Stent-Supported Carotid Angioplasty
Should It Be Done, and, If So, by Whom? A 1998 Perspective
Cardiologists, focusing primarily on the heart and its function, diseases, and treatments, had until recently expressed minimal interest in peripheral vascular diseases. Now, spurred on by their successful endovascular treatments of a variety of cardiac pathologies, a few cardiologists have championed the proposal that endovascular treatments should be disease specific and not site specific; thus, their interventional efforts should not be limited to the heart. This perspective has been perceived by some medical and surgical specialties as self-serving and demeaning of these specialties and their accomplishments. In addition, the structure of cardiology training programs has paid little or no attention to the natural history, pathology, diagnosis, and management of vascular patients, let alone endovascular therapeutic procedures. In addition, many peripheral endovascular publications, written in large part about observational studies by private practice cardiologists, have been viewed with circumspection by others, because the authors focused on clinical issues outside their area of expertise, and their audacity transcended the sacrosanct, but nebulous, specialty boundaries. Although these arenas were often considered to be outside the purview of cardiologists, the results of these investigations motivated other cardiologists to move forward. Now, the efforts of cardiologists to vanguard the evaluation of endovascular revascularization techniques involving obliterative extracranial carotid bifurcation disease have made more transparent the significant interdisciplinary tensions that were created, in part, not only by physicians’ angst about their future but also by the blurring of the distinctions between specialties.
A few cardiologists have made these more taut through their premature conclusions and extrapolations drawn from nonrandomized, observational carotid stent-supported angioplasty studies1 and the pronouncements that this technique is state of the art and does not require any randomized trials. Such statements have resulted in the foisting of this technique as the standard of care on unsuspecting patients. The surreptitious insertion of patients into treatment paradigms the aims of which are ill defined and the methodologies and devices for which have not yet been perfected is unethical. Although the use of unorthodox techniques for unusual and extreme clinical situations may be apropos, the utilization of stent-supported carotid angioplasty outside well-designed and carefully monitored experimental protocols is inappropriate, because carotid endarterectomy, the “gold standard,” is available.
Everyone wants a device or system that can easily be visualized, prevents or eliminates embolization, has elasticity to preclude “crushing,” does not deteriorate with time, does not prevent computerized tomographic and MRI modalities from being used (eg, stainless steel produces a void in MRI), and is safely used or deployed by the average skilled interventionist with as few untoward events as possible. The use of balloon-expandable stainless steel devices presents significant problems, because the incidence of subsequent problems remains cumulative with the demonstration of stent deformation without any external forceful blow.
Although I fervently believe that endovascular therapies will supplant carotid endarterectomy, the course chosen by some cardiologists and a few radiologists and vascular surgeons actuated this article, which addresses carotid stent-supported angioplasty data from Yadav et al,1 the interpolation of observational data conclusions, the purpose of carotid revascularization, and a patient population that might best show extant differences between endovascular revascularization and carotid endarterectomy.
Stent-supported carotid angioplasty is no different than other “recent advances in biomedical technology and therapeutic procedures … [which] have generated a moral … crisis in modern medicine. … With the unfolding of new discoveries and techniques, the scientific and intellectual communities have developed a keen awareness of the ethical issues which arise out of man’s enhanced ability to control his destiny.”2
Physicians want to expeditiously reestablish intracranial blood flow using appropriate technology that eliminates procedural complications and lesion recurrence and thereby prevents ipsilateral stroke. Without exception, patients want to undergo a therapy that is safe, simple, and atraumatic; has few or no complications; will not need to be repeated; and will preclude the occurrence of a future stroke. Uniformly, manufacturers want to produce an economically beneficial device or system that will achieve those aforementioned desires simply, safely, and effectively.
