Stroke, the third leading cause of death in the US, remains one of the most feared conditions of our modern society. Many fear a neurologic catastrophe resulting in a chronic vegetative state more than death. Likewise, much of our modern medical practice seeks to maintain a healthy state free of such diseases. Almost one half of strokes are related to atherosclerotic disease found at the carotid artery bifurcation, while cardiac disease and intracranial vascular disease are responsible for most of the remainder. Likewise, many strokes can be prevented with a healthy lifestyle and medical therapy aimed at reducing risk factors - control of blood pressure, lipid management, smoking cessation, and platelet inhibition. However, when disease of the carotid bifurcation progresses or symptoms cannot be controlled with medical therapy, then surgical treatment, specifically carotid endarterectomy (CEA), has been shown to prevent strokes.1 In recent years, we have witnessed the rapid evolution of endovascular therapy for several vascular diseases where surgery has been routinely used. The clinical results of coronary angioplasty and stenting directed the attention of endovascular therapies toward other vascular territories including the carotid bifurcation. However, while some endovascular procedures have been well accepted, carotid angioplasty and stenting (CAS) remains a controversial therapy among many disciplines that have ventured into the treatment of peripheral vascular disease. Early randomized clinical trials showed the superiority of CEA over CAS, but the recent SAPPHIRE trial showed a similar outcome between the two therapies when treating high risk patients. However, while this rather small trial showed a low rate of strokes when embolic protection was used, the stroke rate in the 200 asymptomatic patients neared 6%, well over the AHA recommended threshold of 3%. There were a reduced number of strokes and sub-endocardial ischemic events in the 50 high-risk patients who underwent CAS compared to CEA. As a result of this trial, an 11-member panel narrowly recommended US Food and Drug Administration (FDA) approval for certain 'high-riskÔÇÖ patients. Full FDA approval has not been received as many device companies are preparing for the coming treatment onslaught. 'OnslaughtÔÇÖ may be the best chosen description, as many carotid physicians fear the inappropriate application of this new technology for those patients best treated with medical therapy or CEA.
At first view, CAS seems to have less physiologic stress than CEA, considering that an uncomplicated groin puncture would be less invasive than a formal neck surgery. Then CAS should be ideal for patients whose physiologic condition is so compromised that surgery would be a hazard. These patients would need to have the similar low stroke and death rates of CEA (3% for asymptomatic and 6% for symptomatic) in order to be clinically effective.
Preliminary reports from SAPPHIRE comparing CAS with distal protection in patients at high risk for endarterectomy trial suggests the stroke rates are similar between CAS and CEA,2 but the myocardial infarction (MI) rate is higher in the CEA group. These findings contrast the non-neurologic problems including cardiac and pulmonary morbidity found in a UAB study3 where there was a 32.8% incidence of non-neurologic problems compared to 17.4% in CEA patients. In a Cleveland Clinic study that included cardiac morbidity and neurologic complications, Ouriel reported in a review of more than 3,000 patients, that medically compromised patients had poorer outcome after CEA. They reported a composite end-point of stroke, MI or death in high-risk patients of 7.4% as compared with 2.9% in the low-risk group.4 The UAB experience did not find a difference in the stroke rate in the high-risk versus low-risk patient, but found almost identical composite endpoints of 7.1% and 2.8% respectively.5
However, most of these cardiac complications were limited to atrial fibrillation of an exacerbation of congestive heart failure, without chemical findings of myocardial ischemia. Based upon those results, we concluded that CAS was unlikely to offer any improvement in stroke risk as compared with CEA , but it could reduce non-stroke morbidity rates associated with some high-risk cases. Lepore et al. reported on the stroke/death rate for patients that were 'high riskÔÇÖ (NASCET/ACAS ineligible) who underwent CEA. There was no significant difference in the stroke/death rate between trial eligible and trial ineligible patients. More importantly, the reported rates for ineligible patients fell well within the AHA suggested guidelines for CEA.6 However, the largest prospective cohort of CAS patients published to date reported composite stroke, MI or death end-points of 8.3%.7 Although these authors did find a very low incidence of peri-procedural MI (.2%) that could support CAS as a less invasive procedure, patients were not subdivided into high or low risk category. In the UAB series, the incidence of cardiac complication was 10% for the high risk and 0.8% for the low risk group. Similarly, Chastain et al reported a composite stroke; MI or death of 10% in patients of different ages (again without specific division in high or low risk); yet, when considering patients older than 80 years the composite stroke and myocardial end-point was 25% (no deaths).8 Another single report analyzing outcomes after CAS in NASCET-ineligible symptomatic patients describes a combined 30-day cardiac morbidity and stroke of 10.7% (plus a 10.7% rate of transient ischemic attacks and 3.6% of acute renal failure).9
Other authors have described a 30-day combined death, stroke, and ipsilateral blindness as high as 27.3% in high risk patients after CAS.10 Therefore, while a single study shows higher non-neurologic morbidity associated with CEA over CAS, the medical community has not yet defined the most appropriate group of patients who should have CAS preferentially over CEA. Obviously, there remains controversy about the definition of the 'high riskÔÇÖ patient. These above-cited studies from Oschner Clinic, University of Michigan, the Mayo Clinic, University of Alabama, and New York University, have all demonstrated that many 'high riskÔÇÖ patients can be treated with the same morbidity as normal risk patients. While the definition of the high risk patient remains in question, some clinical trials are comparing the results of carotid stenting with CEA in the normal risk patient (CREST). Currently, 50 centers across North America are enrolling patients in this NIH-controlled, randomized, clinical trial to compare CEA to CAS. After these trial results are examined along with data about the longer-term efficacy of stents in preventing strokes, then there may be a broader acceptance of the application of this technology across multiple disciplines.
