Informing vs Changing the Practice of Carotid Revascularization

Abstract

The Carotid Revascularization Endarterectomy vs Stenting Trial (CREST) was a multicenter randomized clinical trial directly comparing the risks and benefits of stenting with those of endarterectomy for symptomatic and asymptomatic highgrade stenosis. Involving 2502 patients across the United States and Canada followed-up for up to 10 years, the trial did not detect significant differences in the end point of perioperative stroke, myocardial infarction, or death or ipsilateral ischemic stroke after the perioperative period. 1 Primary results were published in 2010. 2 CREST clearly informed practice guidelines, but the extent to which CREST altered practice is less clear. The effects of CREST on the practice of stenting may have been muted because not everyone accepts the validity of having included perioperative myocardial infarction in the primary composite end point, although the trial showed that perioperative myocardial infarction carried significant morbidity. Perioperative myocardial infarction increased mortality by more than 3-fold. 3 In CREST, there were subtle absolute differences in event rates and opposing trends in perioperative stroke and perioperative myocardial infarction rates.One way to gain insight into how trials alter practice is to perform an ecologic study assessing temporal trends before and after the trial. Otite and colleagues 4 report in this issue of JAMA Neurology an analysis of temporal trends in the use of stenting as a means of carotid revascularization from 2007 to 2014 in patients older than 70 years in the Nationwide Inpatient Sample (NIS). Most of the patients were asymptomatic, and the mean age was 78.1 years. The study epoch is important because it spans multiple pre-CREST years (2007)(2008)(2009)(2010) as well as multiple post-CREST years (2011)(2012)(2013)(2014), thus allowing for insight into the effects of the CREST results on practice. The choice of study population older than 70 years is also important because, in CREST, age acted as a treatment effect modifier for the primary end point, with the efficacy of stenting and endarterectomy showing equivalence at approximately age 70 years and with stenting being relatively more risky at older ages. 5 Otite and colleagues 4 found that the number of revascularizations per million hospitalizations fell significantly, while the proportion of revascularizations done for symptomatic stenosis increased significantly by 55%. These trends suggest that surgeons and interventionists alike are focusing their energies on treating symptomatic rather than asymptomatic stenosis. This is, of course, immanently reasonable. Trials supporting the efficacy of revascularization for asymptomatic patients used highly selected surgeons and outmoded medical therapy and, based on a detailed review of the evidence,