Drug-Eluting Stents: Current Limitations

Mike Cowley: Let’s begin with discussing case selection. Which patients should receive drug-eluting stents (DES)? Should they only receive one stent? Should we reserve DES for a select patient population? What should the approach be for multi-vessel disease patients? Christopher Cates: Those constitute the salient questions regarding DES. In the U.S. today, this issue hinges on the economic factor, which we seem to be tiptoeing around. I suspect that if cost were not an issue, there would be no talk of patient selection. My first comment involves DES and their economics which were discussed at the previous IAGS meeting. Much of that has been modified by the work that Cordis Corporation has done. I think Cordis is to be commended for obtaining a DRG designation for increased reimbursement. Alex, Brian and Paul Marshall, at Cordis, have shown us how industry can work with providers to help alleviate some of the economic issues on the front end. It nonetheless leaves us with issues regarding how to treat these patients with complex multi-vessel disease or those with longer lesions than was initially estimated. The cost issue has been very problematic for us at Emory University, because we have tried to look at it as one of the major factors. In fact, we evaluated our penetration of CYPHER™ sirolimus-eluting stent (Cordis Corp., Miami, Florida) use in our program and were surprised to find that it was only about 40% due to size availability and, more importantly, the cost in multi-vessel lesion cases. It would be very interesting to poll the panel here on what their institutions’ approach is for multi-vessel patients. Emory University has taken the approach of considering the LAD — as Jorge pointed out — as the most important site for CYPHER stent placement, and using other stents in the other arteries of multi-vessel lesion patients. Greg, how do you treat multi-vessel lesions? Greg Braden: We are also interested in cost. When we encounter a patient with two or three-vessel disease, we try to stage the interventions, treating the most significant lesion first, and then bringing the patient back a week or so later to treat the other lesion(s). This approach helps in terms of economics for the hospital, but it doesn’t particularly help the practitioner. Obviously, cost becomes a big issue when we place four or five stents at a time in a patient. In patients with two-vessel disease, we try to get away with placing two or three stents, and complete the treatment at one time, and then we look for lesions that are suboptimally suited for stenting, and avoid placing CYPHER stents in those particular lesions. I tend to avoid stenting some bifurcation lesions or vein grafts — at least the first time around — as a way to keep costs down. Tim Fischell: We are a little more aggressive with CYPHER stent use in most cases at our institution. CYPHER stents are placed in approximately 85% of our patients. I do want to applaud Brian and Cordis for their efforts to increase reimbursement for these stents. At our hospital, we tend to break even in terms of cost when 1.5 drug-eluting stents are used per patient. That 1.5 figure is my personal target as well, and requires that I be selective. For example, in cases where I place three CYPHER stents, I will need to perform three or four other cases where only one stent is placed. There are very few new technologies that are break-even for the hospital, so if your break-even point is 1.5 DES and you’re placing an average of 1.1 DES per patient, then you’re probably under-utilizing the technology. As Greg has suggested, there are ways to keep costs down by returning to good ol’ balloon angioplasty for some of the long 2.0 mm lesions in a diagonal branch, for example. Those lesions can be ballooned and a CYPHER stent can be placed in the proximal LAD, and the postero-lateral artery could be treated with balloon angioplasty as well. The 60% circumflex lesion could thus be staged, as it may not necessarily cause angina. The patient could then be brought back 30 days later for a stress test. If he is ischemic or experiencing symptoms, his circumflex artery could then be treated. I believe that this is an ethical and medically appropriate way to stage procedures and to more effectively use resources, while at the same time protecting the hospital. We all must make trade-offs while also honoring the Hippocratic oath, which dictates that we do what is best for the patient. But that does not mean that we should ignore other factors such as economics. I still have a hard time using bare metal stents; I’d rather balloon the lesion, then place a CYPHER stent three months later, as opposed to facing bare metal in-stent restenosis down the road. The MARSEILLE Registry data are fairly good; it showed a 2.8% six-month target lesion revascularization rate for 430 in-stent CYPHER cases. We have essentially abandoned brachytherapy based on the data now available. Jeffrey Werner: I don’t understand this, so please correct me if my thinking is wrong. I don’t use CYPHER stents in