8. Management of Extracranial Carotid Disease and Intracranial Atherosclerosis
2017 UPDATE October 2017
Note: These recommendations are applicable to ischemic stroke and transient ischemic attack.
8.1 Symptomatic Carotid Stenosis
- Patients with recent transient ischemic attack or non-disabling stroke and ipsilateral 50 to 99 percent symptomatic carotid stenosis should have an evaluation by an individual with stroke expertise and selected patients should be offered carotid endarterectomy (revascularization) as soon as possible [Evidence Level B].
- Carotid stenosis should ideally be measured by CTA to guide surgical decision-making [Evidence Level C].
- In individuals with non-disabling stroke or transient ischemic attack and 70-99 percent symptomatic carotid stenosis, carotid endarterectomy should be performed [Evidence Level A], on an urgent basis.
- Ideally carotid endarterectomy should be performed within the first days following non-disabling stroke or transient ischemic attack [Evidence Level B] and within 14 days of ischemic event onset for patients who are not clinically stable in the first few days [Evidence Level A]. Refer to Table below for summary of recurrent stroke risk at various time points
- Carotid endarterectomy should be performed by a surgeon/centre that routinely audits their performance results, especially perioperative stroke and death rates. The randomized trials upon which these recommendations are based (benefits accrued for patients undergoing surgery within 6 months of symptoms) involved combined perioperative stroke and death rates of 6 – 7 % [Evidence Level A].
- Carotid endarterectomy is generally more appropriate than carotid stenting for patients over age 70 years who are otherwise fit for surgery as current evidence indicates stenting carries a higher peri-procedural risk of stroke and death in older patients. [Evidence Level A].
- Carotid stenting may be considered for patients who are not operative candidates for technical, anatomic or medical reasons [Evidence Level A].
- Carotid stenting should be performed by an interventionist/centre with expertise that routinely audits their performance results, especially perioperative stroke and death rates. The randomized trial upon which these recommendations are based involved combined periprocedural stroke and death rates of 5% [Evidence Level B].
8.2 Asymptomatic and Remotely Symptomatic Carotid Stenosis
- Patients with symptoms of TIA and non-disabling stroke who are found to have an asymptomatic carotid stenosis should be evaluated by a physician with expertise in stroke management [Evidence Level C].
- Stroke patients with asymptomatic carotid stenosis should receive aggressive medical management of risk factors as defined throughout the Secondary Prevention of Stroke Module (for example, blood pressure, cholesterol, antiplatelet therapy lifestyle changes) [Evidence Level B].
- Carotid endarterectomy may be considered for selected patients with 60 to 99 percent carotid stenosis who are asymptomatic or were remotely symptomatic (i.e., greater than six months) [Evidence Level A].
- Patients should be evaluated to determine eligibility for carotid endarterectomy, such as a life expectancy of more than five years, and an acceptable risk of surgical complications [Evidence Level A].
- In carefully selected patients, carotid endarterectomy should be performed by a surgeon who routinely audits their performance results and demonstrates a less than 3 percent risk of peri-operative morbidity and mortality [Evidence Level A].
- Carotid stenting may be considered in patients with 60 to 99 percent carotid stenosis who are not operative candidates for technical, anatomic or medical reasons provided there is a less than 3 percent risk of peri-procedural morbidity and mortality [Evidence Level A].
8.3 Intracranial Stenosis
- Intracranial stenting is not recommended for the treatment of recently symptomatic intracranial 70% to 99% stenosis [Evidence Level B].
- Based on the SAMMPRIS trial (Derdeyn C et al. 2014), the medical management arm included dual antiplatelet therapy with ASA 325 mg and Clopidogrel 75 mg started within 30 days of stroke or TIA and treated for up to 90 days [Evidence Level B], and should be considered for each patient on an individual basis. In addition, there should be aggressive management of all vascular risk factors including blood pressure, lipids, diabetes mellitus, and other at-risk lifestyle patterns [Evidence Level A].
- In patients who have been managed with maximal medical therapy in the presence of intracranial stenosis and experience a recurrent stroke, there is lack of clear evidence to guide further management decisions; intracranial angioplasty (with or without stenting) may be reasonable in carefully selected patients [Evidence Level C].
