Surgical Versus Percutaneous Therapy of Carotid Artery Disease: An Evidence-Based Outcomes Analysis
Elvera L. Baron, MD,Daniel I. Fremed, MD,Rami O. Tadros, MD,Pedro A. Villablanca, MD, MSc,Adam S. Evans, MD,Menachem M. Weiner, MD,Eric H. Yang, MD, FACC,John T. Augoustides, MD, FASE, FAHA,Farouk Mookadam, MD, FACC, FRCP,Harish Ramakrishna, MD, FASE, FACC
STROKE is a major cause of morbidity and mortality worldwide and is a leading cause of long-term, acquired adult disability in most developed countries. In the United States, stroke affects approximately 800,000 people each year and currently is the fifth leading cause of death.
Carotid artery stenosis from atherosclerotic changes of the vessel wall contributes to 10% to 20% of all strokes.
As such, carotid artery stenosis is considered to be a modifiable and treatable factor for the risk reduction of subsequent stroke.
The first successful CEA was performed by DeBakey on August 7, 1953. A 53-year-old bus driver presented with intermittent episodes of right arm weakness associated with expressive aphasia. The diagnosis of carotid stenosis was made based on history and physical examination findings of an extremely weakened left carotid and left superficial temporal pulse. The patient underwent thromboendarterectomy of the left ICA with primary repair of the vessel. The patient experienced an uncomplicated postoperative course with resolution of his neurologic symptoms.
Since that first endarterectomy, several advances have been made in the surgical treatment of extracranial carotid disease. The use of a carotid shunt was first described by Al-Naaman and Cooley in 1956 to maintain cerebral perfusion during carotid cross-clamping. Patch closure of the carotid arteriotomy to preserve luminal diameter was first described in 1965, with the first randomized trial in 1987 showing its benefit over primary closure. Eversion carotid endarterectomy has evolved as an alternative to conventional longitudinal arteriotomy. Etheredge was the first to describe complete transection of the internal carotid to facilitate removal of carotid plaque with end-to-end closure.
Since then, randomized trials have demonstrated comparable results with either conventional or eversion techniques. Several large, randomized trials have been conducted over the last 30 years demonstrating the benefit of CEA in both symptomatic and asymptomatic patients. NASCET was the major trial demonstrating the benefit of CEA in symptomatic patients. NASCET was a randomized, prospective, multicenter trial designed to compare the efficacy of CEA versus medical therapy in patients with symptomatic extracranial carotid disease. The study included 659 patients with hemispheric symptoms and severe ICA stenosis (70%-99%) who were assigned randomly to either CEA or antiplatelet therapy. Another 865 patients with moderate ICA stenosis (50%-69%) were assigned randomly in a similar fashion. At 2-year follow-up, there was a significant reduction in ipsilateral stroke in the group with severe stenosis undergoing CEA versus medical therapy (9% vs 26%, respectively). At 5 years, patients with moderate stenosis demonstrated a reduced ipsilateral stroke rate of 15.7% versus 22.2% in the medical arm (p = 0.45). The benefit seen in both groups persisted throughout the trial’s 8-year follow-up.
ECST was the largest European trial comparing CEA and medical therapy. ESCT was a prospective, multicenter trial that randomly assigned 3,024 patients with symptomatic ICA stenosis to either CEA or ASA therapy. At 3 years, patients undergoing CEA demonstrated a significantly reduced ipsilateral major stroke rate compared with patients randomly assigned to ASA alone (2.8% v 16%, respectively). The frequency of all major stroke or death at 3 years was 14.9% in the CEA group and 26.5% in the medical therapy group, with an absolute benefit of CEA of 11.6%. When stratified by severity of carotid stenosis, the 3-year risk of any major stroke (including surgical events) in patients with 80% to 99% stenosis was 6.8% in the CEA group versus 20.6% in those treated medically (p < 0.0001). At 5 years, the absolute risk reduction (ARR) of stroke within the CEA group was 21.2% (p < 0.0001). The authors concluded that symptomatic patients with high-grade (80%-99%) ICA stenosis benefit more from CEA than from medical therapy alone. The risk of succumbing to major stroke was found to increase based on severity of the stenosis.
