VEITHsymposium: Eight Hot Topics

Cath Lab Digest talks with:

1. Frank J. Veith, MD, and Wesley S. Moore, MD, on carotid stenting and CREST; 2. Michael Dake, MD, on the first DES approved for the SFA; 3. Frank A. Lederle, MD, on diabetes’ negative association with AAAs; 4. Peter Lin, MD, on ultrasound-accelerated catheter-directed thrombolysis in PEs; 5. Michel Makaroun, MD, on endovascular repair of AAAs; 6. Claudio Rabbia, MD, on a newly developed flow diverter developed to treat peripheral and visceral aneurysms; 7. Seshadri Raju, MD, on stenting in chronic venous disease; 8. Gunnar Tepe, MD, on an angioplasty balloon with a unique embolic capture capability.

Held November 17-21, 2010, in New York City, the annual VEITHsymposium was founded by vascular surgeon Frank J. Veith, MD, in 1973. ————————————————————

CREST (Carotid Revascularization Endarterectomy versus Stent Trial)

CREST enrolled 2,502 symptomatic or asymptomatic patients randomized to carotid endarterectomy or carotid artery stenting plus medication. The trial’s primary composite endpoint was stroke, myocardial infarction, or death from any cause during the periprocedural period or any ipsilateral stroke within 4 years after randomization. Frank J. Veith, MD, and Wesley S. Moore, MD, offer their thoughts on the trial. Dr. Veith is Professor of Surgery at New York University and The Cleveland Clinic and The William J. von Liebig Chair in Vascular Surgery at The Cleveland Clinic, Cleveland, Ohio and New York, New York. Dr. Veith, what was your reaction to CREST and how did you feel the trial was received?

That’s a very complex question. There are great expectations for CREST because it’s been talked about and lauded for many years. I thought it was very well done, very carefully done, but it was obvious that it wasn’t perfect. After the trial results were presented and the NEJM publication¹ came out, the conclusions of the trial were ‘spun,’ in the sense that news is spun by various commentators or politicians or data misinterpreted by various specialists. If you were an interventional specialist, even if you were a vascular surgeon but somebody who was in favor of carotid stenting, you said, “This trial is the definitive trial. It proves that carotid endarterectomy (CEA) and carotid stenting (CAS) are equal or equivalent.” And, in fact, the concluding statement in the trial abstract is that the two procedures are equivalent. That happens to be incorrect. That’s the spin put on CREST by interventionalists, be they surgeons or cardiologists or anybody else. If you’re a surgically-oriented person, i.e., in favor of carotid endarterectomy, you spin the trial totally the opposite by saying, “The number of strokes and deaths is greater in the carotid stenting arm, therefore carotid endarterectomy is superior.” Both are true!

The problems with CREST, then, are that, first, it started out to be a trial of carotid stenting versus endarterectomy in symptomatic patients. Yet it ended up, for a variety of reasons, diluted with asymptomatic patients. Symptomatic patients should be treated by an intervention or a procedure. Asymptomatic patients, in my opinion, should be treated primarily by statins and not by either endarterectomy or CAS. Symptomatic patients justify treatment, and to dilute CREST by adding an equal number of asymptomatic patients was a defect in the study. Symptomatic and asymptomatic carotid disease are different disease entities; we are talking about apples and oranges. Most asymptomatic patients should be treated only with statins and other medications. They shouldn’t be subjected either to endarterectomy or carotid stenting. So the asymptomatic arm of CREST, I believe, was very flawed. It’s almost useless. CREST does not contribute anything to the argument about what to do for asymptomatic patients. We need a trial that compares stenting, carotid endarterectomy, and best medical treatment, and that trial doesn’t exist. The reason I feel so strongly about including best medical treatment is that the stroke rate per year in asymptomatic patients is somewhere less than 1%. It’s probably closer to 0.4%.^2-3 Strokes have been decreasing in asymptomatic patients over the last couple of decades.

If we look at the symptomatic patients, what are the flaws in CREST? First, there were not enough symptomatic patients for the study to be statistically significant. Another major flaw in CREST was the use of a composite endpoint, consisting of strokes, deaths and myocardial infarctions (MIs). The purpose of both carotid stenting and carotid endarterectomy is not to prevent MIs, but to prevent strokes and deaths. Adding the MIs, to me, doesn’t make sense. There was a statistically significant greater number of strokes in the CAS arm than in the CEA arm: 11 major strokes in the CAS group and 4 in the CEA group. It was not enough, because of the poor power of the study, to be statistically significant, but still, it’s worth noting. Mortality was also in favor of CEA. Minor strokes were statistically significant less after CEA, than after CAS. If you look at just the strokes and deaths in CREST, which is what the procedures are supposed to prevent, CEA comes out better than CAS. The only way that the two become equivalent is by combining the endpoints of MI and stroke. That’s a flawed combination. There is more disability after strokes than after MIs. Even if it’s a minor stroke and there are no measurable neurologic deficits — in other words, the patient recovers — they are not really the same after a stroke. Whereas if you recover from a minor MI, you are fine. All of the MIs and most of the strokes in CREST were minor. You cannot say that the two procedures are equivalent. There was greater disability in the CAS patients than in the CEA patients.

So the two big flaws in CREST were first, that they combined symptomatic and asymptomatic disease, diluting the power of the study, and second, they used a composite endpoint, where strokes were equivalent to MIs. The third CREST defect is that the CAS patients received a different antiplatelet treatment. It was a more intense antiplatelet treatment than the CEA patients, which may have accounted for the difference in MIs.

Despite these flaws, I believe that CREST was extremely well done. It was carefully analyzed. It was a good randomized study, but in addition to the defects that I mentioned, which are specific to CREST, during any randomized study that involves follow up (CREST follow up went to 4 years), technology evolves. Today, we can see that the way CAS was done in CREST wasn’t optimal. The CAS results would be better, I believe, if done today, because we have better protection devices, with reversal of flow and cessation of flow. These protection devices were not used in CREST. In addition, better stents are appearing and will appear in the future, and there is better patient selection. Everything else I’ve pointed out is against CAS, but this particular aspect is not. CEA isn’t going to change. It’s already highly evolved. CAS is changing all the time and will continue to improve.

