Carotid Stenting: A Gordian Knot to be Unraveled (Part I of II)

Balloons, stents and embolic protection devices, as treatment for extracranial obliterative disease, have created in recent decades a predicament not dissimilar to the percutaneous treatment of coronary angioplasty and mitral valvuloplasty. The latter 2 procedures, using observational data, supplanted standard surgical procedures, and these successions occurred without the imprimatur of a large, expensive, “randomized” trial, which juxtaposed the percutaneous against the surgical procedure. Thus, why the furor over carotid stenting? “Changes in attitude, and changes in latitude” (Jimmy Buffett). During the last two decades, opinions concerning the seminal and creative carotid angioplasty work of Theron1–7 have dramatically changed. Editorials8–17 have been disparate. Some physicians, including myself, spoke of caution, while others advanced passionate diatribes against the “unconscionable and uncontrolled” usage of carotid stenting. A few physicians, loudly and clearly, made unsubstantiated pronouncements of its virtue as the primary treatment for extracranial carotid bifurcation obliterative disease. The subsequent verbal jousts often focused upon why should endarterectomy, “the gold-standard”, be challenged let alone potentially be eliminated from the surgeon’s repertoire; interrogatories often centered about the appropriateness of cardiologists entering into this surgical fiefdom of vascular disease. These interactions were not pleasant; however, the persistence, perseverance, tenacity, openness and forthrightness by these daring physicians has enabled this endovascular therapeutic application to evolve.18–24 Cardiologists, spurred onwards by their successful endovascular treatments of a variety of cardiac and non-cardiac endovascular pathologies, have vigorously championed carotid stenting as an alternative to carotid endarterectomy. Such pronouncements disturbed interventional radiologists and surgeons, who perceived this posture to be self-serving and gratuitous, as well as demeaning of these surgical specialties, their accomplishments, perspectives and concerns. However, the vociferous outcries of foul by these same specialties fortuitously not only failed to acknowledge as well as summarily dismissed the cardiologists’ contributions.25–28 The utilization of carotid stenting was initially considered apropos only for high-risk surgical candidates; its employment in ordinary risk surgical patients, outside of well-designed and carefully monitored experimental protocols, was deemed inappropriate and even unethical. Experience, time, and observational data accumulation now requires a re-consideration of these initial positions. The crucial interrogatory to consider is, “Do data indicate that carotid stenting should be limited to only high-risk surgical candidates, or should carotid stenting also be utilized for patients with severe extracranial carotid bifurcation obstructive disease who are reasonable surgical candidates?” A positive answer to this query has far reaching implications. Data assessment. For the last two decades, interventional cardiologists in conjunction with industry have routinely introduced new, unique, and brilliant therapeutic ideas, many of which supplanted standard procedures or therapies. More often than not, these novel therapies had better immediate results, fewer complications, and superior outcomes than those of standard therapeutic options. Carotid angioplasty’s immediate results and complication rates, even in the most difficult cases, remains equivalent if not superior to those achieved by carotid endarterectomy, and its follow-up results have demonstrated prevention of subsequent ipsilateral neurologic events. Roubin and Iyer’s non-randomized observational data15,16,18–23,30–32 underscore this premise. A total of 528 consecutive patients, in whom 604 hemispheres were treated by stent revascularization of the stenotic carotid artery, underwent the procedure with a mortality of 1.6%, 0.6% fatal stroke and a 1% non-stroke death rate at 30 days. The major stroke rate was 1%, and the minor stroke rate was 4.8%. The overall 30-day stroke and death rate was 7.4%. Over the 5-year study period, the 30-day minor stroke rate improved from 7.1% for the first year to 3.1% for the fifth year (p 80 years. After the 30-day period, the incidence of fatal and nonfatal stroke was 3.2%. The 3-year freedom from ipsilateral or fatal stroke was 92 + 1%. In the past, the quick and flippant dismissal by some cardiologists of carotid endarterectomy, the benchmark, against which carotid angioplasty had to be measured was irritating because of the superficial dismissal of carotid endarterectomy data. Furthermore, the numerous