Seth Bilazarian, MD


July 31, 2014

This feature requires the newest version of Flash. You can download it here.

Editor's Note: On July 1, Dr. Bilazarian's practice was acquired and he became an employed physician. To reflect this change, his series will be renamed Practitioner's Corner.

To download the slides used in Dr. Bilazarian's presentation, click here.

Seth Bilazarian, MD: Hi. Seth Bilazarian on on Medscape for Practitioner's Corner. I was thinking recently about adoption of new therapies and what I've gotten right and what I've gotten wrong, and what my regrets and triumphs are in this area. I did this exercise in which I spent several hours thinking about my 20 years in practice, and I thought it would be an interesting exercise to share with you. Hopefully others will weigh in on what they have gotten right and wrong.

Before adopting, I realized that there are several steps in the adoption process. I think many of these are intuitive things that we don't actually take time to think about, such as, is it effective? Is it better than current therapies, whether those are technologies or drugs? Is it better than nothing (ie, placebo controls)? Is it safe? And, of course, is it reimbursed for the provider and for the institution? We're increasingly thinking about reimbursement from a patient standpoint -- that is, can a patient afford it? Is it in the correct payer category? Can I do it well if it's a new procedure or should I refer for it? And is it really high value for both the patient and the provider? It may seem indelicate to say for the provider, but certainly something that's not reimbursed adequately will be difficult to sustain in the current marketplace in US medicine.

The topic of early vs late adoption came up in a Twitter exchange I had with fellow blogger John Mandrola when the CardioMEMS™ device (St. Jude Medical) got FDA approval. I began thinking about that and about whether that was something to consider adopting. In our 140-character exchange, John expressed his thinking that this was an appropriate therapy to adopt late, to wait 3 years, because then it will become clear whether this technology was worthy of adoption. I had another Twitter exchange with a different user who shared the idea that there's been inadequate or slow adoption of the new dual-antiplatelet therapies to the market, and these things made me think a lot about what we are thinking in terms of early vs later adoption.

What I found when I've talked to different adopters (for disclosure purposes, I'm an early adopter) is that both early and late adopters consider their position to be virtuous or ideal. Being on either end of the spectrum is deemed ideal, and there are arguments to be made for both. Being an early adopter of a therapy that proves to be effective, safe, and of high value is ideal, but if it's really none of these then that's far from ideal. Early adopters are not universal adopters -- that's certainly not my case. Being a late adopter is not ideal if the drug, device, or therapy is proven to be effective, because patients were denied an effective therapy during the period of waiting. Of course it's ideal to be "just in time," to take a term from manufacturing, but that's obviously very difficult to do on a prospective basis. There are also significant geographic differences when I talk to physicians around the country, or my patients who receive care in other parts of the country, in terms of adoption in regional parts of the United States or internationally. There are also differences between suburban and rural centers and urban centers, so there are certainly variables in early vs late adoption.

It's not always a clear case of early or late; there are shades of adoptions. I don't think early adopters use a new device or drug in 100% of their patient population; they use it selectively. But I've noticed differences in that some physicians are early adopters of preventive diagnostics or preventive therapies, but not so for disease treatments. I also have noticed that there are some physicians -- interventionalists, for instance, or sometimes electrophysiologists -- who are early adopters of devices but may not be early adopters in drugs. And, of course, there's also the overlay of adoption that may be influenced by where we practice or by payers that may influence our ability to adopt, even if we're interested or enthusiastic about adoption.

At the end of the day (I don't think this is a Pollyanna or idealistic view), what really drives early vs late adoption is the physician's desire to do the right thing for patients. Patients, of course, have a bias; many patients have the view that newer is better, but many patients have the opposite view and they don't want any new drugs. Also, there are referral provider preferences; some referring physicians would prefer to have early or late adoption.

There may be biases toward areas of commission vs omission. I think that's what I see when I read the statin debates. Some physicians are more willing to commit an error of commission, and they use a statin because there's an idea that statins reduce important events. And others would rather not commit an error of omission and do not use therapies that may not be as beneficial as they had hoped.

There's also a bias towards prevention. In my group, I'm one of the most aggressive prevention-therapy physicians, which may be because as the interventional physician, I see the ravages of coronary disease in patients with ST-segment elevation myocardial infarction (STEMI), shock, and coronary disease.

Another thing that drives early and late adoption is competition -- the view that regional providers are doing a new thing and that we have to do it too in order to catch up with them. That plays a significant part in hospital adoption of therapies, but I think it also plays a role in physician adoption.

What do we adopt? Of course there are drugs. The new oral anticoagulants (NOACs) are the most recent thing that we have spent a lot of time thinking about adopting. Devices like drug-eluting stents or cardiac resynchronization therapy (CRT) devices. Procedures, such as atrial fibrillation ablation, transcatheter aortic valve replacement (TAVR), chronic total occlusion percutaneous coronary intervention (CTO-PCI). Techniques like radial catheterization. Physicians also have to decide whether to adopt BNP-guided heart failure care. A big current controversy is whether to adopt the new lipid guidelines. They're out there and it's recommended that we use them, but many physicians are choosing not to adopt them.