Restenosis
The long-term outcomes of CEA are well documented in large trials of both symptomatic and asymptomatic lesions; but the incidence of restenosis remains undefined for CAS. Currently, accepted restenosis rates after CEA range from 0% to 8% depending on the type of arterial closure. Duplex ultrasonography has been established as the best clinical tool to screen for restenosis after CEA. Given the substantial amount CAS that has been undertaken, there is a limited long-term follow-up duplex data. Stankovic et al. defined restenosis as a recurrent lesion involving >50% of the vessel diameter and reported a restenosis rate of 3.4% at 12 months in 84 of 100 patients who underwent CAS.11 Gray et al. reported 43% of 102 stents had a peak systolic velocity (PSV) of greater than 1.0m/sec at one year and 5% had PSV of >/=2.5m/sec.12 Criado et al. reported a 3% hemodynamically significant (>50% by duplex) restenosis rate at mean follow-up of 16 +/-9 months.13 In addition, Ohki et al. reported significant in-stent restenosis (>70%) in four of 30 patients (13%) at a mean follow-up of 17 months who underwent CAS. Two of these patients required intervention to prevent ICA occlusion within one year.14 When one considers a flow abnormality to be any lesion that results in a PSV of >1.1m/sec (>40% stenosis), a review of the CAS patients at UAB found an 8% restenosis rate at one year, and 33% at five years. Seven patients (5%) had lesions with PSV >/=2.5m/sec (80% to 99% stenosis).
The natural history of these recurrent lesions is as of yet undetermined,yet our practice has been to intervene once stenosis reaches a critical level (>80%) to prevent ICA occlusion even in the asymptomatic patient. Therefore only four patients in our series have required surgical correction,15 while the others remain under medical management with ultrasound surveillance. Patients are typically asymptomatic and continued surveillance is advisable to identify such lesions.
What Kind of Specialist Conducts CAS?
While carotid revascularization has traditionally been undertaken by a variety of surgical specialists (vascular surgeon, general surgeon, cardiac surgeon, and neurosurgeon), the realm of CAS includes many non-surgical specialists. Specifically, radiologist and cardiologists have used their catheter-based skills gained in treating other vascular territories to the carotid bifurcation. Therefore, CAS, unlike CEA, crosses many disciplines. Surgeons have maintained an 'ownershipÔÇÖ of CEA due to its surgical nature. However, CAS uses endovascular skills that are shared by surgeons, radiologists, and cardiologists. All three of these specialists have published reports that show similar stroke rates, with no single discipline maintaining 'ownershipÔÇÖ for the procedure. However, surgeons, based upon the nature of their training, are the single group that has shown the ability to employ both CEA and CAS in treating patients. Surgeons have similarly demonstrated an ability to safely and broadly apply endovascular and open treatment for abdominal aortic aneurysms and occlusive disease in other vascular territories (lower extremity, renal, upper extremity).This broad understanding and ability to apply multiple treatment options for carotid disease likely will position the surgical specialist to excel in grasping and applying new technologies for these complex vascular patients. Initially, CAS was associated with a 10% to 12% stroke rate, well beyond the accepted standard for treating both symptomatic and asymptomatic patients. However, various embolic protection devices have been developed to limit the embolic peri-procedure risks associated with CAS. Subsequent reports have suggested stroke rates of 4% to 6% when protection is used. Therefore, while not proven in a randomized clinical trial, most physicians agree that embolic protection is required to meet the treatment standard that CEA has established in preventing strokes. Consequently, embolic protection devices then have added to the complexity and cost of the procedure.
The cost of CAS has been shown to exceed the cost of CEA. A study at UAB showed a difference in hospital charges of $22,000 for CEA compared to $30,000 for CAS.16 However, these figures represented hospital charges to patients or third-party payers and not direct hospital costs. When adding the actual cost of the protection device, the total cost of the procedure will even more dramatically exceed CEA. Considering that both CEA and CAS are often undertaken without general anesthesia and the patients are commonly discharged on the day following the procedure, both procedures can be considered 'minimally invasiveÔÇÖ. Consequently, the benefit for treating asymptomatic patients likely will not be found when considering the dollar cost and the physiologic cost of a stroke. According to the ACAS data, stroke rates must be less than 3% in the asymptomatic patient to offer a real benefit over a five-year period. Therefore, the ill and infirmed patient who does not have lateralizing symptoms from carotid bifurcation stenosis and faces a limited life expectancy is probably best treated with medical therapy. Regardless, while the cost of CAS may be greater than CEA, there likely will emerge a certain subset of patients who will benefit from CAS.
So, which therapy is best to prevent strokes? Currently, CEA remains the best therapy for preventing strokes in patients with symptomatic >50% carotid stenosis and symptomatic >60% stenosis. CAS also appears to be a promising tool to use for prevention of stroke in carefully selected patients.
To date, most published series show the inferiority of CAS when compared to CEA (see Table 1).