trials and publications concerning carotid endarterectomy data, for symptomatic and asymptomatic patients33–43 employed by opponents of carotid angioplasty, were considered and presented as unimpeachable data. Initially, all attention was directed towards procedural outcomes. When it became apparent that angioplasty could be performed as simply and safely as CEA, then the argument metamorphosed into the fact that angioplasty had not demonstrated the long-term prevention of subsequent ipsilateral neurologic deficit(s). However, the cardiology procedural proponents, in creating their protocols, had demanded independent neurologic patient assessment, which was appropriate as well as clever and fortuitous; however, what was initially not readily gleaned from the surgical literature was the repetitive absence of independent, immediate, and long-term neurologic assessment. Recent publications have indicated that actual endarterectomy complication rates were far different than those quoted. Wennberg44 detailed the variation in 30-day risk of stroke and death associated with carotid endarterectomy among 113,300 Medicare patients treated in 2,699 non-federal hospitals, during 1992 to 1993, of which 86 institutions (TRIAL hospitals), had participated in the NASCET and ACAS trials, and 2,613 institutions (non-TRIAL hospitals) had not been involved. Retrospectively, these two institutional groups’ perioperative mortality data of 113,300 carotid endarterectomy patients was compared. The 30-day mortality (Figure 1) was 1.9% in TRIAL and 1.7% in NON-TRIAL hospitals. NON-TRIAL hospitals were stratified according to procedural volumes, which revealed mortalities of 1.7% in high, 1.9% in average, and 2.5% in low volume cohorts. When comparing TRIAL to NON-TRIAL institutions, the mortality risk reduction was 15% for high volume (1.4% vs. 1.7%), 25% for average volume (1.4% vs. 1.9%), and 43% for low volume (1.4% vs. 2.5%) institutions. TRIAL hospitals had the best results, and procedural volume directly impacted perioperative mortality. However, these mortality (not stroke) statistics were higher than generally quoted by surgeons. Thus, the mortality risks recorded for a patient in the carefully controlled endarterectomy trials probably do not correspond to those mortality risks, which a patient must face when undergoing carotid endarterectomy by low or medium volume surgeons. If these major carefully controlled trial results represented the best surgeons, with the best track records, operating on the most carefully chosen patient cohort, then what makes an “occasional carotid-endarterectomy-surgeon” believe he/she can utilize them when addressing a patient, while failing to disclose his/her data? Furthermore, the generalized quoted statistics rarely are appropriate for the vast majority of potential surgical patients. Goldstein45 assessed the impact of potential perioperative risk factors on carotid endarterectomy results in 12 academic centers. A random sample of 1,160 asymptomatic endarterectomy patients had their charts abstracted with regard to perioperative complications. The perioperative stroke and death rate was 2.8% with the risk being similar for patients with cerebrovascular symptoms (1.8% vs. 4.2%; p = 0.21). However, the postoperative stroke and death rate was more than 3 times higher in women (5.3% vs. 1.6%; p = 0.02), more than 4 times higher in patients > 75 years (7.8% vs. 1.8%; p = 0.01), nearly 4 times higher in patients with congestive heart failure (8.6% vs. 2.3%; p = 0.03), and nearly 9 times higher in patients who had undergone combined endarterectomy and coronary bypass surgery (18.7% vs. 2.1%; p Data “check”. Roubin and Iyer’s numerous and carefully monitored studies, all with independent neurologic assessment, never focused upon or included patient assessment of the cognitive domains. However, their meritorious studies included many patients excluded from the major surgical trials (NASCET, ACAS, ECST). As such, their data demand indagnation with regard the poor surgical candidates category. Observers have assumed that these patients were excluded from the trials because their potential surgical risk was high, and, by implication and extension, that the angioplasty procedure risks were increased. This conclusion is a misinterpretation of the protocols’ intent. These patients were excluded from the surgical trials because their co-morbidity(ies) could potentially result in stroke. The difficulty avoided was the subsequent determination of whether a stroke was clearly attributed and causally related to the concomitant co-morbidity or the endarterectomy. Thus, inclusion of NASCET and ACAS ineligible patients into any study evaluating subsequent occurrences of neurologic events should be