Conversations in Cardiology: In the United States, Should TAVR Be Offered to a Low-Risk Patient?

Morton Kern, MD (University of California, Irvine), often engages his colleagues via email in brief, informal dialogue on clinically relevant topics in interventional cardiology. With permission from the participants, TCTMD presents their conversations for the benefit of the interventional cardiology community. Your feedback is welcome—feel free to comment at the bottom of the page.


William Suh, MD (Ronald Reagan UCLA Medical Center, Los Angeles, CA), asks:

I have a question for the group. We were referred a 77-year-old patient for TAVR, but his STS score is only 2% so low risk. He was offered a mini AVR which he initially agreed to. His cath showed single-vessel CAD with severe stenosis of distal RCA so the surgical plan changed to full sternotomy for AVR and CABG

He has been doing his research, and his impression is that TAVR and PCI would be much easier, and he prefers this route. He saw the data from ACC showing that Sapien 3 (Edwards Lifesciences) was superior to SAVR in intermediate-risk patients. We had informed him that TAVR is only indicated currently for high-risk or inoperable patients and that off-label use would not be reimbursed by insurance. He is also aware of unknown durability issues.

He came back to us and said that he would pay cash for his TAVR.

What have others done in this situation? Take the man’s money and do the procedure? Not do the procedure because of fear of possible Centers for Medicare & Medicaid (CMS) audit in future? Document two-surgeon consultation and patient refusal, and petition insurance company to pay for off-label use? I am very interested in the group’s opinions.

Dean J. Kereiakes, MD (Christ Hospital, Cincinnati, OH), replies:

Unfortunately, it really is not possible to “take the money and do the procedure.” We would offer him randomization into the low-risk TAVR trial (PARTNER 3). Personally, depending on vessel caliber and lesion length, I would stent the RCA either before or after (easier with respect to DAPT timing) the planned “minimal access” SAVR. We have done this multiple times without difficulty. The only way he can “pay” for TAVR as a low-risk patient at this time is to go out of the country. Medicare patients cannot “buy” their way around CMS.

Pinak Bipin Shah, MD (Brigham and Women’s Hospital, Boston, MA), replies:

I personally think there remain enough unanswered questions about TAVR (stroke, durability, etc) that even if he could pay for the procedure out of pocket, I would not feel good about offering it to him. If he is truly low-risk, then he needs to be in a randomized trial. If your surgeons feel he is intermediate-risk for any reason, then perhaps he can wait for intermediate-risk approval or go into an intermediate-risk registry, though his STS seems pretty low to argue being at intermediate risk. I doubt it would be possible for him to pay for it out of pocket at our institution, but even if he could, I would not offer it to him.

William F. Fearon, MD (Stanford University Medical Center, Stanford, CA), replies:

Agree with Binny and Dean. Would take same approach at Stanford.


David J. Cohen, MD (Saint Luke’s Mid America Heart Institute, Kansas City, MO), replies:

I also agree with Binny and Dean. The one option that is potentially available to him if he really wants to pay would be to go to a European site that can accept cash payment. I believe a number of patients have gone to places like Switzerland (Windecker) or Italy (Colombo) and paid cash. I think it’s pretty pricey, though (eg, $150K or so, although the price may have come down).

Suh replies:

Thank you for the quick responses.

US centers cannot accept cash payment because of fear of CMS audit? Our center accepts cash payments for other services (for example, international patients for PCI or CABG). Why would TAVR be different?

He has the resources to pay the $150K, and I suppose could go to Europe.

If we were a PARTNER 3 site, we would offer him randomization. If he agrees to participation in the low-risk trial, I would have to refer to our colleagues at Cedars. I do not believe they are enrolling yet though.

Thanks again for the great dialogue.


Lloyd W. Klein, MD (Rush Medical College, Chicago, IL), replies:

Randomization is the only ethical choice.


Cohen replies:

I’m going to play contrarian here and respectfully disagree with Lloyd’s point.

