How are you taking part in this consultation?

You will not be able to change how you comment later.

You must be signed in to answer questions

    The content on this page is not current guidance and is only for the purposes of the consultation process.

    Interventional procedure overview of transfemoral carotid artery stent placement for asymptomatic extracranial carotid stenosis

    Closed for comments This consultation ended on at   Request commenting lead permission

    Evidence summary

    Population and studies description

    This interventional procedures overview is based on over 300,000 patients from 9 key studies including 1 systematic review and network meta-analysis (Gasior 2023), 2 systematic reviews (Wang 2022 and Müller 2020), 2 randomised controlled trials (Halliday 2021 and Reiff 2022), 1 prospective cohort study (Yang 2021), 2 retrospective cohort studies (Cole 2020 and Jalbert 2015) and 1 retrospective international administrative dataset (Gaba 2021). This is a rapid review of the literature, and a flow chart of the complete selection process is shown in figure 1. This overview presents 9 studies as the key evidence in table 2 and table 3, and lists 154 other relevant studies in table 5.

    Studies took place in Europe (including the UK), North America, Oceania, South America, Asia and Africa. There was a higher proportion of men and the mean or median age in the studies that reported it was above 60 years.

    The 2 randomised controlled trials included only asymptomatic people with carotid artery stenosis, and both had 5-year follow up. One compared CAS against CEA and random allocation was done between 2008 and 2020, with a sample size of 3,625 (Halliday 2021). It included asymptomatic people with severe unilateral or bilateral carotid artery stenosis (generally 60% or higher on ultrasound). This trial was included in the systematic reviews by Gasior et al. (2023) and Wang et al. (2022). The other trial was initially designed to have 3 arms comparing CAS, CEA and BMT. But, it was amended because of slow recruitment to 2 substudies, each with 2 arms, comparing CEA plus BMT with BMT alone and CAS plus BMT with BMT alone. It included asymptomatic people who had at least 70% (using European Carotid Surgery Trial criteria) stenosis and were 50 to 85 years old. Recruitment and survey took place between 2009 and 2019. The original planned sample size was 3,640, but the study was prematurely terminated because of low recruitment and the final sample size was 513 (Reiff 2022). The 1-year results from this trial were included in the systematic reviews by Gasior et al. (2023) and Wang et al. (2022).

    The systematic review and network meta-analysis included data on asymptomatic people with carotid artery stenosis from 13 randomised controlled trials. The trials were published before November 2021 with follow-up data up to 5 years. It was designed to compare perioperative and long-term stroke and mortality risk between CEA, CAS and BMT (before and after 2000; Gasior 2023).

    The systematic review by Wang et al. (2022) included data on people with asymptomatic carotid artery stenosis who had CAS or CEA. The data was from 7 randomised controlled trials published before November 2021, with follow-up data up to 10 years. It described the overall risk of bias among the included trials as low. The systematic review by Müller et al. (2020) included 22 randomised controlled trials comparing CAS with CEA, 7 of which had data on asymptomatic people. The trials were published before August 2018. It described the evidence as high quality in general but noted that more studies are needed on the asymptomatic population.

    The retrospective cohort study by Cole et al. (2020) included data on 435,627 people who had CAS or CEA for asymptomatic (70%) or symptomatic carotid artery stenosis between 2010 and 2015. People were matched based on demographics, comorbidities and severity of illness. Data on periprocedural outcomes and readmissions within 90 days was analysed.

    The study by Gaba et al. (2021) included data from an international administrative dataset on 16,079 people with asymptomatic and 2,918 people with symptomatic carotid artery stenosis, who had CAS or CEA. The primary outcome was in-hospital death within 7 days.

    The study by Yang et al. (2021) was a prospective cohort study of CEA or CAS in China. It included people with both asymptomatic and symptomatic carotid stenosis. Outcomes were presented separately for each group. The primary outcome was stroke, myocardial infarction or death within 1 month of the procedure.

    The study by Jalbert et al. (2015) was a retrospective cohort study of Medicare beneficiaries aged at least 66 years in the US who had CAS. Approximately half were asymptomatic, 91% were at high surgical risk, and 97% had carotid stenosis of at least 70%. There was a high prevalence of ischaemic heart disease, heart failure, diabetes mellitus and peripheral artery disease in the study population. A high proportion of participants had had coronary artery bypass surgery in the past year. The main outcomes were periprocedural (30-day) and long-term risks of mortality and stroke or TIA, as well as periprocedural myocardial infarction.

    Table 2 presents study details.

    There were also a number of case reports of adverse events identified in the literature, which have been included in table 5.

    Figure 1 Flow chart of study selection

    Table 2 Study details

    Study no.

    First author, date

    country

    Patients (male: female)

    Age

    Study design

    Inclusion criteria

    Intervention

    Follow up

    1

    Gasior 2023

    Austria, Germany, Switzerland, US, Canada, Israel, Germany, Russia and 2 studies included 30 or more countries.

    n=14,310

    'The majority of patients in all reports were men'

    Pooled mean 67.9 years (range 40 to 91)

    Systematic review and network meta-analysis

    Search date: November 2021

    13 randomised controlled trials (17 records)

    Published randomised controlled trials comparing at least 2 modalities of intervention (CEA, CAS, and BMT alone) were included. Asymptomatic lesions were defined as image confirmed carotid stenosis 50% or more in the absence of preceding neurological symptoms, indicating a cerebrovascular event in the 6 months before study enrolment.

    • CAS

    • CEA

    • BMT (traditional, which was used with recruitment dates before 2000 or modern, used for trials with recruitment dates after 2000)

    Up to 5 years

    2

    Wang J, 2022

    Country not reported for individual studies

    n=7,117

    Proportion of women was between 30% and 43%

    Mean ranged from 66.6 to 70.0 years

    Systematic review and meta-analysis

    Search date: November 2021

    7 randomised controlled trials (16 articles) were included.

    Only randomised controlled trials evaluating participants having CAS versus CEA in asymptomatic carotid artery stenosis patients, published in English, were included. Outcome events included any stroke (whether disabling, severe, non-disabling, or minor), death, or myocardial infarction.

