Cardiovascular Business invited five luminaries in the field of interventional cardiology to engage in a discussion about stents, including the choice to use PCI over CABG, DES versus BMS, a specific stent over contracting and the radial approach versus the femoral approach—always with an eye toward running a better practice.
|Listen to highlights from the interview|
Participants of the discussion are:
- Gregory J. Mishkel, MD, interventionalist at Prairie Cardiovascular in Springfield, Ill.
- Jeffrey J. Popma, MD, director of innovations in interventional cardiology at Beth Israel Deaconess Medical Center and associate professor of medicine at Harvard Medical School in Boston.
- Gregg W. Stone, MD, director of cardiovascular research and education and an interventionalist at New York-Presbyterian Hospital/Columbia University Medical Center in New York City. Vice chairman of the Cardiovascular Research Foundation, which conducts the Transcatheter Cardiovascular Therapeutics annual meeting.
- Christopher J. White, MD, chairman of the department of cardiovascular diseases at Ochsner Clinic Foundation in New Orleans, Editor-in-Chief of Catheterization and Cardiovascular Intervention.
- Alan C. Yeung, MD, division chief and director of interventional cardiology at Stanford University Medical Center in Stanford, Calif.
- Moderator: Justine Cadet, News & Features Editor, Cardiovascular Business
What is the decision-making process to stratify patients with coronary artery disease (CAD) to receive either PCI or CABG at your facility?
White: At Ochsner, six interventionalists perform about 4,000 caths per year and 1,500 interventions. We make individual choices about how to manage our patients, and do not have an algorithm or a structured decision-making model. We occasionally use the SYNTAX Score to help stratify patients for risk. Generally, we use bypass surgery in patients who have extensive vascular disease—three-vessel complex lesions or chronic total occlusions (CTOs). However, multivessel disease, in which each vessel would be a straightforward single-vessel intervention, is routinely handled with PCI. We average about 2.1 to 2.2 stents per patient which averages high, refl ecting our aggressive interventional approach. While our surgeons are not involved in the decision-making process, we typically present patients with left main disease to them.
Popma: Deciding whether a patient receives CABG or multivessel PCI is a multidisciplinary approach at Beth Israel. We work with our surgeons and ask their opinion on patients who have complex CAD, particularly with respect to comorbidities, such as prior stroke and reduced left ventricular ejection fraction, or the elderly. While we are not routinely using the SYNTAX Score, we apply the concepts of the score, in such cases as complicated left main bifurcation disease or diffuse disease in diabetic patients. If there are no surgical contraindications, those patients are moved to CABG. With respect to the specific subset of left main disease, we also have been using surgery as the default therapy.
Stone: We have a collaborative relationship with our surgeons, but don’t regularly consult them. Surgery could be recommended for left main disease, complex triple-vessel disease or a nonrecanalized CTO. However, it’s only about 10 to 20 percent of the cases that could go either way. In that small population, we stop the procedure, take the patient off the table and call for a formal surgical consult. This is most important in patients with complex triple-vessel or left main disease. All left main cases if they are unprotected would be good surgical candidates, although we continue to perform PCI on unprotected left mains in the post-SYNTAX era. Yet, the patient has to be informed because it hasn’t yet made the guidelines or received widespread acceptance. We do not use the SYNTAX Score because it was only tested in a cohort of patients with triple-vessel or left main disease, and hasn’t been tested against other predictive instruments.
Yeung: We have a long-standing conference on Saturday mornings, which is attended by most of the cardiac surgeons and cardiologists both from the university and community. Generally, we use that forum to discuss patients who could receive either treatment. The discussion is very helpful because it establishes a seamless method to approach diffi cult cases. Instead of the SYNTAX Score, we mainly employ qualitative eyeballing, assessing how many bifurcations, how many CTOs are involved, along with the number of stents required. Also, cases of left main disease, heavily calcified coronaries or a bifurcation are all generally discussed with the surgeons.
Mishkel: We also don’t routinely meet with our surgeons or calculate a SYNTAX Score. According to the SYNTAX Score Web site, the score works best when there is a panel of three MDs calculating the score, which is impossible in a private practice. However, using the ideologies behind the SYNTAX Score, we assess and form a gestalt. In terms of collaborative relationship, we have dramatically reduced the use of ad-hoc PCI for elective patients or stable angina patients. Most often, a patient with known CAD will be taken off the table and all features of their history will be examined, and particularly for complex patients, a surgical consultation will be sought. As a matter of routine, we do not perform left main stenting at this time, but certainly if medical co-morbidities are present we would consider it. We remain concerned about the exact duration of dual-antiplatelet therapy in
these patients and although stent thrombosis in SYNTAX was low, when it occurs, it is catastrophic—all the more so in patients with left main disease.
Discuss factors which have influenced your PCI volume at your institution. Where do you see it going in the next 12 months?
