A Word From an EVA-3S Participant
Alain Branchereau, MD, a Principal Investigator for the EVA-3S trial, discusses the protocol of the trial, its physician experience requirements, and the future of CAS.
Endovascular Today: The EVA-3S trial has received tremendous exposure in the US and has prompted significant debate related to the design of the study. Is it accurate to say that the surgeons in the endarterectomy (CEA) arm were required to have a minimum of 25 cases in the preceding year, but an interventionist could participate in the carotid stenting (CAS) arm with no previous CAS experience if he or she were supervised by a proctor?
Dr. Branchereau: No, that is not exactly correct. The interventionists had to have performed at least 12 CAS cases prior to entering into the trial. Any interventionist who did not meet these criteria intervened with a proctor, who was in the room, scrubbed, and wearing gloves.
At the beginning of EVA-3S, in November 2000, the required previous operator experience for stenting was usual because, at that time, a very small number of interventionists in Europe or the US had experience with more than 100 cases. There were many people, however, who had extensive experience in CEA.
To join the trial, each center was required to form a team of physicians composed of one neurologist, one vascular surgeon, and one interventionist. The neurologist handled the initial evaluation and patient follow-up, the vascular surgeon had to have performed at least 25 CEAs in the year before enrollment, and the interventionist must have performed at least 12 CAS procedures, or at least 35 stenting procedures in the supra-aortic trunks, five of which were in the carotid artery. The radiologist must have, in the 3 previous years, performed at least 50 angiographies of the carotid arteries yearly. The surgeon must have sent the operative protocols and the hospital discharge forms of each of these 25 procedures to the scientific committee of the trial.
Centers that could fulfill all requirements except those with regard to the interventional physician could still join the EVA-3S study and randomly assign patients, but all stenting procedures had to be performed under the supervision of an experienced tutor until the local interventional physician became self-sufficient (according to the tutor) and performed a sufficient number of procedures according to the predefined criteria. These are the facts written in the protocol.
Endovascular Today: Under the protocol of the study, was it possible for the interventionist to be performing his first CAS procedure and still be in the study?
Dr. Branchereau: Yes, it was possible. The interventionist requiring this scenario had to perform at least 10 to 15 cases with a proctor from another center. It was the responsibility of the proctor then to decide if the "interventionist in training" was qualified to enter the trial on his own. Post hoc analyses have shown that there was no one-center effect, meaning no center had impacted the results, and that 15.8% of stent placements were performed by experienced physicians (more than 50 previous cases), 45.4% by physicians having performed 50 or fewer, and 38.8% by physicians still in procedural training. The 30-day stroke plus death rates in these three groups were 12.2%, 11%, and 7.1%, respectively (P=.49). It is much more important that we realize in the analysis of the results that the protocol continued for 5 years, and there was no difference between the beginning and end periods, which is why we cannot say that the results are due to a learning curve.
Endovascular Today: In 2003, the trial started requiring embolic protection. Were you able to determine if that made a difference?
Dr. Branchereau: Yes, there was a trend, but the difference was not significant. We cannot say that there was a difference between the cases with protection and without protection. The question of the benefit of protection is still pending.
Endovascular Today: The high major adverse event (MAE) rate in the CAS arm is one remarkable aspect about EVA-3S. The other remarkable aspect was the low stroke rate for CEA. Can you comment on these two issues?