From my perspective, the only meaningful issue(s) with respect to TAVR in low risk-patients in 2016 are: (1) valve durability, and (2) long-term consequences of complications that are increased with TAVR (ie, paravalvular leak, and maybe need for pacemaker, although that isn’t so clear). While I certainly believe in the PARTNER 3 trial, let’s be realistic and admit that this is really a “regulatory” trial that is designed principally to obtain Food and Drug Administration (FDA) approval for low-risk patients and hence CMS reimbursement.

Based on all the data we have to date, it is very difficult to imagine a scenario in which either the Sapien 3 valve or CoreValve will not be approved for low-risk patients in the next 2 to 3 years—ie, well in advance of any meaningful knowledge about 10-year outcomes, which is where the real clinical questions currently lie. Therefore, in all likelihood, if this patient had come around in another 2 to 3 years, he would have readily been offered TAVR as per his stated (and apparently well-informed) preference with no greater assurance that the true uncertainties would have been resolved.

The bottom line is that we are going to be able to provide TAVR to our low-risk patients long before we really know if it’s a good idea. So, I don’t really see the ethical issue in providing it now (putting reimbursement aside) as long as the patient is very well informed about the uncertainties.

Flame away . . .

Gregg W. Stone, MD (Columbia University Medical Center, New York, NY), replies:

I’ll start Dave. You hit the nail on the head regarding the two main issues of TAVR in low-risk patients: durability and the long-term implications of potential complications. Given these unknowns, I don’t believe we can ethically justify implanting a device in such a scenario not yet supported by evidence-based medicine just because funding is available. While it may be highly likely that TAVR will be shown to be safe and effective in this scenario, if we know that to be the case with certainty, then the trial would not need to be done and FDA would have already given the indication. Now it is true that physicians are free to use approved devices in an off-label fashion if they have adequate justification for such use. In my opinion, however, there are enough uncertainties now facing the long-term durability of TAVR, especially from issues that have been raised in the last few months, that until these are sorted out either through very careful long-term routine echocardiographic surveillance studies (including long-term comparison with surgically implanted valves) or with randomized trial data with long-term follow-up, we should stick to the label. Especially because surgery is of such low risk in patients with a low STS score. Nor should we game the system by trying to justify escalating truly low-risk patients into an intermediate-risk category. Finally, I don’t believe informed patient consent in the situation (assuming such a thing is truly possible) removes our independent ethical obligation as physicians.


Arnold Seto, MD (Long Beach Veteran’s Affairs Medical Center, CA), replies:

I agree with David on the medical/ethical question here.

On the regulatory side, I’m not a TAVR operator, so I don’t know if there are special issues with TAVR and whether CMS will deny you from future reimbursement if you do one TAVR off-label, but from my simplistic understanding:

1) FDA approves drug/device for a specific approved use

2) CMS decides whether approved use is reimbursable, and under what conditions (ie, heart team, two-surgeon consultation, volume, registry enrollment). CMS also decides whether nonapproved use is reimbursable (ie, TAVR in low-risk patient as part of clinical trial)

3) Any licensed practitioner can use a product for off-FDA label use, however, that off-label use may be nonreimbursable by CMS (ie, low-risk TAVR outside of a trial)

Unlike in some other countries, “neither the FDA nor the Federal Government regulate the practice of medicine,” so I don’t think they could go after you. The California medical board legally could, but also probably wouldn’t.

The main risks I see are:

1) Because the current standard of care in the US is to not to perform TAVR in low-risk patients, you put yourself and your program at significant risk from a tort claim if something goes wrong. You could have lots of expert witnesses testifying that TAVR in a low-risk patient was below the standard of care, even if the patient signed a detailed consent.

2) The device manufacturers may reexamine their relationship with you. For example, if you were doing only cash-pay TAVRs without surgical backup in your outpatient surgical center, they probably don’t want to be selling to you.

If the patient was willing to pay cash upfront for medical costs (including implant, hospitalization costs, and follow-up—everything CMS might have covered), signed and notarized a detailed letter understanding the risks and waiving their rights to sue (if legally permissible), and you had clearance from your department/hospital, then I would say it would probably be acceptable. That’s a big headache, though, and the Europeans are likely able to do it for him for less money, so I would send him abroad.