A new stroke results in a neurological deficit in nearly 2000 Americans each day, and as a result of stroke, >500 Americans die daily. When death does not occur, stroke can devastate the patient, who may endure long-term physical and intellectual disability; the families, whose lives are forever altered; and society, which loses a productive individual while incurring enormous expenses. Because most strokes result from arterial blood flow obstruction and/or embolic debris, any treatment that could better solve these impediments while improving on the results of carotid endarterectomy would be welcomed. However, unlike other surgical therapies supplanted by endovascular procedures, carotid endarterectomy is a relatively simple surgical procedure, has excellent immediate (procedural) success and few complications, and has demonstrated a significant reduction in ipsilateral stroke occurrence (the premise for the performance of the procedure).
The North American Symptomatic Carotid Endarterectomy Trial (NASCET), which enrolled symptomatic, generally healthy patients who were randomized to either medical treatment or best medical treatment plus carotid endarterectomy, demonstrated that endarterectomy was superior to medical management in reducing the risk of stroke.3 In NASCET, the surgical group had a 30-day 2.1% major stroke and death rate, while the medical group had a 0.9% incidence. At 24 months, the risk of fatal or nonfatal ipsilateral stroke after randomization was 26% for the medical and 9% for the surgical patients, which is an absolute risk reduction of 17% and a relative risk reduction of 65%.
The eligibility requirements for patient participation were very clear. Patients were excluded from NASCET if they were mentally incompetent or unwilling to give informed consent; had no angiographic visualization of both carotid arteries and their intracranial branches; had an intracranial lesion that was more severe than the surgically accessible lesion; had organ failure of the kidney, liver, or lung; had cancer judged likely to cause death within 5 years; had a cerebral infarction on either side that deprived the patient of all useful function in the affected territory; had symptoms that could be attributed to nonatherosclerotic disease (eg, fibromuscular dysplasia, aneurysm, or tumor); had a cardiac valvular or rhythm disorder likely to be associated with cardioembolic symptoms; or had previously undergone an ipsilateral carotid endarterectomy. Patients were temporarily ineligible for NASCET if they had uncontrolled hypertension, diabetes mellitus, or unstable angina pectoris; myocardial infarction within the previous 6 months; signs of progressive neurological dysfunction; contralateral carotid endarterectomy within the previous 4 months; or a major surgical procedure within the previous 30 days. Such patients could become eligible if the disorder causing their temporary ineligibility resolved within 120 days after their qualifying cerebrovascular event.
The Asymptomatic Carotid Atherosclerosis Study (ACAS)4 5 found that endarterectomy in patients with an asymptomatic stenosis resulted in a 5.8% absolute risk reduction of fatal and nonfatal ipsilateral stroke. Thus, patients who survived endarterectomy without neurological morbidity had fewer subsequent neurological deficits (transient or permanent; major or minor) compared with medically managed patients. The 5-year risk of ipsilateral stroke and any perioperative stroke or death was 11.0% for the medical group and 5.1% for the surgical group. The estimated 5-year event risk of ipsilateral stroke was 8.3% for the surgical group and 19.2% for the medical group. The 5-year event rate was reduced by 66% in men and 17% in women. However, the perioperative complication rate was 3.6% for women and 1.7% for men. In patients without perioperative complication, the 5-year event-free rate was reduced by 56% for women compared with 79% for men. These numbers exclude arteriographic (1.2%) and perioperative complications.
Patient eligibility requirements of ACAS were, as in NASCET, detailed. Patients were excluded from ACAS if they had previous cerebral infarction, previous endarterectomy with restenosis, previous extracranial-to-intracranial bypass, high risk because of associated medical illness, long-term anticoagulation therapy, intolerance of aspirin or long-term aspirin therapy at a high dose, life expectancy <5 years, surgically inaccessible lesion, noncompliance, or refusal to participate in the protocol.