Similarly, another trial defect that applies both to CAS and CEA is that neither of the groups were treated with the optimal statins and other medical management as it exists today. We know, for example, that high-dose statins are more effective than low-dose statins in preventing strokes and heart attacks. The level of medical management back in 1999–2002, when CREST patients were enrolled, was not as good as it is today. Perhaps with better medical management, there wouldn’t be as many MIs and strokes in either group.

There is also the added finding in CREST of a greater number of cranial nerve deficits following CEA than following CAS. This is not surprising, because there’s no neck dissection with CAS. Some have made the point that a cranial nerve deficit is the equivalent of a stroke. That’s not true. A cranial nerve deficit (all of which were minor in CREST) doesn’t affect the brain, and doesn’t have residual effects on mood and intelligence. You can’t equate a cranial nerve deficit and a stroke.

In addition to CREST, there has been another recent trial (which also has its flaws) called the International Carotid Stenting Study (ICSS), which, in addition to showing more strokes with CAS than with CEA, showed a very large number of asymptomatic, diffusion-weighted defects on MRI in the brain.⁴ This means patients are having silent strokes, and these were much more common after CAS than after CEA.

However, CREST was a good trial in some ways. It showed that both procedures could be carried out with a lower than expected morbidity and mortality. But it didn’t add much to the questions regarding asymptomatic patients, and it is not the definitive trial for showing the equivalence of CEA and CAS in symptomatic patients.

Having said that, I believe that CAS will prove equal to or better than CEA for some patients with symptomatic carotid disease. But that hasn’t been shown yet. We still need more trials. CREST doesn’t add a great deal to the asymptomatic disease debate, because in this setting the debate is not, ‘Which form of therapy is best?’ The debate is whether any form of invasive therapy is needed. The unpopular view, which I hold, is that most asymptomatic patients shouldn’t be treated interventionally or with surgery. The idea that treatment prevents strokes in that group is very hard to defend. CREST basically shows that if you have a non-threatening carotid lesion, there’s not much risk to putting in a stent or doing an endarterectomy. I think that asymptomatic carotid disease, in most instances, is the equivalent of not having a disease that really matters.

What about some of the new technology that can identify vulnerable plaques?

We do need a way of identifying vulnerable plaque in an asymptomatic patient that is going to go on and cause a stroke. Dr. Richard Spence has done transcranial Doppler evaluations in asymptomatic patients.⁵ He spoke at the VEITHsymposium and showed that before the statin era, asymptomatic patients had a lot of hits on transcranial Doppler, meaning that when they listen over the artery, there are noises that indicate that there are some embolic particles reaching the brain. After Dr. Spence treated these same patients with statins and other medical treatment, the number of hits decreased and the number of strokes decreased, from about 3.3% per year to 0.7% per year. That’s better than we can do with any carotid endarterectomy or carotid stent. So, I think the best test we have now, although it’s not widely done, is to do transcranial Doppler, measure the number of hits, and the group of asymptomatic patients that continues to have hits despite statins, probably needs an operation or a stent. In addition, there’s little current evidence that stenting works to prevent stroke in asymptomatic disease. The only evidence is that CEA may decrease the number of strokes, based on a recent Lancet article on the Asymptomatic Carotid Surgery Trial (ACST).⁶ This study compared CEA with no treatment, followed the patients and found that there was a decreased stroke rate in the CEA group. Some of those patients got statins, but most of them didn’t get modern statins. However, the ACST study is somewhat flawed, and not fully applicable at present. I think we still need a study that compares interventions (CAS and CEA) with current best medical treatment. In the ACST study, patients received at most 10 mg of simvastatin, which is, by current standards, a suboptimal dose. Nevertheless, this is an area still rife with controversy.

Any final thoughts?

Well, you know, there’s bias in everything, and when I talk about ‘spin,’ I’m talking about interpreting something with a bias. And obviously, I have a bias, although I think I’m an enthusiast for carotid stenting — it’s going to work. I’ve been interested in it for a long time. It was our work that led to all these embolic protection devices. We did some in vitro studies in the late ‘90s showing that CAS caused emboli. Even though I am an endo enthusiast, I don’t believe that carotid stenting is yet justified in most cases. Right now, the evidence doesn’t support it. The purpose of carotid procedures is not to prevent MIs, it’s to prevent strokes. MIs matter. But CREST doesn’t yet prove that carotid stenting is the equivalent of carotid endarterectomy . I believe it will be, but it’s not there yet.

Disclosure: Dr. Veith reports no conflicts of interest regarding the content herein. Dr. Veith can be contacted at fjvmd@msn.com.

References and Recommended Reading

1. Brott TG, Hobson RW, Howard G, et al. Stenting versus endarterectomy for treatment of carotid-artery stenosis. N Engl J Med 2010; 363:11–23. 2. Abbott AL. Medical (nonsurgical) intervention alone is now best for prevention of stroke associated with asymptomatic severe carotid stenosis:results of a systematic review and analysis. Stroke 2009;40:e573-e583. 3. Naylor AR, Gaines PA, Rothwell PM. Who benefits most from intervention for asymptomatic carotid stenosis: patients or professionals? Eur J Vasc Endovasc Surg 2009;37:625-632. 4. Bonati LH, Ederle J, McCabe DJH, et al. Long-term risk of carotid restenosis in patients randomly assigned to endovascular treatment or endarterectomy in the Carotid and Vertebral Artery Transluminal Angioplasty Study (CAVATAS): long-term follow-up of a randomised trial. The Lancet Neurology 2009; 8 (10):908–917. 5 Spence JD, Coates V, Li H, et al. Effects of intensive medical therapy on microemboli and cardiovascular risk in asymptomatic carotid stenosis. Arch Neurol 2010 Feb;67(2):180-186. 6. Halliday A, Harrison M, Hayter E, et al. 10-year stroke prevention after successful carotid endarterectomy for asymptomatic stenosis (ACST-1): a multicentre randomised trial. The Lancet 2010; 376(9746):1074–1084. 7. Paraskevas KI, Mikhailidis DP, Nicolaides AN, Veith FJ. Interpreting the Carotid Revascularization Endarterectomy Versus Stent Trial (CREST): Additional trials are needed. Vascular 2010 Sep-Oct;18(5):247–249. 8. Veith FJ. The spinning of CREST. Medscape General Surgery. Posted June 14, 2010. Available online at https://www.medscape.com/viewarticle/723214. Accessed December 8, 2010.