multi-vessel disease patients — not because I don’t think the stent is effective, but isn’t it the “weakest link” theory if you’re talking about the patient as a whole? In other words, the goal is to eliminate recurrent angina and the need for a repeat procedure. If two or three stents are placed, you might use two Hepacoat or bare metal stents, but there will still be a 25% restenosis rate in the other two non- CYPHER stents. Thus, the benefit of the CYPHER stent used in the one lesion is lost, since the patient still has a 25% chance of needing further intervention because of the non-DES used. In multi-vessel disease patients, thus, I have been using Hepacoat stents, and have used CYPHER stents in the more ideal, or smaller lesions. Tim Fischell: I am trying to adhere to the Hippocratic oath and do what is best for the patient. You can send the patient for bypass surgery or place three CYPHER stents, bite the bullet, and then find four other cases where only one CYPHER stent is required. I know, however, that if I place bare metal stents in those three lesions, the patient will have a 25% chance of restenosing. So I disagree with your approach. Mike Cowley: We posed many similar questions regarding bare metal stents when they first became available. Even today, if you need to do "full metal jackets, bare metal stents can run up costs as well. But in terms of doing what is best for a particular patient, as opposed to considering hospital or population dynamics, there is indeed a conflict. In a perfect world, DES might be placed in every lesion for which appropriate sizes are available. Since this is not the case in the real world, however, there appear to be a variety of approaches to take which can be outlined in the following questions: 1) Does each patient receive just one DES? 2) Should diabetics be given DES, and if so, should diabetics receive multiple DES? 3) Should the physician look at the patient from the possible bypass candidate standpoint, with the goal of reducing the need for bypass surgery? Christopher Cates: Perhaps we could poll our audience to ask what the CYPHER use rate is at your respective in situtions. The question could be divided into three parts: 1) 75% use. (show of hands) Approximately 10–15% of you raised your hand to indicate that your hospital has 75% use of DES. So there are about 30% who did not vote here. Next, I would like to poll the audience about whether it is common practice at your facility to place a CYPHER stent in one artery and bare metal stents in the other? (show of hands) About 20% of you indicated that this is the case. How many of you use CYPHER stents in every artery in multi-vessel lesion cases? (show of hands) None of you do, according to the show of hands. I suppose that in a perfect world, we would all answer that question affirmatively. I would like input now from our European colleagues on these same questions. Alfredo Rodriguez: In Europe and South America, it is impossible to use DES in multi-vessel lesion cases. In Argentina, we conducted registries at five centers on multi-vessel CAD called ERACI III. These registries showed very good results. We performed interventions on two ERACI trials at our institution, one with balloon and one with bare metal stents, and we observed a decrease in the TVR from 30% with balloon only in ERACI I, to 17% in ERACI II with bare stent therapy. And now in ERACI III with DES, we have an 8% TVR rate at one year post-intervention. But it is impossible to use DES in all of our patients because they are obliged to pay for these stents out-of-pocket. I think we need to find alternative solutions. Besides, there are not enough scientific data to support the use of DES in 100% of small vessels, bifurcation lesions, left main artery lesions, and kissing stents — regardless of the cost. Barry George: I would like to return to the notion that we are using mechanical means to treat a biochemical problem. Also, in an ideal world, all of you on the panel would use CYPHER stents in multi-vessel disease patients, and you would pay for them if you had to. We have all been doing this for a long time and have been waiting for a magic bullet to lower restenosis rates to the degree that DES have been able to do. Let’s be doctors and not hospital administrators. We use a novel approach at Riverside Hospital, where a very unique relationship between administrators and cardiologists has been established. Cardiologists are part of the management team and are paid as management team members by the hospital. I advised the administration to put some money into the “war chest” to cover the costs of DES as soon as they were approved. Thus, during the years prior to DES approval, our hospital collected approximately $300 million in cardiovascular services and put some money aside each year in anticipation of this new technology. I told administration there was no way that they could tell us doctors what to do and how to manage DES use. Thus, we are currently weathering the storm well. The TAXUS™ stents (Boston Scientific, Maple Grove, Minn.) will be available