Then there is a variety of processes or workflows that we may choose to adopt or not, but more likely they are going to be forced on us by organizations that we participate in. These include the use of electronic medical records and preprocedure evaluations that are required now by the Joint Commission on Accreditation of Health Care Organizations (JCAHO) before undertaking a catheterization or another invasive procedure. Or the newest enthusiastic redesign of outpatient care, the patient-centered medical home. These are all things that we will be forced to adopt, but I'm focusing mostly on the things that we have a decision about adopting.

What decisions about adopting am I now facing? There is vorapaxar, the Merck drug called Zontivity™, which is a thrombin receptor antagonist, a PAR-1 inhibitor. Should we use that therapy (not that it's FDA approved)? Another device, the Reveal LINQ™ device from Medtronic, is a small implanted loop recorder which is being promoted by the manufacturer for detecting atrial fibrillation in cryptogenic stroke. Should I adopt that? The CardioMEMS device is something I mentioned already; MRI safe pacemakers; the use of scintigraphy to evaluate patients with congestive heart failure; and optical coherence tomography (OCT). Should I adopt these things? These are not all new; OCT has been around for a while, but I'm now faced with that decision in my own community and in my own practice. So we're constantly faced with these new adoption ideas.

The exciting thing about cardiology -- which in my mind makes cardiology the most fantastic of medical subspecialties -- is that these decisions are constantly upon us. This list is all the things that we are on the verge of having to make a decision about adopting in the coming 3 to 5 years (that would be my estimate).

There is another type of adoption which is not as interesting, and that's the "me-too" or iterative adoption, such as another angiotensin receptor blocker (ARB) or another beta-blocker. A fourth NOAC is expected sometime this year. There's a new rotational atherectomy device, or the next next-generation drug-eluting stent (DES). The company Abbott has filed for a trademark name of a DES called the XIENCE ALPINE. I know nothing about it other than the fact that they have filed for a trademark name, so I presume that this is a new drug-eluting stent. For these iterative adoptions, do we really need to think about them in the same way?

As I reflected on which of my early adoption decisions have been wise or unwise in retrospect, I was thinking about the durability of Class 1 ACC/AHA guidelines, and I was somewhat encouraged to see that a consensus of physicians who are reportedly making unbiased decisions, making recommendations for our entire specialty, were wrong 20% of the time. It ranged from 10% to 26% based on their own re-review, but they changed their recommendation from Class 1 to another class 20% of the time. Judging my own adoption, we shouldn't hold ourselves to an expectation that our adoptions will be 100% unchanged.

I went through a list of things that I got wrong. I thought this was a wise idea and would help me make better decisions as I go forward with future adoption. I did adopt niacin; I was a very robust user of niacin for secondary prevention in patients with low HDL-C. I also did homocysteine testing and folate supplementation, and I did use Mucomyst® (acetylcysteine) for prevention of contrast-induced nephropathy. I did all of those things and I've given them all up because the evidence has evolved. I didn't adopt the use of spironolactone or eplerenone, and I've still not fully adopted that for congestive heart failure with systolic dysfunction. I think that's perhaps something that I should have done a better job of adopting.

Areas that I got wrong in devices include directional coronary atherectomy; I was an early user of that and have now completely abandoned it -- of course that was not willful; it was taken off the market. For peripheral arterial devices, I used atherectomy, PolarCath™, and laser, and have been frankly disappointed in their results after my experience with them. I was a user of renal stenting for resistant hypertension as guidelines recommended, and we now know from the CORAL trial[1] that that's no longer recommended. I used AngioJet thrombectomy in primary PCI (percutaneous coronary intervention). I closed patent foramen ovale (PFOs) in patients who had recurrent strokes. I used VerifyNow testing to evaluate patients' clopidogrel response. And I also was using Enhanced External Counterpulsation (EECP) in our practice; that was something that I got wrong -- again, I don't think because of any kind of ineffective therapy, but because it wasn't reimbursed by CMS (Centers for Medicare & Medicaid Services).

It's interesting that there's nothing on the did-not-adopt list. I left that blank because this perhaps says something about my adoption strategy as a person who is an early adopter and commits more errors of commission than omission.

Some things I got right in terms of medicines: I was an early adopter of ACE (angiotensin-converting enzyme) inhibitors and beta-blockers in systolic dysfunction heart failure. And high-intensity statins in secondary prevention -- I used them from the time the PROVE-IT trial[2] was published. (I participated in that trial.)

But there were things that I didn't adopt and I really don't know the theme behind those decisions. Why did I not adopt ARBs? There were certainly many good reasons why they might have been superior to ACE inhibitors, but I was never convinced by the theoretical benefits. I never once prescribed nesiritide. I didn't use echo contrast agents. I didn't use fish oil, I never used vitamin E, I avoided estrogen. I'm still avoiding testosterone replacement. I think fibrates are an enormous waste of money. And I'm a very slow and cautious user of Ranexa®, or ranolazine. If I could identify why I was unconvinced by those new therapies, it might help me going forward, but I don't have a sense about it.