The pioneering work of Roubin1 is exceptional and will ultimately prove to be correct. However, their statement that most of their patients would have been excluded from those surgical trials requires indagation. First, the implication is that, because the vast majority of their patients would have been excluded from these surgical trials as poor surgical candidates, their cohort would be at a higher procedural risk than the surgical cohorts in ACAS and NASCET. This logic is faulty. Patients were excluded from the surgical trials not, in the main, because of potential surgical procedural difficulties but because the study data, if such patients were included, could have become contaminated, because comorbidities might adversely influence the incidence of subsequent neurological events and thereby preclude a successful response to the query as to whether carotid endarterectomy prevented future ipsilateral neurological deficit. Additionally, a large percentage of their patients were asymptomatic women or patients with restenosis after a prior carotid endarterectomy, cohorts that could positively influence their procedural and follow-up conclusions because the occurrence of subsequent neurological events in these groups is small, whether surgically corrected or not.
These data from Roubin indicate the feasibility of endovascular therapy and underscore the necessity for performing a carefully designed and controlled trial with an appropriately designed device (remember their transition from balloon-expandable to the self-expanding stents). However, clinical equipoise can be reached only once an appropriate device has been constructed and pilot studies demonstrate feasibility. Only at this point could a randomized trial of endovascular stent-supported angioplasty and carotid endarterectomy be rationally initiated.
Furthermore, any trial must address the patient population that can best answer the proposed hypothesis. If an alternative revascularization technique to endarterectomy is to be considered acceptable, then it must a priori achieve comparable procedural success and subsequent stroke prevention, which should be tested on a population cohort that can clearly show significant differences, while exposing as few patients as possible to unknown risks. Furthermore, these data points must be gathered by independent neurological evaluation to determine the unbiased incidence of periprocedural complications and freedom from subsequent neurological events.
The retrospective analysis of Sundt et al6 7 8 of carotid endarterectomy has provided a perspective with regard the clinical, neurological, and angiographic variables affecting surgical outcomes and which patient cohorts might best be chosen for inclusion in a randomized trial. Their patients were stratified according to neurological status, comorbid conditions, and extent of carotid disease. Patients were placed in six risk classes by use of medical, neurological, and angiographic variables. Medical parameters included coronary artery disease (angina, myocardial infarction <6 months, or congestive heart failure), hypertension (>180/110 mm Hg), severe peripheral vascular disease, chronic obstructive pulmonary disease, age exceeding 70 years, and severe obesity. Neurological parameters included neurological deficit within 24 hours, general cerebral ischemia, recent cerebrovascular accident (<7 days), and frequent transient ischemic attacks. Angiographic parameters included occlusion of contralateral internal carotid artery; stenosis in siphon; plaque >3 cm distal in internal carotid artery or >5 cm proximal in common carotid artery; bifurcation of common carotid at the level of the second cervical vertebra; a short, thick neck; and soft thrombus extending from an ulcerative lesion.
Sundt et al6 7 8 analyzed 3111 consecutive endarterectomy patients. The combined morbidity and mortality data revealed a very low complication rate for class I and II patients, whereas class III patients had a 3.7% combined risk (with a 1.3% mortality) and class IV patients had an 8.1% risk (with a 2.9% mortality). Surgical morbidity and mortality rose in the presence of comorbid conditions, an unstable neurological status, and age exceeding 70 years and in patients with contralateral internal carotid artery occlusion. However, despite the positive influence surgical experience had on mortality—632 class I patients operated on between 1972 and 1985 had a 0.2% mortality and 0.3% major stroke rate, whereas 267 patients operated on between 1985 and 1992 had a 0% mortality and a 0.4% major stroke rate—the occurrence of procedurally related neurological deficits remained an extant uncontrollable procedural vagary, improvements in skill and anesthesia notwithstanding. The data of Sundt et al are consistent with those of NASCET and ACAS (which excluded the overwhelming majority of Sundt class III and IV patients). The medical community currently is not sufficiently perplexed as to the equivalency of carotid endarterectomy and stent-supported carotid angioplasty. Spokespersons for each method vigorously and passionately detail their observational experiences and conclusions; however, clinical equipoise, a genuine state of doubt regarding the equivalence of each procedure, does not exist.9 10 The lack of equipoise does not preclude continuation of reviewed and monitored studies to evaluate specific types of patients and individual devices; in fact, this is appropriate and ethical. As such, the initiation of a randomized trial appears premature, particularly with regard to devices and patient eligibility. The problems relating to design and performance are extensive and beyond the scope of this article. However, concerning patient eligibility, a randomized trial attempting to show equivalency or superiority should include Sundt class III and IV patients and add patients with radiation fibrosis and contralateral internal carotid occlusion to expeditiously demonstrate such differences.
