Head To Head

Is there a role for revascularisation in asymptomatic carotid stenosis? No

BMJ 2010; 341 doi: http://dx.doi.org/10.1136/bmj.c4900 (Published 15 September 2010) Cite this as: BMJ 2010;341:c4900
  1. J David Spence, director
  1. 1Stroke Prevention andAtherosclerosis Research Centre, Robarts Research Institute, University of Western Ontario, 1400 Western Road, London, ON, Canada N6G 2V2
  1. dspence{at}robarts.ca

    Marco Roffi (doi:10.1136/bmj.c4898) thinks revascularisation should be the strategy of choice for patients with asymptomatic carotid stenosis but J David Spence argues that all but a tiny minority would do better with medical management

    Carotid endarterectomy and stenting for asymptomatic stenosis are based on historical risks that no longer pertain. With more intensive medical management (including lifestyle modification), the risk of stroke or death is now lower than with intervention; patients with asymptomatic stenosis are now more likely to be harmed than helped. In most cases, carotid stenting or endarterectomy for asymptomatic stenosis is inappropriate and unwarranted.

    In the Asymptomatic Carotid Artery Surgery and Asymptomatic Carotid Surgery randomised trials, the extrapolated five year risk of stroke or death was about 10% with medical management versus about 5% with surgery.1 2 Those outdated figures are now being compared with recent results of the Carotid Revascularisation Endarterectomy versus Stenting Trial (CREST),3 to justify routine endarterectomy or stenting. In CREST, patients with carotid stenosis were randomised to endarterectomy or stenting; 47% of participants had asymptomatic stenosis. The periprocedural risk of stroke was 2.5% for stenting and 1.4% for endarterectomy (with four year risks of 4.5% and 2.7%, respectively).3 This is being widely interpreted to mean that endarterectomy and stenting could go full steam ahead in patients with asymptomatic carotid stenosis. However, the risk of stroke or death in these patients is now lower (<1% a year) with medical management than with intervention. These findings were consistent in a recent meta-analysis, a prospective population based study in Oxfordshire, and a prospective study in patients attending a Canadian vascular prevention clinic.4 5 6

    Few benefit from surgery

    The CREST result does not warrant uncritical intervention in asymptomatic stenosis.7 Patients with microemboli on transcranial Doppler may benefit from intervention,8 but these account for less than 5% of patients with asymptomatic stenosis.6 Imaging markers of vulnerable plaque or reduced cerebral blood flow reactivity may slightly increase the proportion of patients with asymptomatic stenosis who benefit from surgical intervention.9 In the absence of microemboli or other indicators of high risk, routine carotid endarterectomy or stenting for asymptomatic stenosis can no longer be justified.

    It is now clear that stenting carries a higher risk of stroke or death than does endarterectomy,10 particularly in elderly people.3 Carotid stenting and endarterectomy are not about increasing blood flow to the brain; they are about preventing embolisation of plaque fragments (atheromatous debris with cholesterol crystals) and platelet aggregates. This is why carotid stenting and endarterectomy should not be called revascularisation; a better collective term would be intervention.

    False beliefs

    Several misconceptions underlie inappropriate endarterectomy and stenting for asymptomatic stenosis.11 These include the unwarranted belief that high grade stenosis justifies intervention, the intuitive beliefs that restoring blood flow to the brain is a good thing and that stenting is less invasive than surgery, and hubris (“my complication rates are lower than those of others”). The chief villain is the outdated belief that endarterectomy or stenting will reduce the risk of stroke. The underlying assumption on which that is based is that risk of stroke or death is lower with endarterectomy or stenting than with medical management. Although that may have been true in the past, it is no longer the case.

    A moratorium on intervention for asymptomatic stenosis outside clinical trials that include a medical arm is certainly justified.12 13 However, a complete moratorium on stenting and endarterectomy, though justified by the facts, is unpalatable to many and may not benefit all patients with asymptomatic stenosis. There are some approaches to identifying the small proportion who may benefit. We reported in 2005 that the 10% of patients with asymptomatic stenosis who had microemboli on transcranial Doppler scanning had a high enough risk to warrant intervention: patients with microemboli had a two year risk of stroke of 15.6%, versus 1.7% without emboli.8 By 2010, however, with more intensive drug therapy guided by imaging of carotid plaques,14 the proportion with microemboli was down to 3.7% and the two year risk of stroke was down to 1%. This is lower than the periprocedural stroke risk in CREST (2.5% for stenting and 1.4% for endarterectomy).

    Nearly all patients (90-95%) with asymptomatic stenosis 6 8 are better off with medical management and lifestyle modification than with endarterectomy or stenting as their risk of stroke or death will be lower.4 5 The very small proportion who may benefit from intervention can be identified by the presence of microemboli on transcranial Doppler imaging .6 8 Markus and colleagues recently confirmed in the Asymptomatic Carotid Emboli Study,15 an international prospective multicentre study, that microemboli detected by transcranial Doppler were associated with higher risk; their two year hazard ratio for ipsilateral stroke was 5·57 (95% confidence interval 1·61 to 19·32). The hazard ratio for the risk of ipsilateral stroke for those who had microemboli in the preceding six months was 6.37 (1.59 to 25.57).

    Routine carotid endarterectomy or stenting for asymptomatic carotid stenosis, without regard to the patient’s risk of stroke with medical managment, should now be regarded as unwarranted. In most patients with asymptomatic carotid stenosis, endarterectomy or stenting is more likely to harm the patient than to help.


    Cite this as: BMJ 2010;341:c4900


    • Competing interests: The author has completed the unified competing interest form at www.icmje.org/coi_disclosure.pdf (available on request from him) and declares no support from any organisation for the submitted work; he has received honorariums for lectures or consulting from Pfizer, Merck, Novartis, Boehringer-Ingelheim, and AstraZeneca; and his research unit has received grants from the Canadian Institutes for Health Research, the Heart and Stroke Foundation of Ontario, and the National Institutes of Health, and has conducted contract research for AGA, NMT medical, Servier, Bristol-Myers Squibb, Sanofi/Aventis, Pfizer, Wyeth, Novartis, SmithKline Beecham, Takeda, ActivBiotics, and Pan American Laboratories.Provenance and peer review: Commissioned; not externally peer reviewed.