Statistics from Altmetric.com
If you wish to reuse any or all of this article please use the link below which will take you to the Copyright Clearance Center’s RightsLink service. You will be able to get a quick price and instant permission to reuse the content in many different ways.
Every year almost 800 000 people in the USA suffer a new or recurrent stroke, and stroke is the third leading cause of death with over 140 000 deaths in the USA in 1995.1 Additionally, stroke is a leading cause of long-term disability with an estimated cost of $68.9 billion in 2009. The relationship between carotid artery disease and stroke was first described by Fisher in 1951,2 and it is estimated that about one-third of ischemic strokes are due to carotid artery thromboembolic disease.3 ,4
Several trials have established carotid endarterectomy (CEA) as an excellent surgical technique for revascularization and prevention of future stroke, with a reasonable safety profile. Over the last 10–15 years, carotid artery stenting has been studied as an alternative and potentially less invasive revascularization method. Early trials of carotid artery stenting struggled with high complication rates but, as experience has grown and techniques improved, more recent trials have shown complication rates comparable to CEA.
The goals of this document are to suggest standards of practice for patients treated with carotid artery angioplasty and stenting (CAS) and to provide a reporting framework for series of patients treated with CAS. Our evidence-based treatment recommendations were assessed according to criteria published by the American Heart Association/American Stroke Association (AHA/ASA) and the University of Oxford's Center for Evidence Based Medicine (CEBM). The development of treatment guidelines for CAS is facilitated by several large clinical trials that have already investigated surgical treatment for extracranial carotid stenosis as well as trials that have compared CAS with the established surgical treatment (CEA). In addition to this standards document, we would also recommend previously published guidelines for the use of CAS.5 Note that this document is not addressing the use of angioplasty with or without stenting in the setting of acute ischemic stroke, as this was discussed in a previous document.6
Identification of evidence
The National Library of Medicine database (PubMed 1966–2011) was searched electronically in order to identify peer-reviewed publications that contained clinical and technical outcome data for treatment of extracranial carotid artery stenosis. Search terms included ‘carotid stenosis’, ‘stroke’, ‘carotid endarterectomy’, ‘carotid angioplasty’ and ‘carotid stenting’.
Over the past 20 years, numerous clinical trials have compared CEA with medical management for extracranial carotid disease. For symptomatic patients with high-grade stenosis, the European Carotid Surgery Trial (ECST), the North American Symptomatic Carotid Endarterectomy Trial (NASCET) and the Veterans Affairs Cooperative Study (VACS) all showed superiority of CEA over medical management in stroke risk reduction.7–9 In particular, for symptomatic patients with >70% carotid stenosis, NASCET showed a 2-year reduction in the risk of stroke or death from 26% to 9% with CEA compared with medical management alone.8 More recently, however, there has been some controversy with respect to the benefit of CEA for patients with asymptomatic carotid disease. While the Asymptomatic Carotid Atherosclerosis Study (ACAS) showed a 5-year reduction in the risk of stroke or death from 11% to 5% with CEA compared with medical management,10 improvements in ‘best medical management’ have led to equivalent reductions in the risk of stroke with medical management alone.11 Future randomized trials comparing present-day ‘best medical management’ with carotid revascularization (CEA and CAS) for asymptomatic patients are being planned.
Carotid artery angioplasty and stenting (CAS)
Charles Kerber performed the first transluminal angioplasty of the carotid artery in 1980.12 The efficacy of CAS has been examined in the context of a well-established open surgical option (CEA) and therefore the majority of prospective randomized clinical trials have compared outcomes in patients randomized to CAS and those randomized to CEA.