What is crystal clear is that before any endovascular approach becomes the standard of care for bifurcation extracranial carotid disease, it must be shown unequivocally that the revascularization procedure not only is safe but also has the same if not reduced occurrence of ipsilateral stroke. Because stent-supported angioplasty in nonrandomized observational studies with varied patient inclusion criteria has an incidence of procedurally related neurological deficits comparable to surgery, inferential or extrapolated conclusions from these data, if wrong, could result in numerous preventable injuries. Therefore, stent-supported carotid angioplasty might best be performed on potentially high-risk carotid endarterectomy patients who are part of an experimental study in which investigators understand that their primary responsibility is following the protocol, adhering to the patient inclusion and exclusion criteria, and obtaining independent neurological assessment. This difficult, trying, and time-consuming approach would guarantee an unbiased methodology of data collection and subsequent assessment that would help resolve this presently contentious situation.
In addition, substantive issues transcend my technical and methodological concerns. Physicians who perform coronary or peripheral interventions have knowledge of the pertinent pathology and anatomy, as well as a complication management paradigm that usually permits correction of a problem and successful procedural outcome. In contrast to coronary and peripheral interventions, neurological interventions may result in complications that, although producing acute cerebral ischemia for only minutes, may result in permanent neurological deficit. Although neurological interventionists performing intracranial procedures want an effective complication management schema that can quickly, effectively, and reliably remedy such problems, no such proven paradigm exists. This is a significant issue.
In this gestational period of endovascular carotid procedures, resolution of the conundrums of whether stent-supported carotid angioplasty should be done as standard of care as well as to whom, by whom, with what indications, with what device, and with what results is insoluble and inappropriate without substantive evidence. Clinical equipoise has not been reached. Furthermore, the image of an inexperienced radiologist, cardiologist, or surgeon who has ill-defined motivations, is fueled by inferential conclusions, and has no peripheral or cerebral diagnostic imaging or interventional skills is disconcerting. The unbridled and cavalier use of stents within the extracranial carotid bifurcation as standard care is inappropriate, cannot and should not be condoned, and should be performed only within carefully constructed trials by physician-scientists, whatever their specialties, who have the background, skill level, dedication, passion, desire, tenacity, facilities, and support personnel to do significant and important clinical experimentation.
Physicians and patients should be reminded that all neurological deficits are permanent when going forward and only transient when looking backward.
This work was sponsored by the Arizona Heart Institute Foundation, Phoenix, and the William Dorros-Isadore Feuer Interventional Cardiovascular Disease Foundation Ltd, Milwaukee, Wis.
Reprint requests to Gerald Dorros, MD, FACC, President, Arizona Heart Institute Foundation, 2632 N 20th St, Phoenix, AZ 85006.
- Copyright © 1998 by American Heart Association
Yadav JS, Roubin GS, Iyer S, Vitek S, King P, Jordan WD, Fisher WS. Elective stenting of the extracranial carotid arteries. Circulation. 1997;95:376–381.
Rosner F. Modern Medicine and Jewish Ethics. New York City, NY: Yeshiva University Press; 1986:7.
Executive Committee for the Asymptomatic Carotid Atherosclerosis Study. Endarterectomy for asymptomatic carotid artery stenosis. JAMA. 1995;237:1421–1428.
Sundt TM Jr, Meyer FB, Piepgras DG, Fode NC, Ebersold MJ, Marsh WR. Risk factors and operative results. In: Weber FB, ed. Sundt’s Occlusive Cerebrovascular Disease. 2nd ed. Philadelphia, Pa: WB Saunders Co; 1994:241–247.
Fletcher J. Should good surgical candidates by operated upon by endovascular procedures? J Intervent Cardiol. 1997;10:497–498.