8.4 Cervicocephalic Artery Dissection
- A diagnosis of carotid or vertebral dissection can be established by CTA, MRA or DSA [Evidence Level C].
Note: CTA or MRA are the preferred non-invasive diagnostic imaging tests for patients with a suspected cervicocephalic artery dissection, as neck ultrasound does not fully visualize the vertebral arteries and can miss distal or carotid dissection originating above the angle of the jaw.
- Antithrombotic therapy for stroke prevention is recommended for individuals with a diagnosis of an extracranial carotid or vertebral artery dissection [Evidence Level B].
- There is uncertainty about the comparative efficacy of antiplatelet therapy vs. anticoagulation with heparin or warfarin; either treatment is considered reasonable and decision should be based on individual risk/benefit analysis [Evidence Level B].
- There is a lack of evidence regarding the optimal duration of antithrombotic therapy and the role of repeat vascular imaging in decision-making. Decisions may be based on individual clinical factors [Evidence Level C].
- Note: There is insufficient evidence at this time to make a recommendation regarding the use of DOACs in patients with arterial dissections.
- There is a lack of evidence regarding the use of anticoagulation in intracranial arterial dissection. Decisions may be based on individual clinical factors [Evidence Level C].
Refer to Section One for recommendations on urgent vascular imaging in patients with acute TIA and non-disabling stroke.
Refer to Section 2 on aggressive prevention management, including lifestyle and antiplatelet therapy.
The recommendations from this module have been published in International Journal of Stroke by SAGE Publications Ltd. Copyright © 2017 World Stroke Organization.
Carotid endarterectomy is a surgical procedure that removes atherosclerotic plaque from the proximal internal carotid artery. Successful carotid endarterectomy substantially reduces the risk of recurrent stroke in patients who present with a hemispheric transient ischemic attack or minor stroke and an ipsilateral high-grade carotid stenosis. One death or severe stroke is prevented for every nine patients with symptomatic severe (70 to 99 percent) carotid stenosis treated with carotid endarterectomy). For selected patients with asymptomatic carotid stenosis, carotid endarterectomy reduces the risk of stroke from about two percent per year to about one percent per year. Aggressive medical management was superior to intracranial stenting for patients with 70 to 99% stenosis of a major intracranial artery.
- Protocols to ensure timely access to diagnostic services for evaluating carotid arteries.
- Development of agreements and processes for rapid access to surgical consults, including a mechanism for expedited referrals as required for carotid interventions.
- Ensure navigation of system is supported increasing patient compliance. Mechanims to increase compliance should be explored and assessed.
- Proportion of stroke or TIA patients with moderate to severe (50 percent to 99 percent) symptomatic carotid artery stenosis who undergo a carotid revascularization procedure following an index stroke/TIA event. (KQI)
- Proportion of stroke/TIA patients with moderate to severe (50 percent to 99 percent) carotid artery stenosis who undergo a carotid revascularization procedure following an index event within 2 weeks of first hospital or SPC assessment. (KQI)
- Median time from onset of index ischemic stroke or TIA symptoms to carotid revascularization (days, hours). (KQI)
- Proportion of stroke patients requiring carotid intervention who undergo the procedure within two weeks of the index stroke event.
- Proportion of stroke patients with moderate carotid stenosis (50 percent to 69 percent) who undergo carotid intervention procedure following the incident stroke event.
- Proportion of stroke patients with mild carotid stenosis (less than 50 percent) who undergo carotid intervention procedure following the incident stroke event.
- Proportion of carotid endarterectomy patients who experience perioperative in-hospital stroke, acute myocardial infarction or death.
- The 30-day in-hospital mortality rate after carotid endarterectomy and stroke rate by carotid occlusion severity.
- Proportion of patients who undergo carotid endarterectomy within two weeks, between two and four weeks, between four weeks and three months, and between three and six months of stroke onset.
- Proportion of patients who wait more than three months for carotid endarterectomy or whose surgery is cancelled because of long wait times. Proportion of patients who experience a subsequent stroke event or death while waiting for carotid endarterectomy.
- Time interval measurements should be taken from the time the patient or family reports as the time of stroke symptom onset to the actual date of surgery.
- The stroke onset time will depend on patient report or that of a reliable observer at the time of the event.