    • CAS (n=3,919)

    • CEA (n=3,198)

    (Note: there is some discrepancy in the total number of patients described throughout the review. The numbers cited here have been taken from the table rather than the text).

    Up to 10 years

    3

    Müller MD, 2020

    Studies were based in UK, Europe, Australia, Canada, US, China, France, Germany, Austria, Switzerland, New Zealand, Czech Republic

    n=3,378 with asymptomatic carotid stenosis (5,396 people with symptomatic carotid stenosis were also included)

    Overall mean not reported

    Systematic review and meta-analysis (Cochrane review)

    Search date: August 2018

    22 randomised controlled trials were identified, 7 of which included data on asymptomatic people.

    Randomised controlled trials comparing CAS with CEA for symptomatic or asymptomatic atherosclerotic carotid stenosis. In addition, randomised controlled trials comparing CAS with medical therapy alone were included.

    Of the 7 trials, 5 included people with asymptomatic carotid stenosis only.

    • CAS (n=2,058)

    • CEA (n=1,320)

    Any acceptable technique for CEA (such as use of a shunt or not, patching or not, local or general anaesthesia) and any acceptable endovascular technique for treatment of carotid artery stenosis (such as simple balloon angioplasty, use of a stent or not, any type of cerebral protection device) were included.

    Up to 5 years

    4

    Halliday A, 2021

    33 countries, from Europe (including UK), North America, South America, Asia and Africa

    n=3,625

    (2,545:1,080)

    1,802 (50%) were below 70 and 1,823 (50%) were 70 or older.

    Randomised controlled trial (ACST-2; international, multicentre)

    Random allocation was done between 2008 and 2020.

    Severe unilateral or bilateral carotid artery stenosis (generally 60% or higher on ultrasound); no relevant neurological symptoms in the preceding 6 months; CT or MRI confirmation of suitability for CAS and for CEA.

    Exclusion criteria included previous ipsilateral intervention, unsuitability for CAS or CEA, high procedural risk, high risk of cardiac emboli, or any major life-threatening condition. Patients likely to need other surgery could not enter the trial until at least 1 month after it.

    • CAS (n=1,811)

    Any CE-approved devices were allowed, and procedural double antiplatelet therapy was usual.

    • CEA (n=1,814)

    Shunting and patching were optional.

    Long-term medical care was to be similar in both groups and generally involved antithrombotic, antihypertensive, and lipid-lowering therapy.

    Mean 4.9 years (range 0 to 12)

    5

    Reiff T, 2022

    Austria, Germany, Switzerland

    n=513

    (381:132)

    Median 70 years

    Randomised controlled trial (SPACE-2; multicentre)

    Recruitment started in 2009 and was terminated early in 2014 because of continued low recruitment.

    People aged 50 to 85 years with asymptomatic carotid artery stenosis of the distal common carotid artery or the extracranial internal carotid artery of at least 70% according to European Carotid Surgery Trial criteria, or equivalent to at least 50% to 99% according to North American Symptomatic Carotid Endarterectomy Trial criteria, using ultrasound criteria.

    • CAS plus BMT (n=197)

    • CEA plus BMT (n=203)

    • BMT alone (n=113)

    CAS was usually done under local anaesthetic. Routine access route was the femoral artery. Before stenting, dual antiplatelet treatment was offered for at least 3 days, and statins (if not already being taken). Dual antiplatelet therapy was continued for 4 to 6 weeks.

    CEA was done under either general or local anaesthetic with optional shunt use, and neurological and cardiopulmonary monitoring. Statins were started 1 week preoperatively, and antiplatelet therapy was offered for at least 3 days preoperatively.

    BMT comprised medication recommended by current treatment guidelines and modification of lifestyle factors.

    Median 59.9 months

    6

    Cole T, 2020

    US

    n=435,627

    70% asymptomatic

    61% male

    Mean 71 years

    Retrospective cohort study (matched)

    Patients who had CEA or CAS for asymptomatic and symptomatic carotid stenosis between 2010 and 2015.

    • CAS (n=57,273)

    • CEA (n=378,354)

    90 days

    7

    Gaba K, 2021

    Australia, Belgium, Denmark, Finland, England, the Netherlands, and US

    n=16,079 asymptomatic

    (2,918 symptomatic)

    35% female

    Median 69 years (CAS), 71 years (CEA)

    International administrative dataset (retrospective)

    All people who had CEA or CAS between 2011 and 2015. They were defined as symptomatic if they had had a stroke, TIA, or amaurosis fugax in the 6 months before CEA or CAS. If they had not, they were deemed asymptomatic.

    • CAS (n=2,777 asymptomatic and symptomatic)

    • CEA (n=16,220 asymptomatic and symptomatic)

    7 days

    8

    Yang B, 2021

    China

    n=1,645 asymptomatic

    83% male

    Mean 65.9 years (CAS), 64.3 (CEA), p<0.001

    Prospective multicentre cohort study

    2013 to 2016

    Age 18 years or over, symptomatic or asymptomatic internal carotid artery stenosis with degree of stenosis 50% or more (moderate to severe stenosis) as assessed by CT angiography or digital subtraction angiography. Patients with a TIA or stroke within 6 months were defined as symptomatic.

    • CAS (n=887)

    • CEA (n=758)

    The choice of embolic protection devices and stents, dilation strategy and balloon size were left to the discretion of the physician.

    1 month

    9

    Jalbert J, 2015

    US

    n=22,516 (11,839 asymptomatic)

    61% male

    Mean 76 years

    Cohort study

    Fee-for-service Medicare beneficiaries at least 66 years old undergoing CAS with embolic protection and continuously enrolled in Medicare for at least 1 year.

    97% of participants had at least 70% carotid artery stenosis.

    • CAS

    Mean 747 days (for mortality)

    Table 3 Study outcomes (results are for asymptomatic people unless otherwise stated)

    First author, date

    Efficacy outcomes

    Safety outcomes

    Gasior S, 2023

    Long-term stroke (13 trials, n=11,693)

    When compared with combined BMT, both CEA (OR 0.35, 95% CI 0.21 to 0.59) and CAS (OR 0.51, 95% CI 0.26 to 0.98) significantly reduced the odds of a minor stroke occurring up to 5 years.