Mishkel: In 2004, we had placed 4,055 stents, 97 percent of which were drug-eluting stents (DES). Last year, we placed 2,466 coronary stents, 70 percent of which were DES. We have seen a very significant reduction in our volumes. The COURAGE trial played a big role in that, along with the economic crisis. Being a rural practice, gas prices also are an inhibitor. Notably, there has been a change in the demographics of CAD. Rather than the epidemic of the past 20 to 30 years, we see more stable, less acute disease, resulting in less PCI volume. This negative trend will probably continue. Private practices have to set themselves apart and not be known only as volume shops. We now are focusing on quality markers, such as adherence to guidelines, as well as lowering readmission rates from MI and congestive heart failure.
Stone: From 2006 to mid 2007, we had the firestorm from studies released at the European Society of Cardiology that raised concerns about the connection of DES to stent thrombosis and increased mortality. People misinterpreted the data of COURAGE and felt they could manage patients without even diagnostic angiography. Some misinterpretation resulted from the trial’s decision to extract high-risk patients from its randomization. Also, current DES data continue to get better. Our interventional level dropped around 20 percent, but in the last year we have seen that rebound by at least half.
Yeung: Our PCI volumes are down approximately 10 percent. However, it appears to be stabilizing, and we can expect a little uptick in the next year. With IVUS and FFR supplementing angiographic judgment, coupled with other noninvasive tests such as stress echo, nuclear or CCTA, we have better ways of defining specified treatment. We certainly treat fewer lesions that are angiographically borderline, and we treat these medically according to COURAGE and FAME. Due to better practices, we don’t see a lot of readmission for CAD patients.
Popma: Our volume is down about 3 percent from last year. I agree with Dr. Yeung that we are going to stabilize, or even grow a bit. PCI volumes will rise a little bit, as we start to understand the limitations of medical therapy and get away from some of the bad press that affected PCI. Our volume downturn has to do with the indications for PCI, and the vagaries of referral networks. For instance, in Massachusetts, a lot of the elective and
STEMI angioplasty is moving out to the community hospitals and away from the tertiary-care centers, resulting in our biggest shift in volume.
White: Over the last 10 years, we have grown our volume each year, but that growth flattened out. As a referral center that performed a lot of brachytherapy for complex restenosis cases, we were greatly impacted most by the dawn of DES. COURAGE didn’t take much air out of our sails, but DES hurt our restenosis volume and made us work much harder to sustain growth. Over the next year, we expect our coronary volume to remain relatively flat, with only a 2 to 3 percent growth in the year-over-year increase. However, we see growth in structural heart disease. We also have a percutaneous left ventricular assist device [Abiomed’s Impella], which allows us to receive patients with very sick ventricles. We hope to make up our growth in areas other than straightforward PCI.
What is your current DES/BMS [bare-metal stent] ratio, and has it changed over the last year?
Yeung: Our current DES usage is 81 percent—it dropped to about 70 percent at its lowest point, compared to 90 percent after their initial introduction. It will probably stabilize around its current ratio. We use BMS in STEMI patients with whom we are unfamiliar or for those who may have dual-antiplatelet compliance issues. Any potential interference with Plavix compliance—either a planned surgery or costs—we use BMS. For STEMI patients, we also use DES after opening the occluded vessel, if the lesion is relatively long, relatively small or for diabetics.
Stone: At the peak, we were around 85 or 86 percent for DES, but that has fallen to about 80 percent, where it will most likely stay.
Popma: The academic centers seem pretty closely aligned, as our rate for DES is between 76 and 80 percent. The newly approved 2.25 millimeter DES [Boston Scientific’s Taxus Express2 Atom] will pick up some of the BMS market for smaller vessels. The larger vessels are just not that common, especially vessels more than 4 millimeters.
White: Our current ratio is around 70 percent for DES. We routinely use DES for STEMI patients.
Mishkel: We swung from about 98 percent utilization in the DES heyday to as low as 64 percent after the negative data, and now we are up around 75 percent. Most likely, 75 to 80 percent will be the ceiling. We have developed a formal Plavix checklist for patients, asking mundane questions like: “Do you fall on the Medicare donut hole?” which can be very illuminating about compliance. In the Midwest, STEMI patients often get flown in and are narcotized, so you can’t obtain their history and we end up defaulting to BMS, unless there are extreme reasons, such as diabetes or small vessels.
What percentage of your PCI procedures use the transradial approach? Do you foresee an increase in the radial approach in the U.S.?
Stone: We use the transradial approach in less than 5 percent of our patients, primarily in patients with little femoral vascular access, who are morbidly obese. However, the radial approach decreases access site bleeding complications, especially for interventionalists who perform them routinely. As bleeding risk data are proliferating, there will be an uptick in the U.S., but it probably will not reach where Belgium and France are-60 or 70 percent. It may, however, climb to 10 percent in the next five years.
Yeung: Previously, we only reserved that approach for vascular access issues. Yet, in the last year, we have been purposefully growing that approach using the 6 French approach because of the data suggesting less vascular complications, especially in women. Patients appreciate the quicker recovery due to less holding, and it potentially could be a major source of cost savings if patients are able to be discharged a few hours after angioplasty.