The Carotid and Vertebral Artery Transluminal Angioplasty Study (CAVATAS) randomized surgical patients who were appropriate for either treatment (as determined by the local investigators) to CEA or endovascular treatment.13 Of the 504 patients, 97% of whom were symptomatic, 251 patients were assigned to endovascular treatment and 253 to CEA. There were no statistically significant differences in the rate of death or stroke at 30 days or in the rate of ipsilateral stroke at 3 years between the two treatment options. However, stenting was used in only 26% of the endovascularly-treated patients and distal protection devices were not used. With advances in technology, the use of stents and distal protection devices is now common in the current practice of CAS.
The Stenting and Angioplasty with Protection in Patients at High-Risk for Endarterectomy (SAPPHIRE) trial randomized symptomatic patients at high surgical risk (>50% stenosis) and asymptomatic patients (>80% stenosis) to CEA or CAS with distal protection.14 The criteria for high risk included clinically significant cardiac disease, severe pulmonary disease, contralateral carotid occlusion, contralateral laryngeal nerve palsy, previous radical neck surgery or neck radiation, recurrent stenosis after prior CEA and age >80 years. A total of 334 patients were evenly divided between CEA and CAS, although 7% did not actually receive their assigned treatment. The rate of death, stroke, myocardial infarction (MI) within 30 days or death or ipsilateral stroke at 1 year was 12.2% for the patients assigned to CAS and 20.1% for those assigned to CEA (p=0.05). By 3 years there was no significant difference between the two treatments. While the major difference seen in the early time points is probably due to rates of periprocedural MI, the SAPPHIRE trial nevertheless supports the position that CAS is not inferior to CEA for high-risk patients.
In order to determine the ‘real world’ application of CAS, the Carotid Revascularization Using Endarterectomy or Stenting Systems (CaRESS) trial was performed.15 This was a multicenter prospective non-randomized trial comparing CAS with distal protection to CEA in symptomatic and asymptomatic patients with carotid stenosis. The choice of treatment (CEA or CAS) was based solely on physician and patient preference. A total of 397 patients were enrolled (254 CEA and 143 CAS), and there was no significant difference in the primary endpoint of death and stroke at 30 days and a 1-year endpoint of death, stroke or MI.
Two randomized European trials of symptomatic patients, Endarterectomy Versus Angioplasty in Patients with Symptomatic Severe Carotid Stenosis (EVA-3S) and Stent-supported Percutaneous Angioplasty of the Carotid artery versus Endarterectomy (SPACE), failed to show non-inferiority of CAS to CEA.16 ,17 Both trials showed a higher 30-day stroke or death rate in patients treated with CAS, although there has been criticism of both trials regarding experience and technical expertise of the treating interventionalists as well as non-uniformity in the use of embolic protection devices (EPDs).18
The International Carotid Stenting Study (ICSS) enrolled 1713 symptomatic patients, with 855 randomized to CAS and 858 to CEA. The likelihood of periprocedural stroke, MI or death was 8.5% in the CAS group and 5.2% in the CEA group (p=0.006).19 The same investigators also examined a subgroup of patients who underwent MRI with diffusion weighted imaging (DWI), and found that the likelihood of new DWI lesions was 50% in the CAS group and 17% in the CEA group.20
The Carotid Revascularization Endarterectomy versus Stenting Trial (CREST) was published in 2010.21 In this study, 2502 patients with symptomatic (>50% stenosis) and asymptomatic (>80% stenosis) disease were randomly assigned to CAS or CEA. The study was initially designed for patients with symptomatic lesions but was later modified to recruit asymptomatic patients as well. Additionally, the degree of stenosis required for enrollment varied according to the imaging modality. The primary endpoint was stroke, MI or death within 30 days or any ipsilateral stroke within 4 years. Stroke was defined as an increased National Institutes of Health Stroke Scale (NIHSS) of at least two points, while MI was defined as an increase in serum troponin, chest pain and ECG changes. Over a median follow-up of 2.5 years there was no significant difference in the primary endpoint between CAS (7.2%) and CEA (6.8%). However, the 30-day rate of stroke was significantly higher with CAS than with CEA (4.1% vs 2.3%) while the risk of MI was lower (1.1% vs 2.3%). This was mainly due to an increased rate of minor stroke (defined as an NIHSS score of ≥9 after 90 days). Major stroke rates were not different between the groups. Additionally, in symptomatic patients CAS had a significantly higher periprocedural stroke and death rate compared with CEA (6% vs 3.2%).22 Health status was measured in this trial using the Short-Form Health Survey (SF-36), and it was found that those with minor stroke had more of an impact on health status compared with MI.23