counterproductive. This is precisely the reason that Roubin and Iyer’s data demands examination, especially with regard to the extremely low incidence of neurologic deficit during the nearly 5 years of follow-up, despite the large number of NASCET excluded patients. The Carotid and Vertebral Artery Transluminal Angioplasty study (CAVATAS),48 involving 504 patients with high-grade symptomatic carotid stenosis who were randomized to carotid angioplasty [with stenting only in the case of a suboptimal result (only 55 /251 patients received a stent)] or carotid endarterectomy demonstrated no difference in procedural and long-term outcomes although, not unexpectedly, fewer complications were recorded with the interventional technique: cranial neuropathy (0 vs. 9%), and major groin or neck hematoma (1% vs. 7%). Amusingly, the angioplasty results were not better than anticipated, but the surgical results were worse than expected. These data were consistent with those of Wennberg and Goldstein. The American Heart Association’s guidelines for performance of carotid endarterectomy61 imply that certain levels of risk are acceptable to reestablish carotid blood flow. However, actual surgical data do not appear to be synchronous to the guideline recommendations. In fact, those of stent-supported carotid angioplasty appear to clearly be better than those of that minimal acceptable values outlined. Since the carotid artery is widely patent using stent-supported carotid angioplasty, and the follow-up data indicate that subsequent ipsilateral neurologic deficit is prevented, why should carotid endarterectomy be used in its place? Randomization is the only true methodology. The premise that randomization is “the only valid and proper scientific” way in which an appropriate conclusion can be determined is nonsense. Many correct conclusions are obvious and or intuitive. Randomized trials’ results cannot and should not be extrapolated as applicable for all patients and all medical situations. In fact, most medical problems can be resolved without the need of a randomized trial. A knife lying within a wound must be removed before repair; a lacerated blood vessel must be repaired to prevent further hemorrhage, and non-obstructed, antegrade arterial blood flow is better than impeded or absent flow. While the idea of not performing a randomized trial for stent-supported carotid angioplasty may be abhorrent to some, to me, today, this requirement is burdensome, absurd, and farcical. The espousement that these trials are absolutely and categorically necessary for the establishment of stent supported carotid angioplasty’s primacy is erroneous; furthermore, the demand that certain trials must be done so as to gain governmental approval for reimbursement for routine physician and hospital payments is horrific. While some small closed-end randomized trials may elucidate specific issues, the pronouncements are flawed that all issues, to be adequately resolved, require randomization. Who is making these demands for us to do these trials, and upon what basis? Are these trials necessary for us to determine what is the best care for our patients? Are the physicians promulgating these positions using this platform for self-aggrandizement, to maintain their patient base, to gain power or to become famous or notorious? Clinical trials, and clinical endpoints: “The Acronym Studies”. Appropriate study endpoints were necessary to be determined so that conclusions based upon data collected appropriately addressed the hypothesis of whether these devices were beneficial. Presently, data indicate that the risks of performing carotid angioplasty in the vast majority of cardiologists’ hands are similar if not superior to that of historical carotid endarterectomy series. Thus, why do we need to have these numerous acronyms entered into our vocabulary: CREST,70–72 SAPPHIRE, ARCHER, CARESS, SPACE, EVA-3S, or CAVATAS-2 or ICSS? For whose benefit are these trials? (Before answering the question, see how they are designed.) CREST (Carotid Revascularization Endarterectomy versus Stent Trial) addresses symptomatic, low-risk, surgical patients. SAPPHIRE and ARCHER are registries that address stenting and angioplasty with distal protection in patients at high risk for endarterectomy. CARESS is an exegesis of asymptomatic, or symptomatic patients within a registry format. SPACE is a randomized trial comparing stent-protected percutaneous angioplasty and carotid endarterectomy. EVA-3S compares endarterectomy versus angioplasty in patients with severe symptomatic carotid stenosis. CAVATAS-2 (or ICSS) compares primary stenting and endarterectomy. However, all these carotid stent studies have defined stroke, myocardial infarction, and death as the appropriate primary