The Asymptomatic Carotid Atherosclerosis Study was the largest American trial investigating CEA for asymptomatic disease. This randomized, multicenter trial enrolled 1,662 patients with 60% to 99% ICA stenosis and compared CEA with medical therapy alone. The degree of stenosis was measured via duplex velocity or by angiographically comparing vessel diameters at the greatest area of stenosis with the distal lumen diameter. Patients were followed up every 3 months for 5 years. At mean follow-up of 2.7 years, the risk of ipsilateral stroke or any perioperative stroke or death was 5.1% in the CEA group versus 11% in the medical group (p = 0.004). Throughout all severity groups (60%-69%, 70%-79%, 80%-99%), there was no statistically significant reduction of 5-year risk of stroke. The perioperative risk of stroke or death in this study was 2.3%. When adjusted for the 7 strokes and 2 deaths that occurred preoperatively, the perioperative risk fell to 1.3%. Although not statistically significant, the benefit of CEA in asymptomatic women was not as apparent (ARR 1.4% v 8% in men). When sexes were combined, the overall 5-year ARR was 5.8%.
The Asymptomatic Carotid Surgery Trial was the largest trial regarding asymptomatic carotid disease. In this study 3,120 patients with significant (60%-99% on duplex) asymptomatic ICA stenosis were assigned randomly to immediate CEA or deferral of CEA until indicated (symptomatic). Patients with deferred CEA underwent antiplatelet, antihypertensive, and statin therapy alone. The first carotid angioplasty was proposed by Mathias in 1977 as an alternative to CEA in high-risk patients. He performed the first in-human carotid angioplasty in 1979 on a 32-year-old woman with severe right distal ICA stenosis secondary to fibromuscular dysplasia. She was deemed too high risk for surgery given the distal extent of her disease. The stenotic segment was dilated successfully with a 4-mm balloon. Currently, the strongest accepted indications for CAS are related to symptomatic patients with unacceptable anatomic or medical risk for CEA. Challenging anatomy can pose added risk for patients undergoing CEA. High lesions above the level of the mandible (approximately C2) are difficult to expose surgically and may add risk to the procedure. History of cervical radiation, prior endarterectomy with restenosis, prior radical neck dissection, or presence of a tracheostomy increases the risks of inadvertent cranial nerve injury and thromboembolism. In these instances, CAS is a reasonable alternative to CEA. Unfavorable physiologic criteria also may favor the use of CAS over CEA. Patients with unstable angina, MI within 30 days, class 3 or 4 congestive heart failure, unreconstructable coronary disease, or severe chronic obstructive pulmonary disease may be considered too high risk for surgery. In these cases, CAS is preferred to minimize periprocedural complications. Certain anatomic restrictions may predispose a patient to a greater risk of complications. These include the following: difficult access to the common carotid artery, excessive ICA tortuosity, dense circumferential calcification of the lesion, and significant disease of the aortic arch.
Current Society for Vascular Surgery recommendations advocate dual-antiplatelet therapy for all patients who will undergo CAS. The antiplatelet regimen consists of ASA 75-to-325 mg daily and clopidogrel, 75 mg daily, beginning at least 4 days before CAS. An alternative to this regimen is a single loading dose of clopidogrel, 300 to 600 mg, 4 hours before the procedure. CAS generally is performed with little-to-no sedation to constantly assess the patient’s neurologic function. Careful attention still must be paid to hemodynamics and anticoagulation status throughout the procedure. Several large-scale RCTs comparing CAS and CEA have been conducted.
The Carotid and Vertebral Artery Transluminal Angioplasty Study (CAVATAS) a multicenter, prospective, randomized clinical trial, randomly assigned 504 patients to CAS versus CEA, with the primary outcome being disabling stroke or death. Secondary analyses were performed for any ipsilateral stroke lasting more than 7 days and for death or ipsilateral disabling stroke. The rates of major outcome events within 30 days were reported as not significantly different between the CAS and CEA groups (6.4% v 5.9%, respectively, for disabling stroke or death; 10.0% v 9.9%, respectively, for any stroke lasting more than 7 days or death). At 3 years’ follow-up, the rate of death or disabling stroke, including treatment-related events, was 14.3% in the CAS group and 14.2% in the CEA group. For any disabling stroke or death, the hazard ratio (HR) for CAS/CEA was 1.03 (95% confidence interval [CI] 0.64-1.64, p = 0.9). No substantial difference in the rate of ipsilateral stroke was noted, with a survival analysis up to 3 years after randomization (adjusted HR = 1.04, 95% CI 0.63-1.70, p = 0.9). The authors concluded that CAS demonstrated similar major risks and effectiveness at stroke prevention as CEA during 3 years’ follow-up.
