Percutaneous Left Atrial Appendage (LAA) Closure Therapy

To whom it may concern,

I am a practicing electrophysiologist with the Orlando Heart Center since 1997. I primarily practice at The Orlando Regional Medical Center (ORMC) and Orlando Health Heart Institute (OHHI). These institutions have a very robust Atrial Fibrillation Center, certified by the Society of Chest Pain Centers and winner of the 2015 Bronze Award from the American Heart Association -Get With the Guidelines Program. ORMC was recognized by the AHA 2014-2015 as one of the busiest centers based on admissions with the primary diagnosis of atrial fibrillation. Therefore, innovations such as the recently FDA approved Watchman LAA Occlusion device is very important in our daily endeavor to protect our AF patients from thromboembolic complications. Many of our warfarin eligible patients are unable to take this lifesaving medication due to bleeding complications or drug intolerance.

For instance, I have a very active and independent 64 yo female with h/o PAF, HTN, IDDM,CVA's (CHA2DS2Vasc- 5) unable to take warfarin therapy due to severe intolerance with long term. She is unable to afford the newer target specific oral anticoagulants. She takes aspirin but is very anxious about the potential for more CVA's.

I have another patient , a less active 75 yo male with h/o PAF, CVA, and recurent falls from early Parkinson's disease resulting in a subdural bleed. (CHA2DS2Vasc - 4/ HASBLED -2) He remains mentally sharp although his body betrays him.

These are but only two examples that come to mind where the new Watchman LAA occluder device can make a difference in their lives and protecting them from a devastating thromboemboic event.

I write this comment to stress to you the importance of this device in our day to day struggle to preserve our patients quality of life and prevent unnecessary suffering. Thank you

Respectfully submitted,
Roland A Filart MD, FACC,FAHA,FHRS

Less

The potential impact of this device on several of my patients can be very useful and helpful. The patient with high stroke risk, who has a low HAS-BLED score, but indeed has had a significant bleed, and likely is at risk for another bleed, and worries all the time about either having a stroke or having a bleeding problem, can be benefitted very much with this device, especially if they have had cardiac surgery, and epicardium is not accessible.

Less

Dear Ms. Syrek-Jensen,

I would like to comment on the proposed national coverage decision to cover percutaneous LAAC therapy using the Watchman device. We are a tertiary hospital in Illinois with a busy electrophysiology program and many patients suffering from atrial fibrillation in dire need for alternatives to anticoagulation therapy using NOACs or Warfarin. I began implanting Watchman devices in 2013 in Germany where the Watchman device has been available for over 2 years. I have learned that patients benefit greatly from LAAC. In particular patients with a high bleeding and stroke risk (HasBled/ChadsVAsc score of 3 and above) who may tolerate only a minimum period of Warfarin therapy may be saved from experiencing another major bleed once only on Aspirin. I do also believe that additional trials are needed to establish the role of "Plavix and Aspirin only" therapy following Watchman implantation in patients with recent bleeding complications or a disease process that forbids the use of NAOCs or Warfarin during the initial period after implantation. Furthermore, since NOACs are now widely used in clinical practice a regimen of Aspirin and NOAC following implantation instead of Warfarin needs to be assessed. Furthermore, I do know of patients who are terrified by the prospect of lifelong Warfarin or NOAC therapy (especially younger patients with many years of anticoagulation therapy ahead) despite a low HasBled score that would rather have a LAAC device implanted. These patients should have a choice between established therapies (anticoagulation versus Watchman), but under the current proposal would no be allowed to undergo LAAC. Finally, I do agree with the proposal to allow LAAC only if patients are included in a national registry in order to gather data on the "real-world" experience using the Watchman device.

I thank you for your consideration of the comments provided herein.

Less

National coverage for LAAC is an important step to improving the care and outcomes of patients with NVAF in the US. Please consider the following points:

1- FDA indication of watchman is for patients who are able to take Anticoagulation. CMS coverage policy should reflect this.

2- risk and benefit of anticoagulation versus watchman is a complex decision making process that is best left to the clinician and patient. Mandating a chads-vasc score consistent with a clinical need for anticoagulation and a documented rationale for avoiding long term antiCoagulation is reasonable. However, Mandating a specific has-bled score is not. For example a patient with recurrent bleed may have a highly compelling reason to come off anticoagulation but a relatively low has-bled

3- patient preference should be adequate on e they have been thoroughly educated on the data. While the majority of patients prefer oral anticoagulation to a procedure, infrequently we deal with patients who refuse long term anticoagulation no matter despite repeated discussions. These patients would be best cared for with LAAC. Furthermore allowing such patients access to LAAC would likely be cost effective by preventing strokes

4- a registry requirement is reasonable but costly to hospitals/providers. If a registry is required, coverage should take into account the added financial support needed for adequately keeping up the registry

5- registry requirements need to be clarified. From the registry goals it appears that control patients may need to be included. Historical controls may be more practical. If controls within the registry is made a requirement, reimbursement for the device should take into account additional support needed to maintain registry

Less

Less

Less

The proposed decision memo for percutaneous LAA Closure Therapy is a thoughtful document that, for the most part, outlines reasonable criteria for coverage of this novel form of therapy.

Less

To whom it may concern,
I am the director of electrophysiology at Emory University Hospital Midtown and a practicing electrophysiologist at Emory Healthcare. I have been treating patients with non-valvular atrial fibrillation since 2008. I have used the Watchman technology as part of clinical trials and commercial setting. I appreciate the opportunity to comment on a well-designed national coverage decision. I specifically would like to comment on the 2nd condition:
A high CHADS score HAS BLED score and contraindication to warfarin.
The last Watchman device I implanted was on a 68 y.o gentlemen with HTN, diabetes, congestive heart failure and prior TIA. He had to have an enucleation of the right eye after an embolic event complicated by multiple surgeries and severe eye pain. He has lost his depth perception after surgery and had multiple falls. After one of his falls he hit his head on his tractor. This patient received a watchman device and is currently on warfarin for 40 days and will be switched to Plavix and aspirin for 6 months followed by aspirin alone. This gentleman did not have an elevated HASBLED score nor an absolute contraindication to warfarin, but does have an elevated CHADS and CHADVASC score and clinically speaking he is at risk of bleeding due to his recurrent falls and head trauma.
Overall, the goal is to provide this therapy safely to appropriate patients. By providing more clarity to condition 2 will allow physicians to select patients who are most likely to benefit from this therapy.
Sincerely,
Mikhael El-Chami, MD, FACC, FHRS

Less

To: Tamara Syrek-Jensen, J.D., (Director of Coverage & Analysis Group)

I am an interventional cardiologist in Columbus, Ohio. I practice clinical cardiology, as well as perform PCI/stents, TAVR, mitraclip and left atrial appendage closure procedures. I practice at Riverside Methodist Hospital (part of the OhioHealth system), which is a large community teaching institution. I am active in clinical research and teaching residents.

I have been very active in clinical research in patients with atrial fibrillation and left atrial closure devices. I see many patients with atrial fibrillation and valvular heart disease that have tremendous difficulty tolerating anticoagulation therapy due to profound bleeding complications. These patients are desperate for alternative therapy, such as Watchman. I have had extensive experience with the Watchman device, especially since commercialization and have excellent clinical outcomes. I have been clinically pleased with the results, and the patients have loved it.

Many of my patients are elderly and have a high risk of bleeding. Many of the patients have already had a significant bleeding event requiring hospitalization and transfusion. I have had one of my partners in practice die from a complication of Coumadin – he was 50 years old. I believe there is a definite need for an alternative to anticoagulation in patients with atrial fibrillation to decrease their risk of stroke when anticoagulation is difficult to take.

My comments on the 7 conditions are:

Condition 2: I believe the language should allow physician discretion for those who have high CHADS2 scores and/or high HASBLED scores. This may include those on Coumadin but are difficult to manage with excess risk of bleeding.

Condition 3,4,5,: Agree that training requirements and facility requirements need to be met with appropriate physician training at each site.

Condition 6. We do need a registry. The cost of the registries is excessive and burdensome.

Condition 7: Agree.

I am very hopeful that CMS provides adequate coverage for this necessary procedure.

Best regards,

Steven J. Yakubov, MD FACC FSCAI
System Chief, Structural Heart Disease, OhioHealth
John H. McConnell Honorary Chair Advanced Structural Heart Disease
Riverside Methodist Hospital
3705 Olentangy River Road
Columbus, OH 43221
Steven.yakubov@gmail.com
Steven.yakubov@ohiohealth.com
Cell 614.738.8917

Less

I am an Interventional Cardiologist who specializes in structural heart disease. I have been performing interventions in the left atrium for mitral valve disease and ASD closure for over 30 years. I have recently begun LAA exclusion procedures for patients with atrial fibrillation who are at high risk for bleeding complications from oral anticoagulants.

This population is typically comprised of elderly patients with fall risks but may also include younger patients with jobs putting them at risk from bleeding due to trauma.

In this population, the Watchman is a particularly attractive device due to its demonstrated efficacy in reducing both bleeding complications in patients with atrial fibrillation but also a reduction in mortality in this population compared to warfarin.

Parenthetically, we were particularly pleased that the FDA, after very careful review of the evidence with the Watchman Device, approved it with a reasonable indication.

We were surprised and disappointed with the proposed CMS coverage proposal and its very restrictive language.

In particular, the language used in Condition 2 suggested that the required CHADS2 scores might be higher than those used in the studies performed in a large population of patients which formed the basis for approval. Additionally the HAS-BLED criteria was not used in the trials and therefor there is no data to support its proposed requirement for reimbursement.

Similarly the contraindication for warfarin stipulation is in exact disagreement with the FDA indication. That seems very odd, if not capricious and arbitrary.

For condition 6, the proposal for a national registry is reasonable but there is no rationale for an OAC arm. This would basically be a requirement for performing another clinical trial with no funding and is simply unreasonable.