———————————————————— Wesley S. Moore, MD, is Professor and Chief Emeritus, Division of Vascular Surgery at UCLA Medical Center, Los Angeles, California. Dr. Moore was a co-principal investigator on CREST. Dr. Veith explained some of the reasons as to why he felt CREST was not the final answer to the CEA versus CAS question. As a co-principal investigator, could you explain your perspective on the trial?

There are several unique aspects of the trial. It is the largest trial involving any kind of carotid intervention, stent angioplasty (CAS) or carotid endarterectomy (CEA). It’s the largest trial of its kind in the world. CREST is the only prospective, randomized trial comparing CAS and CEA that included both asymptomatic as well as symptomatic patients. Another unique feature is the recognition that CAS is relatively new technology. We felt there were probably a number of interventionalists still on the steep part of the learning curve, and therefore, in order to be able to bring the participating interventionalists up to maximum speed by the time that they got to the point of randomization, there was a lead-in phase. Individual interventionalists who passed the initial review by an interventional management committee as being competent individuals were next required to submit up to 20 lead-in CAS cases that were evaluated with the same strict criteria as in the randomized trial. Only those individuals who went through several areas of scrutiny were then approved to begin to participate in the randomized phase of the trial. In this circumstance, we dealt with the best of the best, in terms of interventionalists available to us in the United States. A similar, though perhaps not quite so rigorous, scrutiny was carried out for surgeons who wished to participate in the trial. The surgical committee, which I was chairing, utilized the same criteria from a previous carotid endarterectomy trial, the ACAS trial (Asymptomatic Carotid Atherosclerosis Study), so this method of screening surgeons had already been proven. The scrutiny of both interventionalists and surgeons allowed us to be in a position of comparing two procedures rather than the individuals participating in the procedures.

Can you tell us about the trial endpoints?

CREST trial endpoints in terms of adverse outcome were death, stroke, and myocardial infarction (MI). It was felt that perhaps one of the potential benefits of a less invasive procedure such as CAS would be a lower incidence of MI when compared with an open surgical procedure. Adding MI did deviate somewhat from standard comparisons of death and stroke alone on most of the previous trials comparing carotid endarterectomy with other strategies such as medical management.

The aggregate of all three endpoints in CREST — death, stroke and MI — demonstrated no statistically significant difference between CAS and CEA. However, the more traditional endpoints of death and stroke showed a major difference between CAS and CEA, with CAS having almost twice as many strokes as CEA. On the other hand, CEA had more MIs than CAS. Simply by weighting all three endpoints the same, the two procedures came out equal. We were rather strict in defining an MI. Criteria were chest pain, electrocardiographic change or enzyme elevation. In order for someone to be adjudicated as having an MI, they had to have two of those three being positive. A patient might have had some EKG changes with an elevated troponin level, and so they would have been called an MI without ever having an episode of chest pain. Somebody could have had chest pain with an elevated troponin, or chest pain with an EKG change and not an elevated troponin. We were very, very careful in scrutinizing patients and I don’t think we missed any MIs. In fact, we may have overcalled some.

When you compare the endpoints of stroke versus MI, you have to ask, what is the purpose of doing any intervention on the carotid artery? The purpose, really, is to prevent a stroke. If there are more strokes in one strategy versus another, then that obviously would be a bad thing. On the other hand, there is a price to be paid for a procedure that has fewer strokes, and that price turned out to be a higher incidence of myocardial infarction (in CEA). At this point, the question becomes, how important is MI, even minor MI? In order to try to answer that question, we did a quality-of-life analysis among those trial patients who had a stroke or MI versus those that did not. The quality-of-life assessment, both from the physical and mental component perspective, showed stroke had the most adverse effect on the patients’ lives. Minor stroke had the next most important effect, and MI had the least, and almost overlapped the median, or had no impact at all. In designing CREST, we failed to put a value on each of those three endpoints. For example, you can’t say a minor MI carries the same importance as a death, but when you put stroke, MI and death together in a composite endpoint, that’s what it looks like. As a result, we did a sub-analysis and found that doing a CEA turns out to be a safer procedure with respect to the outcomes of death and stroke.

What about the inclusion of asymptomatic patients in CREST?

We were slow in accruing symptomatic patients, and we needed to increase our numbers in order to bring the trial to a conclusion. Also, we had to recognize the fact that in the United States, in this day and age, the majority of people that are undergoing any sort of intervention on their carotid arteries are asymptomatic. So it made sense to include asymptomatic as well as symptomatic patients.

What about medical therapy for patients in CREST?

The trial began 10 years ago, and statins didn’t exist at that time. The importance of beta blockers or ACE inhibitors in terms of reducing the risk of MI had not been fully evaluated. If we were to design CREST today, the medical management of these patients would have included being on a statin, and certainly on a beta blocker. The other issue is that the patients in the CAS arm of the trial were on dual antiplatelet therapy, aspirin and clopidogrel, and the surgeons were reluctant to put their patients on dual antiplatelet therapy; most of their patients were on aspirin alone, and that may have had some impact on MI rate as well.