For the specific question of Cedars, I sent them a low-risk patient recently for randomization into SAVR versus TAVR, so it’s out there if the patient is willing to be randomized.


Nauman Siddiqi, MD (MedStar Union Memorial Hospital, Baltimore, MD), replies:

At our small but growing program (260-plus TAVRs and counting), we would tell him to save his money as there are options. I would see if he was willing to wait until August when I could enroll him in the low-risk Medtronic trial, assuming he’s ok with being randomized against SAVR and ok with Evolut R/CoreValve over Sapien 3. (We’re not a PARTNER 3 site)

Another option is to send him to our sister institution MedStar Washington Hospital Center where they have a nonrandomized, nonindustry-sponsored low-risk TAVR study with either Medtronic or Edwards. So if he’s willing to fly to DC, contact Ron Waksman. I believe they are currently enrolling.


Klein replies:

I believe that TAVR is a fabulous technique, and I think that your optimistic forecasts are quite likely to be correct. I respect your views of the data, understand that the low-risk AVR study is likely to show good things about TAVR, and recognize that you are correct when you suggest that in 2 years or 4 years, this patient will qualify for TAVR under FDA-approved standards. Nevertheless, that is not today. We don’t have that data yet. I don’t think we can say with certainty what the study will show, or characterize the study as only for regulatory reasons rather than true scientific ones. We don’t know for sure if everyone is better off with TAVR versus SAVR or if there are subgroups. Therefore not only is it unethical to advise a nonproven procedure, but in fact randomization is the only ethical choice (including advising him to go abroad) for this patient and the ones to follow when there is equipoise, which is in fact what exists at this moment in time. Indeed, if we just take your argument to its logical conclusion, then all of the participants in the trial are acting unethically by carrying on a charade when we already “know” the answer. But really, we don’t.

Jeffrey W. Moses, MD (Columbia University Medical Center), replies to Suh:

PARTNER 3 is enrolling. I’d be surprised if Cedars isn’t up. Don’t think self pay is legal if you participate in Medicare. $150K is a bit high for overseas costs, but it’s his best option.

Moses replies to Klein:

It’s clear we don’t know the answer, which is why we are running the trial. But at some point we have to accept patient autonomy and respect their wishes as long as they are fully informed about the current limits of our knowledge.

Frederick Welt, MD (University of Utah Health Care, Salt Lake City, UT), replies:

Yet another example of how a simple question to the group uncovers so many interesting and divergent opinions.

From my viewpoint, I think the ethical question is more nuanced than others. There may be some patients for whom the issues are much more about shorter-term outcomes than 10-year durability. Certainly in Europe, low-risk patients seem to be treated with TAVR routinely. Does that make those physicians all unethical? As we were all taught in medical school, the principles of medical ethics involve autonomy, justice, beneficence, and nonmaleficence. I think we could have a very robust discussion based on those principles in this case.

The payment side is another issue. I am in no way an expert on these issues. Institutions and physicians can’t charge a qualified Medicare beneficiary for a covered procedure. Foreign nationals are a completely different scenario, which is why so many institutions covet these patients. How this would play out in this scenario is unclear to me. Would love to be educated.

Raj Makkar, MD (Cedars-Sinai Medical Center, Los Angeles, CA), replies:

Our institutional review board (IRB) approved the PARTNER 3 protocol last week. Contract issues being finalized, and we should be able to enroll in a couple of weeks. Happy to enroll this patient, Will, if he decides to go that route.


Suh replies:

Thank you all for the interesting dialogue. Based on what has been said, I think the best option will be to encourage patient to participate in low-risk trial. 

Raj, I am glad to hear that your IRB has approved. Will let you know what happens.

David A. Cox, MD (Lehigh Valley Physician Group, Allentown, PA), replies:

Good to hear, Raj. I guess the bottom line, though, is he has to be willing to accept randomization into the surgical arm if he does the trial and be willing to walk that path if that is what the dice throws. We can’t do a randomized trial ethically if those who randomize to SAVR pull out of the trial and go to Europe to get their valve!