- Analysis should be stratified between those patients undergoing carotid stenting and those patients undergoing carotid endarterectomy, by severity of stenosis and by whether the patient had symptomatic or asymptomatic carotid artery disease.
- Data source for surgical date should be surgical note, nurses’ notes and discharge summary.
- In some cases, it may be more appropriate or relevant to record the time interval from the first time the patient has contact with medical care until the time of carotid surgery. This has occurred in cases where the patient was out of the country at the time of the stroke event and chose to return to Canada before seeking definitive medical intervention. It is important to note the nature of the start time when calculating turnaround times or intervention times.
Health Care Provider Information
Carotid endarterectomy (CEA) has been shown to be beneficial for preventing stroke recurrence in patients who have sustained a minor stroke or TIA with ipsilateral high-grade carotid stenosis. There are three large trials comparing endarterectomy for symptomatic stenosis with best medical treatment in such patients: The North American Symptomatic Carotid Endarterectomy Trial (NASCET, 1991), the European Carotid Surgery Trial (ECST, 1998) and the Veterans Affairs Trial (Mayberg et al. 1991). The results of these three trials were pooled in a Cochrane review (Rerkasem & Rothwell 2011). The risk of any stroke or operative death at 5-years in patients with severe stenosis (70–99%) was significantly reduced in patients in the CEA group (RR=0.53, 0.42-0.67, p<0.0001, NNT=6) with an associated absolute risk reduction of 16.0%. For patients with moderate stenosis (50-69%) the risk was also reduced (RR=0.77, 0.63- 0.94, p=0.001, NNT=22). For patients with mild stenosis, there was no benefit of treatment. Perioperative death or stroke incidence was 7.0% (95% CI 6.2 to 8.0). The greatest benefit of treatment was found in men, patients aged 75 years or over, and patients randomised within two weeks after their last ischaemic event.
The use of CEA for asymptomatic carotid artery disease is more controversial, given that it is a lower-risk condition. Significant improvements have been made in the medical management of stroke risk factors during the previous 20 years, including the use of statins, antihypertensive agents, and antiplatelets or anticoagulants. Using data from the Asymptomatic Carotid Emboli Study (ACES), which included 477 patients with at least 70% carotid stenosis and no symptoms in the carotid artery territory for at least the previous 2 years, the use of antiplatelet and antihypertensive agents were both significant independent predictors of lower stroke risk or TIA at the end of the follow-up period (King et al. 2013). There are three large trials that have evaluated the risks and benefits of CEA in the asymptomatic group. The Asymptomatic Carotid Atherosclerosis Study (ACAS) Group, the MRC [Medical Research Council] Asymptomatic Carotid Surgery Trial (ACST) Collaborative Trial and the Veterans Affairs Trial. The results of these trials were pooled in a Cochrane review (Chambers & Donnan 2008). Median duration of follow-up ranged from 2.7-4.0 years. Although the risk of perioperative stroke death was higher in the CEA group (3.0% vs. 0.46%, RR= 6.49, 95% CI 2.53-16.61, p<0.0001), CEA was associated with significant reductions in the risk of perioperative stroke or death or subsequent ipsilateral stroke, (RR=0.71, 95% CI 0.55-0.90, p= 0.0051) as well as stroke or death or any subsequent stroke (RR= 0.69, 95% CI 0.57- 0.83, p<0.0001). The greatest benefits were evident in men and younger patients. There were insufficient data to determine whether increasing degree of stenosis was associated with increasing beneﬁt from surgery. In 10-year follow-up of ACST (Halliday et al. 2010) in which patients were randomized to receive immediate treatment vs. delayed, immediate CEA was associated with a reduced occurrence of stroke at both 5 and 10 years (6.4% vs. 11.8%, p<0.0001 and 10.8% vs. 16.9%, p<0.0001, respectively). The authors concluded that despite a 3% perioperative stroke or death rate, CEA for asymptomatic carotid stenosis reduced the risk of ipsilateral stroke, and any stroke, by approximately 30% over three years, while acknowledging that the absolute risk reduction with carotid endarterectomy is small (1%/year).