I should add that we noted an age effect as to which procedure had better outcomes. If you looked at how patients did approximately 4 years after their treatment, patients over the age of 70 did better with CEA, and patients below the age of 70 did better with CAS. If you have a symptomatic patient of average risk, perhaps 70 years of age or older, then the recommendation should be that they have CEA. If you have an asymptomatic patient, then the recommendation becomes a little less clear, because the stroke differences between CAS and CEA in these patients were not quite as dramatic as they were in the symptomatic patients. On the other hand, with newer methods of medical management, there is now real question as to whether any intervention should be offered to asymptomatic patients. It may be they should just be treated medically. That is probably going to be the basis of the next trial, which will be an asymptomatic group of patients randomized between one or other of the interventions, versus medical management alone. We have just received permission from the National Institutes of Health (NIH) to submit a protocol for their review.

What about advances in technology over the time of the trial?

It used to be mistakenly thought that CEA had reached a plateau and there were no further advances; that turns out not to be the case, because when you compare the stroke and death rates that occurred with CEA in CREST, they were reduced by half, compared with trials done 15-20 years ago. CEA has improved; certainly CAS, which is a newer therapy, is going to improve at even a steeper rate. CAS is going to continue to improve, and there may be better ways of doing CAS; for example, rather than doing it through a transfemoral approach, it might be better to do it through a transcervical approach, and avoid a lot of the complications associated with the aortic arch. The best that you can do is go with the best you have at the time, and try to recruit patients as quickly as you can, so the trial doesn’t go on for a prolonged interval, during which there may be important changes in technology.

Disclosure: Dr. Moore reports that he was the co-PI for surgery in the CREST trial and that his institution receives research support from the NIH for participating in the trial.

Dr. Moore can be contacted at wmoore@mednet.ucla.edu

———————————————————

The First DES Approved for the SFA

Michael Dake, MD, Director, Catheterization and Angiography Laboratory, Stanford University Medical Center, Stanford, California, discusses the Zilver PTX (Cook Medical, Inc., Bloomington, In.), currently CE-marked and investigational in the U.S. Dr. Dake reported on a multicenter, randomized, prospective trial in the U.S., Germany and Japan. What are some of the difficulties in stenting in the superficial femoral artery?

It’s been notoriously a very difficult vascular bed, because of a number of factors. The superficial femoral artery (SFA) can be up to 40 cm long, and depending on the pattern of disease, if it’s diffuse disease affecting that entire vessel, it can be a real challenge. In addition to being long, the SFA has some different forces that don’t really affect other arteries. There are degrees of torque and dynamic muscular contractions that can possibly cause stents to fracture. Opposed to the coronary arteries, which have balloon-expandable stents, the stents in the periphery and the superficial femoral artery are self-expanding stents, so that this muscular contraction or any other external force upon the stent doesn’t compress the stent permanently or plastically deform it, denting it as a can might be dented. In those situations where self-expanding stents resist these forces, and can actually accommodate to cyclical compression, you may end up with wear of the metal and a fracture of the stent, which can affect patency. These are probably the two biggest challenges.

Can you tell us about the Zilver PTX stent?

First, let me say that there are a number of contemporary stents that have solved the issue of fractures. I am not sure we will ever have a fracture-resistant stent, but we certainly have fracture-safe stents. It is not unique to the Zilver. The Zilver PTX stent is different because it is a paclitaxel-eluting stent.

We did a randomized, controlled trial comparing the Zilver PTX to balloon angioplasty and bare metal stenting (BMS) in the treatment of diseased SFA segments. Looking at the group of patients who were in the randomized trial (reported at the VEITHsymposium) as well as those in an earlier, single-arm trial, stent fracture rates were very low, 0.9-1.2% in over 1,500 lesions. There were four fractures in the randomized study and none of them had any effect on vessel patentcy. We are developing stents that can withstand the unique forces that potentially led to fractures in earlier generations. The randomized Zilver trial was the first drug-eluting (DES) stent trial to give us a signal that some of the benefits that DES have provided to coronary arteries are now not only conceivable, but deliverable, in the SFA and peripheral arteries. There are really only two trials before the Zilver PTX in the SFA. Both, at one year, did not show a significant difference in the results with a DES and a BMS. In the Zilver randomized trial, the primary randomization was between balloon angioplasty and a DES, and the DES clearly had a statistically significant benefit. In the secondary randomization, for about 126 patients where angioplasty was initially deemed a failure, these patients were secondarily randomized to either a BMS or the Zilver PTX. In this group, the Zilver PTX provided a 63% reduction in restenosis compared to the BMS.

It isn’t the ultimate answer, certainly, and it is necessary to say that this is something that isn’t going to be used in all patients, but it certainly gives us a signal that we are capable of marrying a drug with a mechanical device like a stent, and delivering incremental value to patients. The reason the randomized trial is not the ultimate study is because the FDA mandated that the lesions had to be a certain length, a maximum of 14 cm. The mean lesion length in this trial was 6 cm, considered by most to be a moderate, modest length. We can look to the single-arm study, which was done outside the U.S. and not randomized. It had 750 patients, so was a very large study, and had real-world, all-comer patients. Patients could have any type of disease, any number of stents could be put in, the mean lesion length was 10 cm, and 22% of the patients had lesions greater than 15 cm. It is important to look for a lack of discrepancy or any sort of results that may be conflicting between studies — and it was found that the results of this single-arm study complement the randomized study. As a result, people are now heartened to go on and do studies that may include the Zilver PTX in longer lesions, diabetic patients, and in-stent restenosis, some of the most troublesome patient groups that we currently face in the SFA. Both of these trials now suggest that now there is something better that can improve the durability of results.

Disclosure: Dr. Dake reports that within the past 12 months, he received research/research grants and clinical trial support from Cook Medical. Dr. Dake can be contacted at mddake@stanford.edu. ————————————————————

Diabetes Reduces AAAs by Half

Frank A. Lederle, MD, Professor of Medicine, VA Medical Center, Minneapolis, Minnesota How did the association of diabetes with a reduced rate of AAAs come to light and what does it mean?