    GRADE assessment established a high degree of certainty that CEA resulted in 20 fewer minor stroke events per 1,000 patients (95% CI -0.03 to -0.01) than modern BMT. CAS resulted in 10 fewer minor stroke events per 1,000 patients (95% CI -0.02 to 0.0) than modern BMT.

    CEA (OR 0.27, 95% CI 0.19 to 0.39) and CAS (OR 0.37, 95% CI 0.22 to 0.62) also reduced the odds of an ipsilateral stroke occurring up to 5 years versus combined BMT alone, a finding consistent with the use of BMT before and after 2000.

    Long-term mortality (9 trials, n=11,101)

    With a moderate level of certainty, the results demonstrated that no treatment was significantly superior in reducing the odds of mortality from 30 days to 5 years.

    Perioperative stroke (6 trials, n=6,855)

    CEA reduced the odds of all stroke events occurring within 30 days perioperatively compared with CAS (OR 1.6, 95% CI 1.1 to 2.2).

    CEA specifically reduced the odds of a minor perioperative stroke (OR 1.7, 95% CI 1.1 to 2.6) but did not affect the odds of a major stroke event occurring (OR 0.89, 95% CI 0.33 to 2.3.) Neither intervention reduced the odds of contralateral perioperative stroke or ipsilateral stroke event (OR 1.7, 95% CI 0.96 to 3.1) GRADE assessment=moderate degree of confidence

    Perioperative mortality (6 trials, n=6,855)

    Neither procedure was favoured to reduce the odds of 30 day perioperative mortality.

    Perioperative myocardial infarction (6 trials, n=855)

    There was moderate certainty that CAS reduced the odds of perioperative myocardial infarction compared with CEA (OR 0.49, 95% CI 0.26 to 0.91). However, the risk difference was 0 fewer cases per 1,000 (95% CI 0 to ‑0.01).

    Cranial nerve injury (5 trials, n=5,360)

    CAS reduced the odds of cranial nerve injury compared with CEA (OR 0.07, 95% CI 0.01 to 0.42). There was moderate certainty that the risk difference was 40 fewer per 1,000 (95% CI 0 to -0.07). Only 1 randomised controlled trial reported a cranial nerve injury following CAS.

    Wang J, 2022

    Long-term outcomes of composite outcome among any perioperative stroke, death, or myocardial infarction and long-term stroke, death, or myocardial infarction

    • CAS=9.8% (309/3,141)

    • CEA=10.9% (264/2,423)

    OR=1.18 (95% CI 0.94 to 1.48), p=0.14, I2=0%

    (4 studies)

    30 days outcome of composite endpoint including stroke, death and myocardial infarction

    • CAS=3.5% (137/3,920)

    • CEA=3.2% (102/3,198)

    OR=1.13 (95% CI 0.87 to 1.47), p=0.37, I2=0% (7 studies)

    30 days outcome of any stroke

    • CAS=3.0% (112/3,735)

    • CEA=1.9% (58/3,010)

    OR=1.62 (95% CI 1.17 to 2.25), p=0.004, I2=0% (5 studies)

    30 days outcome of non-disabling strokes

    • CAS=2.5% (88/3,562)

    • CEA=1.4% (40/2,833)

    OR=1.81 (95% CI 1.23 to 2.65), p=0.003, I2=0% (4 studies)

    30 days outcome of disabling strokes and death

    • CAS=0.7% (24/3,494)

    • CEA=0.8% (22/2,765)

    OR=0.91 (95% CI 0.50 to 1.65), p=0.76, I2=0% (3 studies)

    Müller MD, 2020

    Death or any stroke between randomisation and 30 days after treatment or ipsilateral stroke until end of follow up

    • CAS=45 per 1,000 (95% CI 31 to 63)

    • CEA=36 per 1,000

    OR=1.27 (95% CI 0.87 to 1.84); n=3,315, 6 studies, p=0.22, I2=0%; moderate-certainty evidence

    Severe (70% or more) restenosis during follow up (in people with asymptomatic or symptomatic carotid artery stenosis)

    • CAS=67 per 1,000 (95% CI 43 to 103)

    • CEA=56 per 1,000

    OR=1.21 (95% CI 0.76 to 1.93); n=5,744, 9 studies, p=0.42, I2=61.4%; low-certainty evidence

    Moderate or severe (50% or more) restenosis during follow up (in people with asymptomatic or symptomatic carotid artery stenosis)

    • CAS=30.3% (309/1,019)

    • CEA=21.1% (231/1,096)

    OR=2.00 (95% CI 1.12 to 3.6); n=2,115, 4 studies, p=0.02, I2=43.6%

    Death or any stroke between randomisation and 30 days after treatment

    • CAS=25 per 1,000 (95% CI 14 to 42)

    • CEA=14 per 1,000

    OR=1.72 (95% CI 1.00 to 2.97); n=3,378, 7 studies, p=0.05, I2=0%; moderate-certainty evidence

    Death or major or disabling stroke between randomisation and 30 days after treatment

    • CAS=5 per 1,000 (95% CI 1 to 19)

    • CEA=3 per 1,000

    OR=1.54 (95% CI 0.39 to 6.11); n=2,601, 2 studies, p=0.54, I2=0%; moderate-certainty evidence

    Myocardial infarction between randomisation and 30 days after treatment

    • CAS=0.65% (12/1,844)

    • CEA=1.45% (16/1,102)

    OR=0.53 (95% CI 0.24 to 1.15); n=2,601, 6 studies, p=0.11, I2=0%

    Death or any stroke or myocardial infarction between randomisation and 30 days after treatment

    • CAS=32 per 1,000 (95% CI 20 to 50)

    • CEA=28 per 1,000

    OR=1.16 (95% CI 0.73 to 1.85); n=2,978, 6 studies, p=0.8, I2=0%; moderate-certainty evidence

    Cranial nerve palsy

    • CAS=0.1% (2/1,823)

    • CEA=3.3% (36/1,092)

