We are on the precipice of a major shift in carotid revascularization management. Carotid endarterectomy (CEA) has historically been the gold standard of revascularization for patients with carotid artery stenosis. Since 2005, the Centers for Medicare & Medicaid Services (CMS) have limited coverage of carotid artery stenting (CAS) to patients with symptomatic carotid artery stenosis of 50% or more or asymptomatic carotid artery stenosis of 80% or more who are at high risk of complications with CEA due to significant comorbidities or anatomic risk factors.

In October 2023, the CMS revised its national coverage determination (NCD) for CAS that had been in place since May 2005.¹ The revised NCD expands coverage of CAS by allowing interventionalists to offer CAS to patients with symptomatic carotid artery stenosis of 50% or more or asymptomatic carotid artery stenosis of 70% or more regardless of whether CEA poses a lower or higher risk of complications. It also eliminates previously mandated specific physician training standards, facility support requirements, and data collection to evaluate patient outcomes.

Interventionalists now have extraordinary freedom in offering CAS to patients. Carotid artery stenting can now be used and reimbursed regardless of patient symptom status or risk profile. However, expanding Medicare coverage to a broader group of patients does not mean that CAS is clinically appropriate for all patients who meet these criteria. As a result, the broadened NCD for CAS may result in significant patient overtreatment and harm.

Justification for expanding the criteria for patients eligible to undergo CAS was based on data suggesting that modern-day CAS may be equivalent to CEA. Four large randomized clinical trials showed that CAS and CEA were effectively equivalent with respect to a composite outcome of stroke, death, and myocardial infarction for standard-risk patients with carotid artery stenosis.^2-5 Health-related quality-of-life scores have also been shown to be better after CAS than after CEA in the short term,⁶ and there have been recent advances in CAS techniques and technology, including widespread use of embolic protection devices, that may mitigate the risks of perioperative stroke experienced in historical trials. Patients enrolled in early CAS trials⁷ also were frequently not maintained with currently accepted antiplatelet therapy and underwent excessive use of contrast injections, with an overemphasis on aggressive balloon angioplasty and without other refinements in technique.

The arguments that contemporary CAS is safe overlook many important facts that raise concern for substantial patient harm. Although recent clinical trial outcomes were similar for CAS and CEA using a composite end point, the risk of perioperative stroke with CAS has consistently been shown to be higher than with CEA.^2-5 The higher perioperative stroke risk with CAS is offset by a lower risk of myocardial infarction compared with CEA,² which ultimately results in similar frequencies of composite end points. However, the long-term morbidity and mortality associated with strokes is substantially higher than the long-term morbidity associated with myocardial infarction,⁸ begging the question of whether CAS and CEA are truly equivalent.

It is also important to note that trial results are likely not reflective of outcomes in the general population. The majority of randomized clinical trials used to support the NCD appeal were heavily scripted with regard to the patients who could be enrolled and the physicians who were allowed to treat them. Interventionalists frequently had to perform a large number of procedures (ranging from 20 to 50) to prove their expertise before being allowed to participate in the trials. In reality, the average number of CAS procedures performed by interventionalists in regular practice ranges from 1 to 3 per year.⁹ It is not appropriate to generalize the results of carefully conducted randomized clinical trials with specific inclusion and exclusion criteria to the general population and to all physicians and hospitals irrespective of their experience. In nontrial settings, the risks of stroke and other complications are likely higher. Allowing for broad application of CAS regardless of patient profile and physician experience will almost certainly lead to a higher frequency of poor outcomes.

Expansion of the CAS NCD allows for use of CAS in patients with moderate or severe symptomatic carotid artery stenosis or severe asymptomatic carotid artery stenosis. However, routine use of carotid revascularization for asymptomatic carotid artery stenosis is not generally supported by contemporary data. Even the benefit of carotid revascularization for symptomatic stenosis less that 70% has not been proven compared with optimal medical management alone and is generally not supported by most current guidelines. Medical management for atherosclerotic disease has improved substantially since the original CAS trials were reported. The current average annual ipsilateral stroke rate for patients with asymptomatic carotid artery stenosis treated with optimal medical management alone is estimated to be 1.0% or less, which is lower than the established risk of periprocedural stroke after CEA or CAS.^2-5 This discrepancy was the impetus for the ongoing CREST-2 trial,¹⁰ which is a randomized clinical trial designed to compare outcomes among optimal medical management, CEA, and CAS. Expanding coverage of CAS to all patients with severe asymptomatic carotid stenosis will allow interventionalists to perform a procedure that may be associated with a higher risk of stroke instead of doing no intervention at all.

The updated CAS NCD may result in patients undergoing CAS who should not undergo carotid revascularization at all and interventionalists offering patients CAS when a CEA would be a lower-risk option. Carotid artery stenting utilization may increase now that the treatment is available to standard-risk patients and to institutions regardless of whether they meet prior CMS minimum standards. This increase will likely lead to higher rates of preventable strokes, unnecessary complications, and increased health care costs.

Moving forward, tracking and analysis of CAS outcomes in the general population will be essential to understand practice patterns and prevent patient harm, ideally through mandated registry reporting. Given that the NCD allows unregulated coverage of all carotid revascularization modalities, a shared decision-making tool that incorporates optimal medical management, CEA, and various CAS techniques is necessary to help guide clinicians and patients through the complex decision-making around carotid artery stenosis management. Cost-effectiveness analyses will also be essential given the increased use of CAS; not only is carotid stenting technology more expensive than CEA, but the higher frequency of strokes that may be observed with CAS will likely lead to higher health care costs as a result of complication (stroke) management and rehabilitation.

Overall, the revised CAS NCD allows for expansive availability of CAS but comes with some potentially serious harm. Physicians, interventionalists, and patients alike need to understand the risks and benefits of CAS irrespective of Medicare coverage.

Corresponding Author: Caitlin W. Hicks, MD, MS, Division of Vascular Surgery and Endovascular Therapy, Johns Hopkins University School of Medicine, 600 N Wolfe St, Halsted 668, Baltimore, MD 21287 (chicks11@jhmi.edu).

Published Online: September 30, 2024. doi:10.1001/jamaneurol.2024.3237

Conflict of Interest Disclosures: Dr Hicks reported receiving personal fees for serving on the speaker’s bureau from Silk Road Medical during the conduct of the study and receiving personal fees for serving on the speaker’s bureau from W.L. Gore and Cook and grants from the National Institute of Diabetes and Digestive and Kidney Diseases outside the submitted work. No other disclosures were reported.