soon, and capitalism will once again prevail. But we must stop acting like hospital administrators and economists and do what’s best for the patient. David Sperling: It seems to me that there is a significant difference between a clinical trial and a registry. We have a lot of data from registries, but I’m not sure that we can say that the evidence-based medicine currently available indicate that DES should be placed in every patient. I would be interested to know how many of you here would, without hesitation, use CYPHER stents for left main arteries, saphenous vein grafts and in-stent restenosis. Is this being done uniformly now in this country? Ted Bass: The truth is, we are not going to get well-defined, evidence-based results on every subject. If you look at the results from some of the large registries, about 60% of DES use is off-label, so it is very difficult to capture precise data. Companies are now trying to formulate tighter, more closely-governed registries by setting up QCA arms and stressing deployment techniques, for instance. I do think that we learn a lot from the registries, though. First of all, deployment techniques have evolved. We are using longer stents and improving on sizing, and we are learning about what to avoid. Thus, when practitioners say they are using DES In 100% of their multi-vessel disease caes, I’m not sure how that can be, because at my institution, we don’t have the equipment to address multi-vessel disease with DES. Improvements in the field are coming. Patrick’s group will soon be publishing a paper that will look at overstretching some of these things from their research registry. The data are not as compelling. Is it stent failure, or device failure? Or have we not been given the appropriate-sized stents? The same thing was was observed in the SECURE registry regarding saphenous vein grafts. The results were good against historical controls, but they were somewhat disappointing when compared to the results from tighter, randomized trials such as SIRIUS and TAXUS. I tend to lean more toward the registries now, especially for post-market release stents. The SIRIUS data, with 100,000 patients, will probably offer a good amount of compelling data such as information on efficacy, CTOs, and other new areas. Kenneth Kent: Our hospital performs more than 400 angioplasty procedures monthly. When the CYPHER stent became available, we planned to use 1 CYPHER stent per patient. For the first two months, we were averaging about 1.1 CYPHERs per patient — a number which was driven to some degree by availability of the product. Now that availability is not an issue, that number has edged up to about 1.3 CHYPERs per patient. Over one-half of the angioplasties we perform involve single-vessel lesions, though the patient may have multi-vessel disease. Thus, we are focusing on about 40% of the patients, and of those 40% who undergo multi-vessel stenting, there often is not an appropriately-sized CYPHER stent available, or it cannot be technically delivered. More importantly, some lesions — such as a 4.0 mm lesion in the RCA — do not require CYPHER stents. Furthermore, there are no data on using multiple DES in the same vessel, but fortunately, we have long CYPHERs to treat these lesions. I don’t think that patient care is diminished if CYPHERs aren’t given to all patients. We must first be sure to use these stents properly, to ensure they’re expanded, and perhaps we ought to reconsider using ultrasound routinely in order to use that $3,000 product optimally, because if you’re going to take such an expensive product off the shelf, you should make optimal use of it. And let’s not forget what we’ve learned in terms of cases where a DES is not necessary. I think we can do both: protect the hospital and provide good care. Our goal is to use 1 DES per patient, which we do in about 70% of the patients who undergo angioplasty at our center. But it is a mistake to use multiple DES in a patient without considering if they are truly necessary. Christopher Cates: I want to add a “con” to Barry’s “pro.” When a disconnect between the cost of a technology and patient care exists — a problem that came to head in the early 1990s — we must understand that medical economics is a zero-sum game. If we don’t exercise some constraint and apply evidence-based medicine to the care of our patients — with cost as one of those factors — then we will absolve our responsibility to the economic end for the sake of the patients. Someone will make the decisions, it just won’t be the physicians. If the issue isn’t DES, it will be AICDs, or the next new therapy. Thus, when there is a very expensive, high-penetration technology, we physicians must be the ones to weigh the medical versus the economic considerations. It’s not just the hospital’s problem, it’s ours as well. Brian Firth: If you look at the health economics of DES from a societal perspective, which is actually the same perspective as the payer, our view at this stage is that DES are under-reimbursed. In fact, what is happening is that the payers are saving money at the