limitations because of

ts decreased (long-term) durability, but the use of a stentertainly did not rule out possible danger. Problems re-orted were direct recoil of the vessel wall after dilatation,ncreased embolism caused by catheter manipulation (4imes greater compared with CEA), severe bradycardia, andypotension after balloon dilatation and dissection [21,3,54]. Intermingled with these reports, changes in tech-ique, especially the introduction of embolic-protection de-ices (EPDs), took place.

AS technique and embolic protection devicesPresently, having the patient under local anesthesia is the

referred way of performing CAS because, in this way, theatient’s neurologic condition can be monitored continu-usly [55]. Access is gained by way of the common femoralrtery to perform selective catheterization of the commonarotid artery. Recognition of normal and variant anatomyf the aortic arch and the cervicocerebral circulation isequired for successful performance of angiography andAS. Selective angiography of both carotid arteries is rec-mmended before CAS to evaluate carotid stenosis severity,orphology, carotid tortuosity, calcification, intracranial

irculation stenosis, collateral circulation, and malforma-ions. Because there is a risk of embolization caused byanipulation during the procedure, EPDs are increasingly

eing used [56]. No randomized trials have compared CAS

ical success%)

Morbidity and mortality rate(30-day; %)

Stroke rate (%)

7.4 5.83 NS7.1 17.3 6.52 47.9 8.49 95.5 3.65.2 3

27.3 24

Technrate (

981009799

100100969599

100

ith EDPS versus CAS without EPDs. However, the avail-

Page 4

acib

bp[rciibtcndctavmliOuettbaspsncpatn

P

sisti(tSistsmgCFEu

se3c

ot[fwspfiIdccrm

IC

str

TC

R

BACCCMSCCMP

PSACCEVstnDNP

262 M.G. van der Vaart et al. / The American Journal of Surgery 195 (2008) 259–269

bility of EPDs seems to decrease the risk of embolicomplications as described by the carotid artery stent reg-stries [57,58]. Importantly, many other studies have noteen powered to show a benefit from EPD.

Three different approaches to achieve protection haveeen used: (1) distal balloon occlusion, (2) distal filterlacement, and (3) proximal occlusion with flow reversal56,59]. Although all distal EPDs are able to capture andemove embolic debris, this does not eradicate embolicomplications. Inability to deliver or deploy the EPD, EPD-nduced vessel injury, ischemia caused by occlusion, andncomplete embolic debris removal may all result in em-olic cerebral complications. Microporous filters are posi-ioned in the ICA distal to the target lesion. The filter isonstrained with a delivery sheath to pass the carotid ste-osis. Once in position, the delivery sheath is withdrawn toeploy the filter. Filters offer the advantage of continuederebral perfusion. In contrast, the delivery system is rela-ive large, which may interfere with crossing the stenosis,nd the stiffness of the system may be a problem in tortuousessels, increasing the risk of embolization during place-ent of the filter. Occlusion balloons offer the advantage of

ower device-crossing profiles, but they still require cross-ng the stenosis as well as interruption of cerebral perfusion.nce protection has been secured, the stent is put into placender angiographic control. Stents used are mostly self-xpanding, but balloon-expandable stents can be used whenreating the ostium of the common carotid artery [55]. Afterhe stent is put in place, postdilatation is applied, followedy control angiogram. A perfect anatomic end result atngiography is not pursued (most studies accept residualtenosis �30%) because aggressive balloon dilatation ap-ears to increase the risk of complications, and residualtenosis is mostly related to calcification, which often doesot resolve with repeated dilatations [60]. The entire pro-edure is performed with antiplatelet therapy, which in mostatients is achieved with a combination of acetyl salicyliccid and clopidogrel. Clopidogrel is stopped 6 weeks afterhe procedure, but acetyl salicylic acid is continued indefi-itely thereafter.

rospective multicenter registriesCompared with the early CAS series previously de-

cribed (Table 1), prospective registries with predefinednclusion and exclusion criteria, independent neurologic as-essment, and oversight committees were designed to fur-her assess safety and United States Food and Drug Admin-stration approval of CAS with EPDs in high-risk patientsTable 2). High-risk surgical patients were defined as (eg,he Boston Sci EPI: A Carotid Stenting Trial for High-Riskurgical Patients [BEACH] trial) those with a surgically

naccessible lesion, previous head and/or neck radiation,pinal immobility, restenosis after CEA, laryngeal palsy,racheostoma, contralateral carotid stenosis, age �75 years,evere comorbidity, planned coronary bypass, or history ofajor surgery [61–64]. The most common high-risk sur-

ery categories observed were anatomic criteria or previousEA. Most registries were conducted to acquire United Statesood and Drug Administration or Conformité Européene (CE;urope) approval. The primary safety end point was

sually the combined rate of myocardial infarction, s

troke, and/or death at 30 days. The primary end point offficacy was the incidence of ipsilateral stroke between0 days and 1 year. These registries did not include aontrol group.

Technical success was achieved in most studies in �97%f all patients. The incidence in 30-day myocardial infarc-ion, stroke, and/or death varied between 2.1% and 8.3%61–63,65–67]. Unfortunately, most registries did not dif-erentiate between symptomatic and asymptomatic patientshen analyzing results. However, the BEACH trial did and

howed a composite end point of 7.9% in symptomaticatients (mortality 0.1%, stroke 7.4%, and myocardial in-arction 1.1%) and 5.0% in asymptomatic patients (mortal-ty 1.6%, stroke 3.4%, and myocardial infarction 0.7%).mportantly, other registries showed that independent pre-ictors of stroke or death at 30 days included symptomaticarotid stenosis, duration of filter deployment, and baselinehronic renal failure. Most registries have not yet been peereviewed, but they have been presented at internationaleetings, so results are preliminary.

nitial randomized controlled trials comparingAS with CEA

The Leicester study was the first prospective randomizedinge-center trial investigating CAS versus CEA in symp-omatic patients [21]. The trial enrolled symptomatic low-isk patients with carotid stenoses �70%. However, the

able 2arotid artery stent registries

egistry N (% symptomatic) Combined MI/stroke/death rate (%)

30 d 1 y

EACH 480 (25.3) 5.8* (1.0/4.4/1.5) 9.1# (1.1/7.0/3.2)RCHeR 581 (23.8) 8.3* (2.4/5.5/2.1) 9.6* (0/1.3/0)ABERNET 454 3.8# 11.5#APTURE 3,500 (13.8) 5.7* (.9/4.8/1.8) NAREATE 543 (17.4) 6.2* (1.0/4.5/1.9) NAAVerIC 498 5.3# NA

ECuRITY 398 (21) 8.5# (.7/6.9/.9) NAaRESS 143 (31) 2.1* (0/2.1/0) 10 (1.7/5.5/6.3)REST 749 (30.7) 4.4* (0/4.0/.8) NAO.MA 157 (19.7) 5.7* (0/5.1/.6) NA