endpoints. Why? If the compelling data of Roubin and Iyer are accurate and substantiated by other observational series, and if major and minor stroke occur so infrequent in skilled hands with present carotid stent techniques, why were the above endpoints of these studies chosen and not since redefined? The response to this query of “it was too late and much too costly” seems unacceptable. Stent supported carotid angioplasty non-randomized observational series have shown that their success, morbidity and mortality rates were comparable, if not superior to CEA in all subsets of patients. Since a very large cohort would be required to demonstrate a significant difference between angioplasty and surgery, and if what we know is accurate, then why even do those studies? CREST will incur enormous costs, reaffirm the extant work, at some point “way into the future”, and, most painfully, subject numerous unsuspecting patients unnecessarily to the complications of carotid endarterectomy. Embolic protection devices. The long-term studies regarding neurologic problems resulting from coronary bypass graft surgery (CABG)47–51 provide insight into why the above major studies might reconsider their endpoints, especially with regard small particulate debris embolization. Vanninen’s47 study in which preoperative and postoperative CABG magnetic resonance imaging (MRI) of the brain, quantitative electroencephalography, and detailed neuropsychological and neurologic examinations concluded that small vessel ischemic lesions were detected and that alterations occurred in the quantitative electroencephalograms. Selnes,48–50 utilizing standardized neuropsychological tests preoperatively, and at 1 and 12 months postoperatively, detailed diminution in cognitive function. The decline in the neuropsychological tests, which assessed cognitive domains (attention, language, verbal and visual memory, visuoconstruction, psychomotor and motor speed, and executive function), found a significant late cognitive deterioration with postoperative neurologic and cognitive complications: stroke (2–5%), delirium (10–30%), and cognitive deterioration [short-term, 33–83% (involving primarily memory); and long-term, 42% (involving more subtle problems dealing with following directions, mental arithmetic, and planning complex actions as well as a lower frustration threshold, and broader mood swings)]. The late decline in the cognitive performance was attributed to manipulation and or cannulation of the aorta that resulted in a showering of atheroembolic debris within the cerebral circulation that resulted in occlusion of both small and large vessels. If a significant stroke had not occurred postoperatively, then, the small vessel occlusions, potentially resulting from embolic debris, could have been responsible for the subsequent cognitive decline. The rationale for use of these adjunctive embolic protection devices appears specious unless the neurologic and cognitive data points are included into these randomized study design. Why introduce a protection device into the protocol schema to prevent embolization, if it would or could not have a significant, potential, positive impact upon the defined endpoints? Are there other motivations? Have we forgotten that any additional device introduced into the carotid angioplasty procedure would have its own distinct set of complications? A study focusing upon the clinical impact of embolization after carotid angioplasty would be a long and expensive study, which is not within the plans of any of the medical vendors, any division of the NIH, or device inventors. Why? If embolic protection devices could reduce the incidence of procedurally related stroke (very difficult to detail) and preclude possible late cognitive decline, would that not be an important and crucial reason for a randomized trial? Physician training and experience. Curiously, why did radiologists not enthusiastically or even cautiously support and rally round the seminal works of Theron,1–7 Kachel,62,63 and Mathias64,65? Why have vascular surgeons only recently proclaimed their sovereignty over peripheral vascular disease and endovascular interventions, which can treat these problems, and, which, until recently, they preemptively dismissed? What has prompted them to change their posture? Why do these surgeons now believe that fiat or divine fate has provided them with the requisite skills necessary to readily perform these procedures, with little imaging and angiographic experience as well as minimal catheter-based training? Their dismissal of the guidewire and technical skills as well as the adjunctive catheter-based procedures, — learned by cardiologists and radiologists through performing coronary and peripheral angiography and