Carotid Artery Stenting vs. Best Medical Management
Carotid-artery angioplasty with stenting (CAS) has emerged as an alternative to carotid endarterectomy in patients at high risk for complications for endarterectomy such as contralateral occlusion or severe coronary artery disease. The percutaneous approach also avoids the risks of general anaesthesia and the local complications of neck haematoma, infection, cervical strain and cranial nerve damage associated with endarterectomy and, requires a shorter recovery period. Several large trials assessing the safety and effectiveness of CAS (without the use of embolic protection devices) have been conducted.
The Stenting and Aggressive Medical Management for Preventing Stroke in Intracranial Stenosis (SAMMPRIS) trial, was the first large open-label clinical trial that randomly assigned patients who had a recent transient ischemic attack or stroke attributed to severe stenosis to receive aggressive medical management alone or aggressive medical management plus percutaneous transluminal angioplasty with stenting (PTAS), using the Wingspan stent system (Chimowitz et al. 2011). The primary end point was stroke or death within 30 days after enrollment or after a revascularization procedure for the qualifying lesion during the follow-up period or stroke in the territory of the qualifying artery beyond 30 days. Enrollment was stopped after 451 patients were enrolled because there was a significant increase in the number of patients in the PTAS group had a primary outcome event (20.5% vs. 11.5%, p=0.009). There was also an increased number of patients in the PTAS group who experienced any stroke during the study period (22.3% vs. 14.1%, p=0.03). The final results of this trial have been published recently (Derdeyn et al. 2014). The median follow-up period was 32.4 months. Fewer patients in the medical group had a primary endpoint event (15% vs. 23%) and the cumulative probability of the primary endpoints was significantly smaller in the medical group (p=0.0252). A similar trial, Vitesse Stent Ischemic Therapy (VISSIT) was halted after the recruitment of 112 patients, when the negative results from the SAMMPRIS trial became available (Zaidat et al. 2015). Among patients who had been randomized up to that point, the 1-year primary outcome occurred significantly more frequently in patients in the stenting group (36.2% vs. 15.1%, mean difference=21.1%, 95% CI 5.4-36.8%, p=0.02). The risk of stroke recurrence (but not TIA) within one year was also significantly higher in the stenting group (34.5 vs. 9.4%, mean difference 25.1%, 95% CI 10.5-39.6%, p=0.003).
Carotid Artery Stenting (without embolic protection) vs. Carotid Endarterectomy
The Carotid and Vertebral Artery Transluminal Angioplasty Study (CAVATAS) included 504 patients with symptomatic or asymptomatic carotid artery stenosis of ≥30%, considered to require revascularisation and suitable for surgery or endovascular treatment (Brown et al. 2001). Patients were randomized to endovascular treatment with balloon angioplasty with or without stent insertion or CEA. Stents were used in 55 patients. At the time the trial was conducted, no protection devices were available. The median delay from randomization to surgery was 20 days (endovascular treatment) and 27 days (CEA). Mean length of follow-up was 2 years. There were no differences between groups (endovascular treatment vs. CEA) including death (3% vs. 2%), disabling stroke (4% vs. 4%), non-disabling stroke (4% vs. 4%), death or disabling stroke (6% vs. 6%) or death or any stroke (10% vs. 10%) within 30 days. At one year following treatment, severe carotid stenosis (70%-99%) was more common in patients who had received endovascular treatment (14% vs. 4%; p<0.001). In a long-term follow-up study (Ederle et al. 2009), the 8-year cumulative incidence of disabling stroke or death was non-significantly higher in the endovascular treatment group (45.2% vs. 50.4%, HR=1.02, 95% CI 0.79-1.32) as was the combined outcome of non-perioperative stroke or TIA (HR=1.37, 95% CI 0.95-1.97).
The Stent-Supported Percutaneous Angioplasty of the Carotid Artery versus Endarterectomy (SPACE) Trial included 1,200 patients, with symptomatic carotid artery stenosis, who had experienced TIA or moderate stroke within 180 days and with severe carotid artery stenosis (≥ 50% according to NASCET) (Ringleb et al. 2006). Patients were randomized to receive CAS (27% used embolic protection devices) or CEA after a median delay of 4-5 days. The trial was stopped prematurely due to concerns regarding funding and futility. There were no differences between groups on either any of the primary outcomes of 30-day ipsilateral stroke or death, or any of the secondary outcomes (disabling stroke or death from any cause within 30 days, disabling stroke, or procedural failures).