    OR=0.09 (95% CI 0.03 to 0.27); n=2,915, 5 studies, p<0.0001, I2=0%

    Access site haematoma (needing surgery, blood transfusion, or prolonging hospital stay, where severity was defined)

    • CAS=0.2% (3/1,726)

    • CEA=1.5% (15/993)

    OR=0.14 (95% CI 0.02 to 0.9); n=2,719, 3 studies, p=0.04, I2=34.7%

    Halliday A, 2021

    Any non-procedural stroke during follow up

    • CAS=5.2% (91/1,748)

    • CEA=4.5% (79/1,767)

    RR 1.16 (95% CI 0.86 to 1.57), p=0.33

    Fatal non-procedural stroke (mRS score 6) during follow up

    • CAS=0.9% (16/1,748)

    • CEA=1.1% (20/1,767)

    Disabling non-procedural stroke (mRS score 3 to 5) during follow up

    • CAS=1.6% (28/1,748)

    • CEA=1.4% (25/1,767)

    Non-disabling non-procedural stroke (mRS score 0 to 2) during follow up

    • CAS=2.7% (47/1,748)

    • CEA=1.9% (34/1,767)

    Myocardial infarction within 30 days of procedure

    • CAS=0.3% (5/1,705; none were fatal)

    • CEA=0.7% (12/1,736; 4 were fatal), p=0.15

    Any stroke within 30 days of procedure

    • CAS=3.6% (61/1,705)

    • CEA=2.4% (41/1,736), p=0.06

    Fatal stroke within 30 days of procedure

    • CAS=0.4% (7/1,705)

    • CEA=0.3% (5/1,736), p=0.77

    Non-disabling stroke (mRS score 0 to 2) within 30 days of procedure

    • CAS=2.7% (48/1,705)

    • CEA=1.6% (29/1,736), p=0.03

    Death or disabling stroke within 30 days of procedure

    • CAS=0.9% (15/1,705)

    • CEA=1.0% (18/1,736), p=0.77

    Death, myocardial infarction or any stroke within 30 days of procedure

    • CAS=3.9% (67/1,705)

    • CEA=3.2% (55/1,736), p=0.26

    Cranial nerve palsy within 30 days of procedure

    • CAS=0% (0/1,653)

    • CEA=5.4% (96/1,788)

    Reiff T, 2022

    Cumulative incidence of any stroke or death from any cause within 30 days or any ipsilateral ischaemic stroke within 5 years (primary endpoint):

    • CAS plus BMT=4.4% (95% CI 2.2 to 8.6)

    • CEA plus BMT=2.5% (95% CI 1.0 to 5.8)

    • BMT alone=3.1% (95% CI 1.0 to 9.4)

    The difference between treatment groups was not statistically significant (p=0.62).

    There were 3 strokes in the CAS group after 30 days (all ipsilateral) and none in the CEA group.

    The first primary endpoint event in the BMT group was on day 153 after randomisation (ipsilateral, disabling stroke), with another 2 ipsilateral events (1 disabling) in the 5-year follow-up period.

    Cox proportional-hazard testing

    • CAS plus BMT versus BMT alone: HR 1.55, 95% CI 0.41 to 5.85, p=0.52

    • CEA plus BMT versus BMT alone: HR 0.93, 95% CI 0.22 to 3.91, p=0.93

    • CAS versus CEA: HR 1.66, 95% CI 0.54 to 5.08, p=0.37

    5-year cumulative incidence of ipsilateral ischaemic stroke

    • CAS plus BMT=4.4% (95% CI 2.2 to 8.6)

    • CEA plus BMT=2.0% (95% CI 0.7 to 5.2)

    • BMT alone=3.1% (95% CI 1.0 to 9.4)

    Cox proportional-hazard testing

    • CAS plus BMT versus BMT alone: HR 1.55, 95% CI 0.41 to 5.85, p=0.52

    • CEA plus BMT versus BMT alone: HR 0.74, 95% CI 0.17 to 3.32, p=0.70

    • CAS versus CEA: HR 2.09, 95% CI 0.63 to 6.92, p=0.23

    5-year cumulative incidence of any ischaemic or haemorrhagic stroke

    • CAS plus BMT=9.8% (95% CI 6.2 to 15.3)

    • CEA plus BMT=5.3% (95% CI 2.9 to 9.6)

    • BMT alone=6.5% (95% CI 2.9 to 13.9)

    Cox proportional-hazard testing

    • CAS plus BMT versus BMT alone: HR 1.66, 95% CI 0.65 to 4.20, p=0.29

    • CEA plus BMT versus BMT alone: HR 0.93, 95% CI 0.34 to 2.57, p=0.89

    • CAS versus CEA: HR 1.78, 95% CI 0.81 to 3.88, p=0.15

    5-year cumulative all-cause mortality

    • CAS plus BMT=9.3% (95% CI 5.7 to 15.0)

    • CEA plus BMT=7.6% (95% CI 4.5 to 12.8)

    • BMT alone=8.0% (95% CI 4.1 to 15.4)

    Cox proportional-hazard testing

    • CAS plus BMT versus BMT alone: HR 1.05, 95% CI 0.45 to 2.48, p=0.91

    • CEA plus BMT versus BMT alone: HR 0.87, 95% CI 0.36 to 2.11, p=0.76

    • CAS versus CEA: HR 1.21, 95% CI 0.57 to 2.53, p=0.62

    Restenosis of at least 70%

    • CAS=10.2% (95% CI 6.6 to 15.8)

    • CEA=3.2% (95% CI 1.5 to 7.1)

    HR 3.19, 95% CI 1.27 to 8.04, p=0.0138

    Stroke within 30 days of procedure

    • CAS=2.5% (5/197; all ipsilateral, 1 was disabling)

    • CEA=2.5% (5/203; 4 ipsilateral, 1 contralateral and disabling)

    There were no deaths within 30 days of the procedure.

