Reply

Dr. Roubin's comments concerning the AJNR commentary on carotid artery balloon angioplasty and stenting (CABAS) are welcome (1). Dr. Roubin is known as a thoughtful and innovative investigator, and he is commended for his approach to this evolving technology. He has treated a high-risk patient population, and yet has shown results competitive with surgical carotid endarterectomy (CEA). He understands the risks involved. Recently he has been quoted as saying, “It is important that a realistic and comprehensive classification of the neurologic complications be used when a newly evolving technique such as carotid stenting is evaluated.” He has also asserted that “independent oversight when the incidence of neurologic complications is assessed cannot be overemphasized (2).” Unfortunately, these rules are not uniformly adhered to by all performing either CABAS or surgical endarterectomy.

His summary of events since the formulation of the previous commentary (1) provides an update of this fast-moving field. Dr. Roubin anticipates that the European CAVATAS trial, comparing angioplasty (with or without stenting) and CEA, will show no difference between either treatment approach or minor and major neurologic events. One hopes that publication of those results will include sufficient case material regarding stenting and an adequate categorization of degree of stenosis to provide guidance. There also will be a need to look carefully at patient data to consider whether this study will be applicable to practice in North America.

Dr. Roubin also suggests that the rigor of the well-planned, IRB-approved clinical trials begun at his former institution, the University of Alabama at Birmingham, was demeaned in the commentary (1). This is not what was written. In fact, the data from the referenced study were outlined in the commentary in a factual and favorable manner. That study, which included pre- and postprocedural neurologic evaluation, complete cerebral angiography, and appropriate use of neuroimaging follow-up, serves as a model for similar efforts. It is clear that Dr. Roubin's multidisciplinary group, which included an interventional neuroradiologist, deserves much credit for having the vision and drive to show that CABAS is a competitive technique.

Controversy in the initial phases of development of a new therapy is common and should not dissuade active pursuit of a potentially valuable technique. Nevertheless, the practice that prompted the commentary is the increasingly widespread promotion and performance of CABAS by those less experienced, less thoughtful, and less competent than Dr. Roubin and his colleagues, before there is scientific proof of its safety and efficacy. It is natural that groups of physicians with appropriate individual or team expertise will continue to investigate this procedure, considering encouraging short- and intermediate-term outcome data already available. We must realize, however, that the favorable data to date comes from a few highly skilled teams. Others who also perform CABAS may not have adequate training and experience, and may not have comparable results. One recent report in the Journal of Vascular Surgery suggests that the learning curve for CABAS may be quite steep in some circumstances, despite internal institutional control of a carefully designed study (3). This recent European single-center study, meant to be a randomized comparison of CABAS and CEA, was halted when five of seven patients who underwent CABAS had a stroke.

Dr. Roubin also writes that the “statement that no long-term outcome information is available is incorrect,” and he goes on to say that “late results have been reported and multiple groups now have confirmed the rarity of neurologic events after stent intervention.” In support, he cites one reference to his own published abstract presented in 1996 (4). Despite that study's title, the abstract refers to only 22 cases maximally followed, and these were only followed for 6 months. For stroke disease prevention, this is characterized best as little more than short-term and certainly far from long-term. That this patient group was followed longer term is not indicated in the abstract; if that was done subsequently, the accurate peer-reviewed publication reference is anticipated. It is wondered, in fact, whether the patients were incorporated into a later (1997) report by the same group, discussed in the commentary (5), dealing with Palmaz stent collapse in 11 (16%) of 70 cases, requiring repeat angioplasty in five and stenting with an alternative device in three. The carotid is an artery that is apt to be subjected to a degree of mobility and torsion greater than others for which stents are approved for use, and caution about enthusiastic promotion before completion of randomized trials is appropriate.

The general call by many for long-term follow-up also relates to published data regarding the real incidence of restenosis after stenting in the carotid and elsewhere; an incidence perhaps higher than that observed with CEA (6–13). Also, we should remember that today's “best medical therapy” has advanced beyond that available in the past and that neither CEA nor CABAS have been compared with this advancing target.

It should not be forgotten that the target of CABAS therapy is the brain, not the neck, kidney, or heart. The most important risk of this procedure is atherothrombotic cerebral embolization, whether from clot formation on instruments or debris from treated plaques. Dr. Theron, a neuroradiologist whom Dr. Roubin properly credits as a CABAS innovator (14), has presented results showing that angioplasty and stenting shower particulate material with the potential to damage the brain (Annual meeting of the Radiological Society of North America [abstract], December 1998). From personal observation, intraprocedural and postprocedural neurologic testing is employed variably in the community during the performance of CABAS. Because avoiding, detecting, and treating any complication is a fundamental aspect of the practice of any medical procedure, direct intracranial endovascular rescue, which requires special expertise, should be an available option at any institution performing CABAS. In recognizing the essential role that the neuroradiologist plays in CABAS, Dr. Roubin also clearly appreciates the training and expertise that a neuroradiologist possesses. This is important for performing and interpreting the diagnostic neuroangiogram, and for performing intracranial vascular navigation and intra-arterial stroke rescue.