The Stenting and Angioplasty with Protection of Patients with HIgh Risk for Endarterectomy (SAPPHIRE) trial was a noninferiority, prospective RCT comparing CAS with CEA that examined 334 symptomatic and asymptomatic patients with severe carotid stenosis and coexisting conditions that potentially increased the risk posed by CEA. The short-term primary endpoint was a composite of death, stroke, or MI within 30 days after the intervention or death or ipsilateral stroke up to 1 year.53 The primary endpoint occurred in 12.2% of patients treated with CAS compared with 20.1% of patients treated with CEA (p = 0.004 for noninferiority). In the analysis of patients with symptomatic carotid artery stenosis, the cumulative incidence of the primary endpoint at 1 year was 16.8% in the CAS versus 16.5% in the CEA groups (p = 0.95). For patients with asymptomatic disease, the cumulative incidence of the primary endpoint at 1 year was lower in the CAS group (9.9%) than in the CEA group (21.5%; p = 0.02).53 Therefore, the study concluded that CAS is noninferior to CEA in the prevention of stroke, death, or MI among patients for whom surgery poses an increased risk.
The Stent-Supported Percutaneous Angioplasty of the Carotid Artery versus Endarterectomy (SPACE) trial was a multinational, prospective, randomized, noninferiority trial of CAS versus CEA in patients with severe symptomatic carotid artery stenosis that analyzed 1,196 patients. The primary endpoint was ipsilateral stroke, either ischemic stroke or intracerebral hemorrhage, with symptoms that lasted more than 24 hours, or death from any cause between randomization and 30 days after treatment. Two interim analyses were performed. In the second interim analysis, the primary endpoint event rate was reported to be 6.84% in CAS versus 6.34% in CEA. The French Endarterectomy versus Angioplasty in Patients with Symptomatic Severe Carotid Stenosis (EVA-3S) trial was a multinational, randomized, noninferiority trial comparing CAS and CEA in patients with symptomatic carotid stenosis of at least 60%. The primary endpoint was a composite of any stroke or death within 30 days after treatment. The trial was terminated early after recruitment of 527 patients for reasons of safety and futility—a significantly lower rate of stroke or death was reported in the patients who underwent CEA. The 30-day incidence of any stroke or death was reported at 9.6% in the CAS versus 3.9% in the CEA groups (95% CI 6.4-14.0 v 2.0-7.2, respectively), resulting in a relative risk (RR) of any stroke or death after CAS versus CEA of 2.5 (95% CI 1.2-5.1). The 30-day incidence of disabling stroke or death was reported at 3.4% in the CAS versus 1.5% in the CEA groups (95% CI 1.7-6.7 v 0.5-4.2, respectively), with a resultant RR of 2.2 (95% CI 0.7-7.2).
In 2010, the International Carotid Stenting Study (ICSS) released an interim analysis of its controlled trial, comparing the safety of CEA and CAS. This international, multicenter RCT enrolled 1,710 patients with symptomatic stenosis. At 120 days—a longer definition of “short-term” than in other studies—no significant difference in the rate of disabling stroke or death between groups was found (4% in CAS v 3.2% in CEA). Most strokes within 120 days were nondisabling, ipsilateral to the treated artery and were ischemic. The incidence of stroke, death, or procedural MI after 120 days was found to be 8.5% in the CAS versus 5.2% in the CEA groups (72 v 44 events; HR 1.69, 95% CI 1.16-2.45, p = 0.006). Risk of any stroke (65 v 35 events, HR 1.92, 95% CI 1.27-2.89) and all-cause deaths (19 v 7 events, HR 2.76, 95% CI 1.16-6.56) were found to be higher in the CAS group versus the CEA group. Furthermore, 3 fatal MIs in the CAS group compared with 4 nonfatal MIs in the CEA group were reported. This interim analysis showed that, in the short term, CEA was safer than CAS for the treatment of patients with symptomatic carotid stenosis.