Respectfully submitted,

Richard Smalling, MD, PhD

Less

I am a clinical cardiac electrophysiologist currently practicing in Vancouver, Washington as part of a multi-specialty group The Vancouver Clinic for which a majority of my clinical practice is devoted to the management of atrial fibrillation. I will have on a daily basis conversations with my patients with atrial fibrillation about stroke prevention and attempt to present a balance argument for the type of prevention needed based on the individual risks and goals of therapy. What has been very satisfying over the last 10 years of practice is the increase in options for stroke prevention in atrial fibrillation that had not been present prior to this. I view the use of left atrial appendage closure (LAAC) devices as yet another advancement in this area and am encourage that CMS is in the process of creating a National Determination Coverage for these devices.

I initially submitted a comment on the initial request for National Coverage Determination for the use of a percutaneous, trans-catheter, intraluminal left atrial appendage closure (LAAC) device for use in primary and secondary prevention of thromboembolism in non valvular atrial fibrillation. CMS released a draft proposal last month and are seeking comments on this proposal. I find the proposal helpful in a variety of ways as it adds some clarity to what type of left atrial appendage closure device would be covered, the Watchman device, and what is required of physicians and facilities to successful and safely implant LAAC devices. What I would ask is that CMS in their final determination clarify the appropriate patient criteria that needs to be met to cover the implantation. The Draft proposal correctly stated that individuals with high risk of embolism based on CHADS2 or CHA2DS2VASC2 score be considered with a high HASBLED score and a contraindication to warfarin. What is unclear in this statement is what is meant as a contraindication to warfarin as this was a primary exclusion for the original trials for the Watchman device. As many individuals with a high HASBLED scores could be interpreted as having a contraindications to warfarin clarity on the meaning of this statement based on an actual score would be helpful. In general, trials involving the Watchman device had patients on average with a HASBLED score of >1.

Patients within our practice that are currently being considered for implantation of this device are individuals with documented non valvular atrial fibrillation with a guideline based recommendation for anticoagulation for the prevention of thromboembolism/stroke based on CHADS2 score >1 or CHA2DS2-VASc score >2. They also have to have a reasonable argument for seeking an non pharmacologic alternative to anticoagulation, mainly high risk for bleeding determined using the HAS-BLED score >1. Many of the patients currently being considered have had prior bleeding complications related to anticoagulation, either a life threatening gastrointestinal bleed or intracranial hemorrhage that has resolved to allow resumption of anticoagulation, but remain at risk in the future for recurrent hemorrhage. Other individuals being considered have a combined risk of future bleeding related to multiple comorbidities that include chronic kidney disease, liver disease, or need of additional anti platelet agents for treatment of atherosclerotic vascular disease. Patients that are currently tolerating anticoagulation who have not demonstrated a high risk for complication of oral anticoagulation are being counseled to remain on their anticoagulation until there is data that may support the routine use of this device in lower risk patients.

I do not view LAAC devices as the end all, be all treatment for stroke prevention in atrial fibrillation and feel that patient selection is paramount in the utility of this device. A clear and concise statement on patient indications would advance our ability to apply this technology to the most appropriate patients.

Thank you for your consideration.

Benjamin John, MD
The Vancouver Clinic
700 NE 87th Ave.
Vancouver, Washington 98664
Phone: 360-882-2778

Less

I am an Electrophysiologist directly involved in the care of patients at risk for stroke related to atrial fibrillation. I have cared for these patients for many years both with and without LAAC therapy. Personally, I have cared for patients that have been hospitalized for prolonged periods of time or passed away because of bleeding complications related to oral anticoagulants. I have also cared for patients that have suffered devastating stroke because of the fact that they were not good candidates for oral anticoagulants or refused to take these drugs because of the risk of bleeding. All of these situations are tragic for the individual patient and are very costly to the health care system in the United States.

Billions of dollars are spent each year in the United States caring for patients that have suffered stroke. Many of these strokes occur in patients with atrial fibrillation who are not taking oral anticoagulants because of the associated risk of bleeding. Both the initial hospitalization and the long term care of these debilitated patients are very expensive. Studies done to date have demonstrated the cost effectiveness of LAAC therapy. These results are not surprising given that approximately 30 percent of strokes are related to atrial fibrillation. These strokes tend to be more disabling which further increases the cost because of the long term care that is needed. As we include the amount of money spent on bleeding complications related to oral anticoagulants, the cost saving nature of this technology becomes readily apparent. As this therapy reaches all patients that need it the cost savings are likely to be immense.

The wording of FDA communications should reflect the importance of this technology both to the individual patient and the health care system as a whole. FDA communications should assist with the efforts to bring this therapy to patients that most need it. To this end , I think that it is appropriate to state that this therapy is appropriate for patients with an elevated risk (not highly elevated) of stroke related to atrial fibrillation. I don't think that a specific CHADS VASC score should be mentioned secondary to the fact that these risk scores are continually modified and a new risk score may replace the CHADS VASC in the future. FDA communications also need to note that this technology is appropriate for patients in which there is an elevated risk of bleeding or a rationale for pursuing non pharmacologic stroke reduction strategies. Again, bleeding risk scores change over time and there are some issues related to bleeding that are not accounted for in the HAS BLED score. The wording should be general enough to allow physicians caring for these patients to get this therapy to the patients that most need it.

I would like to thank the FDA for approving this life saving and life altering therapy. Since FDA approval, I have placed this device in approximately 100 patients that had no other good option for stroke prevention. Prior to the FDA approval of LAAC, these patients would have had to accept a significant risk of a major stroke. This situation gave me a distinct helpless feeling prior to the approval of LAAC. LAAC is a much needed advance in the clinical care of patients of atrial fibrillation. It is my hope that the FDA will continue to act through its communications to make this technology available to those patients that need it.

Less

Less

Public Comment on CMS Proposal

I am an Interventional Cardiologist in a large Multi Speciality Group in Illinois. I believe the Watchman device is effective, safe, and backed by solid scientific data. The device is routinely used outside the USA and appropriately received FDA approval after almost a decade of randomized studies.

The proposal contains guidance on physician documentation and consent, as well as registry requirements that are appropriate for this technology. The CMS proposal, however, also contains poorly worded requirements (contraindication to warfarin) that should be clarified and corrected in the final decison.

I echo the comments of the many talented physicians, often nationally and internationally recognized experts in their field, as well as the concise statements from our Cardiovascular Societies. The Watchman device should be available to trained physicians in dedicated centers as an alternative to coumadin therapy in carefully selected patients.

In my own practice, I frequently see patients with bleeding complications related to antiplatelet therapy in combination with coumadin therapy, often leading to emergency room visits, hospitalization and transfusion, and interruption of anticoagulation.

It is important to realize that the CMS payment decision has the potential to accelerate innovation in stroke prevention in patients with atrial fibrillation. Embolic stroke (due to lack of anticoagulation) and hemorrhagic stroke (inheritent risk of any form of anticoagulation) remain the most devastating cardiovascular events, very often leading to disability or loss of quality of life.

Christian Zellner MD

Less

Dear Ms. Syrek-Jensen

I am an electrophysiologist who is treating patients suffering from atrial fibrillation on a daily basis. Atrial fibrillation is the most common cardiac arrhythmia in clinical practice. Affected patients are at increased risk for thromboembolic events. Anticoagulation therapy has proven to be effective in preventing further events in patient with atrial fibrillation and high CHADS2VASC score. Unfortunately, significant number of these patients are not on any form of anticoagulation as a long term due to bleeding , difficulty managing the INR and high risk for injury or other forms of intolerance. Furthermore, some of these patients suffer from embolic events despite appropriate therapy with anticoagulants.

The Watchman Device was recently approved by the FDA as long term alternative for patients with nonvalvular atrial fibrillation who are Warfarin eligible, at high risk for stroke and have medically appropriate rationale to seek an alternative therapy. The Data from several clinical trials demonstrated that Left atrial appendage occlusion using the watchman device to be a safe and effective alternative to anticoagulation.

I strongly urge CMS to provide national coverage for the left atrial appendage closure therapy with the Watchman device to be able to offer this great therapeutic option to our patients.

Sincerely,
Ahmad Abdul Karim, MD, FHRS
Medical director, Cardiac Electrophysiology
ST Joseph Medical Center, Presence Health Group

Less

RE: Percutaneous Left Atrial Appendage (LAA) Closure Therapy

Dear Ms. Syrek-Jensen,

I am an Electrophysiologist in the San Francisco East Bay affiliated with multiple hospital networks in the area. The treatment of patients with atrial fibrillation remains one of the most important parts of my practice, and the Watchman device, as well as the appropriate coverage of this device, remains an important topic of discussion with my patients. The discussion points below were my comments regarding this therapy and coverage on 6/2015. I wanted to resend these thoughts as part of my interest in the discussion regarding left atrial appendage closure devices and the ability of my appropriately indicated patients to continue to access this therapy.

I am president of a large cardiovascular practice comprised of 23 cardiologists and 12 cardiovascular PA’s/NP’s. In this group we oversee and treat patients with a variety of cardiovascular conditions, and one of the largest patient populations we care for are those patients with non-valvular atrial fibrillation. By way of disclosure, I have been an investigator on many device trials with Boston Scientific over the years, but I was not an investigator in the WATCHMAN IDE clinical trial. The comments and opinions expressed in this commentary are representative on my review of the available data on the management of patients with atrial fibrillation, as well as the opinions of the patients I have treated in my 27 years of practice as a clinical Electrophysiologist.

There are approximately 5 million people in the United States suffering from atrial fibrillation, with this number expected to double by the year 2050 (1) . As the data also states, patients with atrial fibrillation have up to a 5-fold increase in their risk of stroke, with over 87% of strokes caused by thromboembolism . For those strokes caused by thromboembolism of clots from the left side of the heart, >90% of the thrombi originate in the left atrial appendage(1,3) . There are multiple pharmaceutical options for our patients at high risk of stroke, but there are patients in my practice who have an appropriate reason to seek a non-pharmacologic alternative to warfarin. Since the FDA approval of the WATCHMAN device, this technology is now a viable option for appropriately indicated patients, taking into account the safety and effectiveness of the device compared to warfarin. The potentially devastating effects from a stroke require careful discussions between patients and physicians regarding appropriate treatment strategies.