It would be hard to offer him randomization if he truly doesn’t believe in . . . randomization!

Will be an interesting discussion for sure.

Stephen Ramee, MD (Ochsner Medical Center, New Orleans, LA), replies to Suh:

Great decision, Will. 

Moses replies to Welt:

Agree with Fred. Obviously I support an RCT as I am on the executive committee of PARTNER 3.

But the larger issue is the concept that there are laypeople utterly incapable of absorbing current medical knowledge and evaluating it in the context of their own needs and priorities. I can say in my life I have met many that can.

While I am not supporting a center embarking on an out of trial procedure that will undermine the two ongoing low risk trials, I still believe in letting patients know about all options, even those overseas. I do not think that is an ethical lapse.


Seto replies:

Quick follow-up on the question of legality of a patient paying cash for a low-risk TAVR that Medicare deems nonreimbursable: short answer is yes, it should be legal, as long as the patient signs an advanced beneficiary notice (ABN) that says they understand that Medicare won’t cover a service or its associated hospital stay. Good to know, for those fortunate enough to be able to pay cash for a TAVR.

A former Department of Justice senior medical consultant feels pretty confident that if something is not covered by Medicare, but the patient is appropriately consented and signs the ABN (Advanced Beneficiary Notice, indicating that the patient understands that the procedure is not covered by Medicare), the doctor should be fine. She gave the example of cosmetic surgery. It’s not covered by Medicare, but it is legal. She also addressed balance billing, which she said only applies to covered services. She agrees with your statement that balance billing is not allowed. If TAVR is not covered in the situation we described, it is not possible to do balance billing.

Finally, she thought it important to mention that, in addition to physician services paid out of pocket, the patient might also need to check on whether they can pay for the associated hospital stay or cath lab services. If TAVR is not covered by Medicare, there is a chance that associated care in the periprocedural period would also not be covered.

Moses replies:

There aren’t going to be hordes of US citizens going to Europe for TAVR. But as Raj knows, it’s key to get the buy in for randomization locked in up front.


Charanjit S. Rihal, MD (Mayo Clinic, Rochester, MN), replies:

Completely in agreement that randomization is best. However, what we don’t want is to enroll patients who will drop out the instant they don’t get assigned to the treatment they were hoping for. 

If after a full discussion, a well-informed patient chooses TAVR, then I don’t believe there is anything to stop you from performing an off-label procedure with a commercially available device. CMS will not pay for it. However, for a self-pay patient you wouldn’t be billing CMS or other third party payer anyway. I would consult your business office and compliance officer.

ps: TAVR costs probably just dropped 30% in the United Kingdom.

Augusto Pichard, MD (Medstar Washington Hospital Center, Washington, DC), replies:

Very good comments from many of you.

Scientific progress has many expressions, not all arising from randomized trials. Ron Waksman in our group has initiated a 250-patient registry of TAVR in low-risk patients. This is an FDA-approved study, with all the needed scientific rigor. It is also approved by CMS for reimbursement. About 20 patients have been enrolled so far.

Kern replies:

The ethical dilemma of using a new interventional methodology when high-quality study data are insufficient or missing is familiar to those of us who were brought up in the time when new coronary device designs came to market every 6-12 months. Recall how directional coronary atherectomy and brachytherapy, to name just a few, were adopted and used widely (and billed for) before randomized study data showed no benefit and the procedures were abandoned.

The use of TAVR has more risk than a new coronary stent but the concept is not that much different. I believe a new procedure should be subject to scientific validation studies but the major validation of TAVR has been done, with refinements ongoing regarding its late complications and durability.

Use of the TAVR will be a standard of care within 5 years for patients with aortic stenosis. The option for participation in studies is my preference, but many operators want to provide the low-risk patient needing AVR with this choice now. The reimbursement issue is separate and not related to the standard-of-care issue.




We Recommend