    Serious adverse events reported over the 5-year follow-up period

    • CAS, n=289; CEA, n=243; BMT, n=151

    Type of serious event (proportion of total events in each treatment group)

    • Cardiovascular events

      • CAS=42% (120/289)

      • CEA=44% (106/243)

      • BMT=50% (75/151)

    • Nervous system and sensory events

      • CAS=17% (49/289)

      • CEA=10% (25/243)

      • BMT=15% (23/151)

    • Musculoskeletal events

      • CAS=11% (32/289)

      • CEA=7% (18/243)

      • BMT=9% (14/151)

    • Urogenital events

      • CAS=6% (17/289)

      • CEA=10% (24/243)

      • BMT=6% (9/151)

    • Digestive events

      • CAS=5% (15/289)

      • CEA=8% (20/243)

      • BMT=7% (10/151)

    • Respiratory events

      • CAS=6% (16/289)

      • CEA=5% (13/243)

      • BMT=6% (9/151)

    • Endocrinopathies, nutritional events, and metabolic events

      • CAS=3% (8/289)

      • CEA=1% (3/243)

      • BMT=2% (3/151)

    • Haematological events

      • CAS=1% (4/289)

      • CEA=4% (9/243)

      • BMT=1% (1/151)

    • Skin events

      • CAS=2% (7/289)

      • CEA=2% (4/243)

      • BMT=1% (1/151)

    • Psychiatric events

      • CAS<1% (1/289)

      • CEA=2% (4/243)

      • BMT=0% (0/151)

    • Allergic events

      • CAS=0% (0/289)

      • CEA<1% (1/243)

      • BMT=0% (0/151)

    • Other events

      • CAS=7% (20/289)

      • CEA=7% (16/243)

      • BMT=4% (6/151)

    Two intracranial haemorrhages were reported, 1 in the CAS plus BMT group (431 days after randomisation) and a fatal event in the BMT alone group that happened 402 days after randomisation. Three hyperperfusion syndromes were reported in the CAS plus BMT group, 1 with cerebral infarction, but none of these led to permanent disability.

    Cole T, 2020

    Length of hospital stay (midnights)

    • CAS=1.9

    • CEA=2.3, p<0.001

    Transfer to a skilled nursing or intermediate

    care facility

    • CAS=2.9%

    • CEA=2.6%, p<0.001

    Discharge with home health care

    • CAS=8.1%

    • CEA=4.8%, p<0.001

    Periprocedural stroke

    • CAS=0.2%

    • CEA=0.3%, p=0.113

    Periprocedural myocardial infarction

    • CAS=0.8%

    • CEA=1.2%, OR 1.58 (95% CI 1.27 to 1.97)

    In-hospital mortality

    There was no difference between the groups regarding in-hospital mortality difference, in asymptomatic patients.

    Readmission within 30 days

    On multivariate analysis, female sex (OR 1.18, 95% CI 1.10 to 1.26, p<0.001) and symptomatic status (OR 1.21, 95% CI 1.11 to 1.33, p<0.001) were associated with a higher risk of readmission within 30 days.

    Gaba K, 2021

    No efficacy data was reported.

    7-day in-hospital mortality

    • CAS=0.7% (18/2,470), 95% CI 0.4 to 1.1

    • CEA=0.1% (21/14,102), 95% CI 0.1 to 0.2

    Length of stay more than 2 days

    • CAS=27.5% (601/2,187), 95% CI 25.6 to 29.4

    • CEA=28.5% (2,864/10,039), 95% CI 27.7 to 29.4

    Head CT within 2 days of procedure

    • CAS=3.2% (71/2,187), 95% CI 2.5 to 4.0

    • CEA=2.1% (207/10,039), 95% CI 1.8 to 2.3

    Yang B, 2021

    No efficacy data was reported.

    Stroke, myocardial infarction, or death within 1 month of procedure (primary outcome):

    • CAS=4.7% (42/887), 95% CI 3.5 to 6.3

    • CEA=4.9% (37/758), 95% CI 3.5 to 6.6, p=0.890

    Stroke within 1 month of procedure:

    • CAS=4.1% (36/887), 95% CI 2.9 to 5.5

    • CEA=3.3% (25/758), 95% CI 2.2 to 4.8, p=0.416

    Ischaemic stroke within 1 month of procedure:

    • CAS=2.9% (26/887), 95% CI 2.0 to 4.2

    • CEA=2.0% (15/758), 95% CI 1.2 to 3.2, p=0.217

    Haemorrhagic stroke within 1 month of procedure:

    • CAS=1.1% (10/887), 95% CI 0.6 to 2.0

    • CEA=1.3% (10/758), 95% CI 0.7 to 2.3, p=0.723

    Death within 1 month of procedure:

    • CAS=0.7% (6/887), 95% CI 0.3 to 1.4

    • CEA=0.5% (4/758), 95% CI 0.5 to 1.2, p=0.699

    Myocardial infarction within 1 month of procedure:

    • CAS=0.5% (4/887), 95% CI 0.2 to 1.1

    • CEA=1.3% (10/758), 95% CI 0.7 to 2.3, p=0.056

    Other complications within 1 month of procedure (including cranial nerve injury, incision haematoma, and pulmonary infection):

    • CAS=2.6% (23/887), 95% CI 1.7 to 3.8

    • CEA=3.6% (27/758), 95% CI 2.4 to 5.1, p=0.254

    Jalbert J, 2015

    All asymptomatic participants (n=11,839)

    • Unadjusted risk of mortality after periprocedural period=27.7% (95% CI 26.4 to 28.9)

    • Unadjusted risk of stroke or TIA after periprocedural period=7.2% (95% CI 6.5 to 7.9)

    National Coverage Determination Indication of asymptomatic carotid stenosis 80% or above (n=9,851)

    • Unadjusted risk of mortality after periprocedural period=29.4% (95% CI 27.9 to 30.9)

    • Unadjusted risk of stroke or TIA after periprocedural period=7.3% (95% CI 6.5 to 8.2)

    All asymptomatic participants (n=11,839)

    • Unadjusted risk of mortality during periprocedural period=1.0% (95% CI (0.9 to 1.2)

    • Unadjusted risk of stroke or TIA during periprocedural period=2.3% (95% CI 2.1 to 2.6)

    • Unadjusted risk of myocardial infarction during periprocedural period=2.2% (95% 2.0 to 2.5)

    National Coverage Determination Indication of asymptomatic carotid stenosis 80% or above (n=9,851)

    • Unadjusted risk of mortality during periprocedural period=1.2% (95% CI 0.9 to 1.4)

    • Unadjusted risk of stroke or TIA during periprocedural period=2.3% (95% CI 2.0 to 2.6)

    • Unadjusted risk of myocardial infarction during periprocedural period=2.4% (95% 2.1 to 2.7)

    Procedure technique

    Most studies did not describe the CAS procedure in detail. There are different stents and different cerebral protection devices available, and table 5 includes a number of studies that investigate different aspects of the technique. These include proximal balloon occlusion compared with distal filter protection, closed-cell stents compared with open-cell stents, and dual-layer stents compared with conventional stents.