Carotid Artery Stenting (with embolic protection) vs. Carotid Endarterectomy
Several randomized trials that directly compared the safety of CEA with CAS (with protection) among patients who were symptomatic and/or asymptomatic have been published. The results from most of them suggest that during long-term follow-up, stenting is as effective as CEA.
The Asymptomatic Carotid Trial (ACT 1) (Rosenfield et al. 2016), a noninferiority trial was stopped early due to slow enrolment. While the protocol aimed to recruit 1,658 patients, data from only 328 patients were available for follow-up assessment at 5 years. At one year, the occurrence of the primary outcome (composite of death, stroke, or myocardial infarction within 30 days of the procedure or ipsilateral stroke within 1 year of the procedure) was 3.8% for stenting group compared with 3.4% for CEA group. The threshold of a 3%-point difference for inferiority was not exceeded (upper 95% CI for difference was 2.27%), suggesting that CAS was not inferior to endarterectomy. Survival from 30 days to 5 years was not significantly different between groups (87.1% stenting group vs. 89.4% CEA group, p=0.21).
The International Carotid Stenting Study (ICSS) trial enrolled 1,713 patients >40 years, with symptomatic carotid artery stenosis ≥50% using the NASCET criteria (Ederle et al. 2010). Between randomization and 120 days, stenting was associated with an increased risk of stroke, death or procedural myocardial infarction, (8.5% vs. 5.2%, HR=1.69, 95% CI 1.16-2.45, p=0.006) any stroke (7.7% vs. 4.1%, HR=1.92, 95% CI 1.27-2.89, p=0.002), any stroke or death (8.5% vs. 4.7%, HR=1.86, 95% CI 1.26-2.74, p=0.001) and all-cause mortality (2.3% vs. 0.8%, HR=2.76, 95% CI 1.16-6.56, p=0.017). In the long-term study analysis Bonati et al. (2015) reported that after a median duration of 4.2 years the risk of any stroke was significantly increased in the stenting group (HR=1.71, 95% CI 1.28 -2.3, p=0.0003), while stenting was not associated with an increased risk of fatal or disabling stroke (HR=1.06, 95% CI 0.72-1.57, p=0.77). There was also a significantly increased risk of the outcome of periprocedural stroke/procedural death or ipsilateral stroke during follow-up (HR=1.72, 95% CI 1.24-2.39, p=0.001).). In both the per protocol and intention-to-treat analyses, the cumulative 5-year stroke risk was significantly higher in the stenting group (HR=1.53, 95% CI 1.02-2.31 and HR=1.71, 95% CI 1.28-2.30, respectively), while the 5-year risk of fatal or disabling stroke was not increased. The distribution of modified Rankin Scores was similar between groups.
The Carotid Revascularization Endarterectomy Versus Stenting Trial (CREST) trial included 2,502 patients with asymptomatic or symptomatic carotid artery stenosis who had experienced a minor stroke or TIA within the previous 180 days (Brott et al. 2010). The primary end point was the composite of any stroke, myocardial infarction, or death during the peri-procedural period or ipsilateral stroke within four years after randomization. There was no significant difference in the estimated four-year rates of the primary end point between groups (7.2% vs. 6.8%); however, the 4-year rate of stroke or death was higher in the stenting group (6.4% vs. 4.7%, HR=1.50, 95% CI 1.05-2.15, p=0.03). During the periprocedural period, there was a significantly increased risk of stroke or death associated with stenting, but no difference in risk for stroke, death or MI between treatment conditions from 31 days to end of follow-up. After the 30-day, periprocedural period, incidence of ipsilateral stroke was similarly low in both groups (2.0 vs. 2.4%, p=0.85). At 10 years, there was no significant difference between groups in the risk of the primary outcome, which included stroke, death or MI (HR=1.10, 95% CI 0.83-1.44, p=0.51), or in the risk of stroke between groups (HR=0.99, 95% CI 0.64-1.52) (Brott et al. 2016).
