Multiple randomized clinical trials and multicenter registries are under way to compare percutaneous with surgical strategies for the treatment of carotid stenosis. These include the evaluation of emboli protection devices, and, to a lesser degree, intravenous glycoprotein IIb/IIIa antagonists. Other clinical trials are aimed towards refining the ability to stratify patients by risk, in order to identify the subsets that would benefit most from these complex and expensive procedures.

Several randomized clinical trials (RCTs) have established the efficacy of endarterectomy for the prevention of stroke from carotid artery stenosis. ^1,2 The accumulated observational data and 1 randomized clinical trial suggest a comparable benefit-to-risk ratio with percutaneous methods. Ongoing basic and clinical trials are moving this field forward rapidly.

To keep things in perspective, it is critical to remember that the rapid pace of advances in drugs and devices constantly renders newly published data somewhat obsolete. Despite the incompleteness or unavailability of data for application to most clinical circumstances, we often cannot suspend making critical decisions or recommendations for the patient in front of us, and we must therefore rely on informed judgment, and on preliminary or indirect data.

There are many excellent published reviews on medical, percutaneous, and surgical approaches for carotid artery stenosis. ^3–7 In this article, we therefore focus on providing the reader a glimpse of developments that are likely to have significant clinical impact in the coming months or years.

The contemporary practice of carotid angioplasty and stenting (CAS) relies heavily on equipment and techniques developed for coronary artery percutaneous interventions. ⁸ This is not an optimal situation from several standpoints. In particular, differences in anatomic conditions mandate different diagnostic and guide catheter designs.

For example, the current practice of seating the guide catheter tip into the common carotid artery (CCA) at a point approximately 1 cm below a typically located carotid bifurcation lesion requires that catheters access and negotiate the origin of the innominate or left common carotid arteries. This can be a problem when the aorta forms an acute angle with the innominate or left CCA. This elderly cohort of patients with carotid artery disease also tends to have tortuous, calcified arteries that are not pliant.

Second, there is a higher absolute plaque burden (in comparison with smaller arteries), and the current non-customized, bulky equipment can dislodge plaque all the way from the aortic arch to the lesion. The brain, of course, is a highly complex organ that is exquisitely sensitive to embolic events.

Third, currently available stents require delivery systems of relatively large caliber that compound the aforementioned problems. And fourth, aside from obstacles to acute technical success, long-term failures in the form of restenosis remain a problem. This section enumerates promising developments that target these issues.

Catheters. Custom-shaped guide catheters are being developed to provide easier access to the origin of the cephalic vessels and to require less manipulation around the aortic arch. It is hoped that these can be tracked easily into the CCA, yet have adequate stiffness to be stable during passage of a stent from the arch into the innominate or left common carotid artery. Ultimately, an ideal catheter system should enable stent delivery with only minimal catheter-tip purchase beyond the aortic arch.

Stents. Balloon-expandable stents were the standard for carotid artery stenting, but concerns about crushing due to external compression have shifted preference to self-expanding stents. ^9,10 Recent iterations of the latter are made with Nitinol, a biocompatible, equiatomic composite of nickel and titanium that has super-elastic properties at body temperature. This means that Nitinol self-expanding stents change crystal structure from Martensite back to Austenite in the human body, thereby recovering their originally engineered shape upon deployment. ^11–13 Aside from crush recoverability, Nitinol-based stents offer dynamic scaffolding, and with certain designs, better wall apposition and conformability (Fig. 1).

Fig. 1 Comparison of Nitinol (A, B, C) with stainless steel stents (D), in regard to their conformability and wall apposition.

Ongoing developments that focus on further miniaturization of equipment have enabled lower-profile stents and delivery systems (Table I). Other experimental stent modifications may offer more favorable biological properties, including mitigation of thrombosis, attenuation of cellular activation and proliferation, and enhancement of endothelial-cell and vascular-wall function (Table II).

Table I. Stent Profiles

Table II. Experimental Stents

Distal Emboli Protection Devices. Plaque débris dislodged during CAS can cause irreversible cerebral infarctions. ¹⁴ These have been documented in percutaneous interventions in different arterial beds; particles liberated during CAS have ranged in size from 16 micrometers to 6.2 millimeters. ¹⁵

Emboli protection devices (EPDs) of several classes are currently undergoing clinical evaluation and should be a major breakthrough in making percutaneous vascular procedures safer. Filters and soft occlusive-balloon catheters are advanced past the lesion to prevent downstream escape of particles (Fig. 2). Embolic débris is retrieved by collapsing the filter after CAS or by catheter aspiration prior to balloon deflation. ¹⁶ Parodi is developing a circulatory control device system that creates temporary blood flow reversal from the carotid artery into a suctioning catheter positioned proximal to the carotid artery target lesion. ¹⁷