For many of my patients, long-term anticoagulation is the conventional treatment, whether it is warfarin or other approved pharmacologic solutions. For patients intolerant to warfarin, I also consider the direct thrombin or Factor Xa inhibitors, which have been shown to provide comparable outcomes with less serious side-effects. These new agents, while attractive and effective, have not translated into widespread adoption or increased treatment rates of anticoagulation therapy. Even within the clinical trials, approximately 1 out of every 5 patients discontinued anticoagulation after 2 years (4,5,6,7). Many reasons for the discontinuation of these medications are present from side effects, lifestyle, and cost of chronic medication. While oral anticoagulants remain an excellent therapy for many patients, there is a need in my practice for an alternative therapy for stroke prevention for patients with non-valvular atrial fibrillation.

The data from the WATCHMAN device has shown in clinical trials to fill an unmet need for many of my patients. For example, the WATCHMAN device is the most studied left atrial appendage occlusion device and the only one proven with data derived from multiple randomized trials as well as multi-center registries. For example, there have been five studies, >2400 patients, and multiple years of patient follow-up regarding the WATCHMAN device. This data shows the device can be implanted safely(1), enables patients to discontinue warfarin(3), and has been shown to reduce AF stroke risk comparably to warfarin(4) .

The data from clinical trials also shows the device implantation success rate to be high5, with 95% of patients implanted successfully with the WATCHMAN device. Further, the trial data showed >92% of patients were able to stop warfarin after 45 days, and >99% of patients were able to stop warfarin at 1 year(5).

The WATCHMAN device also demonstrated comparable stroke risk reduction, and statistically superior reductions in hemorrhagic stroke, disabling stroke, and cardiovascular death compared to warfarin over long-term follow-up (6,7). Also, the WATCHMAN device data shows a significant reduction in the risk of a major bleeding event by 49% for WATCHMAN patients post-procedure (Meta-analysis of the PROTECT AF and PREVAIL data) .

The WATCHMAN device has been extensively evaluated and the endpoint of the largest clinical trial, PROTECT AF Trial, demonstrated statistical superiority for stroke, systemic embolism, and cardiovascular death, with a 39% reduction in events versus the warfarin control group. The data shows, in addition to the effectiveness in appropriately indicated patients, that the WATCHMAN device can be safely implanted by properly trained physicians with varying levels of device procedure experience . In fact, almost 50% of implants in PREVAIL were performed by appropriately trained new operators, demonstrating that the learning curve to implantation can be achieved through appropriate training and experienced-operator guidance.

In short, I respectfully ask you to consider the data in determining appropriate national coverage for percutaneous left atrial appendage closure therapy. The benefit-risk profile of the WATCHMAN device as an alternative therapy to warfarin in carefully selected patients supports the use of this device. In appropriate patients, the WATCHMAN data appears to provide substantial protection against stroke while freeing patients from the risks associated with long-term warfarin use.

Also, this coverage for the WATCHMAN device should be consistent with the patient population studied in the WATCHMAN clinical trials (as this data was the basis for the FDA approval of the WATCHMAN device). The robust long-term data on this device and clinically relevant outcomes provide a compelling case that this device be made available to Medicare patients. Patients at high risk for stroke based on the CHADS2 and CHADS2-VASc scores but have an appropriate rationale to seek an alternative to warfarin deserve access to this device.

I appreciate your prompt attention to this matter as well as your consideration on this subject. The treatment of atrial fibrillation and providing my patients with an individualized, safe approach to stroke-reduction is—and will always be—a cornerstone of my practice. If you have any additional question or I can be of any further help, please do not hesitate to call or e-mail.

Sincerely,
Paul L. Ludmer, MD, FACC
President, Cardiovascular Consultants Medical Group
Stanford Healthcare
Cell: 510-299-6848
E-mail: pludmer@stanfordhealthcare.org

Less

To Whom it May Concern:

I am an interventional and general cardiologist dealing with a large number of patients who suffer from chronic atrial fibrillation. In our aging population, long term anticoagulation is many times difficult due to significant risk of bleeding or falls. These patients are often treated suboptimally by having their anticoagulation stopped or they are put at high risk for bleeding by having their anticoagulation continued. Left atrial appendage occlusion/closure has been shown in the literature to be a safe and efficacious method of treating such patients. I believe that not only Watchman but Lariat suture closure of the left atrium is very beneficial in these patients. The PROTECT and PREVAIL trials specifically support the use of the Watchman device. I support the seven proposed coverage criteria.

Less

Dear Ms. Syrek-Jensen:

I am a clinical cardiac electrophysiologist at Northwestern Memorial Hospital who has implanted the Watchman device in the Prevail trial, the CAP registry, and after FDA approval of the Watchman device. However, our ability to implant the Watchman LAA occlusion device for patients in desperate need of therapy to prevent strokes or systemic embolism (SSE) in non-valvular AF (NVAF) has been greatly hampered by a LCD covering Illinois that has not granted Medicare coverage. Therefore, I strongly encourage CMS to provide a national coverage decision in favor of Watchman LAAC.

The Watchman device has robust clinical trial data from 2 RCT and 1 registry demonstrating safety and efficacy in preventing SSE in patients with NVAF who are also candidates for oral anticoagulation (OAC) with warfarin. The Watchman device is non-inferior and possibly superior to warfarin anticoagulation (as demonstrated by the 4-5 year follow-up of Protect-AF) in preventing SSE and avoiding bleeding complications.

The novel anticoagulants (NOAC) have clearly demonstrated non-inferiority to warfarin but are still plagued by the limitations of daily OAC pharmacotherapy, namely bleeding complications and eventual non-compliance. Across all the NOAC trials, the major bleeding risk is ~2-3%/year with the NOACs. In addition, approximately 20% of the NOAC trial participants eventually stopped their NOAC treatment during the clinical trial follow-up.

There is clearly a need for a non-pharmacologic therapy to prevent SSE in NVAF that will not eventually rely on daily compliance with a medication and for patients unable or unwilling to take OAC.

My concerns regarding the proposed 7 conditions required for Medicare coverage of Watchman LAAC are as follows:

1. The required CHA2DS2-VASc score should be the same score which would trigger the use of OAC. Therefore the CHA2DS2-VASc score should be 2 or greater.
2. The HAS-BLED scoring system should not be used solely to assess the risk of bleeding given that even HAS-BLED patients with a score of 2 have a ~4% risk of bleeding. In addition, 1 criteria, labile INR, may not apply to patients being considered for a NOAC and Watchman LAAC rendering the HAS-BLED score incomplete.
3. The language of "contraindication to warfarin" is problematic as the clinical trials all excluded patients with a contraindication to coumadin and it runs counter to the FDA approval of the device.
4. I have major concerns regarding a OAC "control" arm in a registry. Given the non-randomized nature of such a "control" arm, I am not convinced this will provide useful data due to strong selection bias and highly likely baseline differences in patient characteristics. This is especially true if clinical conclusions will be drawn by comparing such a non-randomized "control" arm with the Watchman LAAC registry data.
5. Many patients that may benefit from Watchman that don't exactly fit the proposed criteria may not receive potentially life saving therapy. For instance, a patient with a stroke while ON OAC (with either coumadin or a NOAC) may benefit from Watchman LAAC as "rescue" therapy. Such an patient may be excluded from coverage with such strict criteria.

The Watchman LAAC is a necessary step in the on-going evolution of therapy to prevent SSE in NVAF. What is absolutely required to continue to evolve our therapies is a NCD to cover Watchman LAAC. I encourage CMS in the strongest terms to approve an NCD for Watchman LAAC with revised criteria based on my concerns listed above.

Sincerely yours,
Albert C. Lin, MD
Associate Professor of Medicine
Associate Medical Director, Center for Heart Rhythm Disorders at the Bluhm Cardiovascular Institute

Less

To: Ms. Tamara Syrek-Jensen, J.D., Director of Coverage & Analysis Group

Dear Ms. Syrek-Jensen:

Thank you for the opportunity to comment on the proposed decision for percutaneous left atrial appendage closure (CAG-00445N). We have reviewed the seven conditions detailed and would like to propose some comments below.

We commend the Centers for Medicare and Medicaid Services (CMS) for their detailed consideration of coverage for percutaneous left atrial appendage closure (LAAC) therapy using an implanted device (CAG-00445N). However, we strongly disagree with the primary conclusion of that analysis, i.e. that such therapy “is not reasonable and necessary to diagnose or treat an illness or injury or to improve the functioning of a malformed body member and, therefore, is not covered under § 1862(a)(1)(A) of the Social Security Act.”

The illness in question is atrial fibrillation (AF). Warfarin and novel oral anticoagulants (NOACs) are widely used for reduction of the risk of stroke and systemic embolism among patients with AF. Such therapy is guideline-driven and evidence-based, and it constitutes an integral part of the medical management of patients with AF. CMS currently covers these therapies for the treatment of AF.

LAAC therapy using the Watchman device, which is currently the only FDA-approved percutaneous LAAC therapy device, was found in a randomized trial to be comparable to warfarin therapy for preventing ischemic strokes and superior to warfarin therapy for a hemorrhagic/disabling stroke and cardiovascular or unexplained death among patients with AF.1 Thus, this therapy has emerged as a potential alternative to oral anticoagulation for the treatment of AF, which is already covered by CMS as a reasonable and necessary treatment for AF. When used according to its FDA labeling, i.e. as an alternative to warfarin, the Watchman device is also reasonable and necessary for the treatment of AF, and it is superior to an already covered therapy. Thus, the primary conclusion of CMS’s present analysis is not evidence based, and LAAC device therapy should be fully covered because it is reasonable and necessary to treat the illness of AF. We respectfully request that this line of reasoning be carefully considered, as it is sound.