Also in 2010, the Carotid Revascularization Endarterectomy versus Stenting Trial (CREST) compared CEA with CAS in 2,502 symptomatic and asymptomatic patients. The primary endpoint was the occurrence of any stroke, MI, or death during the periprocedural period or ipsilateral stroke thereafter up to 4 years. No significant difference in the primary endpoint between the CEA and CAS groups was reported (6.8% v 7.2%, HR 1.11; 95% CI 0.81-1.51, p = 0.51). Although during the periprocedural period the incidence of the primary endpoint was similar for CEA and CAS, differences in the endpoint components were reported (stroke 4.1% v 2.3%, p = 0.012; MI 1.1% v 2.3%, p = 0.032; and death 0.7% v 0.3%, p = 0.18). Outcomes were found to be slightly better after CAS in patients < 70 years old and better after CEA for patients aged > 70 years. The reported 30-day periprocedural incidence of death and stroke was only 1.4% compared with 2.9% in the earlier landmark trials. These improved outcomes of CREST compared with prior studies have been attributed to more rigorous preprocedural and postprocedural evaluations, improved medical therapies, and improved endovascular surgical expertise including the requirement for distal protection in all patients. A slightly increased risk was seen for major ipsilateral stroke in the CAS group compared with the CEA group up to 4 years after treatment (1.4% v 0.5%, p = 0.05).42 CREST was analyzed subsequently for long-term outcomes at 10 years. The study found no significant difference between the CAS (11.8%, 95% CI 9.1-14.8) and CEA (9.9%, 95% CI 7.9-12.2) groups with respect to the primary composite endpoint of stroke, MI, or death during the periprocedural period or any subsequent ipsilateral stroke over the 10-year follow-up. The rates of postprocedural ipsilateral stroke over the 10-year follow-up, which was the primary long-term endpoint, were not found to be significantly different between groups (CEA 6.9%, 95% CI 4.4-9.7 v CAS 5.6%, 95% CI 3.7-7.6). During the postprocedural period, there were nominally more major strokes among patients assigned to CAS than among those assigned to CEA, yet the difference was not statistically significant (12 events CAS v 6 events CEA, p = 0.20). In addition, no significant treatment differences were found according to symptomatic status. This suggests that symptomatic status is of relevance in the context of periprocedural risk but ceases to be significant after revascularization. Secondary analysis revealed that the risk of periprocedural stroke or death and subsequent ipsilateral stroke was 37% higher in the CAS versus CEA group, with the advantage of CEA largely due to differences in the rate of periprocedural events. Most recently, the
Asymptomatic Carotid Trial 1 (ACT 1) examined the safety and efficacy of CEA versus CAS in asymptomatic patients younger than 79 years with severe carotid stenosis. In this study, 1,453 patients with a mean age of 68 years were assigned randomly to CAS (1,089) versus CEA (364); follow-up was for 5 years. A noninferiority margin of 3% was used for the primary composite endpoint of death, any stroke, ipsilateral stroke within 1 year, or MI within 30 days after the procedure in the CAS versus CEA groups. The primary endpoint analysis was a between-group comparison of the rate of death, stroke, or MI within 30 days or of an ipsilateral stroke within 365 days after the procedure. The study concluded that CAS was noninferior to CEA with regard to the primary composite endpoint. Short term, a 30-day rate of stroke or death was reported to be 2.9% in the CAS group compared with 1.7% in the CEA group (p = 0.33). In addition, the rate of MI was low in both study groups, 0.5% in the CAS versus 0.9% in the CEA group. During the long-term 5-year follow-up, all stroke and survival rates were not found to be statistically different between the CAS and CEA groups. These long-term findings were similar to those of CREST, which examined asymptomatic and symptomatic patients with severe carotid stenosis, and for which no significant differences in the rates of the primary endpoint between the CEA and CAS groups over 4 years or over 10 years were found. Furthermore, similar to CREST, the ACT 1 study also found that the rates of stroke, death, and MI overall were low with each intervention.