RIAMUS 416 (63.5) 4.6* (0/4.1/.5) NA

ARCHeR � Acculink for Revascularization of Carotids in High-Riskatients; CABERNET � Carotid Artery Revascularization Using Bostonci EPI Filterwire EX/EZ and EndoTex NexStent; CAPTURE � Carotidcculink/Accunet Postapproval Trial to Uncover Rare Events; CaRESS �arotid Revascularization Using Endarterectomy or Stenting Systems;REATE � Carotid Revascularization With ev3 Arterial Technologyvaluation; MI � myocardial infarction; MAVerIC � Endarteractomyersus Angioplasty in Patients With Severe Symptomatic Carotid Steno-

is; MO.MA � Multicenter Registry to Assess the Safety and Efficacy ofhe MO.MA Cerebral Protection Device During Carotid Stenting; NA �ot available; PRIAMUS � Proximal Flow Blockage Cerebral Protectionuring Carotid Stenting; SECuRITY � Registry Study to Evaluate theeuroShield Bare Wire Cerebral Protection System and X-Act Stent inatients at High Risk for Carotid Endarterectomy.* Data available from publications in peer-reviewed journals.# Data from: www.strokecenter.org/trials or www.cms.hhs.gov/med.

tudy was terminated after allocation of only 17 partici-

Page 5

ptC

pteds(elcCpoCs(cnweits

Cfacsprp

iapyiispnCtp�bcytywpCtiim�ec

tthc

TR

T

L

W

S

K

K

C

S

E

CIAA

T Intern

263M.G. van der Vaart et al. / The American Journal of Surgery 195 (2008) 259–269

ants. Interim analysis showed that 70% of all patients inhe CAS arm had neurologic complications. In contrast,EA was performed uneventfully (Table 3).

The Carotid and Vertebral Artery Transluminal Angio-lasty Study was an international multicenter randomizedrial with 504 patients, but it lacked strict inclusion andxclusion criteria [27]. Major outcomes within the first 30ays of treatment, defined as any disabling stroke or death,howed no significant difference between CAS and CEA10.0% vs 9.9%). Noteworthy, only 26% of patients treatedndovascularly received a stent. At 1-year ultrasound fol-ow-up, severe restenoses (70% to 90%) occurred signifi-antly more in the endovascular-treated group (CAS 14% vsEA 4%). The incidence of recurrent ipsilateral stroke ap-eared to be higher in the first year in cases of stenosesccurring after CAS compared with stenoses occurring afterEA. However, survival analysis at 3-year follow-up

howed no difference in the occurrence of ipsilateral stroke14.2%) between both groups [27,68]. The investigatorsoncluded that there was a similar major risk and effective-ess with endovascular treatment of ICA stenosis comparedith CEA, but minor complications were avoided with

ndovascular treatment. Notably, the wide 95% confidencentervals in this study for stroke rate make interpretation ofhe data even harder. Results were certainly not up to thetandard advocated by the American Heart Association.

The Kentucky randomized trials comparing CAS withEA were published in 2001 and 2004. The first publication

ocused on symptomatic patients; the latter focused onsymptomatic patients. Both studies reported low compli-ation rates for either treatment and challenged the “goldtandard” of CEA [22,23]. However, the small number ofatients in each group makes the extraordinarily low riskate difficult to interpret. Afterward, interest shifted to those

able 3andomized trials of CAS versus CEA

rial n Patients

eicester 17 Low-risk symptomatic

allstent 219 Low-risk symptomatic

APPHIRE 334 High-risk (a)symptomatic

entucky 1 104 Low-risk symptomatic

entucky 2 84 Low-risk asymptomatic

AVATAS 504 Low-risk (a)symptomatic

PACE 1,183 Low-risk symptomatic

VA-3S 527 Low-risk symptomatic

REST 2,500 Low-risk (a)symptomaticCSS 1,500 Low-risk symptomaticCT 1,540 Low-risk asymptomaticCST 5,000 Any risk asymptomatic

ACT � Asymptomatic Carotid Stenosis Versus Endareterectomy Triaransluminal Angioplasty Study; CEA � carotid endarterectomy; ICSS �

atients who might benefit most from CAS. t

The run-in phase analysis from the Carotid Revascular-zation Versus Stent Trial (CREST) focused on patient agend periprocedural risk for patients receiving CAS. Fouratient-age categories were created: �60 years, 60 to 69ears, 70 to 79 years, and �80 years. Risk of stroke or deathncreased with age, but this was seen mainly in octogenar-ans (12.1%) [24]. Risk was not mediated by adjustment forymptomatic status, use of antiembolic devices, sex, orercentage of stenosis. Notably, patients �80 years wereot excluded from actual randomization within CREST.arotid artery stenosis is relatively frequent in older pa-

ients. Large population-based studies indicated that therevalence of carotid stenosis increases to 10% in persons80 years old [69]. In a subgroup analysis of NASCET, the

enefits of CEA in patients �75 years with symptomaticarotid stenosis was compared with the benefit seen inounger patients [70]. Among medically treated patients,he highest risk of stroke at 2 years was in patients �75ears (36.5%). The perioperative rate of stroke and/or deathas not higher in patients �75 years (5.2%) compared withatients �65 years (7.9%). The absolute risk reduction byEA in patients �75 years was 28.9% (number needed to

reat 3% of patients). The ECST data also indicate thatncreasing age is associated with greater benefit from CEAn patients with symptomatic carotid stenosis [71]. Further-ore, Miller et al [72] showed, in a prospective analysis of300 CEAs performed in patients �80 years, that periop-

rative risk is increased, but outcomes remain within ac-eptable guidelines [72].

In the Stenting and Angioplasty With Protection in Pa-ients at High Risk for Endarterectomy (SAPPHIRE) trial,he hypothesis was that CAS was not inferior to CEA inigh-risk patients [29]. Both surgeons and interventionalardiologists had to meet certain procedural criteria to par-

y end point Results (%)

roke and/or death CAS 70CEA 0

oke and/or death CAS 10.4CEA 4.4

I, stroke, and/or death (1-y stroke or death) CAS 12.2CEA 20.1

roke and/or death CAS 1.8CEA 1.9

roke and/or death CAS 0CEA 0

roke and/or death (3-y stroke) CAS 10.0CEA 9.9

roke and/or death CAS 6.8CEA 6.3

roke and/or death (4-y stroke) CAS 9.6CEA 3.9

I, stroke, and/or death (4-y stroke) Active enrollmentI, stroke, and/or death (3-y stroke) Active enrollmentI, stroke, and/or death (1-y stroke) Active enrollmentI, stroke, and/or death (1-y stroke) Active enrollment

� carotid artery stenting; CAVATAS � Carotid and Vertebral Arteryational Carotid Stenting Study; MI � myocardial infarction.

