The Growing Role of Carotid Artery Stenting
An innovative procedure is ready for prime time.
Carotid artery stenting (CAS) for occlusive disease is growing at a rate of 20% per year. Unfortunately, there are fewer than 60 centers performing the procedure in the US and as a result, the number of CAS procedures being performed is almost 4 years behind schedule in terms of earlier predictions.
It is only recently that Medicare’s revised policy has agreed to consider reimbursement for CAS, and will do so only when the procedure is performed under an approved investigational device exemption (IDE)-sponsored trial. Currently, there is one NIH-funded trial and eight industry-sponsored high-surgical-risk registries, one of which included a randomized high-risk arm. Trial data, however, do not tell the whole story regarding the success of CAS versus the risks of surgery.
COMPARING CEA AND CAS
When carotid endarterectomy (CEA) was initiated in 1976, the procedure was reimbursed and performed in the absence of any trial. The North American Symptomatic Carotid Endarterectomy Trial (NASCET) was initiated in 1988 and completed in 1991. Investigators enrolled symptomatic patients with >70% stenosis of the internal carotid artery and randomized them to receive either CEA or best medical management. The periprocedural stroke and death rate in the trial was 5.6%. There was a cumulative 2-year stroke rate of 9% in CEA patients versus 26% in medically managed patients.
What is less well understood from the NASCET trial was that a 7.8% incidence of cranial nerve palsy developed in the surgical group. Significant neck hematomas were also described in 5.5% of the patients. This was a low-risk trial considering that there was an extensive list of exclusion criteria. The only high-risk criteria included patients with ipsilateral stenosis and contralateral occlusion. In that relatively high surgical subset, investigators noted a 14% perioperative stroke and death rate.1,2
The European Carotid Surgical Trial (ECST) described a perioperative stroke and death rate of 7.1% and a rate of 6.4% for cranial nerve palsy. In a multicenter study, Chaturvedi et al reported an 11% stroke and death rate following the procedure, describing those results as the “real world” of CEA.3 During this same period, single-institution centers began reporting their data following CAS in a high-risk surgical patient subset. Most of these patients would not have been eligible for enrollment in either the NASCET trial or the Asymp-tomatic Carotid Atherosclerosis Study (ACAS).4 In our experience with 640 patients, only 9% stented at the Pittsburgh Vascular Institute (PVI) would have been eligible for enrollment in either the NASCET or the ACAS groups.
Both single- and multi-institutional studies are now reporting a perioperative stroke and death rate <4% utilizing CAS. At PVI, we reported a perioperative stroke and death rate of 3.8%. In our most recent 200 patients, the stroke rate has been reduced to 2% when utilizing distal protection.
RESTENOSIS NOT A CONCERN
Restenosis for endovascular carotid stents is not an issue because the vessel is ideally suited to a high flow rate, excellent outflow, and a low resistive system. The restenosis rate at 1 year is now less than 5% in most centers.5–7 Only 3% of patients at PVI require repeat angioplasty for restenosis, based on a mean follow-up of 26 months.
A multicenter carotid World Registry has enrolled 8,600 patients from 42 centers in the US, Europe, and South America. This biannually updated study currently reports a 30-day stroke and death rate of 5.8%, of which 1.02% was non–procedure-related deaths. Technical success in the study was 98.5%, with a rate of 1.3% for major strokes and 2.7% for minor strokes.8
Although the World Registry lacks the validation of a randomized trial, a comparison with the NASCET data suggests that in the stenting subgroup, an event rate of 5.8% compares closely with the NASCET 5.6% stroke and death rate. The study was further divided into the symptomatic and asymptomatic groups; the asymptomatic group experienced a 2.9% event rate that compares favorably with the ACAS data of 2.3%. These data are even more significant when one considers that most of the patients who underwent CAS had significant high-surgical-risk criteria in contrast to those in the low-risk ACAS and NASCET studies.
If the 8,600 patients from the World Registry data were combined with more than 3,000 patients from the carefully monitored high-surgical-risk CAS registries, data would be available in more than 11,000 patients. This amount of data should be sufficient to overcome the resistance to CAS and allow full reimbursement for the procedure as a service-covered admission. Yet resistance continues.