    Efficacy

    Non-procedural stroke

    Stroke during follow up beyond 30 days was reported as an outcome in 3 studies. In the network meta-analysis of 14,310 people, there was a statistically significant lower odds of a minor stroke occurring up to 5 years after the procedure when CAS was compared with BMT (OR 0.51, 95% CI 0.26 to 0.98; n=11,693, 13 trials). The odds of an ipsilateral stroke were also reduced (OR 0.37, 95% 0.22 to 0.62; Gasior 2023). In the randomised controlled trial of 3,625 people comparing CAS with CEA, the rate of non-procedural stroke during follow up was the same in both groups (5% [91/1,748 and 79/1,767]; RR 1.16, 95% CI 0.86 to 1.57, p=0.33). The rate of fatal non-procedural stroke (mRS score 6) was 1% in both groups (16/1,748 for CAS and 20/1,767 for CEA). The rate of disabling non-procedural stroke (mRS 3 to 5) was 1.6% (28/1,748) for those who had CAS and 1.4% (25/1,767) for those who had CEA. For non-disabling non-procedural stroke (mRS 0 to 2) the rate was 2.7% (47/1,748) in the CAS group and 1.9% (34/1,767) in the CEA group (p value not reported; Halliday 2021).

    Mortality

    Mortality during follow up beyond 30 days was reported as an outcome in 3 studies. In the network meta-analysis of 14,310 people, comparing CAS, CEA and BMT, no treatment was statistically significantly superior in reducing the odds of mortality from 30 days to 5 years (n=11,101; 9 trials; Gasior 2023). In the randomised controlled trial of 513 people, 5-year cumulative all-cause mortality was 9% in those who had CAS plus BMT (95% CI 6% to 15%), 8% in those who had CEA plus BMT (95% CI 5% to 13%) and 8% in those who had BMT alone (95% CI 4% to 15%; Reiff 2022). In the cohort study of 22,516 people (11,839 asymptomatic), the risk of mortality after the periprocedural period was 28% (95% CI 26% to 29%; Jalbert 2015).

    Composite outcomes of stroke, death and myocardial infarction

    Composite outcomes were reported in 3 studies, all of which showed similar rates for CAS and CEA within the studies. In the systematic review of 7,117 people, long-term outcomes of any perioperative stroke, death or myocardial infarction and long-term stroke, death or myocardial infarction was 10% for CAS (309/3,141) and 11% for CEA (264/2,423), OR 1.18 (95% CI 0.94 to 1.48, p=0.14, I2=0%, 4 studies; Wang 2022). In the systematic review of 3,378 people with asymptomatic carotid artery stenosis, the rate of death or any stroke between randomisation and 30 days after treatment or ipsilateral stroke until the end of follow up was 45 per 1,000 in those who had CAS and 36 per 1,000 in those who had CEA (OR 1.27, 95% CI 0.87 to 1.84, p=0.22, I2=0%, n=3,315, 6 studies, moderate-certainty evidence).

    In the randomised controlled trial of 513 people, the cumulative incidence of any stroke or death from any cause within 30 days or any ipsilateral stroke within 5 years was 4% for CAS plus BMT (95% CI 2 to 9), 3% for CEA plus BMT (95% CI 1 to 6) and 3% for BMT alone (95% CI 1 to 9), p=0.62 (Reiff 2022).

    Restenosis

    Restenosis was reported as an outcome in 2 studies. In the systematic review by Müller et al. (2020), severe restenosis (70% or more) during follow up in people with asymptomatic or symptomatic carotid artery stenosis was 67 per 1,000 (95% CI 43 to 103) for CAS and 56 per 1,000 for CEA (OR 1.21, 95% CI 0.76 to 1.93; n=5,744, 9 studies, p=0.42, I2=61%, low-certainty evidence). The rates of moderate or severe restenosis (50% or more) were 30% (309/1,019) for CAS and 21% (231/1,096) for CEA (OR 2.00, 95% CI 1.12 to 3.6, n=2,115, 4 studies, p=0.02, I2=44%). In the randomised controlled trial of 513 people by Reiff et al. (2022), restenosis of at least 70% within 5 years was 10% in the group who had CAS (95% CI 7 to 16) and 3% in the group who had CEA (95% CI 2 to 7), HR=3.2 (95% CI 1.3 to 8.0, p=0.014).

    Safety

    Stroke within 30 days of procedure

    Periprocedural stroke was reported as an outcome in 7 studies. Several studies showed a statistically significant increased risk in periprocedural non-disabling stroke associated with CAS when compared with CEA. In the network meta-analysis by Gasior et al. (2023), CEA reduced the odds of all stroke events within 30 days of the procedure compared with CAS (OR 1.6, 95% CI 1.1 to 2.2). It specifically reduced the odds of a minor perioperative stroke (OR 1.7, 95% CI 1.1 to 2.6) but did not affect the odds of a major stroke event (OR 0.89, 95% CI 0.3 to 2.3).

    In the systematic review by Wang et al. (2022), the rate of any stroke within 30 days was 3% (112/3,735) for CAS and 2% (58/3,010) for CEA (OR 1.62, 95% CI 1.17 to 2.25, p=0.004, I2=0%; 5 studies). For non-disabling stroke, the rate was 3% (88/3,562) for CAS and 1% (40/2,833) for CEA (OR 1.81, 95% CI 1.23 to 2.65, p=0.003, I2=0%; 4 studies).

