The Carotid Revascularization Endarterctomy versus Stenting Trial (CREST), published in 2010, generated a fair amount of controversy and was expected to inform clinical guidelines.
But a new study published in JAMA Neurology suggests clinicians have opted to perform carotid artery stenting (CAS) more often since the trial has been published, even in patient populations for which CREST demonstrated the benefit of carotid endarterectomy (CEA).
Lead researcher Fadar Oliver Otite, MD, ScM, with the department of neurology at the University of Miami, and colleagues studied 494,733 procedures for carotid revascularization in patients older than 70 from 2007 to 2014. The study population was 58.2 percent men and an average of 78.1 years old.
The proportion of patients who received CAS increased from 11.9 percent pre-CREST (2007-10) to 13.8 percent post-CREST (2011-14). In multivariable-adjusted models, the odds of receiving CAS increased by 13 percent in all patients older than 70 after the trial’s publication.
These findings contrast with Otite et al.’s expectations because CREST showed favorable results for CEA over CAS in patients older than 70. Stenting was slightly more beneficial in younger patients, with the entire CREST cohort showing a nonsignificant difference between the procedures in meeting the primary endpoint of any stroke, MI or death during the periprocedural period or ipsilateral stroke on follow-up.
A CREST subanalysis found women with symptomatic stenosis had a 7.5 percent chance of periprocedural stroke with CAS versus 2.7 percent following CEA. Despite this result, Otite and colleagues noted the odds of symptomatic women receiving CAS increased by 31 percent following the publication of the trial, with CAS use increasing at a rate of 5.6 percent per year across the entire study period.
“Some physicians view CREST simply as an indication of efficacy equivalence between CAS and CEA and give no strong consideration to the subtle differential treatment effects by age, sex, or possibly symptomatic status,” Otite and co-authors wrote. “It is also possible that many U.S. physicians still rely on evidence from the stent-favorable Stenting and Angioplasty With Protection in Patients at High Risk for Endarterectomy (SAPPHIRE) trial.”
In addition, the researchers said CAS technology has improved since CREST, which may give operators more confidence to opt for stenting procedures. They also pointed out changes in which physician subspecialties perform revascularization could affect utilization; cardiologists and interventionalists may prefer stenting, while neurologists and surgeons could be more likely to use CEA. However, the authors didn’t have the referral data necessary to test this theory.
“In CREST, distal emboli protection devices were utilized, but newer stents with improved deliverability and using proximal emboli protection devices that may allow for easier navigation of tortuous extracranial vessels and lower likelihood of distal embolization are now available,” Otite and colleagues wrote. “It is arguable that stroke risk associated with newer devices and stenting techniques may approximate the stroke risk associated with CEA and hence positively influence CAS utilization.”
Although the researchers found an overall rise in CAS procedures post-CREST, the rate of CAS declined by 5.6 percent from 2012 to 2014. That was after a steady increase of 6 percent annually from 2007 to 2012.
In an accompanying editorial, James F. Meschia, MD, said this late dropoff of CAS use suggests maybe CREST is changing practice after all.
“One possible interpretation of NIS (National Inpatient Sample) trends is that the results of CREST initially changed practice by slowing a rise rather than initiating a fall in the proportion of older patients receiving stenting,” wrote Meschia, with the Mayo Clinic’s department of neurology. “Without a control group, there is no way to be certain. …
“The preference for stenting over endarterectomy may have peaked in 2012 and there now appears to be a downward trend in carotid artery stent utilization. Full translation of trial results into practice may take years, particularly when operators need to be persuaded rather than regulated into changing behavior.”