Several factors account for the problem of resistance. In the earliest stages, the FDA declared CAS a risk procedure that should not be performed without an approved IDE trial. A feasibility study was initiated and followed by a randomized clinical trial enrolling defined high-surgical-risk patients to CAS versus CEA. This was then followed by six registries also based on high surgical risk entry criteria. The high-risk criteria included (1) lesions that were surgically inaccessible, (2) prior radiation, tracheotomy, or radical dissection, (3) ipsilateral stenosis and contralateral occlusion, (4) prior CEA with restenosis, (5) significant carotid artery occlusive disease and an urgent need for aortocoronary bypass surgery, (6) internal carotid artery tandem lesions, and (7) a number of medical high-risk comorbidity factors that included Class III or Class IV CHF, LVEF of <30%, open heart surgery within the past 6 weeks, recent MI, unstable angina, chronic obstructive lung disease with an FEV<1, and patients with renal failure undergoing hemodialysis.
In fact, these high-risk medical comorbidity criteria represent extreme risk. Consequently, there are few patients that meet the comorbidity criteria that would allow enrollment under the previously mentioned categories. Most patients selected for the high-risk registries are entered for anatomical indications. The randomized component has difficulty in enrolling patients because most either refuse surgery or are not candidates for CEA. Only 20% to 25% of the CAS patient population has high-risk criteria that would enable enrollment in one of the high-risk registries.
One exception to the high-risk registries is the recently introduced Carotid Revascularization With Endar-terectomy or Stenting Systems trial (CARESS). The CARESS trial represents a parallel registry allowing either CEA or CAS in a study comprising 2,000 patients with 1,000 patients in each arm. Initial enrollment is based on a 2:1 ratio of CEA versus CAS for the first 350 patients. Because there are no major restrictions, this registry represents another “real world” look at CEA and the alternative, CAS. The remaining 15% to 20% of patients may be eligible to participate in the symptomatic Carotid Revascularization Endarterectomy vs. Stent Trial (CREST).9
RESISTANCE FROM WITHIN
Additional resistance to CAS comes from delays in the Internal Review Board processing of submissions, a limited number of CAS centers, and the early absence of distal protection devices as an adjunctive component for the endovascular stenting procedures. The acceptance of CAS has also been held up by vascular surgeons’ reluctance to enroll in either the high-risk trial or the relatively low-risk CREST trial. If the randomized trials are to be successful, the surgeons must be willing to enroll their patients and, to date, this has not occurred.
The need for randomization is an admirable but rarely attainable goal. Device developments, particularly those pertaining to carotid stents and distal protection devices, are in continual evolution. This is especially true in device trials. There is little question that in the high-surgical-risk subset, it may be the preferred method. By the time a randomized trial is completed, it would already be historically obsolete. The technology is not stabilized, learning curves will influence the results, and case selection is frequently biased. Carefully monitored, multicenter observational studies correlated closely with randomized clinical trials could enroll larger numbers in a more timely fashion, and would be considerably less expensive.
IMPROVED DISTAL PROTECTION
Despite the delay in the spread of CAS, significant progress has occurred in the profile of the delivery systems. Stents have improved from 7F catheter systems to currently 5F, and third-generation stents are now in a sub-5F platform. Either a 6F catheter sheath or an 8F catheter guide can be used, allowing adequate space for injection of contrast media. Other recent developments include guiding catheters that the physician can position at the carotid ostium. Such catheters eliminate the need to pass the guide across the common carotid to the bifurcation and avoid embolic events secondary to guide manipulation.
Distal protection devices have been developed in either a balloon configuration or a filter design in order to minimize embolic events during CAS procedures. Few interventionalists would proceed without distal protection, and protection devices are mandatory for all carotid trials.
The World Registry described a 4.2% perioperative stroke and death event rate in 1,596 patients without distal protection. This organization noted a 1.7% incidence of stroke with distal protection in 771 patients.10 At the PVI, we observed a 3.5% stroke rate in 444 patients without distal protection compared with a 1.3% event rate in 78 patients with distal protection.