Based on the results of multiple prospective randomized studies, the following guidelines are proposed for the use of CAS:
For patients with carotid stenosis <50%, there is no indication for CEA or CAS (AHA/ASA Class III; Level of Evidence B, CEBM Level 1b; Grade B).24
For patients with symptomatic carotid disease and 50–69% stenosis, CEA is recommended. CAS is considered an alternative to CEA for patients at average or low risk for CAS or for those who are at high risk (>6% morbidity and mortality) for CEA (AHA/ASA Class I; Level of Evidence B, CEBM Level 1b; Grade B).25–29 While age >80 years was considered high risk for CEA in the SAPPHIRE trial, the CREST trial showed a benefit for age <70 years with CAS, perhaps due to anatomic factors such as more favorable aortic arch anatomy in the younger cohort. We suggest that age alone should not be considered in determining CEA versus CAS and that other patient-specific factors should be considered.
CAS may be considered for patients with symptomatic severe stenosis (>70%) who are at high risk for CEA (AHA/ASA Class IIb; Level of Evidence B, CEBM Level 2b, Grade B).24
The benefit of CAS in asymptomatic patients is less clear and there is uncertainty regarding CAS over CEA (AHA/ASA Class IIb; Level of Evidence B, CEBM Level 2b, Grade B).30
In order to facilitate comparison among series of patients treated with CAS, a common set of reporting data is suggested. We have outlined particular areas in greater detail below.
It is well-established that there are certain non-modifiable and modifiable risk factors for stroke.30 Demographic information regarding patient age, gender and race contains useful information about non-modifiable risk factors. Documentation of medical history to include hypertension, cigarette smoking, diabetes and dyslipidemia are also important in order to identify modifiable risk factors.
We recommend that data regarding demographic information include well-established non-modifiable and modifiable stroke risk factors.
Patients are considered to have symptomatic carotid stenosis based on the criteria from NASCET.8 Symptomatic patients must have had a hemispheric transient ischemic attack (TIA), with distinct focal neurological dysfunction or monocular blindness persisting <24 h or a non-disabling stroke with persistence of symptoms for >24 h within the previous 180 days.
We recommend that data regarding symptoms include cerebral or retinal TIAs or non-disabling stroke within the previous 180 days.
The original criteria for patients to be considered high risk for surgery were defined in the SAPPHIRE trial. These included clinically significant cardiac disease, severe pulmonary disease, age >80 years and certain anatomic factors.14 Criteria for high risk have been further defined by the AHA/ASA and include both medical and anatomic factors.30 Medical factors are class III/IV congestive heart failure (CHF), class III/IV angina, left main coronary artery disease (CAD), ≥2-vessel CAD, left ventricular ejection fraction (EF) ≤30%, recent MI, severe lung disease or severe renal disease.
We recommend that data regarding medical comorbidities include the presence of severe CHF, angina, CAD, low EF, recent MI and severe lung or renal disease.
As with medical factors that make patients high risk for surgery, anatomic factors were also defined by the SAPPHIRE trial.14 These were contralateral carotid occlusion, contralateral laryngeal nerve palsy, previous radical neck surgery or neck radiation and recurrent stenosis after prior CEA. The AHA/ASA has further defined these factors and the current anatomic factors are prior neck surgery (such as radical neck dissection), neck irradiation, post-endarterectomy restenosis, surgically inaccessible lesions, contralateral carotid occlusion, contralateral vocal cord palsy or the presence of a tracheostomy.30
We recommend that data regarding these anatomic factors be reported.