    In the randomised controlled trial by Halliday et al. (2021), the rate of any stroke within 30 days was 4% (61/1,705) for CAS and 2% (41/1,736) for CEA, p=0.06. The rate of fatal stroke was less than 1% in both groups (7/1,705 and 5/1,736, p=0.77). The rate of non-disabling stroke (mRS 0 to 2) was 3% (48/1,705) for CAS and 2% (29/1,736) for CEA, p=0.03. In the randomised controlled trial by Reiff et al. (2022), there was a similar rate of stroke within 30 days for CAS (3% [5/197]) and CEA (3% [5/203]).

    In the retrospective cohort study by Cole et al. (2020), the rate of periprocedural stroke was similar for CAS (0.2%) and CEA (0.3%), p=0.113. In the prospective cohort study by Yang et al. (2021), the rates of stroke within 1 month of the procedure were also similar for CAS (4% [36/887]) and CEA (3% [25/758]), p=0.416. In the retrospective cohort study by Jalbert et al. (2015), the risk of stroke or TIA within the periprocedural period was 2% (95% CI 2 to 3).

    Mortality within 30 days of procedure

    Perioperative mortality was reported as an outcome in 6 studies. In the network meta-analysis by Gasior et al. (2023), neither procedure was favoured to reduce the odds of 30-day mortality (6 trials, n=6,855).

    In the randomised controlled trial by Reiff et al. (2022), there were no deaths within 30 days of the procedure.

    In the retrospective cohort study by Cole et al. (2020), there was no difference between CAS and CEA regarding in-hospital mortality. In the retrospective dataset by Gaba et al. (2021), 7-day in-hospital mortality was 1% (18/2,470) for CAS (95% CI 0.4 to 1) and less than 1% (21/14,102) for CEA (95% CI 0.1 to 0.2). In the prospective cohort study by Yang et al. (2021), mortality within 1 month of the procedure was 1% for both CAS (6/887) and CEA (4/758), p=0.699. In the retrospective cohort study by Jalbert et al. (2015), the risk of mortality after CAS during the periprocedural period was also 1% (95% CI 0.9 to 1.2).

    Perioperative myocardial infarction

    Perioperative myocardial infarction was reported as an outcome in 5 studies.

    In the systematic review by Müller et al. (2020), the rate of myocardial infarction between randomisation and 30 days after treatment was 1% for CAS (12/1,844) and 2% for CEA (16/1,102), OR 0.53 (95% CI 0.24 to 1.15, n=2,601, 6 studies, p=0.11, I2=0%).

    In the randomised controlled trial by Halliday et al. (2021), the rates of myocardial infarction within 30 days of the procedure were less than 1% for CAS (5/1,705) and 1% for CEA (12/1,736), p=0.15.

    In the retrospective cohort study by Cole et al. (2020), the rate of periprocedural myocardial infarction was 0.8% for CAS and 1.2% for CEA (OR 1.58, 95% CI 1.27 to 1.97). In the prospective cohort study by Yang et al. (2021), myocardial infarction within 1 month of the procedure was 0.5% for CAS (4/887) and 1% for CEA (10/758), p=0.056. In the retrospective cohort study by Jalbert et al. (2015), the risk of myocardial infarction after CAS during the periprocedural period was 2% (95% CI 2.0 to 2.5).

    Composite outcome of periprocedural death, stroke and myocardial infarction

    A composite outcome for periprocedural death, stroke and myocardial infarction was reported by 4 studies.

    In the systematic review by Wang et al. (2022), the 30-day outcome for stroke, death and myocardial infarction was 4% (137/3,920) for CAS and 3% (102/3,198) for CEA (OR 1.13, 95% CI 0.87 to 1.47, p=0.37, I2=0%; 7 studies). The 30-day outcome for disabling stroke and death was 1% in each group (24/3,494 for CAS and 22/2,765 for CEA), OR 0.91 (95% CI 0.50 to 1.65, p=0.76, I2=0%; 3 studies). In the systematic review by Müller et al. (2020), the rate of death or any stroke between randomisation and 30 days after treatment was 25 per 1,000 for CAS and 14 per 1,000 for CEA (OR 1.72, 95% CI 1.00 to 2.97, p=0.05, I2=0%, n=3,378; 7 studies; moderate-certainty evidence). The rate of death or major or disabling stroke between randomisation and 30 days after treatment was 5 per 1,000 for CAS and 3 per 1,000 for CEA (OR 1.54, 95% CI 0.39 to 6.11, p=0.54, I2=0%, n=2,601; 2 studies; moderate-certainty evidence). The rate of death or any stroke or myocardial infarction between randomisation and 30 days after treatment was 32 per 1,000 for CAS and 28 per 1,000 for CEA (OR 1.16, 95% CI 0.73 to 1.85, p=0.8, I2=0%, n=2,978; 6 studies; moderate-certainty evidence).

    In the randomised controlled trial by Halliday et al. (2021), the rate of death or disabling stroke within 30 days of the procedure was 1% in both groups (p=0.77). The rate of death, myocardial infarction or any stroke was 4% (67/1,705) for CAS and 3% (55/1,736) for CEA (p=0.26).

    In the prospective cohort study by Yang et al. (2021), stroke, myocardial infarction or death within 1 month of the procedure was 5% for both CAS (42/887) and CEA (37/758), p=0.89.

    Cranial nerve injury

    Cranial nerve injury was reported as an outcome in 3 studies, all of which showed a reduced risk associated with CAS compared with CEA.

    In the network meta-analysis by Gasior et al. (2023), CAS reduced the odds of cranial nerve injury compared with CEA (OR 0.07, 95% CI 0.01 to 0.42). In the systematic review by Müller et al. (2020), the rate of cranial nerve injury was less than 1% (2/1,823) after CAS and 3% (36/1,092) after CEA (OR 0.09, 95% CI 0.03 to 0.27, p<0.0001, n=2,915, I2=0%; 5 studies).

    In the randomised controlled trial by Halliday et al. (2021), the rate of cranial nerve palsy within 30 days of the procedure was 0% (0/1,653) after CAS and 5% (96/1,788) after CEA (p value not reported).