However, we recognize that the proposal to cover LAAC therapy using an implanted device may ultimately be considered by you to be reasonable and necessary only for research under §1862(a)(1)(E) of the Social Security Act using the Coverage with Evidence Development (CED) paradigm. Although we would consider this an incorrect conclusion, we do wish to comment upon the present proposal for such coverage.

In particular, item #2 of the proposal, which addresses patient characteristics, requires revision. Most importantly, it lists “a contraindication to warfarin” as a necessary patient characteristic for use of LAAC therapy. However, this directly contradicts the FDA labeling of the Watchman device, which states that it is indicated only for patients who “Are deemed by their physicians to be suitable for warfarin”, and also the inclusion criteria for the pivotal trial which led to Watchman device approval.1 The reason for this is that the device requires short-term post-procedure anticoagulation for safe use, as there is a risk of thromboembolism prior to endothelialization.

We believe that this language “a contraindication to warfarin” perceived as CMS’s intention to limit use of the Watchman device to patients with a relative contraindication to long-term anticoagulation, for example due to bleeding complications. We consider this highly restrictive and non-evidence-based, and we request that it be removed. However, failing removal, it would of course be essential to change the wording to “a relative contraindication to long-term anticoagulation”, to allow for safe use and to avoid a contradiction with FDA labeling.

Finally, we request that the items in section #2 regarding bleeding risk and stroke risk be revised. LAAC device technology has not been shown to be more effective or safe for any subgroups of warfarin-indicated patients. Similarly, there is no available data suggesting that the Watchman device is more, or less, beneficial among particular HAS-BLED subgroups. Furthermore, the pivotal trial utilized a CHADS2 score of one or more as an inclusion criterion, and there were no requirements for particular bleeding risk profiles other than a requirement for eligibility for long-term anticoagulation.1 Finally, the FDA labeling of the Watchman device has no such requirements other than the above-mentioned suitability for warfarin.

Therefore, based upon clinical trial data and FDA labeling, it is medically appropriate for clinicians to consider LAAC device technology for warfarin eligible patients who have strong reasons to seek an alternative therapy because of long-term risks associated with warfarin. The CHADS2, CHA2DS2 VASC, and HAS-BLED requirements in the current proposal should be revised pending availability of new relevant evidence to support such requirements.

Thank you for your consideration.

Respectfully,

Gunjan J Shukla, MD
Director, Atrial Fibrillation Center
Hackensack University Medical Center

John Zimmerman, MD
Chief, Cardiac Electrophysiology
Hackensack University Medical Center

Less

December 10, 2015

Tamara Syrek Jensen, JD
Director, Coverage & Analysis Group
Office of Clinical Standards and Quality
Centers for Medicare & Medicaid Services
Mailstop C1-09-06
7500 Security Blvd
Baltimore MD 21244

Dear Ms. Syrek Jensen:

We are pleased that CMS is proposing coverage for Percutaneous Left Atrial Appendage Closure (LAAC) and appreciate the thought and consideration reflected in the proposed coverage policy. LAAC [1] is an important treatment option for patients with non-valvular atrial fibrillation (NVAF) who are at high risk of stroke yet have relative contraindications to long-term therapy with oral anticoagulants (OACs).

Coverage with evidence development would allow access to this life-saving therapy for Medicare beneficiaries. WATCHMAN is available in 75 countries, including many in Europe. This effective intervention has been prescribed for more than 10,000 patients worldwide over more than a decade.

Boston Scientific agrees with most of the conditions of coverage in the proposed National Coverage Determination (NCD). We support finalizing Conditions 3, 4, 5 and 7 as proposed. We recommend an editorial revision to Condition 1, as described below. For Condition 2, we have substantive recommendations related to the proposed patient eligibility criteria. And for Condition 6, we have substantive recommendations and significant reservations regarding the proposed inclusion in the LAAC patient registry of a control arm consisting of patients taking oral anticoagulants (OACs). We believe that the data collection goals of the proposed OAC study arm can be better and more efficiently achieved using existing data sources as described below. Our comments on each of these conditions are detailed below.

As a threshold matter, it is important to clearly distinguish three distinct populations of NVAF patients who are at high risk of stroke (as predicted by an elevated CHADS2 or CHA2DS2-VASc), because each group has very different options for stroke risk reduction.

• Population A: Patients with Absolute Contraindications to OAC. Despite being at high risk for stroke, these NVAF patients have an absolute contraindication to taking oral anticoagulants for any period of time. Because these patients cannot take warfarin for any period of time, they are not candidates for WATCHMAN LAAC based on the current WATCHMAN FDA label.
• Population B: Patients able to take Long-Term OACs. This group of NVAF patients is able to tolerate long-term OACs and their need for stroke risk reduction is either being well-managed or is predicted to be well-managed on warfarin or NOACs (new oral anticoagulants). These patients are not the target population for LAAC.
• Population C: Patients with Relative Contraindication(s) to Long-Term Use of OACs, but who are able to take short-term warfarin. This group of NVAF patients at high risk of stroke is not absolutely contraindicated to OACs, but the benefit of long-term OAC therapy is outweighed by the risks associated with lifelong exposure to OACs, particularly with respect to the risk of bleeding complications. Because of these risks, these patients may not take any OAC on a long-term basis and therefore would be left unprotected from stroke. (Importantly, these patients can use warfarin for short periods.) Such patients are in need of an alternative, non-pharmacologic, lifelong stroke risk reduction therapy. These patients are the target group for LAAC as long as they can tolerate a defined period of short-term warfarin therapy following LAAC implantation (e.g., at least 45 days, as required per the current WATCHMAN FDA labeling).

Our comments and recommendations for modifications to Conditions 2 and 6 in the proposed NCD are based on the different clinical characteristics of these three populations of patients and their distinct treatment options. Our recommendations are as follows:

We request that CMS make an editorial revision to the language regarding the FDA approval requirement to clarify that the device has been approved by FDA pursuant to a Premarket Approval (PMA) Application.

As proposed, the language reads:

Our suggested language would read:

The addition of the acronym “PMA" to the proposed language is important to make it clear that CMS coverage is limited to devices that have undergone the rigorous scientific and regulatory review by the FDA that is required for a premarket approval as opposed to a 510(k) clearance. We believe current and future coverage of LAAC devices should be limited to devices that have received PMA approval so as to ensure they meet the highest standards of safety and efficacy prior to being covered for Medicare beneficiaries.

CMS has proposed the following requirements for patient eligibility for LAAC:

We are in full agreement with CMS’ proposed requirement for a high CHADS2 or CHA2DS2-VASc score for assessing level of stroke risk, consistent with the entry criteria for the WATCHMAN clinical trials. This requirement is consistent with current professional society guidelines that define NVAF patients with a CHA2DS2-VASc score of 2 or greater as high risk patients for whom stoke prevention is recommended. [2] 96% of patients enrolled in the four WATCHMAN clinical trials leading to FDA approval had a CHA2DS2-VASc score of 2 or greater. Further subgroup analysis of the WATCHMAN trials using other CHA2DS2-VASc cut points showed that those with and without higher CHA2DS2-VASc scores benefited equally from the WATCHMAN Device (e.g., using a cut point of > 3, the p-value for interaction was 0.176). [3] We applaud CMS for recognizing both professional society guidelines and the strong WATCHMAN clinical data in proposing a requirement for a high CHADS2 or CHA2DS2-VASc (score of 2 or more) in the proposed eligibility policy.

However, we urge CMS to revise the proposed eligibility policy to permit LAAC coverage for patients who, in addition to having a high CHADS2 or CHA2DS2-VASc score, are able to take short-term warfarin, and have either a high HAS-BLED score or have another reason for which the clinical risk of long-term anticoagulation outweighs the benefit.

If CMS adopts our recommendations, Condition 2 would read as follows (“CAPS” shows recommended changes):

We provide our rationale in the sections below.

As proposed, the word “contraindication” is confusing because it could be misinterpreted to mean an absolute contraindication to warfarin. Patients with an absolute contraindication to warfarin fall into Population A of NVAF patients (as described above). The FDA has not approved WATCHMAN use for this population because these patients were not studied in any of the four trials that formed the basis of the FDA approval for WATCHMAN. However WATCHMAN was studied and approved in the U.S. for patients in Population C: those who have a relative contraindication to long-term warfarin (i.e., patients for whom the risk of long-term oral anticoagulants outweighs the benefit).

Use of WATCHMAN in Population A would be off-label and should not be covered by Medicare. Restricting coverage only to this population would be overly restrictive and deny access to WATCHMAN LAAC for the population that has been studied and for which FDA approval has been granted. Those in Population C, however, should be candidates for coverage if they meet the other criteria in the NCD. [4] If this point is not clarified, however, we are concerned that the word “contraindication” by itself will create confusion for clinicians, Medicare contractors, the Department of Justice, and others.

The second issue concerns the requirement that a patient have both a high HAS-BLED score and a contraindication for warfarin. The HAS-BLED scoring system was developed to assess the one-year risk of major bleeding in patients with atrial fibrillation (AF). The most recent European Society of Cardiology guidelines for AF categorize a HAS-BLED score of >=3 as high. [5]

HAS-BLED does not capture all of the relevant clinical considerations that may make a patient a poor candidate for long-term OAC therapy (e.g., frailty), and is known to be imperfect. In an evaluation of HAS-BLED and ATRIA bleeding scores, HAS-BLED demonstrated only modest ability to differentiate bleeding risk for patients with low-to-moderate (0-2) vs. high (3+) scores, with a c-statistic of 0.68. [6] As applied to ROCKET-AF, HAS-BLED had a remarkably poor ability to discriminate bleeding risk between moderate and high values with a c-statistic of only 0.59 (the c-statistic for a coin flip = 0.50). [7] In addition, HAS-BLED is only one tool for assessing major bleeding risk, and would become obsolete as a requirement for coverage if replaced by a more accurate scoring system.