Primar

30-d st

1-y str

30-d M

30-d st

30-d st

30-d st

30-d st

30-d st

30-d M30-d M30-d M30-d M

l; CAS

icipate. Surgeons were required to have performed an av-

Page 6

eprprt(spstawagoisndrdtbh7

wf1dvTa

CrinbCwd

asTpa

r[sbtotl“

R

rvaS(pwnwbstdfitcv

wgdosfmrmdnfpwtsERisaderCpioaooebtn

T

264 M.G. van der Vaart et al. / The American Journal of Surgery 195 (2008) 259–269

rage of 30 CEAs/y, with low corresponding major com-lication (eg, death, stroke, and/or myocardial infarction)ates of �1%. The interventionalists were required to haveerformed an average of 64 interventions/y, with low cor-esponding complication (eg, stroke, TIA) rates of �2%. Aotal of 723 symptomatic (stenosis �50%) or asymptomaticstenosis �80%) patients–normally deemed high risk forurgery because of concomitant morbidity, such as cardio-ulmonary disease or previous surgery–were considereduitable for entry to either the endovascular- or surgical-reatment arms. Consensus agreement by a multidisciplin-ry team of neurologists, surgeons, and interventionalistsas required for a patient’s enrollment into the randomized

rm of study. EPDs were used in the endovascular-treatedroup. The primary end point was the cumulative incidencef major cardiovascular events at 30 days and at 1 year afterntervention. The study was stopped prematurely because oflow enrollment: Most patients initially included were fi-ally excluded because perioperative risk with CEA waseemed too high (n � 409). Finally, 317 patients wereandomized to CEA or CAS. Among patients in the ran-omized study, a significantly higher number of patients inhe stenting arm had undergone previous coronary arteryypass (CAS 43% vs CEA 31%, P �0.05) and also hadigher history of cardiovascular disease (CAS 85% vs CEA4%, P �.05).

The 30-day myocardial infarction, stroke, and death rateas 4.8% in the CAS arm versus 9.8% in the CEA arm, thus

avoring endovascular treatment (P � .09) [29]. Results atyear (eg, myocardial infarction, ipsilateral stroke, and/or

eath) were also in favor of endovascular treatment: 12.2%ersus 20.1% for patients treated by surgery (P � .048).he 3-year incidence of stroke was similar between bothrms (7%).

Importantly, differences in this trial between CAS andEA treatment at the composite 1-year end point were

elated to a greater association of CEA with myocardialnfarction. Without the inclusion of myocardial infarction,o statistical differences in rates of stroke and/or deathetween both groups would have been noted (CAS 5.5% vsEA 8.4%, P � .4). The majority of myocardial infarctionsere non–Q-wave events identified by routine postproce-ural laboratory studies.

The majority of patients in the SAPPHIRE trial weresymptomatic. In the CAS arm, only 30% of patients wereymptomatic; in the CEA arm, only 28% were symptomatic.he primary end point did not differ in these symptomaticatients. In asymptomatic patients there was a differencefter one year in favor for those treated with CAS.

A Cochrane review published in 2005 showed only 5andomized controlled trials comparing CAS with CEA28]. The combined primary outcomes, defined as anytroke or death within 30 days of intervention, did not differetween treatment arms. The meta-analysis was limited byhe premature ending of 3 trials because of inconsistent usef stents and EPDs and heterogeneity of groups with regardo symptomatic status and surgical risk. Because of theseimitations, this review concluded that CEA remained the

gold standard” of treatment. f

ecent publicationsRecently, results of 2 independent randomized noninfe-

iority controlled trials, the Stent-Protected Angioplastyersus Carotid Endarterectomy (SPACE) trial and the End-rterectomy Versus Stenting in Patients with Symptomaticevere Carotid Stenosis (EVA-3S) trial, were publishedTable 3). The SPACE trial included 1,183 symptomaticatients with a �70% stenosis of the ICA [25]. Patientsere randomly allocated to either CAS or CEA. The tech-ique used by the interventional physician (ie, type of stent,hether or not to use a protection device) was not restrictedy protocol. Primary outcome was ipsilateral stroke, withymptoms lasting �24 hours or death between randomisa-ion and 30 days after treatment. The complication rate at 30ays was 6.8% for CAS versus 6.3% for CEA. The nonin-eriority test was not significant (P � .09). In this study, thenvestigators concluded that CEA remains the preferredreatment for patients with symptomatic ICA stenosis be-ause evidence is lacking for equivalent or superior endo-ascular treatment.

The EVA-3S trial included 527 symptomatic patientsith an ICA stenosis of 60% to 90% according to NASCETuidelines [26]. The primary end point was any stroke oreath within 30 days after intervention. The systematic usef a CPD was instituted during the trial on instigation of theafety committee. Analysis showed a significant higher riskor death or any stroke at 30 days for endovascular treat-ent (9.6%) compared with CEA (3.9%), with a relative

isk of 2.5% and an absolute risk of 5.7%. Although moreinor and systemic complications occurred after CEA, this

id not reach significance, except for patients with cranialerve injury. Noteworthy, the trial was ended prematurelyor safety reasons. Inclusion stopped after enrollment of 527atients, although power analysis indicated that 872 patientsere needed to reach a statistical power of 80%. Because

he study was ended prematurely, the inferiority questiontill remains. Nevertheless, the results of CAS in theVA-3s study differed from those in the SAPPHIRE trial.easons are probably multifactorial and may have been the

nclusion of more symptomatic patients in the EVA-3study, the use of a protection device and antiplatelet ther-py, the EVA-3s study patients not being at high risk ofeveloping coronary artery disease, and varying levels ofxperience in performing CAS. Furthermore, a systematiceview showed that the 30-day rate of death or stroke afterAS was 5.8% among patients treated without EPD com-ared with 1.8% among those treated with EPD [58]. Earlyn the EVA-3S trial, EPDs were not used, and the incidencef stroke was �25%. The study was even briefly stoppednd later on restarted with the incorporation of routine usef cerebral-protection devices. Nevertheless, the incidencef stroke remained higher compared with CEA. Furthervidence must be awaited from a meta-analysis, which haseen planned from the combined results after completion ofhe SPACE and EVA-3S trials and the still-ongoing Inter-ational Carotid Stenting Study [73].

he High-Risk PatientSome conclude that CAS may be an excellent procedure

or high-risk patients who are not fit for surgery. Is CEA per

Page 7

smarMbpaCoihrhtctpnrttr1eofui3a

nrpit[mapmgnclpfriOnccB

C

mR

voCmliamaC“tf

crWteaina

sCNpitsb

pemwshivysntwCtce

pHBcypiwo

265M.G. van der Vaart et al. / The American Journal of Surgery 195 (2008) 259–269

e harmful in high-risk patients? It seems that when patientseet NASCET or ACAS exclusion criteria, they are marked

s “high-risk.” However, complication rates in these high-isk patients are not per se increased when performing CEA.

ozes [74] and Mozes et al [75] analyzed their CEA resultsy stratifying according to SAPPHIRE criteria for high-riskatients [74,75]. Such criteria included positive stress test,ge �80 years, contralateral carotid occlusion, and repeatedEA. There were no statistical differences in either stroker death rate between low- and high-risk patients. Thenvestigators showed that CEA can be performed in suchigh-risk patients with acceptable standard complicationates. Ballotta et al [76] and Nguyen et al [77] showed thatigh-risk patients are more common than previouslyhought. Their perioperative neurologic and cardiac out-omes are comparable with those reported in other pa-ients [76,77]. The idea that operative risk is higher inatients excluded from NASCET or ACAS has not beenot supported. Definite accepted criteria to identify high-isk patients have not yet been developed. A study fromhe Cleveland Clinic attempted by retrospective analysiso identify a subgroup of patients who were at increasedisk for CEA. From a prospective database covering a0-year period, 3,061 patients with histories of CEA werexamined. High-risk patients were identified on the basisf presence of coronary artery disease, congestive heartailure, severe chronic obstructive disease, or renal fail-re. The composite risk for stroke, death, and myocardialnfarction was 7.4% in high-risk patients compared with.8% in others. Perhaps such patients would benefit fromlternatives to CEA.