In the early 1990’s, we began the Aneurysm Detection and Management (ADAM) trial, a cross-sectional screening study. The trial involved 125,000 veterans, who were brought in to fill out a questionnaire and then have an ultrasound screening to see if they had an aneurysm. We reported the first 75,000 in 1997,¹ and found that diabetic patients were only half as likely to have an aneurysm. This was in all of our models, with very tight confidence intervals, and completely unexpected. We were mystified. Previous screening studies showed a trend, but were too small to give much indication. We first thought, well, we must have made a mistake. Is there some way that people with both diabetes and aneurysm didn’t turn up in our cohort, i.e., is it a very lethal combination? Were those folks dead? We looked at indicative studies and didn’t find anything. Our next thought was, did we exclude these patients, because we only looked at screening aneurysms? If you already had your aneurysm diagnosed, we excluded you, so perhaps diabetics got a lot more health care and their aneurysms were already diagnosed, meaning they weren’t in our cohort. This would mean that patients with aneurysms that were already diagnosed and referred to our study should have a high rate of diabetes, and they did not, they had a low rate, just like the screened aneurysms. It was looking to be a true finding. Finally, we found a Massachusetts surgical study that was quite different. The study looked at racial differences and found that diabetics were less likely to have aneurysm surgery than they were to have the control, which was appendectomy. It had a significant p-value. That convinced us that we were on to something, and we went with it, devoting a fair amount of our 1997 paper to describing the association, possible mechanisms and why they didn’t seem to hold up.

Since then, a number of large studies have looked at this issue, like our second set of 50,000 veterans,² a Western Australia study, the REACH vascular registry, women’s health initiative data, and also now the Life Line screening of 3 million people (https://www.lifelinescreening.com). All have found significantly lower rates of aneurysm in people with diabetes, with an odds ratio of between 0.3 to 0.8, but all showing it is less. The finding is looking very reproducible and very real.

What does it mean? People have speculated whether aneurysm disease is a manifestation of atherosclerosis, and I think aneurysm researchers would say probably not. This finding really drove a wedge between the two conditions. Diabetes is predictive of atherosclerosis and protective of aneurysm disease. There are very few negative associations between diseases. There are some between risk factors, such as smoking, which might lower the rate of ulcerative colitis, but there are very few where one disease is protective against the other. Probably the most important thing is that it gives laboratory researchers something tangible on which to focus. We know aneurismal disease is smoking-related, but now this notion of protection from diabetes gives researchers another angle. Several papers have been coming out and showing that diabetes is associated with lower matrix metalloproteinase levels, for one reason or another, which are instrumental in forming aneurysm. There is some evidence that the vessel wall is thicker in diabetes, and maybe that relieves stress. There are studies that suggest that glycogen products related to high sugar crosslink the collagen, and make it resistant to aneurysm formation. I don’t know that we have the answer, but there is a lot of interest now. If we see a mechanism explaining how smoking promotes aneurismal disease and diabetes reduces it, and perhaps why women are less apt to have it, we would know we were really on to a solid mechanism. Those would all be confirmation that you were on the right track. I am not a laboratory researcher, but do more epidemiological research. Right now, we are looking at a cohort of enlarging aneurysms in the VA electronic medical records where we can look at more co-factors. Studies have also shown that aneurysm enlargement rate is slower in patients with diabetes.

Can we take away anything for day-to-day practice?

If a patient is an older male smoker, their risk for AAAs would be somewhere around 6%, and if you cut that in half with a diabetes association, that’s 3%, which is not negligible. In fact, it’s a higher risk than in male nonsmokers or in women who have smoked. I don’t see a tremendous clinical importance; it’s more that it informs future laboratory studies. If a diabetic person was very resistant to being screened, they have a little more reason to not be screened, but I don’t see that as a headliner. It is a curious thing that will play out to be important, but it is not a clinical lesson.

What are some of the questions we still have about AAAs?

It’s not clear whether it’s all one disease, or whether it’s a collection of diseases. There is still some debate as to what the relationship is with atherosclerosis. There is some kind of relationship, but it doesn’t seem to be part of atherosclerosis. Is it largely genetic? Is it largely autoimmune? I think we have some big answers in that smoking is clearly the main promoter, so efforts to stop smoking are very positive in reducing the disease. Another question is when we say diabetes reduces the prevalence of aneurysm, it could be diabetes, or it could be the treatment for diabetes. It would have to be an older treatment for diabetes, because in our original study, there were none of the new treatments, just glipizide, glyburide, metformin and insulin. A few studies have suggested it is probably not the treatment; it is probably the disease. One showed that in people who don’t have diabetes, and so who don’t have any treatment, blood sugar is still inversely related to diameter. And another in mice showed that it’s the blood sugar that caused the aneurysm and insulin would prevent the aneurysm. The treatment would be less surprising, it would just mean you found a drug with a second use. But it seems not to be the case.

Disclosure: Dr. Lederle reports no conflicts of interest regarding the content herein. Dr. Lederle can be contacted at frank.lederle@va.gov

References

1. Lederle FA, Johnson GR, Wilson SE, et al; Aneurysm Detection and Management (ADAM) Veterans Affairs Cooperative Study Group. Prevalence and associations of abdominal aortic aneurysm detected through screening. Ann Intern Med 1997;126:441–449. 2. Lederle FA, Johnson GR, Wilson SE, et al. The aneurysm detection and management study screening program: validation cohort and final results. Aneurysm Detection and Management Veterans Affairs Cooperative Study Investigators. Arch Intern Med 2000 May 22;160(10):1425–1430.

————————————————————

Ultrasound-Accelerated Catheter-Directed Thrombolysis in Pulmonary Embolism

Peter H. Lin, MD, Chief of Vascular Surgery of the Michael E. DeBakey Department of Surgery at Baylor College of Medicine, Houston, Texas What have we learned about ultrasound-accelerated catheter-directed thrombolysis for the treatment of pulmonary embolism?

