Poor outcomes after endovascular treatment of symptomatic carotid stenosis: time for a moratorium

Peter M Rothwell a

The long-term outcomes of the CAVATAS trial, reported in the October issue of The Lancet Neurology,1, 2 add substantially to the published data from trials on the durability of endovascular treatment versus endarterectomy for symptomatic carotid stenosis.

Brown and colleagues should be congratulated for continuing to follow up patients for 10 years (median follow-up was 5 years). Such follow-up is vital to ascertain the effectiveness of procedures intended to reduce the long-term risk of stroke. Comparable data on outcomes beyond the immediate post-procedural period come from the SAPPHIRE,3 EVA-3S,4 and SPACE5 trials, which had median follow-up periods of 3, 2, and 3·5 years, respectively, although the SAPPHIRE trial had substantial loss to follow-up.

Endovascular treatment in CAVATAS resulted in a three-fold greater incidence of severe restenosis during follow-up (hazard ratio 3·14, 95%CI 1·87—5·26; p<0·0001), with a higher risk of recurrent transient ischaemic attack and stroke in patients with restenosis and a non-significant higher long-term risk of non-perioperative stroke in the endovascular treatment arm overall (1·66, 0·99—2·80). The worse outcomes in the endovascular group were partly accounted for by the patients treated with angioplasty alone, and the results might not be entirely generalisable to current clinical practice. However, the SPACE trial, in which stenting was mandatory, also reported a higher incidence of 70% or more restenosis after endovascular treatment (11·1% vs 4·6%, p=0·0009).5 Ongoing trials of carotid stenting must therefore follow-up patients in the long term to ascertain reliably the clinical consequences of these high rates of restenosis. In the large randomised trials of endarterectomy versus medical treatment alone for symptomatic carotid stenosis, nearly 20 000 patient—years of follow-up were available on the 3248 patients who underwent endarterectomy; therefore, the long-term durability of endarterectomy was reliably established.6, 7 CAVATAS, SAPPHIRE, EVA-3S, and SPACE together provide about 3500 patient—years of follow-up after endovascular treatment.

More worrying, however, is the increasingly consistent evidence of substantially higher procedural risks of stroke after endovascular treatment than after endarterectomy in patients with symptomatic carotid stenosis. The recently presented International Carotid Stenting Study (ICSS),8 the largest trial of endarterectomy versus endovascular treatment to date, reported a nearly two-fold higher procedural risk of stroke in the endovascular treatment group—which is consistent with the results of meta-analysis of all previous randomised trials—and a five-fold excess of new infarcts seen on brain imaging. The initial result of the CAVATAS trial was an outlier in the previous meta-analyses of procedural risk; CAVATAS initially reported similar risks for endovascular treatment and endarterectomy, albeit with wide confidence intervals. However, the latest report includes data on all strokes within 30 days of the procedures,1 rather than just data on those associated with neurological impairment that lasted for longer than 7 days. These additional strokes (1 in the endarterectomy arm vs 8 in the endovascular arm; Fisher's exact test p=0·02) bring the results of CAVATAS more in line with other trials. The inclusion of these so-called minor strokes in the procedural risk is essential for several reasons. First, although the neurological impairment resolves quickly, the associated risks of cognitive decline, fatigue, depression, and epilepsy do not. Second, the most reliable data we have on risk of stroke with medical treatment alone,6 and this is what the risks of interventional treatments must be compared with, included all strokes with symptoms that lasted longer than 24 h. Third, most of the strokes that are prevented by interventional treatment for symptomatic carotid stenosis are non-disabling.6 Endarterectomy for severe symptomatic carotid stenosis (and hence endovascular treatment, if it were equivalent) results in only a small absolute reduction in disabling stroke compared with medical treatment alone and no significant reduction in fatal stroke.6

Even without the inclusion of the results of ICSS and the other trials that reported only the periprocedural risks, a meta-analysis of all the available data on long-term outcome in randomised trials of endovascular treatment versus endarterectomy for symptomatic carotid stenosis now shows a significantly worse outcome after endovascular treatment (figure). Carotid stenting should still be used in patients with inoperable stenosis or if patients strongly prefer endovascular treatment, although the recent finding of the GALA (general anaesthetic versus local anaesthetic for carotid surgery) trial—that endarterectomy can be done at least as safely under local anaesthetic as it is under general anaesthetic—should influence patients' preferences.9 There is also a need for pooled analyses of trial data to ascertain whether or not there are subgroups of patients for which the procedural risk of stenting is not higher than that of endarterectomy. In view of the clear-cut differences in the procedural risk of endarterectomy between subgroups of patients10 and in the benefit of intervention versus medical treatment alone,11, 12 such subgroup-treatment effect interactions are possible. However, pending the full report of the results of ICSS and the presentation of the results of the equivalent trial in the USA (CREST [Carotid Revascularization Endarterectomy versus Stent Trial]),13 the routine use of stenting in patients with recent symptoms of carotid stenosis who are suitable for endarterectomy can no longer be justified. In particular, the use of vague and non-evidence-based categorisations, such as “high risk for surgery”, which have been systematically misused to justify the uncontrolled roll-out of carotid stenting in many centres, must stop. Most patients who have undergone stenting for symptomatic carotid stenosis outside of randomised controlled trials would have met the eligibility criteria for CAVATAS, EVA-3S, SPACE, and ICSS, and must therefore have faced a greater procedural risk of stroke, a greater risk of severe restenosis, and a worse long-term outcome than if they had had endarterectomy.