Popma: We use the radial approach about 10 to 20 percent of the time. I try to do one case a day radially. I choose this method for patients who may have groin bleeds, and it’s a good alternative for peripheral vascular disease and obese patients. I use it for routine interventions in appropriate patients, using a rotor blade with a 6 French guiding catheter. We perform ad-hoc angioplasties in the vast majority. I don’t hesitate about using the radial approach when not knowing the anatomy. The national levels are probably 1 or 2 percent of total procedures, but it will grow. It might not grow up to 20 percent, but some practices will start to endorse it, especially with the looming issues of outpatient APC reimbursements for elective PCI. The only plausible way to discharge patients in the same day is through a radial approach.
Stone: With these advantages, why not perform 90 percent of cases radially?
White: In my personal practice, about 90-plus percent of my cases are transradial and that includes both coronary and peripheral intervention. I do everything I can through the radial approach. Every week, a leading journal links periprocedural bleeding and bleeding complications in the cath lab to mortality.
Yeung: It’s a matter of experience. We would try to grow that percentage over time as physicians and staff become more comfortable with the radial approach. We also perform ad-hoc angioplasties. If there is multivessel disease, bifurcations or CTOs, we might do simple lesions through the radial first and then bring them back to the CTO, because they need bilateral groin access or we can get access in the groin using the radial access for the contralateral injection.
Stone: The Lahey Clinic has converted almost exclusively to radials, and begun a randomized trial assessing STEMIs with the approach. It would be a logistic challenge for us to convert to 90 percent right now because the nurses aren’t ready to handle this change.
Mishkel: The logistics are one of our biggest impediments, where we have 15 to 20 interventionalists. Although five of us may use it on a regular basis, we would not try to convert the other interventionalists to a radial approach, like they do in Europe or the Lahey Clinic, where you can mandate behavior. Also, in practices with younger cardiologists, the conversion may be easier.
How important is the specific DES versus contracting?
Popma: We view all DES as generics, BMS as generics, balloons as generics and wires as generics. With all the marketing efforts to profile clinical data or stent superiority, I am struck with the resounding thud the data make when purchasing agents are making deals. In this process, we need to stop the physician incentives. For instance, I will use any stent if it is $2 cheaper. However, we still need longer term outcome data and some better differentiators between the stents, which may come when Dr. Stone presents SPIRIT IV at TCT.
Stone: We don’t see everything as a commodity. I see differences in DES, BMS and guidewires, yet you can use most of them to accomplish the procedure. The completed SPIRIT or TAXUS trials, for example, show differences between the two stents involved. There are not huge differences and Dr. Popma has a valid cost-effectiveness argument. However, if the price difference is modest, then I would prefer to evaluate the data.
Mishkel: I agree with Dr. Popma that each stent is a commodity. Also, you have to commit to a certain volume to get the price with contracts. We have begun to disallow the vendor to dictate the price, but instead, we submit our price to the vendor. We would be willing to pay a little bit more for Promus or Xience than Cypher, but not on an order of magnitude. If you want to negotiate a lower-priced contract with a vendor, you will have to make major volume commitments. Practices also need to educate themselves on the average selling price for each stent nationally.
Yeung: The difficulty in this stent contracting process is that with a volume-based discount, you are committed to one particular vendor. For example, you select a stent that may be the cheapest, but you have to agree to use it 45 percent of the time. If you use it less, the price probably will go up and may even be more expensive than the other stents. The difficulty is that the vendors are trying to maximize their returns. If you sign up a particular company as your sole provider, you probably would get the best price for any stent.
White: In my practice, I require some choice and do not believe that all DES are created equal. It’s not so much that there are huge differences in the drug that is deployed, but there still are significant differences in deliverability, trackability and the interaction of a certain stent and a lesion. There are times when I prefer one type of device or one stent system platform. We contract for all of our stents, but it doesn’t diminish our choice.
How many cardiologists are performing peripheral interventions?
White: Our six interventionalists perform all forms of intervention, including peripheral, renal, carotid and neurovascular interventions. About a third of our volume is non-coronary work. Also, our cardiologists perform acute stroke interventions, typically two or three strokes a month.
Mishkel: We have about 25 percent of our interventionalists doing peripheral vascular work. The volume is no more than 10 to 20 percent of our interventional work. While coronary work remains the bullwork of the lab, interventional electrophysiology is becoming popular, especially from an economic standpoint. Interventional electrophysiology might drive the economics of the cath lab in the future.
Stone: Most of the 15 interventionalists in our practice regularly do endovascular work, but it still accounts for 10 percent of our interventions.
Yeung: We have six people and about 1.5 of them concentrate on non-coronary vascular work.
Popma: About 20 percent of our interventionalists do peripheral work, which takes place about 20 to 25 percent of the time. However, our vascular surgeons are quite busy with endovascular work, consuming about 40 percent of the business with the cardiac surgeons.