A Cochrane review (Bonati et al. 2012) included 16 trials of patients with symptomatic stenosis, who had experienced a minor stroke, asymptomatic stenosis or both asymptomatic and symptomatic carotid stenosis. The treatment contrasts included any CEA procedure vs. any endovascular technique. Endovascular therapy was associated with a higher risk of death or any stroke within 30 days of treatment (OR=1.72, 95% CI 1.29- 2.31, p<0.0003), and death or any stroke between randomization and 30 days after treatment or ipsilateral stroke until the end of follow-up. The risk was highest in the group of studies with follow-up of 2.4 years, or longer. There was no difference between treatment groups for the outcomes of death or major or disabling stroke between randomization and 30 days after treatment (OR=1.28, 95% CI 0.93-1.77, p=0.13) or any stroke between randomization and 30 days after treatment (OR=1.21, 95% CI 0.36- 4.04, 9=0.76), although the risk was increased among the subgroup of patients at standard surgical risk, who received endovascular treatment. Endovascular therapy was associated with a reduced risk of cranial nerve palsy within 30 days of surgery and access site hematoma (OR=0.37, 95% CI 0.18- 0.77, p= 0.0082). The authors suggested that while endovascular treatment was associated with an increased risk of peri-procedural stroke or death compared with endarterectomy, the excess risk may be limited to older patients. The results from another meta-analysis, examining the same treatment contrasts, and using the results from 13 RCTs reported that compared with CEA, stenting was associated with an increase of 19 strokes and 10 fewer MIs for every 1000 patients treated (Murad et al. 2011).
Cervical Artery Dissection
While the incidence of cervical artery dissections (CAD) is relatively low, estimated to be between 2.6 to 2.9 per 100,000, CAD is over-represented among persons less than 45 years (Weimar et al. 2010). Given the increased risk of recurrent stroke associated with CAD, treatment with either antiplatelets or anticoagulants for at least 3 months is recommended. Based on the results of the Cervical Artery Dissection in Stroke (CADISS) Study (2015), treatment with either agent appears to be equally effective for the prevention of recurrent stroke. In this trial, 250 patients with extracranial carotid or vertebral artery dissection were randomized, within 7 days of the event, to receive antiplatelet agents (dipyridamole, aspirin or clopidogrel, alone or in combination) or anticoagulant therapy (UFH, LMWH, followed by warfarin, with a target INR of 2-3), for the study duration. At the end of 3 months, the frequency of the primary outcome (stroke or death), was similar between groups. There were 4 recurrent strokes (3 antiplatelet vs. 1 anticoagulant) and no deaths in either group. There was a single case of major bleeding in the anticoagulant group. Similar findings were reported in a meta-analysis including the results of 34 non-randomized studies examining the same treatment contrast (Menon et al. 2008). There were 13/185 (7.0%) in the antiplatelet group and 17/447 (3.8%) in the anticoagulant group who suffered a TIA or stroke. The risk difference between groups was not significant (5%, 95% CI -1% to 11%, p = 0.11). The use of novel oral anticoagulants (NOAC) for the prevention of recurrent stroke following CAD has not been well studied. There are no RCTs to date. In a retrospective study (Caprio et al. 2014) including 149 patients with CAD, who were prescribed antithrombotic medication at hospital discharge, there were 2 recurrent strokes during a median of 7.5 months follow-up in the NOAC group compared with one each in the anticoagulant (AC) and antiplatelet (AP) groups. There were significantly fewer major hemorrhagic events in the NOAC group (0 vs. 8 [AC] and 1 (AP], p=0.034).
Johansson E, Cuadrado-Godia E, Hayden D, Bjellerup J, Ois A, Roquer J, Wester P, Kelly PJ. Recurrent stroke in symptomatic carotid stenosis awaiting revascularization A pooled analysis. Neurology. 2016;86(6):498-504.
Johansson EP, Arnerlöv C, Wester P. Risk of recurrent stroke before carotid endarterectomy: the ANSYSCAP study. International Journal of Stroke. 2013;8(4):220-7.
Marnane M, Chroinin DN, Callaly E, Sheehan OC, Merwick A, Hannon N, Horgan G, Kyne L, Moroney J, McCormack PM, Dolan E. Stroke recurrence within the time window recommended for carotid endarterectomy. Neurology. 2011;77(8):738-43.
Ois A, Cuadrado-Godia E, Rodríguez-Campello A, Jimenez-Conde J, Roquer J. High risk of early neurological recurrence in symptomatic carotid stenosis. Stroke. 2009;40(8):2727-31.