The first-generation distal protection devices represented major advances. Unfortunately, the 4F and 5F catheter profiles were technically limited in accessing across tight internal carotid target lesions. Second- and third-generation distal protection devices now being entered into clinical practice have solved most of these access problems. These devices are currently available in 2.5F and 3F catheter configurations that track quite easily and have satisfactory particle capture efficiency.
FILTERS AND ANTICOAGULANTS
Filters may decrease flow in an already compromised carotid artery. Therefore, it is important that careful anticoagulation is observed with an ACT of 300 or above.11,12 During elective CAS it is rarely necessary for the physician to add a glycoprotein receptor antagonist to the anticoagulation regime. However, the impressive results with IIb IIIa receptor antagonists in acute coronary syndromes may not transfer to carotid application.13–15 The effects of aging, hypertension, and prior hemispheric strokes, as well as poor vessel intracranial adventitial support, the fragile blood/brain barrier, and the potential for reperfusion intracranial bleeding are all variables that affect the use of receptor antagonists in CEA procedures.16,17
Although the risk of intracranial bleeding is less than 1%, it does not justify the reward. In our experience with 202 patients receiving either abciximab (ReoPro, Eli Lilly and Company, Indianapolis, IN) or eptifibatide (Integrilin, Millennium Pharmaceuticals, Inc., Cam-bridge, MA), we noted a 4.4% stroke and TIA incidence versus 3% without the use of platelet receptor antagonists. Furthermore, there were two intracranial bleeds in the IIb IIIa subset. In the World Registry, 1,335 cases were reviewed from six centers describing six intracranial bleeds. The overall complications rate in that series including hematomas was 11%. Our policy is to consider platelet receptor antagonists only in nonhypertensive patients with extremely complex types of lesions who have not had a prior stroke.
Since the advent of distal protection devices and the improvement in dedicated carotid artery stents and delivery systems, complications have become less frequent (Figures 1 through 3). When they do occur, a basic understanding of neurovascular anatomy and interventional neurorescue techniques is necessary.18,19
Complications are frequently related to the duration of the procedure. This was again confirmed in a multicenter World Registry suggesting that neurologic events could be reduced from 10% to 4% by experience with at least 50 patients.8 These data, however, include patients treated prior to the availability of distal protection.
Certain types of lesions carry a higher risk of complications. At the PVI, we have noted that lesions >2 cm in length result in three times more neurologic events than lesions measuring <1 cm in length (personal observation over 4 years, from 1998 to 2002). Polypoid or globular lesions rarely result in an acceptable stenting procedure, and
in our experience are best managed surgically. This is also true when there is a suspicious clot.
The algorithm that we follow in managing neurologic complications is as follows: If a TIA or minor stroke occurs and the angiogram is normal in both the arterial and venous phases, patients are treated conservatively with a conventional anticoagulin.7 Most of these patients will recover and return to baseline within 24 hours, assuming intracranial circulation is also normal.
Patients who have had a minor or major stroke with positive occlusion noted on the angiogram are managed more aggressively. The physician may utilize thrombolytics via the positioning of a microcatheter at the clot site, IV platelet receptor antagonist IIb IIIa, or a low-profile 1.5-mm angioplasty balloon (assuming the thrombolytics were unsuccessful in recanalization).
STENTING IS PREFERRED
CAS is ready for prime time. There is little question that, in the high-surgical-risk subset, it may be the preferred method. With the addition of distal protection and improved stent configurations, CAS is also a viable alternative in the presence of most other indications. An increasing number of patients are requesting CAS rather than CEA.
Several issues surrounding CAS remain unanswered. The procedure is not yet available at community hospitals, and reimbursement is denied to a significant number of patients who are not surgical candidates. Sending these patients home because the procedure is a non–service-covered admission is assigning them a stroke warrant.
Finally, carotid randomized clinical trials in the US have been unable to enroll the necessary patients. Patients in the high-risk category either refuse surgery or are too high-risk to undergo the procedure. For low-risk patients, surgeons are not convinced that CAS equals CEA.