Fig. 2 Composite schematic of a filter-based emboli protection device

There are important pitfalls to the current devices under investigation. None of these devices offers protection from embolic phenomena during the diagnostic catheterization phase, where up to 1% of strokes occur. ^1,18 Also, to be positioned, these devices themselves must traverse the aortic arch and the proximal great vessels, which may likewise be diseased. ¹⁹ Even when these devices are in place, the particle capture efficiency is far from perfect. ²⁰ Finally, the use of EPDs usually adds procedural time and complexity to a technique in which manipulation should be minimized. The potential advantages and disadvantages of each class of EPD are tabulated in Table III.

Table III. Embolic Protection Devices: Potential Relative Advantages and Disadvantages

Randomized clinical trials in the coronary literature have established the central role of antiplatelet agents in decreasing peri- and post-procedural adverse events. Intravenous oral glycoprotein IIb/IIIa antagonists decrease 30-day thrombotic events by approximately 30%. These results may not be amenable to extrapolation to CAS because large-vessel behavior may differ from that of coronary arteries, and the proportion of thrombus-based (versus plaque-based) emboli may be less in CAS (as indicated by material obtained from EPD studies). Furthermore, clinical observations have noted that peri-procedural stroke almost always occurs in the laboratory rather than at a later time. ^21,22 On the basis of a small consecutive series of 70 CAS patients who received adjunctive abciximab bolus and infusion, Kapadia has suggested that periprocedural events in high-risk patients might be mitigated without an increase in bleeding complications. ²³ Larger studies are needed to see if the benefits of this strategy outweigh the higher risks of intracerebral hemorrhage in this older population with a high prevalence of cerebrovascular comorbidities.

Similarly, the oral antiplatelet regimen for CAS is directly adapted from what is known from coronary stenting. ^24,25 Whether this dual coverage with aspirin and short-term clopidogrel or ticlopidine is optimal, or indeed overdone, with regard to drugs, dosing, and duration remains to be established. There is even evidence to suggest that oral antiplatelet drugs might, at early stages of carotid artery disease, attenuate the progression of carotid artery intima-media thickness.

A large potential exists for neuroprotective agents to play a role in the management of stroke patients, and possibly as adjunctive prophylaxis for the particulate emboli that are almost an inevitable consequence of CAS. When ischemia occurs, breakdown of ionic gradients results in membrane depolarization and an outpouring of excitatory amino acid neurotransmitters, particularly glutamate. Glutamate interacts with NMDA and non-NMDA receptors, which increases calcium conductance and raises intracellular calcium concentration. A series of cellular events results in activation of lipases, proteases, and nitric oxide synthetase, and in generation of nitric oxide and oxygen free radicals, which eventually leads to cellular death. Neuroprotective agents interfere with 1 or more steps in this ischemic cascade. Several neuroprotective agents target the NMDA receptors either by inhibition of glutamate binding or blockade of the glycine site, a cofactor in activation of these channels. Others interfere with the nitric oxide pathway (lubeluzole) or act through stabilization of membranes (citicoline). In animal models of stroke, neuroprotective agents have been effective when administered as long as 6 hours after onset of stroke, but are most effective when given immediately or within a few hours of onset. ²⁶ Several agents showed promise in phase II trials, ^27–30 but to date all phase III randomized controlled trials of neuroprotective agents have failed to significantly improve outcome. ^31–34

There are many potential reasons for the failure of these studies to demonstrate a significant clinical benefit, despite promising preclinical and phase II results. One of the reasons is the long interval before treatment, which in most cases was 6 hours or later. The use of neuroprotective agents in the setting of carotid stenting might provide an optimal application for this approach to treatment, because treatment can be given almost immediately after the onset of neurologic deficits or potentially can be given prophylactically and be on board when the stroke occurs. The preclinical studies suggest that this very early treatment might reduce neurologic deficits from stroke, even though such a benefit has not been demonstrated in randomized controlled trials of patients treated hours after the onset of stroke.

Since minimal manipulation is a central principle for avoiding embolic complications, there are advocates of direct or primary stenting, which entails stenting without preceding balloon dilation. ^35,36 This strategy tests whether the avoidance of balloon predilation can improve procedural outcomes, and at the same time offer cost savings by skipping this step. A case series of carotid stenting, performed by European vascular surgeons, used this strategy with apparently acceptable 30-day outcomes. However, without vascular preparation with predilation, a relatively high profile stent may be expected to incur more vascular trauma, and indeed dissection (2%) and vasospasm (1%) were observed in this study. ³⁷ One RCT that routinely applied this technique was prematurely terminated because of unusually high event rates in the stent group, although the roles of final balloon sizing, other technical factors, learning curve, or simple chance are not clear. ³⁸ Whether there may be more or less embolic débris liberated using this approach needs to be cautiously studied now that lower-profile stents are available. Even then, this should start with the few patients who have carotid artery lesions with lumina that are unquestionably wider than the stent profile.