The above-mentioned risk factors, (ie, degree of stenosis,eurologic symptoms, etc) do not sufficiently identify theeal risk presented by the patient. In contrast, plaque mor-hology may identify patients at risk for stroke duringntervention [78–82]. The risk of rupture is strongly relatedo plaque composition and degree of carotid stenosis80,83,84]. Gray-scale measurements (GSM) of intima–edia thickness using ultrasound have been studied to an-

lyze vulnerable plaques. GSM is an overall measure oflaque echogenicity in which low-GSM plaques generateore embolic particles [85]. The Imaging in Carotid An-

ioplasty and Risk of Stroke study showed that the onset ofeurologic deficits during and after intervention signifi-antly increased in patients with low GSM values [86]. Aow GSM is not a contraindication to CAS but rather aredictor of increased stroke risk. Low GSM values areurther related to future coronary events, higher rate ofestenosis, positive brain computed axial tomography forschemic lesions, and rapid plaque progression [87–91].ther modalities, such as high-resolution magnetic reso-ance imaging, have also been tested as measures of plaqueomposition [92–97]. These imaging techniques may be-ome important in the planning of future clinical trials andMT modalities.

ommentsThe longevity of CEA predominantly has been deter-

ined by comparison among large-scale randomized trials.

andomized trials comparing CEA with BMT have con- a

incingly proven that CEA significantly decreases the riskf subsequent stroke in patients with severe carotid stenosis.urrently, surgery remains the “golden standard” of treat-ent, but CAS has progressed in recent years and chal-

enged CEA. Despite many trials, only a few methodolog-cally correct randomized trials compared CAS with CEA,nd they failed to establish consensus. Using predefinedargins of noninferiority, recent trials–such as the SPACE

nd EVA-3S trials–indicated that CAS is not as good asEA. Proponents of CAS responded by focusing on theinterventionists’ experience and CAS methods” in bothrials. Consequently, many have been left questioning theuture of CAS compared with CEA.

It is important to realize that most randomized trialsomparing CEA with CAS did not succeed in achievingecruitment as determined before the study. The Leicester,

ALLSTENT, SPACE, and EVA-3S trials specified thathe total intended number of patients should be 3,772. How-ver, only a total of 1,989 patients (52%) were randomizeds a result of early trial completion because of excess in riskn the CAS arm. The expanded use of CAS outside orga-ized randomized clinical trials further threatens studies oflternatives to CEA.

Failure to achieve a study or meta-analysis with adequateize will not produce convincing evidence of the value ofAS in stroke prevention. An important reason thatASCET, ECST, ACST, and ACAS influenced clinicalractice and proved the importance of CEA is that theyncluded large numbers of patients. It is therefore importanto limit the use of CAS to randomized trials to ensuretatistical power to produce a consensus in best evidence-ased therapy in patients with severe carotid stenosis.

The technical expertise required from interventionistsarticipating in trials comparing CAS with CEA may in partxplain the excessive risk of CAS procedures. The require-ents stipulated for interventionists in the EVA-3S trialere having performed 12 previous CAS or 35 previous

upra-aortic stenting procedures, and interventionists whoad not met these requirements still were allowed to partic-pate in the study when their CAS procedures were super-ised. Furthermore, CEA has evolved during the last 30ears and has been widely used by experienced vascularurgeons. In contrast, CAS is still in development and mayot so easily be generalized. It is important to realize thathe EVA-3S trial demonstrated what happens if CAS isidely implemented by showing the results achieved whenAS is performed outside of the “top” CAS units. Again,

his calls for exclusive performance of CAS procedures inontrolled clinical trials using standards of practice andxpert technical skills.

Some have concluded that CAS may be an excellentrocedure for high-risk patients who are not fit for surgery.owever, both CAS and CEA are being compared with theMT of more than a decade ago. Notably, the cumulativeomplication risk in SAPPHIRE is striking: 17% after 3ears for both CAS and CEA. As pointed out by others,erhaps such high-risk patients are better off without stent-ng or endarterectomy [77]. Medical treatment has evolvedith modern angiotensin-converting enzymes inhibitors,ther antihypertensive drugs, statin medications, and newer

ntiplatelet therapies [98–102]. In the NASCET trial, only

Page 8

5lrrmlmawtthslHtostkfc

tamrdbrntC

arisddtcg[dsC[pmi[seaCwf

bs

pTmafla

R

266 M.G. van der Vaart et al. / The American Journal of Surgery 195 (2008) 259–269

0% of patients with increased lipid levels received lipid-owering medication. A recent meta-analysis analyzing theelationship between statins and the risk of stroke showed aelative stroke reduction rate �21% [103]. The anti-inflam-atory effects of statins seem as important as their lipid-

owering effects [104,105]. For antithrombotic therapy,ost patients in the NASCET and ACAS trials were taking

spirin only. In a large study, clopidogrel was comparedith aspirin, and clopidogrel conferred an 8.7% risk reduc-

ion for the prevention of stroke and an even greater reduc-ion in high-risk patients [106,107]. Increased levels ofomocysteine have also been associated with increasedtroke risk, and lower homocysteine levels have been re-ated to a lower risk of cardiovascular restenosis [108–110].owever, high levels of homocysteine seem not to increase

he risk of restenosis after CEA [111,112]. Finally, the usef angiotensin-converting enzyme inhibitors also decreasestroke risk, as demonstrated by the Heart Outcome Preven-ion Evaluation Study [113]. In this study, patients notnown to have low ventricular ejection fraction or heartailure derived benefit from using ramipril, not only fororonary events but also for ischemic strokes.

In contrast, one should consider patient commitment, pa-ient potential to receive lesser therapy due to randomization,nd cost when advocating repetition of previous trials. Further-ore, despite the proven efficacy of BMT, it is important to

ealize that only a small number of patients actually take theirrugs; it takes sometimes several years of treatment beforeenefit in risk reduction is reached; and the mentioned riskeductions are sometimes “misleading” [114–117]. Fortu-ately, a new study–Transatlantic Asymptomatic Carotid In-ervention Trial, a randomized trial–is currently comparingEA, CAS, and current BMT [118].