Thus, the proposed eligibility criteria would effectively deny LAAC coverage to NVAF patients eligible for WATCHMAN according to the FDA label and in need of an alternative to OAC therapy. Specifically, there are patients who have valid clinical reasons (e.g., history of GI bleeding) not to take lifelong OAC therapy but who may not have these valid clinical reasons manifested in a high HAS-BLED score. In essence, this speaks to the need for additional language to cover patients with “another reason for which the clinical risk of long-term anticoagulation outweighs the benefit.” This addresses specific situations where patients without a high HAS-BLED score require access to LAAC therapy.

The following profiles are examples of patients who are eligible for WATCHMAN (according to the label) because of a high risk of stroke and clinical risks where long-term OAC use outweighs the benefits, but who have moderate HAS-BLED scores. These patients would not be eligible for WATCHMAN under proposed Condition 2:

Typical Patient Example 1: High stroke risk and prior bleed:
A 77 year-old female with NVAF, diabetes, prior myocardial infarction, prior life-threatening GI bleed requiring ICU hospitalization and transfusion, and osteoarthritis. She has a high CHA2DS2-VASc score of 5, demonstrating her increased stroke risk, for which she takes warfarin. She has a HAS-BLED score of 2, which means she has a moderate risk of a major bleeding event. She needs to take a daily NSAID (nonsteroidal anti-inflammatory drug) for osteoarthritis, but because of bleeding risk, her doctor is reluctant to allow her to do so in combination with warfarin. However, she is not a patient who is absolutely contraindicated for OACs. Based on her individual risk-benefit profile, her doctor is willing to prescribe warfarin for 45 days after LAAC implantation (consistent with current labeling) but not for long-term use.

Left without any reasonable stroke risk reduction treatment option in the absence of LAAC, she has an annual stroke risk of 6.7%, representing a greater than 1 in 4 chance of a stroke before her 82nd birthday (5-year stroke risk of 29%).

This Medicare beneficiary would be ineligible for coverage based on the proposed language in Condition 2 that requires a high HAS-BLED score.

Typical Patient Example 2: High stroke risk and dual antiplatelet therapy (DAPT)
A 76 year-old male with NVAF, diabetes, congestive heart failure, and coronary artery disease has a high stroke risk CHA2DS2-VASc score of 5, requiring stroke risk reduction OAC therapy, and a moderate HAS-BLED bleeding risk score of 2. His recent acute coronary syndrome was treated with angioplasty and drug eluting stent (DES) implant, requiring dual antiplatelet therapy (DAPT).

Unfortunately, the combined use of OAC with DAPT results in a very high risk of major bleeding 12.3% per year [8] that is not fully represented by the HAS-BLED score. However, without stroke risk reduction therapy, this patient has an annual stroke risk of 9.8%.

This Medicare beneficiary would be ineligible for coverage based on the proposed language in Condition 2 that requires a high HAS-BLED score.

These exemplar patient profiles demonstrate that there are many other risk factors that are not accounted for by the HAS-BLED score alone, and WATCHMAN eligible beneficiaries in Population C would be excluded from coverage based on the proposed NCD language despite compelling clinical reasons for seeking a non-pharmacologic alternative for stroke prevention. Patients without high HAS-BLED scores should not be excluded from coverage, but rather this score should be considered as one of several relative contraindications, and the language should be revised to indicate coverage for those with high HAS-BLED scores or another reason for which the clinical risk of long-term anticoagulation outweighs the benefit.

Finally, the use of a HAS-BLED score to limit the covered population is not supported by the clinical study data. At the time of the trials the HAS-BLED scoring system did not exist, so analyses of the data using HAS-BLED have been conducted retrospectively using conservatively estimated scores (i.e., giving zero points for the two HAS-BLED elements that were not collected in the case report forms). Subgroup analysis of the WATCHMAN trials using these conservatively estimated HAS-BLED scores shows that patients with and without “high” HAS-BLED scores benefited equally from the WATCHMAN device (p-value for interaction 0.098). [9]

We urge CMS to revise the patient eligibility criteria as suggested (deletions in [BRACKETS] and additions in “CAPS”).

In aggregate, these changes would allow coverage (under CED) of patients who have been shown in clinical trials to benefit from WATCHMAN LAAC.

Boston Scientific supports the national registry for LAAC as a condition of CED. A registry will allow tracking long-term clinical outcomes of patients with WATCHMAN implants. However, we strongly disagree with the proposed requirement of a contemporaneous OAC control arm and provide our rationale in the subsequent sections.

We urge CMS to revise the language of Condition 6 as follows: (deletions in [BRACKETS] and additions in “CAPS”).

We believe these changes will allow CMS to accomplish its goal of comparing the outcomes of patients in the WATCHMAN registry to a contemporaneous group of patients taking OACs or not receiving therapy. We explain why below.

The proposed language is inconsistent with respect to the length of follow-up to be required. In some cases, CMS has indicated >=5 years and sometimes indicated >5 years, suggesting that certain data collection and patient follow-up beyond 5 years would be required. We believe that 5 year follow-up is sufficient for the required analyses and therefore recommend that 5 years be used consistently within the document.

The proposed CMS LAAC coverage policy would require a registry-based, multi-year, prospective, observational clinical study in which all LAAC eligible patients, their treating physicians and institutions would have to participate in order to receive coverage for LAAC therapy. Boston Scientific understands and supports the need for additional data on outcomes for on-label LAAC therapy. We have partnered with the American College of Cardiology’s (ACC) National Cardiovascular Data Registry (NCDR) Program to establish the Left Atrial Appendage Occlusion (LAAO) Registry to follow LAAC patients as a means to collect and expand real-world LAAC outcomes data on Population C (as described above).

Additionally, CMS has proposed a contemporaneous OAC control arm in the registry to which LAAC outcomes can be compared. Including a required OAC arm in the registry would pose scientific and operational challenges that could jeopardize Medicare LAAC coverage because of failure to adequately execute the terms of the NCD.

The key objections to an OAC control arm are as follows:

As proposed, the analysis would compare outcomes of LAAC and OACs in distinctly different patient populations. Patients eligible for LAAC therapy are in Population C, those for whom long-term OAC therapy is relatively contraindicated but are able to take short-term warfarin therapy (as described above in our discussion of Condition 2). Patients doing well on long-term OACs are an entirely different patient population described above as Population B. The two populations are intrinsically different and are not comparable. Physicians would not be able to recommend treatment to any one patient based on the resulting comparisons because the treatments are not therapeutic alternatives for a common patient profile.

Furthermore, collecting outcomes data on OAC patients within the context of an LAAC registry raises the following selection bias, operational, and recruitment concerns that will severely limit potential enrollment and reduce the scientific validity of the effort:

• Patients with a history of long-term adherence to OACs are self-selected for better outcomes because they have been on OACs and have not suffered bleeding or ischemic stroke. So these patients would be inappropriate to use as a group intended to provide an appropriate comparison to those receiving LAAC.


• OAC patients (Population B) are generally cared for by family practitioners, internists and cardiologists. They are also cared for in many sites of service (e.g., physician offices, community hospital clinics, large hospital systems with anticoagulation clinics) and in many settings (e.g., rural, urban). These are all factors that must be taken into account because they all may affect risk. For example, family practitioners may be more likely to treat compliant patients with an easy to manage INR (International Normalized Ratio), and cardiologists may be more likely to treat complex patients with a difficult to manage INR . Any of these factors could bias the proposed control arm and provide an invalid comparison to LAAC.

  In addition, Boston Scientific and other industry stakeholders do not have access to the physicians who treat patients with OACs, making both physician and patient recruitment difficult. The vast majority of NVAF patients on long-term OACs are treated by medical cardiologists and primary care physicians not interventional cardiologists (IC) and electrophysiologists (EP), who treat LAAC patients. In fact, these IC and EP specialists only see highly selected patients on long-term OAC therapy, and are generally not involved in the long-term management of these patients.

• Because OAC treatment is already covered by Medicare Part D plans and likely by most Medicare Advantage plans, physicians and the vast majority of patients have no incentives to take the time and effort, and expend the resources required to enroll in a control arm. In contrast, LAAC patients would be entered into the registry consecutively and completely as a condition of coverage. Differences in coverage and rationale for enrollment could create selection bias and limit enrollment in the OAC arm.


• While the LAAO Registry has received a waiver for IRB approval and informed consent for the LAAC arm, it is not clear whether such a waiver could be granted for an OAC control arm. Therefore, enrollment in a control arm could be problematic.


• The size of the OAC population is so large that it would not be feasible to enforce consecutive and complete enrollment of these patients across such a massive physician network. Choosing a subset of providers and sites will introduce selection bias in contrast to the LAAC arm where all sites will be required to submit data. Even if a subset of sites are chosen and OAC patients could be recruited to participate, there could be an additional selection bias produced if physicians selectively enroll patients in the registry (as opposed to comprehensively enrolling all patients consecutively which is what is required for LAAC patients). The selection bias inherent in non-consecutive enrollment of patients in a limited set of sites would make any comparison to LAAC either invalid, or at best, limited in utility.

There is no way to adequately adjust for selection bias between the two disparate patient groups and this situation would result in an ‘apples to oranges’ comparison. This, along with the serious structural and methodological issues discussed above, makes an OAC control arm within the LAAC registry unworkable, creating another reason why the results will not be scientifically valid for treating physicians, researchers, or policy makers.

Stakeholders such as Boston Scientific have generally relied on CMS guidance on CED when formulating product development programs and designing clinical trials. Specifically, the medical device industry has designed FDA pivotal trials with CMS coverage criteria in mind. The guidance provided by CMS since 2006 on CED is clear.