Vincent et al in 201554 recently reported the results of a large meta-analysis that included 4 RCTs with symptomatic patients,44,46,49,55 3 RCTs with asymptomatic and symptomatic patients,42,45,47 and 1 RCT with merged symptomatic and asymptomatic patients for long-term analysis. Of note, Vincent et al did not include results of ACT 1,2 which were reported later. The total population studied was 7,091 patients, with 3,561 in the CAS arm versus 3,530 in the CEA arm, with a mean age of 68.1 to 72.6 years. This meta-analysis reported that during the periprocedural period (defined as 30 days after the intervention), the CAS group was associated with a higher risk of stroke outcomes. Furthermore, the CAS group exhibited a higher risk of any stroke across trials (RR 1.49, 95% CI 1.11-2.01). Results from individual trials have attributed this increase in perioperative risk in the CAS group to a higher incidence of periprocedural nondisabling or minor strokes. The risks of periprocedural disabling or major stroke were similar between the CAS and CEA groups. In most trials, Vincent et al found a higher incidence of a composite outcome of periprocedural stroke or death in the CAS arm (RR 1.5, 95% CI 1.12-2.02). On the other hand, the CAS group also was found to have a significantly lower risk of MI in the periprocedural period (RR 0.47, 95% CI 0.29-0.78). The meta-analysis reported pooled risk differences for MI and stroke in the CAS group patients; CAS was associated with a 0.4% decreased risk of MI and a 1.7% increased risk of stroke. When examining long-term follow-up outcomes, with data pooled across these 8 RCTs, the risk of any long-term stroke was significantly higher with CAS versus CEA (RR 1.36, 95% CI 1.16-1.61). In addition, CAS was associated with a significantly increased risk of long-term ipsilateral stroke compared with CEA. Specifically, the incidence of disabling or major stroke also was higher in CAS versus CEA in all RCTs except SAPPHIRE. The authors of this meta-analysis concluded that “although CAS had more favorable periprocedural outcomes with respect to MI [since] CEA had more favorable effects on periprocedural and long-term outcomes. It should remain the treatment of choice for patients with carotid stenosis.These results were similar to those reported in prior meta-analyses,57,58 with confirmation that the higher cumulative incidence in stroke-related events during follow-up was likely to be driven by an increased risk during the periprocedural period. Outcome data comparing CEA vs CAS in Medicare patients have been an issue that has needed more granularity, and consequently there has been no significant change to the national coverage determination provided by the CMS for patients with carotid disease requiring revascularization over the past few years.
A recently published study assessed the comparative effectiveness of CAS versus CEA among Medicare beneficiaries. The investigators linked Medicare data for patients treated with CAS or CEA over a 9-year period (2000-2009) with 2 large registries, the Society for Vascular Surgery’s Vascular Registry (2005-2008) and the National Cardiovascular Data Registry’s Carotid Artery Revascularization and Endarterectomy Registry (2006-2008/2009). After adjusting for patient and provider factors, the overall effectiveness of CAS and CEA in the Medicare population studied was similar for all outcomes/endpoints—
Lastly, the conclusion was similar to the 2016 Lancet pooled meta-analysis (referred to earlier),59 which demonstrated that older Medicare patients (aged>80) with symptomatic carotid disease may have better outcomes with CEA than with CAS. The 2011 American College of Cardiology/American Heart Association guidelines on the management of patients with extracranial carotid and vertebral artery disease have been endorsed by 14 professional societies and remain the current standard. Per these guidelines, CAS is recommended as alternative to CEA (class I, level of evidence [LOE] B) in average surgical risk patients with symptomatic carotid disease and a < 6% risk of mortality or periprocedural stroke. These guidelines also recommend CAS over CEA in patients with anatomically unfavorable carotid lesions such as high cervical locations, after radiation therapy, and in restenosis (class IIa, LOE B) and in patients with>60% angiographically proven carotid stenosis (class IIb, LOE B). As previously noted, the short-term results of RCTs comparing CEA and CAS have been conflicting, whereas the long-term results are similar, with no reported differences in rates of strokes or death. What is not debatable is that CAS has demonstrated clinical equipoise with CEA, as evidenced by Food and Drug Administration clearance of 7 carotid stent systems in the past few years.