expense of the hospitals. That situation needs to be re-balanced. The game is not over as far as Cordis is concerned. Our objectives at Cordis are three-fold: 1) Ensure that DES DRGs are kept separate from bare metal stent DRGs; 2) increase the reimbursement for DES; CMS will do that based on their MedPAR charge data; 3) pursue different DRGs for multi-vessel stents. The PREMIER hospitals database has excellent data on the charges for single-vessel versus multi-vessel stenting, as well as the utilization rate of CYPHER stents versus bare metal stents in single-vessel versus multi-vessel cases. There is a much greater mix, as you might assume, of bare metal stent use with DES in multi-vessel procedures. The PREMIER database shows a 2.3 DES per patient average in multi-vessel cases, and 1.45 DES use per patient in single-vessel cases. Thus, the utilization rate has not changed much in single-vessel cases. The charge difference, however, is approximately $10,000 between patients who undergo multi-vessel stenting versus those who undergo single-vessel stenting, not withstanding the mixed use of stents in multi-vessel cases. We will provide the charge data to justify it and have Medicare evaluate this. Second is the clinical piece of the equation. CMS has already stated that they will not do anything in the multi-vessel area until compelling clinical data are provided. Fortunately, the ARTS II registry is nearly completed, but data won’t be available for another year or so. These data will be an important piece of the puzzle in terms of building our case. When the economics of multi-vessel DES versus CABG are evaluated, the payer and society win hands down. Christopher Cates: Brian, one of the things I jotted down while we were talking, and one of the issues that has come up in Atlanta, was the letter from Cordis about a year ago regarding off-label use and the avoidance of certain patient classifications or patient characteristics when using CYPHER stents. This letter was followed up in the National Register from the FDA regarding off-label use and liability if, for example, there is a problem when placing DES in an AMI or saphenous vein graft (SVG) patient. Obviously, that emanated from MDR reporting of untoward events, but of course, prior to release, there were no MDR reports, so any new MDR reporting raises questions. How is Cordis handling that? And what are you recommending to us, because there is a wide variability of CYPHER stent use in subsets of patients who fall outside of the stent’s indication criteria. Brian Firth: Let’s step back from this. When people start talking about off-label use of CYPHER stents, what do we think they are doing with bare metal stents? To the best of my knowledge, there isn’t a bare metal stent that is currently approved for SVG, AMI, or bifurcation lesions. We went to the trouble to get SVG indications for the PALMAZ® and CROWN stents (Cordis Corp.), but we haven’t sold those in years. Thus, it’s ironic that people talk about off-label use of the CYPHER stent when bare metal stents are used off-label every day. Christopher Cates: But there is no letter saying, “Don’t use it in this patient subset.” Brian Firth: That’s correct. That emanates from the controversy surrounding thrombosis. Frankly, we think that the FDA jumped the gun by issuing a notice on thromboses with the CYPHER stent that they subsequently retracted. But unfortunately, they are like the traffic cop who says, “You were going 70 mph,” and you have to say, “Yes sir, how can we work with you?” It was a very frustrating time for us because non of our data suggested that there was a problem with stent thromboses with CYPHER, and I know it was very frustrating for physicians as well.. I do understand your question. I think that you need to consider the follow-up letters from the FDA which state that they did not think there was any issue, and perhaps use those letters in concert with the earlier one if a legal issue were to arise. Ted Bass: There is also a legal website for people who think they’ve been harmed by DES. In terms of off-label use, it is important to consider — as ultrasound has shown us — that these are all very different stents. We are going to soon have choices, and those choices may involve understanding the platform, the polymer, and the different diffusion properties of the drug — not just off-label use. It would appear that one stent’s platform may not be as optimal as another’s. Perhaps one stent’s drug is slightly less effective than another’s drug in terms of its mode of delivery. Thus, we will have to integrate this knowledge, as well as the lessons learned form ultrasound and from the randomized trials, to optimally use the DES that will become available for off-label use. This will be very challenging indeed. Mike Cowley: The real world, to a large extent, is off-label! Jamey Jacobs: I would like to comment on the reduction of in-stent restenosis rates which significantly raises the bar on some of these performance attributes. Guidant Corporation looks at the DES program as a partnership in which our