In July 2006, CMS CED guidance stipulated that “CED take place in the context of existing data systems when feasible.” Furthermore, in November 2014, additional CMS CED guidance stipulated that “CED will lead to the production of evidence complementary to existing medical evidence.”

The proposed LAAC study OAC arm ignores the fact that there is already a wealth of comprehensive and varied OAC treatment outcomes data that is more scientifically valid than what could possibly be produced through an OAC control arm that is part of an LAAC registry. Establishing another warfarin or NOAC registry would be duplicative of data being collected in existing registries and would, as described above, at best, yield limited new information that would fail to justify the time, effort, and expense of such an undertaking.

In summary, including a control arm in a registry required under CED would be contrary to longstanding CMS policy and guidance and would result in creating a new, scientifically questionable registry study arm that, at best, would duplicate data from existing registries.

The desire for OAC outcomes data to provide context for LAAC performance can be satisfied using data from one or more existing sources. We list several below. These sources contain data that may be more useful than would data produced by an OAC control arm in the WATCHMAN registry.

• The Swedish National Hospital Discharge Registry (Friberg; doi:10.1093/eurheartj/ehs031) has helped clinicians and policy makers to better understand the outcomes of NVAF patients treated with OACs compared to those untreated by OACs. This full population registry includes all individuals diagnosed with AF at any Swedish hospital (July 2005 – December 2008). They were prospectively followed for an average of 1.5 years (260,000 years at risk). Its objective was to investigate risk factors for stroke and bleeding in AF and evaluate CHA2D2-VASc and HAS-BLED scores for outcomes of stroke and bleeding. Its endpoints include ischemic stroke, transient ischemic attack, systemic embolism, intracranial hemorrhage, and major bleeds.

  These data were the basis of the 2014 National Institutes of Clinical Excellence (NICE) AF patient decision aid. [10] In addition, Boston Scientific performed a comparison using this registry versus ischemic stroke rates observed in WATCHMAN Device trials. This analysis demonstrated that LAAC provided a 65-81% relative risk reduction in ischemic stroke over no therapy in an unselected, similar population. [11] As part of our reporting to the FDA, our analysis is regularly updated and demonstrates similar performance of WATCHMAN to OACs in reducing the risk of ischemic stroke as compared to patients who are not taking OACs.

• The Outcomes Registry for Better Informed Treatment of Atrial Fibrillation II (ORBIT-AF II), currently recruiting patients in the U.S., Puerto Rico, and U.S. Virgin Islands, is a prospective, observational study of outpatients with the objective of evaluating target-specific antithrombotic agents and outcomes associated with these agents (https://orbit-af.dcri.duke.edu/). Eligible patients are those with new onset/first detected AF diagnosed within the preceding 6 months, or with AF who had initiation or transition to a Factor Xa inhibitor or a direct thrombin inhibitor within the preceding 3 months. Over two years of follow-up, the primary endpoints will include bleeding events, stroke, or systemic embolism; secondary endpoints will include major cardiac events. The primary effectiveness endpoint events of stroke and systemic embolism are the only events adjudicated by review of primary documentation.


• The Global Registry on Long-Term Oral Anti-thrombotic TReatment In PAtients with Atrial Fibrillation (GLORIA-AF), currently in Phases II/III, is collecting real world data on important outcome events in newly-diagnosed patients with NVAF over three years of OAC use (https://clinicaltrials.gov/ct2/show/NCT01937377?term=gloria-af&rank=2). Phase II, beginning early after approval of dabigatran, monitors dabigatran safety and addresses potential channeling across treatment options based on propensity scoring to assess comparability of baseline characteristics of patients treated with dabigatran or VKA. Phase III entails analysis of large treatment groups, adjusting for differences in propensity score, to provide information about the relative effectiveness and safety of NOACs and VKA in routine clinical care. Endpoint events not adjudicated include cardiovascular accident, transient ischemic attack, systemic embolism, myocardial infarction, pulmonary embolus, death, and bleeding.


• PINNACLE-AF Registry (http://cvquality.acc.org/ncdr-home/data-collection/pinnacle-data-collection.aspx) is an outpatient cardiology registry managed by NCDR that collects information on OACs for AF patients in the U.S. Data collection began in early 2012 with summary reports available after Q1 2012. The targeted practices are general cardiologists in the outpatient setting who voluntarily enter OAC data. The data is pulled from electronic health records into the existing construct of the PINNACLE Registry. Currently, there is no adjudication process for the data collected and it does not capture data related to stroke or systemic embolism.

Another possible source of comparator information is the use of Medicare claims data to track patients and outcomes. Conceptually this is similar to the construct of the Swedish AF Registry, but the Swedish registry did not use claims data. While the use of Medicare claims data to track outcomes of patients with NVAF on OACs is theoretically possible, there are challenges that make it unlikely that the aggregated data will be useful to provide a cohort comparable to the WATCHMAN registry.

To initiate this effort, it is likely that the Medicare Part B 5% file would be used as a source to identify a cohort of NVAF patients who developed new onset NVAF in 2016. Once this cohort has been identified, one would link this group to patients with the Part A and Part D databases to obtain information on hospitalizations, SNF stays, and use of OACs and other medications. These patients would then need to be followed from a claims perspective for at least five years (or until death).

The Part A and B information in these files contains all items and services billed to Medicare, the payment amounts for those items and services and the diagnosis code(s) for each item and service. The Part D data will include the information submitted by pharmacies to the Part D plan.

There are a number of important limitations and challenges when it comes to using Medicare claims to track patient outcomes. These include:

• Diagnosis coding by physicians is acknowledged to be inaccurate and non-specific
• Lack of familiarity with ICD-10 which is new starting Oct 2015
• ICD-10 is entirely different from ICD-9 and contains many more codes and is much more granular
  • There are many codes for cerebral infarction and hemorrhage that relate to different territories and different etiologies. Will physicians code specifically enough to provide information or just use a general code for infarction that doesn’t say whether the infarction was due to thrombus or embolism?
• Hospital diagnosis coding is more accurate but patients may never be in the hospital
• Deriving an accurate CHA2DS2-VASc score from claims data may be difficult
• Even if one can accurately identify patients with new onset NVAF (which may be difficult), one may not know whether the patient stays in AF or reverts to normal sinus rhythm (NSR) as physicians may continue to use an NVAF diagnosis code due to a “history” of NVAF
• Claims data will let us know how often the INR was checked but not the results of the test
• Part D data will let us know when scripts were filled but not whether the patient was actually taking the OAC
• Patients may stay on OACs even if no longer in NVAF
• Severity of stroke will be very hard to determine from claims data
• Encounter data not available for the approximately 30% of patients enrolled in Medicare Advantage plans

Based on the challenges identified above, the utilization of Medicare claims data as a stand-alone is unlikely able to construct a meaningful comparative cohort for NVAF patients in the WATCHMAN registry.

We recommend that CMS permit the use of the Swedish data. Alternatively, CMS could permit the use of one or more of the other data sources discussed above to provide context for the outcomes of patients in the LAAC registry. To ensure a scientifically rigorous and meaningful comparison, it will be important that the analytic plan ensures:

1. that comparative data are derived from “de novo” OAC patients;
2. to the extent feasible, the source data includes adjudication of events similar to what will occur in the LAAC registry;
3. to the extent feasible, the same follow-up and duration of follow-up is available as required in the LAAC registry; and
4. if the other data source(s) are not population-based, patients from these other data source(s) should be propensity matched to potentially minimize biases.

CMS indicated in the proposed decision memo that the population that may benefit the most from percutaneous LAAC therapy are those high-risk NVAF patients with an absolute contraindication to OACs (i.e., Population A described above) and expressed a desire for additional clinical data on this group. Boston Scientific agrees and has begun working closely with the FDA to design a prospective, randomized controlled multi-center, global clinical trial (ASAP TOO) to establish the safety and effectiveness of WATCHMAN for stroke risk reduction in patients who have an absolute contraindication to OAC therapy (i.e., patient Population A described above) . The data from this study will help provide important evidence to address the comparative effectiveness of WATCHMAN in this patient population.

Boston Scientific recommends that CMS retain the LAAC registry and proposed study outcomes, including the five-year follow-up, but we urge the elimination of the proposed OAC arm of the registry study. We have provided above a strikeout of proposed Condition 6.

In order to answer CMS questions and provide better context around the comparator group of interest, we further recommend utilization of existing sources of OAC outcomes data. Finally, Boston Scientific’s upcoming ASAP TOO IDE randomized trial will provide important information about safety and efficacy of LAAC in NVAF patients with absolute contraindications to OACs.

The changes discussed and recommended above will have the effect of ensuring that the final NCD covers only on-label patients. (As discussed above, the proposed policy could be misinterpreted to cover off-label patients.) If changes are made to the proposal so that the final NCD covers only on-label patients, the final NCD should explicitly make clear that coverage of future IDE studies will be provided under CMS’ existing IDE trial policy regulations and will not be precluded by the final NCD. This is especially important because the Category B IDE ASAP TOO trial mentioned above is expected to include outcomes in Population A, which is of interest to CMS.

We recommend the following language:

• Boston Scientific supports Conditions 1, 3, 4, 5, and 7.
• We request CMS add “PMA” before “approved” in Condition 1 to clarify that coverage is for FDA “PMA” approved devices for patients with NVAF.
• We urge CMS to make changes to Conditions 2 and 6 detailed below. The specific changes to these conditions are identified in [DELETIONS] and “ADDITIONS.”