angioplasty seems incredulous. Why have surgeons perfunctorily minimized the conclusions reached by interventional cardiology and radiology that excellent imaging systems with the facile ability to achieve oblique views and road-mapping, along with specialized training is necessary to attain those interventional skill sets so as to assure patients minimum levels of competence and expertise? How can physicians utilize imaging equipment in the operating theater, which would be unacceptable in any modern cardiac catheterization facility or radiology-imaging special procedures suite? Operating theater imaging systems exist, including cantilevered, moveable and rotational, radiolucent operating tables, which can provide the same level of quality as found in the catheterization laboratories and radiology special procedure suites. Why are they not being put into the operating theaters? Should patients be subjected to procedures using inferior equipment because of politics or finances? Furthermore, there is no justification for the routine use of general anesthesia to perform any of the standard interventional procedures including carotid angioplasty. Video transmitted live procedures should allow teaching-leaders to display to their audience readily available interventional equipment, standard techniques and methodologies, as well as the methods of successful extrication from an unforeseen occurrence, i.e. complication. Guidewire techniques should be emphasized: steering rather than pushing the guidewire through severe, tortuous, ectatic lesions; guidewire’s whose tips are bent or kinked or otherwise damaged should not be advanced, but rather removed; knowledge and vision of distal guidewire position must be emphasized before insertion of a catheter or stent delivery system; avoidance of procedures or techniques which have an associated unsatisfactory complication record, i.e., direct carotid cannulation. These are some of the obligations that the physicians who teach procedures must abide by, and they must forgo the opportunity of primarily demonstrating their unusual prowess and skill often to unsophisticated or novitiate audiences. Experimental or unusual devices may be displayed, but not at the expense of showing standard methodologies. All physicians are not equally gifted, and this fact was underscored at the 2001 American College of Cardiology Congress. A surgeon66 detailed his carotid angioplasty experience (mortality, 4.4%; stroke rate, 10.0%; TIA, 7.8%), and stratified his data with regard procedures in which distal protection devices were used (mortality, 2.5%; stroke rate, 7.5%; TIA, 5.0%) and those procedures without their use (mortality, 5.0%; stroke rate, 12.0%; TIA, 10.0%); the conclusion advanced was that carotid angioplasty was more dangerous than carotid endarterectomy (mortality, 0%; stroke rate, 2.0%; TIA, 1.0%). An alternative interpretation of these data, when comparing it to those presented by other cardiologists,67,68 might have been that the requisite skills necessary to perform carotid angioplasty were lacking. Shawl’s67 study of 225 patients under the age of 80 years had a mortality of 0%, a minor stroke rate of 3.0%, and a major stroke rate of 0.9%; and, for the 74 patients > 80 years, the mortality was 0%, the minor stroke rate was 1.3%, and the major stroke rate was 0%. Henry68 reported on 164 protected interventional procedures on 148 high-risk patients that there were 3 neurological complications (1.8%) of which two (1.2%) were major, and no procedurally related deaths. Since these data further demonstrate that physician skill levels may be very disparate, should not patients be advised of the physician’s actual level of expertise? Patients have the absolute right to know if someone else can perform a procedure safer and more effectively. These facts underscore all our fears about inadequately trained and educated physicians performing procedures. Physicians must have not only the necessary cognitive skills to determine which patients are procedural candidates, but a level of technical proficiency necessary to adequately perform such interventions, the capacity to anticipate, manage, and/or overcome the majority of difficulties encountered,69 and dedication. Somehow, a few unschooled physicians, for whatever motivation and without a skill set or knowledge base, including a paradigm to successfully extricate their patient from a predicament, have declared their intention to obtain privileges to perform carotid angioplasty, regardless of their capabilities and their patients’ needs. Perhaps they believe that a celestial warrant rather than “training” would convey upon them with necessary technical and cognitive abilities. Should we permit this? (Continued - See Part II of II)