The use of ultrasound-related technology ((EKOS Corporation, Bothell, Wash.) is being studied as a way to complement thrombolytic therapy to break down the clots in the lung — pulmonary embolism. In 2009, we reported on the use of this technology in massive pulmonary embolism (PE), which means that the patient has a significant amount of clot burden. [Ed. See our interview with Dr. Lin in the January 2010 issue, at https://tinyurl.com/MassivePE] Massive PE patients are clinically unstable. Their blood pressure is low and their heart is failing. Since reporting that data, we have had additional cases in patients that have what is called “sub-massive” PEs. These patients are not quite as severe. They do have clot, and it’s likely that if nothing is done, these patients will progress further down the road and turn into massive PEs. We found that ultrasound-accelerated thrombolysis is equally effective in treating patients with so-called sub-massive PE as it is massive PE.

We have done over 32 patients with a combination of massive and sub-massive, symptomatic PE using ultrasound-accelerated thrombolysis (UAT), and we have seen that once a catheter is placed, noticeable hemodynamic improvement occurs within 3 hours after the initiation of UAT therapy. This is a significant finding, because compared to the other group that simply received the catheter-directed thrombolysis without the ultrasound component (the conventional method of doing thrombolytic therapy), the clinical evidence, hemodynamic improvement, in the massive PE patients did not occur until 8-9 hours later. Sub-massive patients are usually more stable, so they don’t have hemodynamic alteration, but the massive PE patients are the patients where their blood pressure is very low and they are just very sick. They are always in the ICU. In this specific subgroup of patients, we noticed a more rapid improvement within 3 hours of some sort of restoration of hemodynamic stability compared to the catheter-directed thrombolysis patients.

Has there been any additional development in the design of the Ekos ultrasound catheter?

Yes, some modifications have been made over the last 1-2 years. The catheter emits non-cavitational ultrasound energy, which means that it doesn’t cause any bodily harm. The purpose of the ultrasound energy is to loosen the bonding of fibrin. Fibrin is one of the essential components of thrombus. If ultrasound can loosen the fibrin, it will expose greater surface area of the fibrin to greater amounts of the thrombolytic agent. If you have a setting where the fibrin are tightly bound to each other within the clot, the medication can only act on the surface region of this tightly bound thrombus, or tightly bound fibrin group. The ultrasound disassociates the fibrin so the medication can penetrate not only on the outer surface, but within. It accelerates the thrombolytic process. The company has changed the technology so that it is preprogrammed with certain settings, meaning that the catheter can communicate various different waves of ultrasound. There are various certain ranges of ultrasound that will be effective in disassociating fibrin. The catheter is not just emitting one wavelength of ultrasound; it actually emits various preprogrammed frequencies of ultrasound energy. It increases the efficacy of the disassociation of the fibrin.

What about the thrombolytic being used?

tPA is currently the preferred drug of choice. There are competing drugs, but there is no other competing technology using ultrasound to complement thrombolytic therapy.

What type of thrombus do you normally see in PE?

It’s not considered primary thrombus, meaning that the clot formed at that site, in its original setting. This is a pulmonary embolism, emboli, meaning they broke off from somewhere. Most of the time, if not 95% of the time, these clots in the lung came from the leg. They are mostly unstable thrombus from the leg that broke off and travel from the leg upward through the inferior vena cava (IVC) and lodge somewhere in the lung. These emboli are more unstable and also larger in quantity because in order for them to break off and run through all these different routes — the IVC, the right heart, the atrium, the right ventricle — and dislodge, they have to have a certain amount of force to be able to cause this type of systematic hemodynamic alteration. They are much larger in size and far more unstable compared to what you see in the coronary setting or the arterial ischemic thrombi in the leg.

Are any additional clinical trials looking at UAT?

Right now there is an ongoing European multicenter study, including centers in Germany and Switzerland, which is comparing the usage of UAT in the setting of PE. There is an effort underway to set up a similar clinical investigation in the United States.

Disclosure: Dr. Lin reports no conflicts of interest regarding the content herein. Dr. Lin can be contacted at plin@bcm.edu ————————————————————

The Endurant Stent Graft for AAA Repair

Michel Makaroun, MD, Professor and Chief of Vascular Surgery, University of Pittsburgh School of Medicine, Pittsburgh, Pennsylvania How frequently is endovascular aneurysm repair (EVAR) being performed compared to surgical repair?

EVAR has taken over from open techniques in the majority of patients who are candidates (there are still some anatomic limitations). In the U.S., EVAR is done in close to 60% of patients. Worldwide, in the countries that can typically afford the increased costs of this technology, it is similar to the U.S., with nearly 50-60% of abdominal aortic aneurysms (AAAs) patients being treated by EVAR.

Can you tell us about the Endurant stent graft?

The Endurant is a third-generation stent graft from Medtronic intended for the treatment of AAAs. It combines several advantages over previous versions. First, it is a lower profile device that can advance better in diseased arteries. It also has a modified stent in the proximal portion, which allows it to adjust very nicely to the irregular necks of the aorta. It can treat shorter necks than we what have been accustomed to in the past. The Endurant has, in addition, a unique deployment mechanism that allows us to accurately place the first portion with pinpoint accuracy. The graft contains a suprarenal stent without fabric that is intended to deploy above the renal arteries. The more accurately we can place the graft itself at the level of the renal arteries, the better off we are in terms of sealing and durability of the procedure. When the graft is opened, the proximal portion, which is the suprarenal stent, remains constrained, allowing us to position the device exactly where we want it. There is a release mechanism at the bottom of the delivery system, which, when turned, delivers and opens the stent very gradually, so we can be extremely accurate in the placement of the proximal system. Previous devices have some imprecision in how well we can accurately place the proximal system. This Endurant is extremely precise and has improved on this aspect of delivery.

So improved deliverability is a very important modification.