Degree of stenosis
Carotid stenosis is usually found either as the result of a screening test in an asymptomatic patient or as the result of a work-up for stroke or transient ischemic events. Doppler ultrasonography, CT angiography (CTA) and magnetic resonance angiography (MRA) are all capable of providing information on the degree of stenosis, but the gold standard test remains catheter angiography. The NASCET method for calculating the degree of stenosis measures the luminal diameter at the level of the stenotic lesion (R) compared with the diameter of the normal distal internal carotid artery beyond the carotid bulb (D) where the walls of the internal carotid artery first become parallel to each other.8
Although degree of stenosis is the most studied independent risk factor, certain features of plaque morphology such as irregularity and ulceration have been identified in the ECST and NASCET trials as associated with an increased risk of ipsilateral stroke at all degrees of stenosis.31 ,32 However, catheter-based angiography may be limited in identifying ulcers.33 Non-invasive imaging techniques such as MRI may allow the pre-procedure identification of unstable plaque features like ulceration or disrupted fibrous plaque.34 ,35 We recommend reporting of plaque features such as irregularity and ulceration identified on angiography or non-invasive imaging.
We recommend that the degree of stenosis be calculated by the NASCET method applied to either non-invasive imaging (CTA, MRA), inferred from ultrasound data or applied to catheter angiography and be reported both pre- and post-treatment.
Use of embolic protection devices (EPDs)
Embolic stroke is a clear concern when performing CAS. Distal protection devices were developed to minimize the chance of emboli reaching the cerebral circulation during CAS; however, there is no clear consensus regarding the use of these devices.36 In the SAPPHIRE trial, CAS was performed with the use of EPDs14 but, in the CREST trial, an EPD was used ‘whenever feasible’.21 This acknowledges that the feasibility of EPD use, technical success, is variable depending upon patient anatomy and operator experience. Although the goal of limiting emboli is foremost, the use of an EPD increases the complexity of the procedure for several reasons including the operator learning curve, patient intolerance to flow arrest or reversal and device-induced vasospasm or difficult retrieval with distal filter systems.37 ,38 Proximal protection includes both circulatory control in the form of flow arrest or flow reversal paradigms and is an alternative to distal occlusion or filter protection in which the counterembolic condition is established prior to lesion crossing.39
We recommend that data regarding the endovascular technique include whether or not an EPD was used and the type of system that was used.
The goal for CAS is reduction of ipsilateral stroke and/or death. To date, studies have used perioperative (within 30 days) events of minor stroke, major stroke, MI and death as outcomes. Long-term events (ie, ≥1-year risk) have included ipsilateral stroke, ipsilateral disabling stroke, any stroke and death. Long-term follow-up of patients in the CREST trial with Doppler ultrasound showed a 6% rate of restenosis or occlusion 2 years after CAS.40 Female sex, diabetes and dyslipidemia were independent predictors of restenosis. Monitoring for restenosis may allow intervention prior to a new stroke. Patients have typically been followed annually for at least 2–5 years.
We recommend that data regarding clinical follow-up include perioperative (within 30 days) assessment for stroke, MI or death. In addition, we recommend that patients be followed annually and data regarding restenosis, stroke and death be collected. We suggest a minimum of 2 years of clinical follow-up. Follow-up imaging with ultrasound is recommended, with consideration for alternative modalities if an abnormality is seen on ultrasound.
Carotid atherosclerosis is a risk factor for stroke, especially in symptomatic patients with >50% ipsilateral stenosis. CEA is a proven surgical therapy. CAS has evolved over the past decade with a favorable safety profile and is an appropriate choice for selected patients. Standardized reporting of series of CAS will facilitate comparison.
Correction notice This article has been corrected since it was published Online First. The author Joshua A Hirsch has been added to the author list and the affiliations updated.
Competing interests None.
Provenance and peer review Not commissioned; internally peer reviewed.