Time to surgery after the first event in carotid stenosis isnother important aspect to consider when comparing theesults and effectiveness of different interventions. Delay-ng surgery in patients with symptomatic carotid stenosisignificantly decreases the aimed-for long-term stroke re-uction [119–121]. Delaying CEA for �12 weeks almostecreased a positive effect on stroke prevention in the long-erm, still putting these patients at risk for perioperativeomplications [15]. Unfortunately, systematic delay in sur-ery �12 weeks seems currently to be the common practice120]. A recent population based study in the United King-om showed that only 43% of symptomatic patients withevere carotid stenosis and who had a stroke risk beforeEA of 32% at week 12 underwent surgery at �12 weeks

120]. The highest risk of stroke in the first weeks after therimary event may be related to plaque vulnerability and/ororphology. Plaque in the early period has been character-

zed by thrombosis formation and spontaneous embolization122–124]. Classical opinions in CEA have been that earlyurgery is associated with increased perioperative risk. Nev-rtheless, delaying surgery may seem to decrease perioper-tive risk but at the expense of long-term benefits: DelayingEA can be accompanied by stroke risks up to 20% at 4eeks [120,125]. Overall, many patients may benefit from

ast-track CEA regarding prevention of stroke.In conclusion, CAS performed with embolic EPDs can

e an effective treatment for patients with carotid artery

tenosis. However, presently there is no evidence that CAS

rovides better stroke prevention compared with CEA.herefore, CEA still remains the “gold standard” of treat-ent. Furthermore, evolving BMT, timely intervention, and

nalysis of plaque composition may have an important in-uence on the future treatment of patients with carotidrtery stenosis.

eferences[1] Moore WS, Barnett HJ, Beebe HG, et al. Guidelines for carotid

endarterectomy. A multidisciplinary consensus statement from theAd Hoc Committee, American Heart Association. Circulation 1995;91:566–79.

[2] Rosamond W, Flegal K, Friday G, et al. Heart disease and strokestatistics–2007 update: a report from the American Heart Associa-tion Statistics Committee and Stroke Statistics Subcommittee. Cir-culation 2007;115:e69–e71.

[3] Ghatnekar O, Persson U, Glader EL, et al. Cost of stroke inSweden: an incidence estimate. Int J Technol Assess Health Care2004;375–80.

[4] Kolominsky-Rabas PL, Heuschmann PU, Marschall D, et al. Life-time cost of ischemic stroke in Germany: results and national pro-jections from a population-based stroke registry: the ErlangenStroke Project. Stroke 2006;37:1179–83.

[5] Meyers PM, Schumacher HC, Higashida RT, et al. Use of stents totreat extracranial cerebrovascular disease. Annu Rev Med 2006;57:437–54.

[6] Sauerbeck LR. Primary stroke prevention. Am J Nurs 2006;106:40–8.

[7] Weinberger J. Prevention of ischemic stroke. Curr Treat OptionsCardiovasc Med 2002;4:393–403.

[8] Weinberger J. Stroke and TIA. Prevention and management ofcerebrovascular events in primary care. Geriatrics 2002;57:38–43.

[9] Hemphill JC III. Ischemic stroke. Clinical strategies based on mech-anisms and risk factors. Geriatrics 2000;55:42–52.

[10] Ad Hoc Committee on Guidelines for the Management of TransientIschemic Attacks, Stroke Council of the American Heart Associa-tion. Guidelines for the management of transient ischemic attacks.Stroke 1994;25:1320–35.

[11] Moore WS. The American Heart Association Consensus Statementon guidelines for carotid endarterectomy. Semin Vasc Surg 1995;8:77–81.

[12] Goldstein LB, Adams R, Alberts MJ, et al. Primary prevention ofischemic stroke: a guideline from the American Heart Association/American Stroke Association Stroke Council: cosponsored by theAtherosclerotic Peripheral Vascular Disease Interdisciplinary Work-ing Group; Cardiovascular Nursing Council; Clinical CardiologyCouncil; Nutrition, Physical Activity, and Metabolism Council; andthe Quality of Care and Outcomes Research Interdisciplinary Work-ing Group. Circulation 2006;113:e873–e923.

[13] Randomised trial of endarterectomy for recently symptomatic ca-rotid stenosis: final results of the MRC European Carotid SurgeryTrial (ECST). Lancet 1998;351:1379–87.

[14] North American Symptomatic Carotid Endarterectomy Trial Col-laborators. Beneficial effect of carotid endarterectomy in symptom-atic patients with high-grade carotid stenosis. N Engl J Med 1991;325:445–53.

[15] Rothwell PM, Eliasziw M, Gutnikov SA, et al. Analysis of pooleddata from the randomised controlled trials of endarterectomy forsymptomatic carotid stenosis. Lancet 2003;361:107–16.

[16] Ederle J, Brown MM. The evidence for medicine versus surgery forcarotid stenosis. Eur J Radiol 2006;60:3–7.

[17] Moore WS. Resolved: NASCET and ACAS need not be repeated:the affirmative position. Arch Neurol 2003;60:775–8.

[18] Mullenix PS, Tollefson DF, Olsen SB, et al. Intraoperative duplexultrasonography as an adjunct to technical excellence in 100 con-secutive carotid endarterectomies. Am J Surg 2003;185:445–9.

[19] Bettendorf MJ, Mansour MA, Davis AT, et al. Carotid angioplastyand stenting versus redo endarterectomy for recurrent stenosis. Am JSurg 2007;193:356–9.

[20] Satler LF, Hoffmann R, Lansky A, et al. Carotid stent-assisted

angioplasty: preliminary technique, angiography, and intravascularultrasound observations. J Invasive Cardiol 1996;8:23–30.

Page 9

267M.G. van der Vaart et al. / The American Journal of Surgery 195 (2008) 259–269

[21] Naylor AR, Bolia A, Abbott RJ, et al. Randomized study of carotidangioplasty and stenting versus carotid endarterectomy: a stoppedtrial. J Vasc Surg 1998;28:326–34.

[22] Brooks WH, McClure RR, Jones MR, et al. Carotid angioplasty andstenting versus carotid endarterectomy: randomized trial in a com-munity hospital. J Am Coll Cardiol 2001;38:1589–95.

[23] Brooks WH, McClure RR, Jones MR, et al. Carotid angioplasty andstenting versus carotid endarterectomy for treatment of asymptom-atic carotid stenosis: a randomized trial in a community hospital.Neurosurgery 2004;54:318–24.

[24] Hobson RW, Howard VJ, Roubin GS, et al. Carotid artery stentingis associated with increased complications in octogenarians: 30-daystroke and death rates in the CREST lead-in phase. J Vasc Surg2004;40:1106–11.

[25] Ringleb PA, Allenberg J, Bruckmann H, et al. 30 day results fromthe SPACE trial of stent-protected angioplasty versus carotid end-arterectomy in symptomatic patients: a randomised non-inferioritytrial. Lancet 2006;368:1239–47.

[26] Mas JL, Chatellier G, Beyssen B, et al. Endarterectomy versusstenting in patients with symptomatic severe carotid stenosis. N EnglJ Med 2006;355:1660–71.

[27] Endovascular versus surgical treatment in patients with carotid ste-nosis in the Carotid and Vertebral Artery Transluminal AngioplastyStudy (CAVATAS): a randomised trial. Lancet 2001;357:1729–37.