  Condition 2:

  The patient [HAS]:
  • “HAS” a high CHADS2 (Congestive heart failure, Hypertension, Age >75, Diabetes, Stroke/transient [ISCHEMIA] “ISCHEMIC” attack/thromboembolism) or CHA2DS2-VASc score (Congestive heart failure, Hypertension, Age >= 65, Diabetes, Stroke/transient [ISCHEMIA] “ISCHEMIC” attack/thromboembolism, Vascular disease, Sex category); and
  • “IS ABLE TO TAKE SHORT-TERM WARFARIN; AND”
  • “EITHER HAS”
    • A high HAS-BLED score (Hypertension, Abnormal renal function and/or liver function, Stroke, prior Bleeding, Labile anticoagulation range, Elderly age>65, Drug therapy such as antiplatelet drugs), [AND] “OR”
    • [A CONTRAINDICATION TO WARFARIN] “ANOTHER REASON FOR WHICH THE CLINICAL RISK OF LONG-TERM ANTICOAGULATION OUTWEIGHS THE BENEFIT.”

  Condition 6:

  The patient is enrolled in, and the treating physician team is participating in a prospective national registry that consecutively enrolls LAAC patients and tracks the following annual outcomes at the patient data level for a period of [AT LEAST] five years from the time of the LAAC procedure.

  • Operator-specific complications
  • Device-specific complications including device thromboses
  • Stroke, adjudicated, by type
  • TIA
  • Systemic embolism
  • Death
  • Major bleeding, by site and severity

  The registry must be designed to permit identification and analysis of patient, practitioner and facility level factors that predict patient risk for these outcomes. [THE REGISTRY MUST INCLUDE CONTEMPORANEOUS PATIENTS FOLLOWED ON ORAL ANTICOAGULANT (OAC) THERAPY TO SERVE AS NON-INTERVENTIONAL CONTROLS.] CMS will review the qualifications of candidate registries to ensure that the approved registry follows standard data collection practices and collects data necessary to evaluate the patient outcomes specified above.

  The “PROSPECTIVE” registry must collect all data necessary to conduct multivariable adjusted analyses and have a written executable analysis plan in place to address the following questions:

  [IN A PROSPECTIVE, CLINICAL STUDY, WHEN] “WHEN” percutaneous LAAC therapy using an implanted device is performed outside a randomized, controlled clinical trial [, COMPARED TO NON-INTERVENTIONAL CONTROLS FOLLOWED ON ORAL ANTICOAGULANT (OAC) THERAPY]:

  • How do the outcomes listed above compare to outcomes in the pivotal clinical trials in the short term (<12 months) and in the long term ([>=] 5 years)?
  • How do the outcomes listed above differ from “OUTCOMES FROM OTHER DATA SOURCES CONTAINING DATA ON PATIENTS TAKING OACS OR NOT TAKING ANY THERAPY” [ OUTCOMES IN COMPARABLE CONCURRENT OAC CONTROLS] in the short term (<12 months) and in the long term ([>]5 years)?
  • What is the long term ([>=]5 year) durability of the device?
  • What are the short term (<12 months) and the long term ([>]5 years) device-specific complications including device thromboses?
• Lastly, we recommend CMS provide clarification for coverage of future IDE trials for LAAC therapy in the final NCD.

While Boston Scientific is pleased that CMS is proposing to cover percutaneous LAAC therapy under CED, we were dismayed by the characterization of the WATCHMAN clinical evidence within the CMS proposal. The CMS analysis highlights data interpretation challenges that occurred throughout the 10 year history of research, but does not recognize that these questions were ultimately addressed and resolved via three successful FDA advisory panels leading up to FDA approval. The WATCHMAN IDE clinical trials were well-defined, well-run and produced sound clinical evidence, adjudicated by independent review boards and accepted by the FDA as evidence to support safety and efficacy.

The long-term follow up (of more than 7,000 patient years) from the WATCHMAN clinical trials continues to demonstrate that WATCHMAN is as effective as warfarin in providing a non-pharmacologic alternative for NVAF patients at high stroke risk and, therefore, that WATCHMAN can be expected to provide a significant benefit to patients who otherwise face the choice between accepting a therapy for which they have a significant relative contraindication or to be unprotected from a high risk of stroke. The real world outcomes data evidenced by our commercial experience continue to support this premise. [12,13,14] The Appendix provides specific clarification of the trial data, statistical methods, and analytical plans.

We are pleased with the CMS proposal to cover WATCHMAN under coverage with evidence development and generally support the CMS proposed approach with the exceptions discussed in this letter. As we discussed, we have two major concerns that, if not addressed in the final NCD, have the potential to severely limit (1) the covered patient population and exclude many patients who could benefit from WATCHMAN and (2) the availability of this therapy in the United States by imposing unworkable post-market clinical research requirements as a condition of Medicare coverage. We look forward to revisions that will ensure appropriate access to this important therapy for Medicare beneficiaries with NVAF at high risk for stroke, who are eligible for WATCHMAN and require a non-pharmacological treatment option for stroke risk reduction.

Sincerely,

Parashar B. Patel
Boston Scientific
Vice President, Global Health Economics & Reimbursement

References:

In addition to the coverage recommendations provided above, Boston Scientific would also like to take an opportunity to comment on concerns we have about the CMS analysis of the WATCHMAN clinical data. While CMS has done a commendable job in consolidating the chronological history associated with approval of WATCHMAN, we believe CMS has taken an unbalanced view of the 10 years of robust clinical data collected on the WATCHMAN device. Explanations and analyses provided by Boston Scientific were compelling enough to persuade three Circulatory Device Advisory (CDRH) Panels to vote in favor of WATCHMAN, ultimately leading to FDA approval. We appreciate the opportunity to address CMS’s discussion of the data in this portion of our comment letter.

While we do not wish to address each point raised in the decision memo, we would like address three main areas of concern and provide context on the following:

Boston Scientific believes the studies, their associated statistical models, and the results produced are of the highest quality and were rigorously executed. We would also note that all of the study designs were reviewed and approved by the FDA. All four IDE trials conducted in support of approval were run according to FDA regulations and requirements. The trials:

• included strict site monitoring and source validation practices;
• used validated statistical models developed in conjunction with FDA that were designed to adequately assess the safety and efficacy of the therapy;
• had oversight from both an independent Clinical Events Committee and Data Safety Monitoring Board for adjudication and safety oversight; and
• had all results evaluated by expert physicians and statisticians as a part of the FDA advisory panel review process.

We disagree with repeated concerns raised by CMS as to the quality and design of these trials.

For instance, the proposed NCD quotes the FDA Executive summary from 2013: “The primary effectiveness endpoint [from PROTECT AF] was met for statistical noninferiority. However,…the model [for PROTECT AF] used to estimate the event rates is not accurate and sole reliance on the statistical results from the primary analysis may be problematic.” CMS has taken this quote out of context. The FDA’s statement was not referring to the model’s accuracy with respect to estimating event rates to calculate the primary effectiveness endpoint for non-inferiority. The non-inferiority findings from PROTECT-AF were not in question, and the results of the PROTECT AF study prove WATCHMAN is a comparable, similarly efficacious alternative to warfarin. The statistical results and model, as it relates to evaluation of non-inferiority, are accurate.

After questioning the statistical validity of the studies, CMS then turns its attention to the biases it asserts were introduced by the patients enrolled, the patients excluded, and the withdrawal rate of the warfarin control group. CMS suggests that PROTECT AF “had…major issues that create low confidence in the ability to determine patient benefit”, including patient population. Specifically, CMS states that “the patient population was weighted toward lower risk patients based on the CHADS2.” This assertion is incorrect. As we discussed in our NCD request, analysis with the more contemporary CHA2DS2-VASc score demonstrated that, in fact, over 93% of the patients in PROTECT AF, 96% in CAP, and 100% in PREVAIL and CAP2 were at high risk of stroke (CHA2DS2-VASc = 2) and all were eligible for anticoagulation based on current ACC/AHA/HRS guidelines. Patients enrolled in these trials were high risk, and any assertion to the contrary is incorrect.

In addition, because of inclusion/exclusion restrictions from the trial, CMS states that “[o]f all patients screened, 86% were deemed ineligible, calling into question the applicability of results to real world patients in [sic] seen in clinic.” Because of the rigors required with any clinical trial, additional exclusion criteria are always conservative so as to remove multiple confounding factors from clouding the results. The FDA-approved WATCHMAN label reflects the exclusion criteria in the trial; therefore, the trial population is a reflection of real-world patients at high risk of stroke. In short, the patients in the trials were in fact at high risk of stroke and are representative of the population for whom we are seeking coverage.

CMS calls into question the quality of the warfarin control arm in the PROTECT AF study, both in terms of the rate of events experienced and withdrawal rates. In particular, with reference to INR measures CMS states, “Changes in trial conduct and execution appear to affect the trial results.” In both PROTECT and PREVAIL, the Time in Therapeutic Range (TTRs) (used to assess adequacy of anticoagulation) were 68-70%, demonstrating an extremely well-managed drug arm, with better warfarin management than that achieved in any NOAC study warfarin arm. And while hemorrhagic stroke rates in PROTECT AF seem, on the surface, higher than other contemporary trials, all were adjudicated by an independent panel of physicians, including neurologists, using the same definition for both randomized arms. As discussed at the most recent FDA advisory panel, the rate of bleeding in the WATCHMAN trials most likely reflects appropriate selection of patients with elevated bleeding risk to participate in the trials. In summary, the magnitude of the benefit cannot be dismissed by trial conduct or anticoagulation management as CMS incorrectly implies.

While CMS is correct in noting the high dropout rate, CMS proceeds to surmise that this “could lead to bias against the control group and favoring of the device group for the long-term event rate comparisons.” CMS has incorrectly characterized the likely influence of the dropout rate on the non-inferiority and superiority outcomes of the trial. As described in the 4-year PROTECT AF results published in JAMA (2014) by Reddy et al., “Additional sensitivity analyses revealed that as assessed by TTR prior to dropout, the higher-risk warfarin patients withdrew from the study, thereby biasing the study against the device group.” In that same publication, a vital status update obtained on a portion of the patients that left the study revealed the overall mortality even more strongly favored the device group (HR 0.63; 95% CI 0.15-0.90; p=0.01). The effect of differential withdrawal actually biases the results against the device, not against the warfarin arm as CMS has concluded.