Dr. Branchereau: It is important to note the stroke rate in the angioplasty arm. It is not surprising if you notice that in most trials, not all lesions are atherosclerotic. SAPPHIRE, for example, includes patients with recurrent stenosis, or post-radiation stenosis, and, like in other registries, asymptomatic and symptomatic patients. If you look at the trials that performed only on symptomatic atherosclerotic patients—like those in EVA-3S and SPACE—you have approximately the same rate of accident. In 2001, the Wallstent trial, sponsored by the Schneider Company before it was acquired by Boston Scientific Corporation (Natick, MA), performed on symptomatic patients and was interrupted due to an excessive risk of stroke in the CAS arm. After randomization of 219 patients, the results showed a 30-day stroke plus death rate of 12% in the CAS arm compared to 4.5% in the surgical arm. These results were only published as an abstract in Stroke in 2001. In 1998, the Leicester trial on symptomatic patients conducted by Ross Naylor, MD, was rapidly abandoned due to an unacceptable stroke rate in the CAS group (5/7). The problem is that symptomatic atherosclerotic stenosis is a much more severe disease than nonatherosclerotic disease and asymptomatic stenosis. The point is that many trials present a mix of cases, but if you only perform angioplasty and stenting in cases of severe symptomatic atherosclerotic stenosis, these lesions are very dangerous, and the rate of accident will be the same as observed in EVA-3S, SPACE, and the Wallstent trial. In addition, if you look at trials such as CAPTURE, the pilot study of CREST, and the registry from Michel Makaroun, MD, et al at the University of Pittsburgh School of Medicine, all observed a higher rate of accident in the elderly patient. Why? The elderly patients have more severe atherosclerotic lesions than others. We know that the rate of accident is significantly higher in this population, and now everyone recommends to not perform CAS in patients older than 75 years because a high-risk lesion will make the procedure risky.
Endovascular Today: Compared to the SPACE study, your CEA arm had a much lower MAE rate. Can you comment?
Dr. Branchereau: In France, more than 90% of vascular surgery procedures and most CEAs are performed by vascular surgeons who specialize in vascular surgery. We have had a specific board of vascular surgery for approximately 15 to 20 years. This is one good reason. All vascular surgeons would agree and tell you that the MAE risk is higher if you have a CEA performed by a surgeon operating only on two or three patients a year than if you are operated on by a surgeon who performed 25, 50, or 100 cases a year. It is evidence-based.
Endovascular Today: Do you know the percentage of general surgeons performing CEA in Germany?
Dr. Branchereau: No, but I know in Germany, like in Britain and the US, there are a lot of vascular surgery procedures performed by general surgeons. Five or 6 years ago, the US vascular surgeons fought to separate from the American Board of Surgery, but unfortunately, they did not win the battle. I think that at least the results from this trial will help them to continue the fighting. The minimum amount of 25 CEA cases a year was difficult, because in France there are some vascular surgeons doing fewer than 25 cases who could not participate in the trial.
Endovascular Today: Do you think that many of the surgeons who participated in EVA-3S did more than 25 procedures in the year?
Dr. Branchereau: Absolutely. For instance, in my group, two vascular surgeons participated in EVA-3S. One was doing approximately 80 cases a year, and the other performed 50. I think this is the case in most other centers. Most of the vascular surgeons who participated were doing 50 cases a year; they were highly skilled. Moreover, in France, a great number of vascular surgeons perform a completion angiogram at the end of the procedure. In my opinion, it may play a role, but it is very difficult to prove.
Endovascular Today: What do you see as the future of carotid stenting in France?
Dr. Branchereau: This is a difficult question and a problem we now face. We are embarrassed because the insurance system and the Ministry of Health are very skeptical after this trial. We have some cases, for example, of very high-risk patients or patients with restenosis, or post-radiation stenosis, for which, in my opinion, stenting is a good solution. However, it is true that it represents maybe 300 to 500 cases in France a year.
Stenting will evolve and improve. The improvement in guidewires and embolic protection systems will continue and allow us to decrease the stroke rate. I do not know the future, but currently, CEA remains the gold standard in France. There will be other trials and other data. SPACE and EVA-3S are not in favor of stenting, but we will gather more data and continue the evolution of the technique. If the trial had been done in 1998 and 2001, the results of stenting would have been worse. The Wallstent and the Leicester trials were stopped quickly due to the stroke rate with CAS. The EVA-3S results show that stenting has improved since then. Nevertheless, it is a reality that CAS is less developed in France than in some other European countries due to the limitations issued by the National Health Insurance System.
Endovascular Today: Do you perform stenting?
Dr. Branchereau: Yes, in my department, we operate on approximately 150 to 200 carotid lesions a year, and of those, 15 to 20 are CAS procedures.
Alain Branchereau, MD, is Professor of Vascular Surgery at the Université de la Méditerranée, in Marseille, France. Dr. Branchereau may be reached at +33 491 38 57 62; firstname.lastname@example.org.