[28] Coward LJ, Featherstone RL, Brown MM. Safety and efficacy ofendovascular treatment of carotid artery stenosis compared withcarotid endarterectomy: a Cochrane systematic review of the ran-domized evidence. Stroke 2005;36:905–11.

[29] Yadav JS, Wholey MH, Kuntz RE, et al. Protected carotid-arterystenting versus endarterectomy in high-risk patients. N Engl J Med2004;351:1493–501.

[30] Debakey ME. Carotid endarterectomy revisited. J Endovasc Surg1996;3:4.

[31] Debakey ME. Successful carotid endarterectomy for cerebrovascularinsufficiency. Nineteen-year follow-up. JAMA 1975;233:1083–5.

[32] Mayberg MR, Wilson SE, Yatsu F, et al. Carotid endarterectomyand prevention of cerebral ischemia in symptomatic carotid stenosis.Veterans Affairs Cooperative Studies Program 309 Trialist Group.JAMA 1991;266:3289–94.

[33] Barnett HJ, Taylor DW, Eliasziw M, et al. Benefit of carotid end-arterectomy in patients with symptomatic moderate or severe steno-sis. North American Symptomatic Carotid Endarterectomy TrialCollaborators. N Engl J Med 1998;339:1415–25.

[34] Fox AJ, Eliasziw M, Rothwell PM, et al. Identification, prognosis,and management of patients with carotid artery near occlusion.AJNR Am J Neuroradiol 2005;26:2086–94.

[35] Endarterectomy for asymptomatic carotid artery stenosis. ExecutiveCommittee for the Asymptomatic Carotid Atherosclerosis Study.JAMA 1995;273:1421–8.

[36] Halliday A, Mansfield A, Marro J, et al. Prevention of disabling andfatal strokes by successful carotid endarterectomy in patients with-out recent neurological symptoms: randomised controlled trial. Lan-cet 2004;363:1491–502.

[37] Taylor DW, Barnett HJ, Haynes RB, et al. Low-dose and high-doseacetylsalicylic acid for patients undergoing carotid endarterectomy:a randomised controlled trial. ASA and Carotid Endarterectomy(ACE) Trial Collaborators. Lancet 1999;353:2179–84.

[38] Cronenwett JL, Birkmeyer JD, Nackman GB, et al. Cost-effective-ness of carotid endarterectomy in asymptomatic patients. J VascSurg 1997;25:298–309.

[39] Kuntz KM, Kent KC. Is carotid endarterectomy cost-effective? Ananalysis of symptomatic and asymptomatic patients. Circulation1996;94:II194–II198.

[40] Moore WS, Barnett HJ, Beebe HG, et al. Guidelines for carotidendarterectomy. A multidisciplinary consensus statement from thead hoc Committee, American Heart Association. Stroke 1995;26:188–201.

[41] Bockenheimer SA, Mathias K. Percutaneous transluminal angio-plasty in arteriosclerotic internal carotid artery stenosis. AJNR Am JNeuroradiol 1983;4:791–2.

[42] Courtheoux P, Theron J, Tournade A, et al. Percutaneous endolu-minal angioplasty of post endarterectomy carotid stenoses. Neuro-

radiology 1987;29:186–9.

[43] Theron J. Protected carotid angioplasty and carotid stents. J MalVasc 1996;21(suppl A):113–22.

[44] Wholey MH, Wholey MH, Jarmolowski CR, et al. Endovascularstents for carotid artery occlusive disease. J Endovasc Surg 1997;4:326–38.

[45] Kachel R. Results of balloon angioplasty in the carotid arteries. JEndovasc Surg 1996;3:22–30.

[46] Diethrich EB, Ndiaye M, Reid DB. Stenting in the carotid artery:initial experience in 110 patients. J Endovasc Surg 1996;3:42–62.

[47] Yadav JS, Roubin GS, Iyer S, et al. Elective stenting of the extracra-nial carotid arteries. Circulation 1997;95:376–81.

[48] Criado FJ, Wellons E, Clark NS. Evolving indications for and earlyresults of carotid artery stenting. Am J Surg 1997;174:111–4.

[49] Bergeron P, Roux M, Khanoyan P, et al. Long-term results ofcarotid stenting are competitive with surgery. J Vasc Surg 2005;41:213–21.

[50] Roubin GS, New G, Iyer SS, et al. Immediate and late clinicaloutcomes of carotid artery stenting in patients with symptomatic andasymptomatic carotid artery stenosis: a 5-year prospective analysis.Circulation 2001;103:532–7.

[51] Henry M, Amor M, Masson I, et al. Angioplasty and stenting of theextracranial carotid arteries. J Endovasc Surg 1998;5:293–304.

[52] Teitelbaum GP, Lefkowitz MA, Giannotta SL. Carotid angioplastyand stenting in high-risk patients. Surg Neurol 1998;50:300–11.

[53] Jordan Jr WD, Voellinger DC, Doblar DD, et al. Microembolidetected by transcranial Doppler monitoring in patients during ca-rotid angioplasty versus carotid endarterectomy. Cardiovasc Surg1999;7:33–8.

[54] Crawley F, Clifton A, Buckenham T, et al. Comparison of hemo-dynamic cerebral ischemia and microembolic signals detected dur-ing carotid endarterectomy and carotid angioplasty. Stroke 1997;28:2460–4.

[55] Stockx L. Techniques in carotid artery stenting. Eur J Radiol 2006;60:11–3.

[56] Macdonald S. The evidence for cerebral protection: an analysis andsummary of the literature. Eur J Radiol 2006;60:20–5.

[57] Wholey MH, Al Mubarek N, Wholey MH. Updated review of theglobal carotid artery stent registry. Catheter Cardiovasc Interv 2003;60:259–66.

[58] Kastrup A, Groschel K, Krapf H, et al. Early outcome of carotidangioplasty and stenting with and without cerebral protection de-vices: a systematic review of the literature. Stroke 2003;34:813–9.

[59] Fanelli F, Bezzi M, Boatta E, et al. Techniques in cerebral protec-tion. Eur J Radiol 2006;60:26–36.

[60] Vitek JJ, Roubin GS, Al Mubarek N, et al. Carotid artery stenting:technical considerations. AJNR Am J Neuroradiol 2000;21:1736–43.

[61] White CJ, Iyer SS, Hopkins LN, et al. Carotid stenting with distalprotection in high surgical risk patients: the BEACH trial 30 dayresults. Catheter Cardiovasc Interv 2006;67:503–12.

[62] Safian RD, Bacharach JM, Ansel GM, et al. Carotid stenting with anew system for distal embolic protection and stenting in high-riskpatients: the carotid revascularization with ev3 arterial technologyevolution (CREATE) feasibility trial. Catheter Cardiovasc Interv2004;63:1–6.

[63] Gray WA, Hopkins LN, Yadav S, et al. Protected carotid stenting inhigh-surgical-risk patients: the ARCHeR results. J Vasc Surg 2006;44:258–68.