Most trials include only patients at high risk for CAS. The data for high-risk asymptomatic patients are basically the equivalent of the low- to moderate-surgical-risk data from either NASCET or ACAS. A randomized clinical trial for low- or moderate-risk patients is unlikely to succeed without surgeons who are willing to enroll their patients. Therefore, it would appear reasonable to accept a carefully monitored muticenter registry to gather this information.
Mark H. Wholey, MD, is Chairman of the Pittsburgh Vascular Institute in Pittsburgh, Pennsylvania. Dr. Wholey may be reached at (412) 623-2083; firstname.lastname@example.org.
1. Taylor DW, Barnett HJM. North American Symptomatic Carotid Endarterectomy Trial Collaborators. Beneficial effect of carotid endarterectomy in symptomatic patients with high-grade stenosis. N Engl J Med. 1991;325:445-453.
2. Moore WS, Barnett HJM, Beebe HG, et al. Guidelines for Carotid Endarterectomy: A multidisciplinary consensus statement from the ad hoc committee, American Heart Association. Stroke. 1995;26:188-200.
3. Chaturvedi S, Aggarwal R, Murugappan A. Results of carotid endarterectomy with prospective neurologist follow-up. Neurology. 2000;55(6):769-772.
4. Asymptomatic Carotid Atherosclerosis Study Group. Endarterectomy for asymptomatic carotid artery stenosis. JAMA. 1995;273:1421-1428.
5. Yadav JS, Roubin GS, Iyer S, et al. Elective stenting of the extracranial carotid arteries. Circulation. 1997;95:376-381.
6. Wholey MH, Wholey M, Bergeron P, et al. Current global status of carotid artery stent placement. Cathet Cardiovasc Intervent. 1998;44(1):1-6.
7. Dietrich EB, Ndiaye M, Reid DB. Stenting in the carotid artery: Initial experience in 110 patients. J Endovasc Surg. 1996;3:42-46.
8. Wholey MH, Wholey M, Mathias K, et al. Updated global review of carotid artery stent placement. Cathet Cardiovasc Intervent. 2000;50:160-167.
9. Hobson RW, Brott T, Ferguson R, et al. CREST: Carotid revascularization endarterectomy versus stent trial. Cardiovasc Surg. 1997;5(5):457-458.
10. Wholey MH, Wholey M, Mathias K, et al. Global experience in cervical carotid artery stent placement. Cathet Cardiovasc Intervent. 2000;50(2):160-167.
11. Al-Mubarak N, Roubin GS, Vitek JJ, et al. Subarachnoidal hemorrhage following carotid stenting with the distal balloon protection. Cathet Cardiovasc Intervent. 2001;54:521-523.
12. Reimers B, Corvaja N, Moshiri S, et al. Cerebral protection with filter devices during carotid artery stenting. Circulation. 2001;104(1):12-15.
13. Hoffman LV, Razavi M, Arepalli A, et al. IIb IIIa Receptor Inhibitors: What the interventional radiologist needs to know. Cardiovasc Intervent Radiol. 2001;24:361-367.
14. Chamorro A, Vila N, Obach V, et al. A case of cerebral hemorrhage early after carotid stenting. Stroke. 1999;31:792-793.
15. McCabe DJH, Brown MM, Clinton A. Fatal cerebral reperfusion hemorrhage after carotid stenting. Stroke. 1999;30:2483-2486.
16. Morie T, Fukuoka M, Kazita K, et al. Intraventricular hemorrhage after carotid stenting. J Endovasc Surg. 1999;337-341.
17. Schoser BG, Heesen C, Eckert B, Thie A. Cerebral hyperperfusion injury after transluminal angioplasty of extracranial arteries. J Neurol. 1997;244:101-104.
18. Satler LF, Laird JR. Limiting the complications of carotid stenting. Cathet Cardiovasc Intervent. 2001;54:524-525.
19. Wholey MH, Wholey M, Tan WA, et al. Management of neurological complications of carotid artery stenting. J Endovasc Ther. 2001;8:341-353.