The most stable part of the area we are using to treat the aneurysm is the part that is not dilated, right next to the arteries of the kidneys, which is in the infrarenal neck. In that portion, we would like to get as close to the kidneys as possible, because that is the portion that is the most stable. When the device deploys lower, typically it is not as stable. Over time, it has more chance of enlarging and actually losing the seal that we are trying to achieve at that level. Being able to place the device exactly at the level of the renal arteries carries a lot of advantages in terms of durability.

Can you tell us about the two studies, both the U.S. regulatory study and the international registry?

Since this stent graft is not yet approved in the U.S., we don’t have a post-market experience, but internationally, close to 900 patients have been enrolled into a registry to identify outcomes. In the worldwide registry, we have 30-day data available that has been partially presented (the registry is not yet complete). The outcomes have been excellent, and very similar to what we experienced in the actual regulatory trial. The regulatory study involved 150 patients at 26 U.S. medical centers and met its primary endpoints. The Endurant was associated with no post‐operative aneurysm ruptures or aneurysm‐related mortalities at one year, and there were no mortalities from any cause at 30 days. Through one year post‐implant, we found no Type I or III endoleaks. None of the stent grafts migrated from their original placement. Five-year follow up is required by the FDA on all patients enrolled in the regulatory trial. At this point, we start looking at durability of the graft, migration and the ruptures that may happen occasionally. Since the Endurant is a very new graft, we still have only the one-year data to report, but as of now, it is looking extremely positive.

Disclosure: Dr. Makaroun has a consulting agreement with Medtronic for serving as the Principal Investigator on the regulatory trial. Dr. Makaroun can be contacted at makarounms@upmc.edu ————————————————————

A New Flow Diverter to Treat Peripheral and Visceral Aneurysms: The Cardiatis Multilayer Stent

Claudio Rabbia, MD, Chief of the Department of Interventional Radiology at San Giovanni Battista University Medical Center, Torino, Italy What are the limitations of stent grafts in treating aneurysms?

First, let me point out that stent grafts are not the only way to treat an aneurysm. Coils may be used for treating selected aneurysms, for instance, although to treat such aneurysms, you have to coil the entire sac, which may not be so easy in some anatomies.

It may be that collateral branches originate from the neck of the aneurysm, and if a stent graft is used, these branches will be occluded. This is the main reason why sometimes a stent graft cannot be used. For instance, when you have an aneurysm involving the common iliac artery, if the distal neck of the sac is very close to the hypogastric artery or if there is no distal neck at all, sometimes we have to sacrifice the hypogastric artery in order to put in a stent graft. Other limitations include the fact that most stent grafts have a limited flexibility: for this reason they cannot be used in very tortuous arteries. Another limitation is the need of large introducers. Finally, stent grafts are expensive — but the Cardiatis stent is also expensive, at least for now.

Can you tell us about the Cardiatis?

The Cardiatis (Cardiatis SA, Brussels, Belgium) is a multi-layer, bare stent made with a nitinol filament, creating a very dense mesh. The dense mesh allows us to maintain the patency of collaterals and simultaneously create a lowering or dampening of flow inside the aneurysm. There is a double mechanism of action: the endothelialization stops at the level of the collateral branches, while the mesh induces a dampening of flow inside the aneurysm, what has been described as “velocity vortex control.”

What studies have been done?

So far, no large studies are available, because the Cardiatis stent has only recently been introduced in the Italian market (it is not yet available in the U.S.). We began using this stent in June 2009, in 32 centers, and so have only preliminary experience.

Generally speaking, there are many issues of interest: first of all, the patency of the stent. We have had some stent thrombosis, maybe due to inappropriate antiaggregation, but at this point, we don’t know what the proper drug regimen is after Cardiatis stent delivery. The second issue is the patency of the collateral branches, and, so far, we have observed a maintained patency of all the collateral branches arising from the aneurysm or the necks of the aneurysm. The third issue is thrombosis of the aneurysm. We were able to obtain, in almost all cases, thrombosis of the aneurysm. It takes a varying amount of time. In some cases, the aneurysm thromboses in one day, perhaps in as little as a few hours. Sometimes it may take a few weeks or months.We don’t fully understand why. The fourth interesting feature is the shrinkage of the aneurysm. We documented aneurysm shrinkage in several cases, which is very surprising, considering that this is not a covered stent, but an open stent with dense mesh. With the Cardiatis stent, a Venturi effect is created that reduces the tension inside the aneurysm. This tension inside the aneurysm is due to turbulent rotational flow. If you are able to dampen or slow this rotational flow, probably the pressure is reduced, and for this reason, the aneurysm may shrink. The question is, why do some aneurysms shrink and some do not? It may depend on the amount of thrombus existing before the procedure, but thus far, we don’t know.

Any final thoughts?

This is a very preliminary experience, and so we will continue to collect data in Italy. In France, there is a very preliminary experience using this stent for aortic pathology, especially for aneurysms involving the origin of celiac artery, superior mesenteric and renal arteries, and for those aneurysms in which currently available surgical/endovascular repair is not convenient in terms of risk/benefit balance. Thirty cases have been done, but we have no follow up at this time. Moreover, it has been supposed that Cardiatis stent may be used for treating type A and B aortic dissections. Treating these severe aortic diseases may become a very exciting field of investigation.

Disclosure: Dr. Rabbia reports no conflicts of interest related to the content herein. Dr. Rabbia can be contacted at rabbiaclaudio@yahoo.it ————————————————————

Stenting in Chronic Venous Disease

Seshadri Raju, MD, Professor Emeritus of Surgery, University of Mississippi Medical Center, Flowood, Mississippi Can you tell us about your work with venous disease?

Our center has been focusing on venous disease since late 1970’s. Treatment used to consist of open venous reconstruction, meaning valve surgery with bypasses on the venous side, but within the last 10-12 years, it has become increasingly endovenous, both for the deep venous and superficial venous systems.