[64] Gonzalez A, Gonzalez-Marcos JR, Martinez E, et al. Safety andsecurity of carotid artery stenting for severe stenosis with contralat-eral occlusion. Cerebrovasc Dis 2005;20(suppl 2):123–8.

[65] Gray WA, Yadav JS, Verta P, et al. The CAPTURE registry: resultsof carotid stenting with embolic protection in the post approvalsetting. Catheter Cardiovasc Interv 2006;69:341–8.

[66] CaRESS Steering Committee. Carotid Revascularization UsingEndarterectomy or Stenting Systems (CaRESS) phase I clinical trial:1-year results. J Vasc Surg 2005;42:213–9.

[67] Dudek D, Bartus S, Rakowski T, et al. MO.MA–a new cerebralstroke protection system during carotid artery stenting. Kardiol Pol2005;62:559–70.

[68] McCabe DJ, Pereira AC, Clifton A, et al. Restenosis after carotidangioplasty, stenting, or endarterectomy in the Carotid and VertebralArtery Transluminal Angioplasty Study (CAVATAS). Stroke 2005;

36:281–6.

Page 10

[

[

[

[

[

[

[

[

[

[

[

[

[

268 M.G. van der Vaart et al. / The American Journal of Surgery 195 (2008) 259–269

[69] Prati P, Vanuzzo D, Casaroli M, et al. Determinants of carotidplaque occurrence. A long-term prospective population study: theSan Daniele Project. Cerebrovasc Dis 2006;22:416–22.

[70] Alamowitch S, Eliasziw M, Algra A, et al. Risk, causes, and pre-vention of ischaemic stroke in elderly patients with symptomaticinternal-carotid-artery stenosis. Lancet 2001;357:1154–60.

[71] Rothwell PM, Warlow CP. Prediction of benefit from carotid end-arterectomy in individual patients: a risk-modelling study. EuropeanCarotid Surgery Trialists’ Collaborative Group. Lancet 1999;353:2105–10.

[72] Miller MT, Comerota AJ, Tzilinis A, et al. Carotid endarterectomyin octogenarians: does increased age indicate “high risk”? J VascSurg 2005;41:231–7.

[73] Featherstone RL, Brown MM, Coward LJ. International carotidstenting study: protocol for a randomised clinical trial comparingcarotid stenting with endarterectomy in symptomatic carotid arterystenosis. Cerebrovasc Dis 2004;18:69–74.

[74] Mozes G, Sullivan TM, Torres-Russotto DR, et al. Carotid endar-terectomy in SAPPHIRE-eligible high-risk patients: implications forselecting patients for carotid angioplasty and stenting. J Vasc Surg2004;39:958–65.

[75] Mozes G. High-risk carotid endarterectomy. Semin Vasc Surg 2005;18:61–8.

[76] Ballotta E, Da Giau G, Baracchini C, et al. Carotid endarterectomyin high-risk patients: a challenge for endovascular procedure proto-cols. Surgery 2004;135:74–80.

[77] Nguyen LL, Conte MS, Reed AB, et al. Carotid endarterectomy:who is the high-risk patient? Semin Vasc Surg 2004;17:219–23.

[78] Verhoeven B, Hellings WE, Moll FL, et al. Carotid atheroscleroticplaques in patients with transient ischemic attacks and stroke haveunstable characteristics compared with plaques in asymptomatic andamaurosis fugax patients. J Vasc Surg 2005;42:1075–81.

[79] Trivedi RA, King-Im JM, Graves MJ, et al. Non-stenotic rupturedatherosclerotic plaque causing thrombo-embolic stroke. Cerebro-vasc Dis 2005;20:53–5.

[80] McCarthy MJ, Loftus IM, Thompson MM, et al. Angiogenesis andthe atherosclerotic carotid plaque: an association between symptom-atology and plaque morphology. J Vasc Surg 1999;30:261–8.

[81] Carr SC, Farb A, Pearce WH, et al. Activated inflammatory cells areassociated with plaque rupture in carotid artery stenosis. Surgery1997;122:757–63.

[82] Balzer K, Boesger U, Muller KM. Plaque morphology of the carotidbifurcation and incidence of embolisms in relation clinical stage ofcerebrovascular insufficiency. Kongressbd Dtsch Ges Chir Kongr2002;119:631–4.

[83] Fisher M, Paganini-Hill A, Martin A, et al. Carotid plaque pathol-ogy: thrombosis, ulceration, and stroke pathogenesis. Stroke 2005;36:253–7.

[84] Rothwell PM, Gibson R, Warlow CP, on behalf of the EuropeanCarotid Surgery Trialists’ Collaborative Group. Interrelation be-tween plaque surface morphology and degree of stenosis on carotidangiograms and the risk of ischemic stroke in patients with symp-tomatic carotid stenosis. Stroke 2000;31:615–21.

[85] Ackerstaff RG, Jansen C, Moll FL, et al. The significance of mi-croemboli detection by means of transcranial Doppler ultrasonog-raphy monitoring in carotid endarterectomy. J Vasc Surg 1995;21:963–9.

[86] Biasi GM, Froio A, Diethrich EB, et al. Carotid plaque echolucencyincreases the risk of stroke in carotid stenting: the Imaging inCarotid Angioplasty and Risk of Stroke (ICAROS) study. Circula-tion 2004;110:756–62.

[87] Flach HZ, Ouhlous M, Hendriks JM, et al. Cerebral ischemia aftercarotid intervention. J Endovasc Ther 2004;11:251–7.

[88] Wolf O, Heider P, Heinz M, et al. Microembolic signals detected bytranscranial Doppler sonography during carotid endarterectomy andcorrelation with serial diffusion-weighted imaging. Stroke 2004;35:e373–e375.

[89] Johnsen SH, Mathiesen EB, Fosse E, et al. Elevated high-densitylipoprotein cholesterol levels are protective against plaque progres-sion: a follow-up study of 1952 persons with carotid atherosclerosis:the Tromso study. Circulation 2005;112:498–504.

[90] Honda O, Sugiyama S, Kugiyama K, et al. Echolucent carotid

plaques predict future coronary events in patients with coronaryartery disease. J Am Coll Cardiol 2004;43:1177–84.

[91] Setacci C, de Donato G, Setacci F, et al. In-stent restenosis aftercarotid angioplasty and stenting: a challenge for the vascular sur-geon. Eur J Vasc Endovasc Surg 2005;29:601–7.

[92] Touze E, Toussaint JF, Coste J, et al. Reproducibility of high-resolution MRI for the identification and the quantification of carotidatherosclerotic plaque components. Consequences for prognosisstudies and therapeutic trials. Stroke 2007;38:1812–9.

[93] Liu F, Xu D, Ferguson MS, Chu B, et al. Automated in vivosegmentation of carotid plaque MRI with morphology-enhancedprobability maps. Magn Reson Med 2006;55:659–68.

[94] Wolf RL, Wehrli SL, Popescu AM, et al. Mineral volume andmorphology in carotid plaque specimens using high-resolution MRIand CT. Arterioscler Thromb Vasc Biol 2005;25:1729–35.