company needs to provide devices that are much less injurious — not only adjacent to the stent, but also in getting to the lesion. The path to the lesion will become a more important issue now that in-stent restenosis has been addressed so effectively. The implanting procedure, with respect to whether IVUS is used or balloons outside the stent, etc., is now at the forefront. Industry must partner with the medical community to determine the appropriate procedural techniques that will minimize restenosis rates so that we can return to the fervor created by RAVEL and other trials. This will help to eliminate arguments over such issues as whether we are just treating disease within the stent. Another performance issue that has not yet been addressed is that of traditional, difficult lesion subsets. Are diabetics just an injury issue, or do they truly require a different pathway? Long, diffuse disease presents issues such as avoiding stent overlap, which comes back to providing the right device that will minimize injury and overlap and that enhances capabilities. The same goes for bifurcation lesions; the right device is needed for these lesions. The pharmacokinetics change dramatically when there is a drug on a stent in a bifurcation lesion. These are traditional performance/device/procedural issues that raise the bar in terms of developing optimal procedures and tools. William O’Neill: This discussion reminds me a little of the history of this meeting and the history of angioplasty in general. I clearly recall the first time I heard Jeff Hartzler speak about multi-vessel balloon angioplasty for the treatment of complex lesions. His ideas were radical then, whereas the Gruentzig school was quite conservative and Germanic in its approach. Much controversy arose — and many of us in the field pushed the envelope. It is fun to observe the tension at this meeting between those who prefer to go “by the book” and those who use a more radical approach to treatment. But that’s how the field will advance. I want to comment on what Barry George said about the conflict of interest for physicians. We must all be honest with ourselves about the enormous conflicts of interest in this domain. Some recent data suggest that at the end of 2004, CYPHER stent use in American for-profit, physician-owned hospitals will be 22%, whereas elsewhere, that number is 60%. So where are the ethics? Where is the Hippocratic oath? Or are they being more judicious? In our own practice, diabetics and chronic total occlusion patients represent a huge imperative for DES use because the failure mode for those patient subsets is restenosis with reocclusion, which is often very difficult to treat. We have not discussed our relationship with our surgical colleagues. In Southeast Michigan, there has been a 30% reduction in the number of open-heart operations from one year to another. The hospitals are getting very nervous, along with our cardiac surgical colleagues. Tension has begun to build, and it will be very interesting to see what happens if we do away with surgery to a large extent. What will the hospitals do? And what will our cardiac surgical colleagues do? The relationship could become very adversarial. Tom McNamara: A scenario I have observed in my own practice involves a patient who has a second restenosis in a stented renal artery; the stent was dilated to 6.0 mm, and the patient only has the one kidney. The standard approach, whether repeat balloon angioplasty or placement of another stent, has not worked. Does anyone know what happens if a 4.0 mm coated stent is dilated to 6.0 mm? Does it work as well if it hadn’t been over-dilated? And secondly, what about reimbursement for this procedure? Does that patient need to be told up front that this procedure costs more, and that if the hospital bills correctly for it — that is, it indicates that the stent is not being placed in the coronary arteries — that the patient may have to pay the difference? Most hospitals, unfortunately, would take that difference — let’s say $1,000 wholesale — and will charge $3,000 or $4,000. And in my experience, the hospital will be heartless about letting the patient off the hook on paying the difference on that inflated price. Tim Fischell: Currently available in the U.S. are the 2.5 mm, 2.75 mm, and the 3.0 mm, which is a 7-cell stent. The 7-cell stent is the same as the 4.0 mm, which is now the subject of a registry trial aimed at obtaining FDA approval. The 4.5–5.0 mm 9-cell stent would be the most ideal for use in the renal arteries. The 9-cell stent, as far as I am aware, is not in trial yet and is not currently available. Trials are planned for the GENESIS™ balloon-expandable stent (Cordis Corp.), which is intended for renal-type use. The answer to your first question is: The 7-cell 3.5 mm CYPHER stent can probably be pushed up to about 4.75 mm, but I would not recommend expanding it beyond 5.0 mm, as it could result in strut fracture. Also, it won’t be reimbursed because I don’t think it’s indicated on the product label, and the patient may not receive an adequate dose of the