    Access site haematoma

    Access site haematoma was reported as an outcome in 1 study. In the systematic review by Müller et al. (2020), the rate of access site haematoma (needing surgery, blood transfusion or prolonged hospital stay, where severity was determined) was less than 1% (3/1,726) after CAS and 2% (15/993) after CEA (OR 0.14, 95% CI 0.02 to 0.9, p=0.04, I2=35%, n=2,719; 3 studies).

    Intracranial haemorrhage

    Intracranial haemorrhage was reported as an outcome in 1 study. In the randomised controlled trial by Reiff et al. (2022), there was 1 report of intracranial haemorrhage in the CAS plus BMT group that happened 431 days after randomisation.

    Hyperperfusion syndrome

    Hyperperfusion syndrome was reported as an outcome in 1 study. In the randomised controlled trial by Reiff et al. (2022), there were 3 reports of hyperperfusion syndrome in the CAS plus BMT group, 1 with cerebral infarction. None of them led to permanent disability.

    Other

    A number of case reports describing adverse events associated with CAS are listed among the studies in table 5. These include acute hemifacial ischaemia, postoperative dislocation of stent, flushing, dyspnoea and palpitations, iliac artery injury, dislodged stent delivery sheath, stent shortening, retinal artery occlusion, massive non-aneurysmal subarachnoid haemorrhage, carotid free-floating thrombus, pseudoaneurysm, and late thrombosis.

    Anecdotal and theoretical adverse events

    Expert advice was sought from consultants who have been nominated or ratified by their professional society or royal college. They were asked if they knew of any other adverse events for this procedure that they had heard about (anecdotal), which were not reported in the literature. They were also asked if they thought there were other adverse events that might possibly occur, even if they had never happened (theoretical).

    They listed the following adverse events that were not categorised as anecdotal or theoretical:

    • access complications (bleeding, limb ischaemia)

    • coronary complications, including bradycardia

    • vascular injury

    • occlusion of the carotid artery

    • in-stent stenosis

    • reperfusion bleed

    • contrast reaction

    • hypotension caused by vasovagal stimulation

    • vasospasm

    • protection device complications

    • increased risk of new white matter lesions on brain DW MRI, of uncertain clinical consequence

    • issues with stent deployment, filter wire insertion and removal.

    Six professional expert questionnaires for this procedure were submitted. Find full details of what the professional experts said about the procedure in the specialist advice questionnaires for this procedure.

    Validity and generalisability

    • There is a large body of evidence, including data from the UK.

    • There were 2 recent randomised controlled trials identified that focused on people with asymptomatic carotid artery stenosis.

    • One randomised controlled trial that was designed to compare CAS and CEA against BMT was stopped prematurely because of low recruitment and funding restrictions. This led to a much smaller sample size than planned, which limits the validity of the results. The authors noted that the study design was complex and that structural reasons may have contributed to the low recruitment.

    • The patient population of the randomised controlled trials is likely to differ from real-life clinical practice and BMT is likely to be implemented more consistently. The expertise of the centre and strict patient selection criteria may lead to lower complication rates than would be expected outside a trial.

    • Mortality was considerably lower in the randomised controlled trial by Reiff et al. (2022) than the cohort study by Jalbert et al. (2015). The authors noted that some of this could be attributed to sufficient implementation of BMT targets, a lower proportion of high-risk patients and the selection of participants with a life expectancy of at least 5 years.

    • The systematic reviews include follow-up data up to 10 years.

    • There is a lot of variation in procedure technique for CAS, including different kinds of stent and different embolic protection devices.

    • There may be variations in the definition of asymptomatic between studies.

    • BMT has improved over time, which has reduced the risk of stroke in this population. Older trials are therefore less generalisable than those that recruited people more recently. Some of the trials included in the systematic reviews enrolled participants in the early 2000s. It is likely that technical developments in CEA and CAS have also led to improvements, lowering the periprocedural risks.

    • Trials included in the systematic reviews differed in assessment and definitions of myocardial infarction. Definition of stroke may also differ between studies.

    • Of the 9 studies included in the key evidence, 5 declared no conflicts of interest. Cole et al. (2020) declared that 1 author was a consultant for several companies and had stock in GT Medical Technologies, none of which were related to the topic. Wang et al. (2022) declared that 1 author was a consultant for Medtronic, MicroVention Inc, and Penumbra Inc. Jalbert et al. (2015) declared that 1 author was a non-compensated adviser to Abbott Vascular, Boston Scientific, Cordis Corporation, Covidien Vascular, and Medtronic Vascular; and was a board member at VIVA Physicians, a not-for-profit education and research consortium; 1 author was a paid consultant to Optum Insight Epidemiology and World Health Information Science Consultants, LLC; 1 author was being supported by the pharmacoepidemiology program at the Harvard School of Public Health, funded by Pfizer and Asisa; 1 author was being supported by a midcareer development award grant from the Agency for Healthcare Research and Quality, US Department of Health and Human

    Services and was also receiving research support from Johnson & Johnson and personal income for consulting from Sanofi; and 1 author had made available online a detailed listing of financial disclosures (http://www.dcri.duke.edu/about-us/conflict-of-interest/).

    • Ongoing studies

      • NCT04900844: Safety and Efficacy of the CGuard™ Carotid Stent System in Carotid Artery Stenting (C-Guardians); single-arm pivotal study; US; n=315; estimated study completion date Oct 2025

      • NCT03121209: Carotid Revascularization and Medical Management for Asymptomatic Carotid Stenosis Trial - Hemodynamics (CREST-H); prospective observational study; US & Canada; n=500; estimated study completion date March 2027

      • NCT02089217: Carotid Revascularization and Medical Management for Asymptomatic Carotid Stenosis Trial (CREST-2); 2 parallel multi-center randomized controlled trials; US, Canada, Australia, Israel & Spain; n=2,480; estimated study completion date February 2026

      • NCT04271033: MicroNet-covered Stent System for Stroke Prevention in All Comer Carotid Revascularization (PARADIGM-EXT); single-arm registry; Poland; n=550; estimated study completion date June 2026