[95] Luo Y, Polissar N, Han C, et al. Accuracy and uniqueness of threein vivo measurements of atherosclerotic carotid plaque morphologywith black blood MRI. Magn Reson Med 2003;50:75–82.

[96] Estes JM, Quist WC, Lo Gerfo FW, et al. Noninvasive character-ization of plaque morphology using helical computed tomography.J Cardiovasc Surg (Torino) 1998;39:527–34.

[97] Gronholdt ML, Wagner A, Wiebe BM, et al. Spiral computedtomographic imaging related to computerized ultrasonographic im-ages of carotid plaque morphology and histology. J Ultrasound Med2001;20:451–8.

[98] Adams Jr HP, del Zoppo G, Alberts MJ, et al. Guidelines for theearly management of adults with ischemic stroke: a guideline fromthe American Heart Association/American Stroke AssociationStroke Council, Clinical Cardiology Council, Cardiovascular Radi-ology and Intervention Council, and the Atherosclerotic PeripheralVascular Disease and Quality of Care Outcomes in Research Inter-disciplinary Working Groups: the American Academy of Neurologyaffirms the value of this guideline as an educational tool for neurol-ogists. Stroke 2007;38:1655–711.

[99] Touze E, Mas JL, Rother J, et al. Impact of carotid endarterectomyon medical secondary prevention after a stroke or a transient isch-emic attack: results from the Reduction of Atherothrombosis forContinued Health (REACH) registry. Stroke 2006;37:2880–5.

100] Frawley JE, Hicks RG, Woodforth IJ. Risk factors for peri-operativestroke complicating carotid endarterectomy: selective analysis of aprospective audit of 1000 consecutive operations. Aust N Z J Surg2000;70:52–6.

101] Somerfield J, Barber PA, Anderson NE, et al. Changing attitudes tothe management of ischaemic stroke between 1997 and 2004: asurvey of New Zealand physicians. Intern Med J 2006;36:276–80.

102] Idris I, Thomson GA, Sharma JC. Diabetes mellitus and stroke. IntJ Clin Pract 2006;60:48–56.

103] Amarenco P, Labreuche J, Lavallee P, et al. Statins in stroke pre-vention and carotid atherosclerosis: systematic review and up-to-date meta-analysis. Stroke 2004;35:2902–9.

104] Davignon J. Beneficial cardiovascular pleiotropic effects of statins.Circulation 2004;109:III39–III43.

105] Nissen SE, Nicholls SJ, Sipahi I, et al. Effect of very high-intensitystatin therapy on regression of coronary atherosclerosis: theASTEROID trial. JAMA 2006;295:1556–65.

106] CAPRIE Steering Committee. A randomised, blinded, trial of clo-pidogrel versus aspirin in patients at risk of ischaemic events(CAPRIE). Lancet 1996;348:1329–39.

107] Bhatt DL, Chew DP, Hirsch AT, et al. Superiority of clopidogrelversus aspirin in patients with prior cardiac surgery. Circulation2001;103:363–8.

108] Hankey GJ. Secondary prevention of recurrent stroke. Stroke 2005;36:218–21.

109] Smith TP, Cruz CP, Brown AT, et al. Folate supplementation in-hibits intimal hyperplasia induced by a high-homocysteine diet in arat carotid endarterectomy model. J Vasc Surg 2001;34:474–81.

110] Southern F, Eidt J, Drouilhet J, et al. Increasing levels of dietaryhomocystine with carotid endarterectomy produced proportionateincreases in plasma homocysteine and intimal hyperplasia. Athero-sclerosis 2001;158:129–38.

111] Assadian A, Rotter R, Assadian O, et al. Homocysteine and earlyre-stenosis after carotid eversion endarterectomy. Eur J Vasc Endo-vasc Surg 2006;33:144–8.

112] Samson RH, Yungst Z, Showalter DP. Homocysteine, a risk factor

for carotid atherosclerosis, is not a risk factor for early recurrent

Page 11

[

[

[

[

[

[

[

[

[

[

[

[

[

[

[

269M.G. van der Vaart et al. / The American Journal of Surgery 195 (2008) 259–269

carotid stenosis following carotid endarterectomy. Vasc EndovascSurg 2004;38:345–8.

113] Yusuf S, Sleight P, Pogue J, et al. Effects of an angiotensin-converting-enzyme inhibitor, ramipril, on cardiovascular events inhigh-risk patients. The Heart Outcomes Prevention EvaluationStudy Investigators. N Engl J Med 2000;342:145–53.

114] Hansson L. ‘Why don’t you do as I tell you?’ Compliance andantihypertensive regimens. Int J Clin Pract 2002;56:191–6.

115] Jones JK, Gorkin L, Lian JF, et al. Discontinuation of and changesin treatment after start of new courses of antihypertensive drugs: astudy of a United Kingdom population. Br Med J 1995;311:293–5.

116] Hamilton-Craig I. The Heart Protection Study: implications forclinical practice. The benefits of statin therapy do not come withoutfinancial cost. Med J Aust 2002;177:407–8.

117] Naylor AR. Does the modern concept of ‘best medical therapy’render carotid surgery obsolete? Eur J Vasc Endovasc Surg 2004;28:457–61.

118] Gaines PA, Randall MS. Carotid artery stenting for patients withasymptomatic carotid disease (and news on TACIT). Eur J VascEndovasc Surg 2005;30:461–3.

119] Rothwell PM. Symptomatic and asymptomatic carotid stenosis:how, when, and who to treat? Curr Atheroscler Rep 2006;8:

290 –7.

120] Fairhead JF, Mehta Z, Rothwell PM. Population-based study ofdelays in carotid imaging and surgery and the risk of recurrentstroke. Neurology 2005;65:371–5.

121] Johnston SC, Nguyen-Huynh MN, Schwarz ME, et al. NationalStroke Association guidelines for the management of transient isch-emic attacks. Ann Neurol 2006;60:301–13.

122] Molloy KJ, Thompson MM, Jones JL, et al. Unstable carotidplaques exhibit raised matrix metalloproteinase-8 activity. Circula-tion 2004;110:337–43.

123] Loftus IM, Naylor AR, Goodall S, et al. Increased matrix metallo-proteinase-9 activity in unstable carotid plaques. A potential role inacute plaque disruption. Stroke 2000;31:40–7.

124] Molloy KJ, Thompson MM, Schwalbe EC, et al. Elevation inplasma MMP-9 following carotid endarterectomy is associated withparticulate cerebral embolisation. Eur J Vasc Endovasc Surg 2004;27:409–13.

125] Baron EM, Baty DE, Loftus CM. The timing of carotid endarterec-tomy post stroke. Neurol Clin 2006;24:669–80.

126] Waigand J, Gross CM, Uhlich F, et al. Elective stenting of carotidartery stenosis in patients with severe coronary artery disease. EurHeart J 1998;19:1365–70.

127] Vozzi CR, Rodriguez AO, Paolantonio D, et al. Extracranial carotidangioplasty and stenting. Initial results and short-term follow-up.

Tex Heart Inst J 1997;24:167–72.