Figure Full-size image (14K)

Meta-analysis of data on long-term risk of any stroke after randomisation and non-stroke periprocedural death from all published randomised trials of endovascular treatment versus endarterectomy for symptomatic carotid stenosis

The SAPPHIRE trial composite outcome also included periprocedural myocardial infarction and death and only ipsilateral ischaemic stroke at more than 30 days after the procedure.

I was a researcher on the European Carotid Surgery Trial, coordinated the Carotid Endarterectomy trialists' collaboration, am on the steering committees of the GALA trial and the UK national carotid endarterectomy audit, the data monitoring committee of the SPACE-2 Trial, the endpoint adjudication committee of the ACST-2 trial, run a clinical service that includes carotid stenting and endarterectomy, and am a lead-investigator on the Vertebral Artery Ischaemia Stenting trial.

References

1 Ederle J, Bonati LH, Dobson J, et alfor the CAVATAS investigators. Endovascular treatment with angioplasty or stenting versus endarterectomy in patients with carotid artery stenosis in the Carotid And Verterbral Artery Transluminal Angioplasty Study (CAVATAS): long-term follow-up of a randomised trial. Lancet Neurol 200910.1016/S1474-4422(09)70228-5. published online August 29. PubMed

2 Bonati LH, Ederle J, Dominick JH, et alfor the CAVATAS Investigators. Long-term risk of carotid restenosis in patients randomly assigned to endovascular treatment or endarterectomy in the Carotid And Verterbral Artery Transluminal Angioplasty Study (CAVATAS): long-term follow-up of a randomised trial. Lancet Neurol 200910.1016/S1474-4422(09)70227-3. published online August 29. PubMed

3 Gurm HS, Yadav JS, Fayad P, et alfor the SAPPHIRE Investigators. Long-term results of carotid stenting versus endarterectomy in high-risk patients. N Engl J Med 2008; 358: 1572-1579. CrossRef | PubMed

4 Mas JL, Trinquart L, Leys D, et alfor the EVA-3S investigators. Endarterectomy Versus Angioplasty in Patients with Symptomatic Severe Carotid Stenosis (EVA-3S) trial: results up to 4 years from a randomised, multicentre trial. Lancet Neurol 2008; 7: 885-892. Summary | Full Text | PDF(216KB) | CrossRef | PubMed

5 Eckstein HH, Ringleb P, Allenberg JR, et al. Results of the Stent-Protected Angioplasty versus Carotid Endarterectomy (SPACE) study to treat symptomatic stenoses at 2 years: a multinational, prospective, randomised trial. Lancet Neurol 2008; 7: 893-902. Summary | Full Text | PDF(197KB) | CrossRef | PubMed

6 Rothwell PM, Gutnikov SA, Eliasziw M, et alfor the Carotid Endarterectomy Trialists' Collaboration. pooled analysis of individual patient data from randomised controlled trials of endarterectomy for symptomatic carotid stenosis. Lancet 2003; 361: 107-116. Summary | Full Text | PDF(380KB) | CrossRef | PubMed

7 Cunningham E, Bond R, Mehta Z, Warlow CP, Rothwell PM. Long-term durability of carotid endarterectomy in the European Carotid Surgery Trial. Stroke 2002; 33: 2658-2663. CrossRef | PubMed

8 Brown MM, Ederle J, Bonati LH, et al. Safety results of the International Carotid Stenting Study: Early outcome of patients randomised between carotid stenting and endarterectomy for symptomatic carotid stenosis. European Stroke Conference 2009; May 27, 2009: Stockholm, Sweden.

9 Lewis SC, Warlow CP, Bodenham AR, et alfor the GALA Trial Collaborative Group. General anaesthesia versus local anaesthesia for carotid surgery (GALA): a multicentre, randomised controlled trial. Lancet 2009; 372: 2132-2142. Summary | Full Text | PDF(201KB) | CrossRef | PubMed

10 Bond R, Rerkasem K, Rothwell PM. A systematic review of the risks of carotid endarterectomy in relation to the clinical indication and the timing of surgery. Stroke 2003; 34: 2290-2301. CrossRef | PubMed

11 Rothwell PM, Eliasziw M, Gutnikov SA, Warlow CP, Barnett HJMfor the Carotid Endarterectomy Trialists Collaboration. Effect of endarterectomy for symptomatic carotid stenosis in relation to clinical subgroups and to the timing of surgery. Lancet 2004; 363: 915-924. Summary | Full Text | PDF(115KB) | CrossRef | PubMed

12 Rothwell PM, Mehta Z, Howard SC, Gutnikov SA, Warlow CP. From subgroups to individuals: general principles and the example of carotid endartectomy. Lancet 2005; 365: 256-265. Summary | Full Text | PDF(209KB) | CrossRef | PubMed

13 Hobson RW. Update on the Carotid Revascularization Endarterectomy versus Stent Trial (CREST) protocol. J Am Coll Surg 2002; 194: S9-S14. CrossRef | PubMed

a John Radcliffe Hospital, Oxford, UK

Article Options