Dr. Roubin provides a welcome announcement of the recent approval of funding for the Carotid Revascularization Endarterectomy versus Stenting Trial (CREST) in North America. This landmark event, important and welcome, is in no small part because of the efforts of Dr. Roubin.

One hopes that within fewer than the 10 years required for NASCET, CREST will produce solid, randomized, clinical outcome data comparing appropriate parameters of nondisabling stroke, disabling stroke, and death for CABAS and CEA. The CREST trialists have an important responsibility to assure that their case material is at least as rigorous as the landmark CEA trials, with patient populations clearly defined.

With the CREST approval, there is the opportunity to caution that protocol methods be very precise in the definition of carotid stenosis measurement and other clinical parameters, striving to go beyond some limitations inherent in previous studies. For example, despite a well-organized and detailed method for verifying duplex sonographic assessment for carotid stenosis in centers participating in the ACAS trial, the ACAS data indicate that 8% of patients were declared ineligible for CEA after angiographic verification of the degree of stenosis (15). Most of this discrepancy was likely because of inadequate correlations between sonography and angiography, despite detailed planning.

For those patients who are excluded from conventional angiographic study in the CREST, it is hoped that an improved correlation with angiographic stenosis assessment beyond that of ACAS is being developed by CREST organizers for two reasons. First, improved correlation would enable the sufficient recruitment of CREST collaborators. Second, this improvement would help clarify the ambiguities others have highlighted in the multiple methods available for stenosis measurement. Although CREST investigators expect a “NASCET-like” patient population, how these patients will be categorized in CREST for various parameters, including stenosis measurement, needs to be clear. Many authors during the last 5 years have referred to the so-called “NASCET method” of stenosis assessment from angiography without attention to the details of how the measurement was actually carried out in NASCET (6–20).

A commonly overlooked detail in NASCET methods is the assessment criteria of “near occlusion.” Accurate assessment of “near occlusion” is important. This group of patients is at the highest risk for intervention, whether by CEA or CABAS. Skewing the patient population in either arm can alter the results of any trial. Measurement of the distal “normal” internal carotid, of course, will be misleading if the diameter has begun to decrease because of a critical proximal stenosis. When deemed appropriate to measure, the distal normal carotid was measured in NASCET “where the walls [were] parallel,” another detail often not adhered to in publications claiming to use the “NASCET protocol.”

Although there has been some lack of attention to detail in the literature regarding stenosis measurement, one hopes that CREST trialists will be very rigorous in detailing stenosis categorization. The collaboration of physicians from multiple disciplines, including some without the tradition of detailed evaluation of cerebral and cervical vasculature, demands a precise method of stenosis measurement. This precision should ensure that “NASCET protocol” stenosis measurement begins with a thorough assessment of the intracranial circulation from all potential routes in order to uncover subtle signs of near occlusion that are frequently encountered. Without this, the future results of CREST cannot be considered a true comparison and extrapolation from NASCET. Now that the NINDS has decided to support CREST, all those who have an interest in CABAS and CEA need to support the trial, and join to maximize its validity.

Dr. Roubin provides details concerning the preliminary registry studies for CABAS that have supported the planning and acceptance of the randomized CREST. He points out the importance of the basic data provided by the North American Cerebral Percutaneous Transluminal Angioplasty Registry (NACPTAR), but provides no references indicating where readers may review the data. Such studies are important but do not replace the need for the CREST, as Dr. Roubin is well aware.

According to the American Heart Association document, coauthored by Dr. Roubin and referred to in the initial commentary, “the technology and techniques for carotid angioplasty and stenting are available, as are a limited degree of experience and a high level of interest (21).” The relatively promising data this procedure yields at certain institutions is commendable and encouraging. Nevertheless, it must be appreciated that these results generally have been achieved by using numerous physical resources and involving investigators from many disciplines, and that there is no clear indication yet of safety and efficacy without the results of controlled trials. That CREST will perform a randomized trial is an important step forward. Experienced interventional neuroradiologic teams need to gear up to support this trial. At the same time, the patients for study need to undergo rigorous assessment and categorization within the trial so that its results will have broad-reaching application, as a natural extension of the recent, exemplary endarterectomy studies.