The concept of plaque trapping, sealing, and stabilization independent of relief of stenosis may eventually prove to be important. ⁸ We have hypothesized that plaque stabilization is initially attained via coverage with the stent, with eventual quiescence of the lesion achieved by fibrosis. In the future, operators may become more tolerant of post-CAS residual stenosis, especially with the use of self-expanding stents, in exchange for safer acute outcomes.

Randomized clinical trials now under way will provide us valuable information regarding the appropriate role of CAS relative to carotid endarterectomy (CEA) or medical therapy (Table IV). Just as important will be prospective registries that include the large majority of patients, who otherwise do not qualify for RCTs. These are the Carotid Revascularization with Stenting Systems (CARESS), which is a nonselective registry sponsored by the International Society of Endovascular Specialists (ISES), and the Acculink™ (Guidant) for Revascularization of Carotids in High Risk Patients (ARCHeR) study, which tracks the high-risk patients who were not eligible for CREST (Carotid Revascularization Endarterectomy versus Stenting Trial). Many different subanalyses will be performed to identify the patient subsets that derive the greatest benefit and the clinical parameters that predict high risk for procedure-related complications. Rothwell has shown that most of the long-term benefits from carotid endarterectomy (CEA) are clustered in only a fifth of the patients in the European Carotid Surgery Trial (ECST), i.e., in those with 70% to 99% stenosis. ³⁹ This implies that up to 80% of patients subjected to the perioperative risks of CEA did not derive a net benefit. Refinement of our knowledge of prognostic factors will be critical in assisting both clinicians and patients to ascertain individualized benefit-to-risk ratios, and will help determine the most beneficial strategy (medical therapy, CAS, or CEA) for each patient. A prospectively validated model for stratifying patients according to risk of periprocedural death or stroke is already available for CEA. ⁴⁰ Once large data-sets are available from both registries and randomized clinical trials, prediction models can be constructed to identify those patients at high risk for complications with either CEA or CAS, leading to appropriate consideration of the alternative treatment.

Table IV. Randomized Clinical Trials Comparing Carotid Angioplasty and Stenting to Carotid Endarterectomy

The availability of the genetic sequences for man (the Human Genome Project) and for key model organisms such as Escherichia coli, the zebra fish (Danio rerio), the fruit fly (Drosophila melanogaster), and the mouse (Mus musculus) will stimulate the development of new paradigms for the prevention, diagnosis, and treatment of human diseases. ⁴¹ Early identification of people at high risk for atherosclerotic disease may help target efforts at prevention. For instance, there is evidence that suggests a higher prevalence of severe carotid disease among patients with renovascular hypertension who carry the DD genotype of the angiotensin-converting enzyme (ACE) gene. ⁴² For arterial restenosis, studies have focused on polymorphisms of genes encoding for proteins that interfere with lipid metabolism, hemostasis, nitric oxide production, inflammatory mechanisms, smooth-muscle-cell proliferation, and matrix production. ^43–46 Genes encoding for angiotensin-converting enzyme, platelet glycoprotein-IIIa, transforming growth factor β (TGF-β), and stromelysin-1 are genetic candidates suspected to play a role in restenosis. ^47–53 Despite recent setbacks in human trials, proof of principle has been achieved for gene therapy in the abrogation of smooth-muscle-cell growth in the rat model for carotid artery balloon injury, which foreshadows the immense promise of this field. ^54–57

Ultimately, prevention is always the best approach. Initiatives to educate patients and their families about the crucial role of proper diet, exercise, and abstinence from smoking remain a priority, not only for public health agencies but for the individual health-care practitioner. Technological development of less invasive testing methods will also play an increasing role in health care. For instance, noninvasive indicators of early carotid artery atherosclerosis, such as the evaluation of intima-media thickness by B-mode ultrasound, have been associated with an elevated risk for future ischemic stroke. ⁵⁸ This finding also serves as a correlate of atherosclerosis in other vascular trees. ⁵⁹ Indeed, the major cause of death in patients with carotid artery stenosis is myocardial infarction and heart failure, and intima-media thickness has been shown to be predictive of cardiovascular events as well. ⁶⁰ Our approach to carotid artery stenosis and stroke will someday be only a component of a global approach to atherosclerotic disease, in our efforts to offer our patients a comprehensive perspective on preserving their health and longevity.