Systematic review of registry data suggests stroke or death rate after carotid stenting “significantly higher” than after endarterectomy



The systematic review of administrative dataset registries, published in the January 2016 issue of the European Journal of Vascular and Endovascular Surgery (EJVES), has further concluded that stroke/death rates after carotid stenting often exceed accepted American Heart Association (AHA) thresholds. Based on this study, there was also no evidence of a sustained decline, over time, in the procedural risk after carotid stenting.

As reported in EJVES, K I Paraskevas, St George’s Vascular Institute, St George’s Healthcare NHS Trust, London, UK and colleagues EL Kalmykov and AR Naylor from the Department of Vascular Surgery, Leicester Royal Infirmary, Leicester, UK, write that randomised trials have reported higher stroke/death rates after carotid artery stenting vs. carotid endarterectomy.

“Despite this, the 2011 American Heart Association (AHA) guidelines expanded carotid artery stenting indications, partly because of the Carotid Revascularization Endarterectomy versus Stenting Trial (CREST), but also because of improving outcomes in industry sponsored carotid artery stenting registries,” the authors write.

This systematic review set out to compare stroke/death rates after carotid stenting or endarterectomy in contemporary dataset registries. The investigators also sought to examine whether published stroke/death rates after stenting fall within AHA thresholds, and to see if there had been a decline, over time, in procedural risk after stenting/endarterectomy.

The investigators systematically searched PubMed/Medline, Embase, and Cochrane databases according to the recommendations of the PRISMA statement from January 1, 2008 until February 23, 2015 for administrative dataset registries reporting outcomes after both endarterectomy and stenting.

“Twenty-one registries reported outcomes involving more than 1,500,000 procedures. Stroke/death after stenting was significantly higher than after endarterectomy in 11/21 registries (52%) involving “average risk for endarterctomy” asymptomatic patients and in 11/18 registries (61%) involving “average risk for endarterectomy” symptomatic patients. In another five registries, stenting was associated with higher stroke/death rates than endarterectomy for both symptomatic and asymptomatic patients, but formal statistical comparison was not reported. Carotid artery stenting was associated with stroke/death rates that exceeded risk thresholds recommended by the AHA in 9/21 registries (43%) involving “average risk for endarterectomy” asymptomatic patients and in 13/18 registries (72%) involving “average risk for endarterectomy” symptomatic patients. In 5/18 registries (28%), the procedural risk after carotid artery stenting in “average risk” symptomatic patients exceeded 10%,” the investigators write.

These findings led the researchers to conclude that “while carotid stenting has advanced significantly over the last decade, evidence suggests that most contemporary administrative dataset registries (predominantly sourced from the USA) still report procedural stroke/death rates following the procedure that are significantly higher (when compared with endarterectomy), particularly in ‘average risk for endarterectomy’ symptomatic patients.”

Most importantly, said the authors, almost three quarters of registries reported procedural risks after stenting that were well in excess of the 6% AHA/ASA recommended risk threshold, with 28% of registries publishing death/stroke rates in excess of 10% in “average risk for endarterectomy” symptomatic patients. They also wrote that this systematic review found no evidence that procedural risks after carotid artery stenting “in the real world” have diminished with time.

Naylor told Interventional News: “Just as with carotid endarterectomy, it is essential that anyone performing carotid artery stenting should be aware of their procedural risks. In asymptomatic patients (even if they are ’high risk for endarterectomy’), there is no justification for accepting procedural risks in excess of 3% (ie, the AHA recommended level). In symptomatic patients, the key issue now is to offer surgery (or stenting) as soon as possible after onset of symptoms. Delay does not benefit the patient. At present, most patients will probably benefit from surgery in the first seven to 14 days after onset of symptoms but this may change as carotid stenting technology improves. One important observation from this systematic review was the lack of any apparent clinical governance feedback in registries reporting excessively high procedural risks.”


The authors point out in EJVES that this systematic review does have limitations. “First, some registries had potential selection or reporting bias, while others conceded potential coding errors.” The authors noted that it would be expected that these errors/biases would be consistent across both carotid stenting and endarterectomy data.

Paraskevas and colleagues further write that protection devices during carotid artery stenting were not routinely used, or information about their usage was not routinely available.” Finally, it is possible that some registries may have sampled patients from overlapping years,” they write.