By Anne Abbott
The history of the evidence-base regarding stroke prevention associated with carotid stenosis is straightforward:
- Trials of patients with “average” carotid endarterectomy-risk symptomatic or asymptomatic carotid stenosis randomised 20–30 years ago showed that endarterectomy plus medical treatment (compared to medical treatment alone) could reduce overall average annual stroke rate by 3% and 0.5–1%, respectively, if patient life expectancy was at least about three years.
- Results from routine practice indicated that this benefit was often not achieved.
- Randomised trials of carotid endarterectomy versus stenting and registries have shown that stenting causes about twice as many strokes or deaths as surgery. This excess risk is not compensated for by peri-procedural myocardial infarction risk, at least for patients with symptomatic carotid stenosis who have been more thoroughly studied.
- Randomised comparisons of carotid endarterectomy vs. stenting vs. medical treatment alone have not been performed for patients considered “high” endarterectomy-risk due to vascular anatomy or major medical comorbidities (conditions which reduce life expectancy).
- The risk of stroke associated with asymptomatic carotid stenosis has fallen at least 60–80% with medical treatment alone since randomised trials of carotid endarterectomy vs. medical treatment indicating endarterectomy is no longer likely to benefit patients with asymptomatic carotid stenosis, and new trials of carotid endarterectomy vs. medical treatment alone for patients with symptomatic stenosis are needed.
Why do international guideline recommendations differ given that the evidence-base findings are clear? In a search for recommendations regarding carotid endarterectomy and/or stenting for asymptomatic and/or symptomatic stenosis and published January 2008–August 2014, my colleagues and I identified 33 relevant guidelines from 20 different regions written by 28 different groups in five languages (systematic review in preparation). There were notable differences in guideline recommendations. For instance, of 27 guidelines giving procedural recommendations for average carotid endarterectomy-risk asymptomatic carotid stenosis, in seven (26%) cases recommendations for endarterectomy were strong compared to 16 (59%) cases where endarterectomy recommendations were weaker. A total of 85% of guidelines gave at least some support for endarterectomy for average endarterectomy-risk asymptomatic carotid stenosis. Strong recommendations for carotid artery stenting for average endarterectomy-risk asymptomatic carotid stenosis patients were few (two or 7%), contrasting with 11 (41%) giving weaker stenting recommendations. A total 48% of guidelines gave at least some support for stenting for average endarterectomy-risk asymptomatic carotid stenosis, in contrast to seven (26%) which specifically and consistently recommended stenting not be used. A large minority of guidelines, 12 (44%) and 10 (37%), respectively, gave at least some support for stenting in asymptomatic carotid stenosis patients considered high-endarterectomy risk due to vascular anatomy or co-morbidities. Medical treatment alone for average endarterectomy-risk asymptomatic stenosis was recommended in only one guideline, while five (18%) gave weak recommendations for it in high endarterectomy-risk asymptomatic carotid stenosis patients due to vascular anatomy or medical comorbidities.
There were also notable differences in guideline recommendations for symptomatic carotid stenosis. For instance, of 32 guidelines giving procedural recommendations for “average endarterectomy-risk” symptomatic stenosis patients, 27 (84%) gave strong recommendations for endarterectomy, whereas three (9%) gave weaker endarterectomy recommendations. In total 93% of guidelines gave at least some support for endarterectomy. Strong recommendations for carotid artery stenting for average endarterectomy-risk symptomatic stenosis patients (six or 19%) were outnumbered by those which gave weaker stenting recommendations (nine or 28%). In total, 47% gave at least some support for stenting in contrast to six (19%) which specifically and consistently recommended stenting not be used. The majority of guidelines, 27 (84%) and 19 (60%), respectively, gave at least some support for stenting in patients with symptomatic stenosis considered high-endarterectomy risk due to vascular anatomy or medical comorbidities. Medical treatment alone for average endarterectomy-risk symptomatic stenosis was recommended by no guidelines, while three (9%) gave weak recommendations for medical treatment alone for high endarterectomy-risk symptomatic stenosis due to vascular anatomy or medical comorbidities.
Despite differences, of more concern, is what these guidelines have in common. They are still based on 20–30 year randomised comparisons of medical treatment alone versus additional endarterectomy in patients with average endarterectomy-risk asymptomatic or symptomatic carotid stenosis. This means they do not take into account major advances in stroke prevention efficacy with medical treatment since then and they no longer reflect the whole evidence-base. In addition, procedures are being recommended in patients with major medical comorbidities (and reduced life expectancy) despite evidence they are unlikely to benefit. Further, many of the most recent guidelines have adopted new recommendations for carotid artery stenting for patients with average endarterectomy-risk asymptomatic and symptomatic stenosis, despite clear evidence stenting carries a higher risk of stroke/death or periprocedural death/myocardial infarction or stroke than endarterectomy. These guideline limitations are supportive of a strong and inappropriate bias towards use of carotid procedures in routine practice. While some of variability in guideline recommendations is likely due to variation in publication date, much is evidently due to bias with respect to what evidence is considered and how this evidence is utilised.
Anne L Abbott is a neurologist, School of Public Health and Preventive Medicine, Monash University, Melbourne, Australia. Abbott’s work is supported by a grant from the BUPA Health Foundation of Australia
Acknowledgements
My collaborators in a systematic review of international carotid stenosis management guidelines (alphabetical order): Marije Bosch (Australia), Longxing Cao (China), Martine Dennekamp (Australia), Larry J. Diaz (USA), Henning Eckstein (Germany), Qiang Fu (China), Jonathan Golledge (Australia), Stavros K Kakkos (Greece), Thomas W Leung (China), Miguel Montero-Baker (USA), Kosmas I Paraskevas (UK), Sabine Pircher (Australia), Peter Ringleb (Germany), Tissa Wijeratne (Australia).
