Invited commentary

Abstract

After surviving earlier attacks on its efficacy through the conduct of well-designed randomized trials, carotid endarterectomy (CEA) is again being challenged by the purported “less-invasive” alternative of carotid artery stenting (CAS). The study by Stoner et al provides important data for the current debate but has limitations that leave many questions unanswered. The SAPPHIRE trial injected the issue of cardiac morbidity into the CEA vs CAS debate. Perioperative cardiac events can be compared between studies only if similar, standardized protocols are used for surveillance for myocardial damage. Unfortunately, the National Surgical Quality Improvement Program database includes only Q-wave myocardial infarction (a small subset of all perioperative myocardial infarctions). However, the observation in this study that local or regional anesthesia was associated with better outcomes at least raises the question of whether the differences in cardiac morbidity seen in the SAPPHIRE trial were anesthetic rather than procedure related. An important issue often neglected in the CEA vs CAS debate is whether neurologically asymptomatic, medically high-risk patients are likely to benefit from either procedure. The relatively benign natural history of asymptomatic carotid stenosis established in the Asymptomatic Carotid Atherosclerosis Study trial should not be forgotten. The limitations of the National Surgical Quality Improvement Program database with regard to the collection of specific cardiac risk factors such as prior myocardial infarction or the presence of angina limit the usefulness of this study in guiding the selection of neurologically asymptomatic patients who are more likely to benefit from CEA. Many of the risk factors identified in this study (diabetes, age, and smoking) are surrogate markers for coronary artery disease. A specific cardiac history combined with noninvasive cardiac screening of selected individuals is likely a more important risk assessment approach for the practicing surgeon than using the risk factors identified in this study alone. Clinical studies like this one are as important as randomized trials in establishing the efficacy of procedures. Community-wide studies are necessary to demonstrate that the procedural outcomes observed in a randomized trial can be achieved across the entire spectrum of patients likely to undergo the procedure and physicians likely to perform the procedure. This study adds to the literature that supports the fact that CEA can be performed across this spectrum with perioperative adverse event rates comparable to those achieved in the randomized trials. It remains to be seen whether this will be as true for CAS as it is for CEA.