What I got right (at least so far) is early adoption of TAVR. I was early to refer patients for clinical trials in TAVR, ICD and CRT, and FFR. I've never been an adopter of intraluminal imaging like intravascular ultrasound or OCT, and I never was a user of cutting balloons and AngioScore.

There are things I adopted and thought I got right, but they are now in question. I was a very early adopter of bivalirudin and would have listed that as something that was a good or a successful adoption, but now we have HEAT-Primary PCI trial,[3] which suggests that there may be an absolute hazard increase with this therapy. I've been a long user of statins in primary prevention, and that's another area that is up in the air at this point. NOACs: I use them as a class and I prescribe whichever one is covered by a patient's insurance. Drug-eluting stents for off-label indication is something that I think I got right using drug-eluting stents in primary PCI and saphenous vein grafts, but those are off-label indications. I do coronary intervention of chronic total occlusions. I use Impella support for primary PCI in shock patients. I supplement patients with vitamin D when their vitamin D level is low. I use aspirin in high-risk primary prevention patients. And I use the EKOS Ultrasound Assisted Catheter, directed thrombolytic catheter, for pulmonary embolism (PE) in submassive PE patients. Did I get these right? I don't know. I think the data will evolve, and I might be able to update my prior slides on what I got wrong in the future.

I didn't adopt microvolt T-wave alternans, and I vividly remember trying to make a convincing argument to justify its use because I thought it was a useful therapy or that it could be useful to help avoid implantation of defibrillators in patients who might not benefit by them. But ultimately I did not adopt it because of a lack of randomized trials.

This retrospective review was a really valuable exercise for me. It was interesting that there's a fair amount of confirmation bias. I persuaded the hospital that I have some influence with to use Brilinta® (ticagrelor) for STEMI therapy, whereas another hospital that I don't influence uses Effient® (prasugrel). I was very concerned that use of prasugrel might be problematic because of its black box warnings. I felt a certain amount of validation when a recent PINNACLE database[4] showed that there was a substantial number of patients, 1 in 5, as reported here by Shelley Wood, who did in fact receive prasugrel inappropriately. That validated my view that I got that one right, but that may be just a confirmation bias and I may avoid criticisms on Brilinta. This is one of those things that we ought to be cautious about when we re-review our adoption of decisions. And I mentioned this already in relation to current agents for PCI: The hazards of inappropriate early or late adoption can be mitigated by continuingly reassessing these prior adoption decisions. It is a valuable thing, and I got a lot out of this exercise.

I learned that I'm somebody who would prefer to commit errors of commission rather than errors of omission, and others may be opposite. I tend to believe in class effects for pharmacologic therapies and even for devices. When a problem is vexing, like superficial femoral artery disease for revascularization of peripheral artery disease patients, I do have a low threshold for adoption of new devices when they're FDA approved, but I've been continually disappointed.

I think that I am influenced by my own values of cost and convenience in chronic therapies. I know personally that I would never want to take warfarin if there was an alternative, and that may have influenced my adoption of the NOACs. I have a prevention bias, as I mentioned earlier, so I think that when it comes to the statin controversy, I was a user of statins and also many nonstatins until it became clear that nonstatins were something to be avoided, and I have pulled away from essentially all nonstatin lipid-lowering therapies. I rely a lot on physiologic plausibility and anatomic plausibility to make adoption decisions. I think this is something that both science and I are becoming increasingly cautious about -- the so-called debate over whether to use surrogate endpoints. But in fairness, many times we don't have large enough datasets to make decisions beyond surrogate endpoints.

I think concern about malpractice issues may dissuade some physicians from adopting early. We see that with the adoption of NOACs; malpractice lawyers came and saw a small window in which they could exploit side effects that may occur that were really no worse than those of standard therapies (ie, warfarin in the case of NOACs).

My final thoughts would be that when thinking about the new thing (device or medication), we should be deliberate about adoption: Check with our colleagues and subspecialties. For me as a non-EP or heart failure doctor, I use colleagues and my old fellow network -- I find that very valuable. Spaces that are virtual sharing places, like Medscape or other electronic media, really allow us to vet these ideas and share our thoughts and experiences on bulletin boards. And, of course, industry input is especially valuable in early adoption; we can't do it without them. One way to help patients and assuage their concern about adopting new therapies when we decide not to adopt is to say something like, "I know you may have heard about this new therapy (drugs or a device), but I don't recommend it." That helps patients know that we've considered it and are making a conscious decision.

That's what I got right and wrong on adoption, and hopefully we'll hear from others about what they're doing in this area. Thanks. Until next time, I'm Seth Bilazarian.


Comments on Medscape are moderated and should be professional in tone and on topic. You must declare any conflicts of interest related to your comments and responses. Please see our Commenting Guide for further information. We reserve the right to remove posts at our sole discretion.