Venous stenting has some surprising aspects. On the arterial side, stents, of course, work reasonably well. On the venous side, one is always afraid it might produce thrombosis, yet that has not turned out to be the case. In what is called non-thrombotic disease or primary disease, stent thrombosis has turned out to be extremely rare, less than 1%. On our post-thrombotic patients, it’s slightly higher, as you may expect, but still less than 10%, and the clinical benefit is quite impressive. Resolution of venous symptoms, including healing of ulcers, resolution of dermatitis, and relief of swelling and pain, is perhaps as good or better than any other technique addressing deep venous insufficiency. What’s more, it seems to be effective in combined obstruction/reflux. This is preliminary data, yet to be confirmed by other centers, but venous stenting seems to provide clinical relief even when there is valve reflux. Traditionally, most venous symptoms have been viewed as originating from venous reflux. Many of these patients have an obstructive component, observed with intravascular ultrasound. If you correct the stenosis, there seems to be clinical relief, even if you don’t do anything about the reflux component, which is quite surprising. We ourselves were quite surprised about it. Some of our data has been published in the Journal of Vascular Surgery,¹ and it remains to be confirmed by other centers. If it is confirmed, it is going to represent a major paradigm shift in the way patients with venous disease are treated.

While you are reporting very good patient outcomes, others have argued that stenting is not appropriate.

You have to choose your patients carefully. You don’t do prophylatic stenting and stent venous disease that’s not appropriate. On the arterial side, it’s slightly different. On the venous side, you stent only when there are severe symptoms and when conservative therapy has failed. Take leg ulcers, for example. This is a serious quality of life issue. The only conservative treatment currently available is compression therapy. There are no known medications that in and of themselves are effective. Patients may perhaps receive some ancillary help with medications, but the basic treatment is compression. Yet a substantial number of patients with compression will fail to heal their ulcers, and in others, the ulcers will reoccur. These are the appropriate patients in which to consider vein stenting. The same is true for other severe symptoms. Venous pain can be quite severe. Venous leg swelling can be disabling and in some patients, can be a consideration for nursing home placement. Of course, you always do elastic stockings and compression first, and if it doesn’t work, or when it fails, these patients would be appropriate candidates for looking at other alternatives, including vein stenting.

How does vein stenting differ from arterial stenting?

The stent delivery is the same, but the technical details are entirely different, because the pathology is different. In arteries, we are dealing with athereosclerotic plaque pathology. In veins, it is fibrotic stenosis; therefore, the technique is different. It is a high-flow, low-pressure system, and the technique, in many ways, has to be more exacting than in the arterial side. The arterial side is a high-flow, high-pressure system. Minor technical imperfections are well tolerated. On the venous side, they are not. The stent may stay open, but you will fail to get decompression of the leg; you will fail to relieve the symptoms. So there are significant differences in the technique. Access is quite different in the arterial side. You want superficial locations where you can directly compress the artery. On the venous side, you can access the vein at deep locations, because it is a low-pressure system. Access complications are rare, especially if you use a sealing device. There are multiple differences in the way venous stenting is approached as compared to arterial stenting. Both self-expanding and balloon-expandable stents can be used. Stents are being used off-label, so it is not an approval indication as such, even though many are working towards getting approved for venous use.

Disclosure: Dr. Raju reports that he holds stock in Veniti, Inc. Dr. Raju can be contacted at rajumd@earthlink.net.

Reference

1. Raju S, Darcey R, Neglén P. Unexpected major role for venous stenting in deep reflux disease. J Vasc Surg 2010 Feb;51(2):401-408.

————————————————————

A Peripheral Angioplasty Balloon with an Embolic Capture Capability: The Proteus Device

Gunnar Tepe, MD, Head of Diagnostic and Interventional Radiology, Academic Hospital of Rosenheim, Rosenheim, Germany Can you tell us about the Proteus device?

The U.S. FDA has cleared this device for peripheral transluminal angioplasty (PTA) and for capture and containment of embolic material during angioplasty in the lower extremities. Initially, the device functions as a normal angioplasty balloon: the balloon is semicompliant and is inflated to a nominal pressure of 8 atm. After angioplasty, the balloon is deflated to 2 atm. At this point, arterial flow remains occluded, and the balloon folds inward, essentially “sucking” in potential embolic debris as it rolls into itself.

What do we know about its effectiveness thus far?

The Proteus (Angioslide, Inc., Minneapolis, Minn.) has not yet been compared clinically to other devices, but we know that it captures a great deal of embolic material. Currently, it is unknown how important embolic capture is in the peripheral arteries. It might be best used in those patients where embolic protection would normally be used, or where thrombus formations might be likely to occur, such as with atherectomy.

Can tell us about the Multi-Center Study for Lower Extremity Angioplasty with Debris Removal (MC-LEADER) clinical trial?

The study, now final, enrolled 123 superficial femoral artery and popliteal disease patients. MC-LEADER was the largest study of the Proteus device, done at three hospitals in Germany. The purpose of the study was to look at the safety of the balloon and its efficacy related to embolic protection. There was a one-year follow up endpoint. The device showed to be effective and safe. Significant amounts of particles have been captured. After one year, the restenosis rate was at the low end of what is published in literature.

Are there any concerns with artery dissection?

The study showed a pooled clinically significant vessel dissection rate of Any final thoughts?

It’s slightly more time consuming to remove, compared to traditional balloon use, although the time is much shorter compared to use of a separate embolic protection device. Sheath size is one French size bigger than for a regular balloon, but if stents or an atherectomy device are used, the device is actually a smaller sheath size.

A new possible indication for the balloon might be to use it as a drug-eluting balloon, but currently the FDA has not accepted drug-eluting balloons for approval or studies, because embolic particles might travel downstream. With the use of this balloon, this problem might be less of an issue.

Disclosure: Dr. Tepe was part of the MC-LEADER study. Dr. Tepe can be contacted at gunnar.tepe@ro-med.de