drug. It would constitute, thus, an off-label use in that case. We do need larger stent sizes which will be useful in the renal, iliac and SFA arteries. But we do not have the tools needed in terms of 9-cell self-expanding stents and next-generation stents from other companies. A decade from now, this debate will seem silly, just like we now view the debate in 1995 over stenting lesions versus ballooning them. We know that stents are better for just about every type of case, but that doesn’t mean balloon angioplasty will no longer be performed. Bare metal stents will also still be used in some patients ten years from now, but DES are better and will be the tool of choice. Also, as always happens in the marketplace, DES prices will come down, making them more affordable. Brian Firth: To answer Tom’s question about reimbursement, there are double-0 codes for peripheral reimbursement, but they don’t map to any higher-paying DRG at this point because there are no data to support them. Thus, the payment will be the same, although the first step — separate codes — is in place. Jeffrey Werner: Does it make sense to use one CYPHER stent, and then several other non-DES if a patient has at least 25% restenosis, or a chance of returning for treatment anyhow? Other practitioners have told me that they use all CYPHERs and their patients show no restenosis, but I have not heard anyone say that mixed stent use is an effective and rational strategy. Tim Fischell: Even from a hospital administration’s standpoint, you are better off placing one DES because the hospital will be paid $2,000 more, and it will only cost the hospital an additional $1,000 to $1,500 for that DES. You can thus add $300 to $500 to the hospital budget’s bottom line by using at least one DES. I don’t necessarily agree with that strategy, however. Rather, I would choose to implant a DES in the highest-risk lesion that supplies the most myocardium and that has the greatest risk of restenosis. In other words, place the DES in a patient’s proximal LAD or the mid-LAD that feeds 25% of the myocardium, not in the tiny postero-lateral RCA, which should instead be ballooned or given a bare metal stent. Thus, the DES should be placed where it will benefit the patient most. The money can then be collected by the hospital under that DRG. Jorge Belardi: I think it’s fair to put one DES in a patient as long as it’s being used in one vessel. What is not fair, however, is to use one bare metal stent and one DES in a vessel. It is reasonable to use a DES in the LAD as long as the entire lesion is covered with that DES. Then, a bare metal stent can be placed or conventional angioplasty can be performed in the other vessels. Alfredo Rodriguez: Combination therapy can be done by placing a DES in the vessel with the most severe stenosis and bare metal stents in the other vessels, along with the administration of oral rapamycin. Jeffrey Werner: Obviously, we all want to do what’s best for the patient, and we are all appreciative of the effort Cordis has made in increasing reimbursement to help with the costs of this technology. When we polled the audience earlier — and correct me if I’m wrong — it showed that most of you who use DES for everything, including multi-vessel disease, come from larger hospitals where there’s a certain break-even point beyond which the hospital begins to make the same or greater profit due to its DES volume. Is the situation different at smaller hospitals that cannot achieve that break-even point and actually lose money on DES use? Is it a question of large versus small hospitals, and if so, how do we solve that problem? Alex Zapolanski: I believe I am the only heart surgeon here and thus I feel obligated to say that you are all doing a wonderful job! Cardiac surgeons need not get upset about this…I didn’t get upset 20 years ago when Richard Myler performed a double-kissing balloon procedure. The cardiac surgon of the 21st century will have to be “reinvented.” Only 25% of the Cleveland Clinic’s cardiac surgery pathology studies involve coronary work; the rest involves valves, arrhythmia surgeries, and so forth. The only thing we can do is improve operation techniques in areas such as beating-heart for those select patients who are sent to the O.R. I have one question: If a patient’s creatinine is 3 and she needs double-vessel revascularization, is she better off undergoing beating-heart surgery or intervention? I don’t think anyone has the answer to this question, but in some subsets of patients, surgery is probably the better option. Chrisropher Cates: The patient may also need a renal arteriogram! Paul Overlie: The last word here should have something to do with off-label versus on-label use. Many of you here have advised me through the years to keep tack of my own statistics, because if I don’t, someone else surely will. If we are intervening on left main arteries, diagonals, bifurcations — with or without new devices — we need to track what we are doing and have statistics available to report. Registries are important, as are multi-center randomized trials.