Next, CMS reported the limitations and short-comings of the PREVAIL study. Missing from this narrative is acknowledgement of a key component of the statistical design: PREVAIL was not designed as a stand-alone assessment of LAAC because it only represents 25% of the randomized trial experience. CMS correctly notes the imbalance of ischemic strokes in the device arm, but does not provide context for this performance. First, the WATCHMAN stroke rates were consistent with all previous trials and commensurate with the increased CHA2DS2-VASc scores of the trial patients. Second, CMS did not mention that out of the 138 patients in the warfarin arm (mean CHA2DS2-VASc = 4.0), only 1 patient had a stroke after 2 years of follow-up a complete outlier when compared to all contemporary NOAC warfarin control arms. Finally, by the very nature of the Bayesian design and the informative prior from PROTECT AF, this study must be viewed in the context of the totality of long-term follow-up data available from all randomized subjects. While pointing out the challenges and findings of PREVAIL, CMS has not put the PREVAIL results in context of the totality of the data.

In order to evaluate the full totality of the data, the most complete and comprehensive assessment comes from the patient-level meta-analysis (Holmes et al., JACC 2015) that CMS has chosen to dismiss as it “adds little to the previous evidence base.” However, it is indisputable that this study encompasses the complete evidence base and thus represents the most robust evaluation of the totality of randomized data available. It definitively shows that, considering the totality of the data, LAAC with WATCHMAN as a comparable, non-inferior alternative to warfarin for efficacy and for prevention of all strokes. While there is an imbalance of ischemic strokes favoring warfarin, this was offset by a highly significant reduction in hemorrhagic strokes favoring the WATCHMAN device (78%, p=0.004). Since hemorrhagic strokes are typically catastrophic, this is an important point in favor of WATCHMAN.

In addition, it is important to consider the analysis of whether any strokes in this high risk population were disabling (Reddy et al., JAMA 2014). Although strokes are generally associated with loss of brain function, not all strokes impact patients’ lives the same way. For some patients, a stroke may result in death or permanent disability while other patients may have loss of function that can be completely regained or remediated by treatment such as physical and/or occupational therapy. WATCHMAN was associated with a significant 56% (p=0.044) reduction in the rate of disabling strokes compared to warfarin. This is of particular significance given the appropriate emphasis CMS places on “health outcomes actually experienced by patients, such as quality of life, functional status, duration of disability, morbidity and mortality…The direction, magnitude and consistency of the risks and benefits across studies are also important considerations.” It would appear that these compelling outcomes should have been mentioned as consistent with CMS priorities.

Finally, CMS includes no meaningful mention of the significant cardiovascular mortality benefit associated with WATCHMAN in the trial population. The aforementioned meta-analysis shows WATCHMAN provides a significant reduction in cardiovascular mortality over warfarin (52%, p= 0.006). No other stroke risk reduction therapy offers such a magnitude of benefit for, arguably, the most important outcome to Medicare beneficiaries.

In conclusion, Boston Scientific would like to reiterate that the statistical models and resulting outcomes of the FDA pivotal trials and registries were robust; patients studied in the trials were representative of general population in need of an alternative to OACs; and the key clinical results support benefit for Medicare patients.

Less

December 10, 2015

Ms. Tamara Syrek-Jensen, JD
Director of Coverage & Analysis Group

Joseph Chin, MD, MS
Deputy Director, Coverage and Analysis Group

Lori Ashby, MA
Director, Division of Medical and Surgical Services

Kimberly Long
Lead Analyst

Jyme Schafer, MD, MPH
Lead Medical Officer

Rosemarie Hakim, PhD
Epidemiologist

Re: CMS Proposed Coverage Decision for Percutaneous Left Atrial Appendage Closure (LAAC) Therapy

Dear Ms. Syrek-Jensen,

I am an Electrophysiologist practicing at a tertiary care medical center, and have been involved with left atrial appendage closure both as a practicing clinician taking care of patients, and as an investigator in several of the clinical trials related to the Watchman device. It is within this context that I would like to make a few comments with regard to the Proposed Coverage Decision for Percutaneous Left Atrial Appendage Closure (LAAC) Therapy.

First, I would like to make a few comments regarding the scientific evidence behind LAAC. I will focus primarily on the Watchman device since:

Second, I want to address safety. It is certainly true that LAAC is a procedure, and as with all procedures, there is a risk for complications with Watchman implantation. However, there have been important trial-level learnings that have allowed us (electrophysiologist and interventional cardiologists) to improve the safety of the procedure. Briefly, if one looks at the most dangerous complication of Watchman implantation, pericardial tamponade, the rate decreased significantly over the course of the clinical trial experiences from as high as ~5% in PROTECT AF to 1.9% in PREVAIL; this is particularly striking since ~40% of the operators and procedures in PREVAIL had been performed by completely naïve operators – that is, operators who had never before performed an LAAC procedure (JACC 64:1-12, 2014). Finally, there is recent real-world data from the large (102 patients) prospective, multicenter, European EWOLUTION study – in this trial, the cardiac tamponade rate was 0.5% and the procedure-related 30-day mortality rate was 0.1% (presented at the late-breaking scientific sessions of the American Heart Association in November 2015; see http://my.americanheart.org/idc/groups/ahamah-public/@wcm/@sop/@scon/documents/downloadable/ucm_479168.pdf). One cannot reasonably contest that these safety data are favorable.

Thus, I strongly believe that an honest look at the data indicates clear superiority of Watchman over warfarin. However, it is also certainly true that warfarin is no longer the only pharmacological option– NOACs are clearly excellent drugs and have become first-line therapy. And there is no randomized data comparing Watchman with NOACs (just as there are no randomized trials comparing one NOAC vs another NOAC). Thus, it is difficult to compare how the Watchman device would perform against NOACs. However, the best comparison that can be made would be to determine the imputed benefit in preventing stroke (as compared with Placebo) based on the various trials (SPAF/BATAF/SPINAF for warfarin, RELY for dabigatran, ROCKET-AF for rivaroxaban, ARISTOTLE for Apixaban, ENGAGE AF for edoxoban, and PROTECT AF/PREVAIL for Watchman LAAC). In this analysis, it turns out that the expected stroke reductions are: 67% for warfarin, 71% for NOACs, and 66% for Watchman. Again, I want to underscore that the purpose of this analysis is not to somehow equate and encourage Watchman vs NOACs. However, this seeming equivalence surely buttresses the argument that for patients at significant risk for stroke who are not good candidates for long-term oral anticoagulation (both warfarin and NOACs), LAAC would be a reasonable option. After all, what else can be done for these patients?

Next, I want to briefly address the issue of cost. While LAAC is a procedure, and all procedures are saddles with a price tag, we recently published our Time to Cost-Effectiveness analysis in which we compared warfarin to NOACs (as a group) to Watchman LAAC closure (Reddy VY, Akehurst RL, Armstrong SO et al, JACC 24:2728-2739, 2015). Unlike most cost effectiveness analyses on stroke prevention therapies which only look at a lifetime horizon (typically defined as 20 years or 30 years), we performed the analysis in a cumulative year-by-year basis; we also took into account the level of disability associated with stroke (after all, the cost for disabling stroke is worse than a non-disabling one). Using the most recent 2015 cost data (including the new DRGs 273/274), this analysis demonstrated that relative to warfarin, LAAC was cost-effective at 7 years ($42,994/QALY), and NOACs were cost-effective at 16 years ($48,446/QALY). Indeed, LAAC was dominant over NOACs by year 5 and warfarin by year 10 – that is, not only cost-effective, but actually cost-saving. Furthermore, in sensitivity analyses, LAAC remained cost-effective relative to warfarin ($41,470/QALY at 11 years) and NOACs ($21,964/QALY at 10 years), even if procedure costs were doubled. Again, I do not imply from this evidence that LAAC should be considered instead of warfarin or NOACs … just simply supportive economic evidence that LAAC should be considered a reasonable alternative if long-term treatment with oral anticoagulation is not feasible.

Based on this evidence, I think it quite reasonable to view Watchman implantation as a reasonable therapeutic option for patients who are poor candidates for long-term oral anticoagulation. Now, a few specific comments regarding some of the individual Conditions.

Condition 2: While I certainly agree that it is necessary to point out that Watchman LAAC should not be used as a broad replacement for oral anticoagulation, I find the language as currently stated somewhat problematic. First, I do agree that a high stroke risk score (CHADS or CHADS-VASc) is necessary to consider Watchman implantation. After all, it only makes sense (and in fact is necessary) to consider Watchman implantation for patients who are atsignificant risk for stroke. And there is clear guidance from the recent ACC/AHA AF guidelines as to who is considered high enough risk to warrant treatment with a stroke prevention therapy (J Am Coll Cardiol 64:2305-7, 2014). On the other hand, the inclusion of the HAS-BLED score is extremely problematic. First of all, the c-statistic (the metric used to measure the ability of a score to predict the intended clinical event, wherein a score of 0.5 is equivalent to complete chance, and a score of 1.0 is perfect) for HAS-BLED in predicting bleeding is estimated in various studies to be ~0.65 with a lower 95% confidence interval of 0.60. And in fact, because of the significant overlap in the risk score criteria, the c-statistic for CHADS-VASc in predicting bleeding (not stroke) is 0.56 with an upper 95% confidence interval of 0.59 (see Table 2, Clin Cardiol 38:555-561, 2015). It is because of the minimal additional predictive power of this score to identify patients prone to bleeding that the HAS-BLED score was not incorporated into the recent 2014 ACC/AHA AF Guidelines. Furthermore, the HAS-BLED score was never used in any of the clinical trials (either the NOACs trials or the Watchman LAAC trials). Thus, I would strongly urge that the HAS-BLED criterion be either completely removed (preferable), or instead be stated as either high CHADS-VASc or high HAS-BLED (not both). Finally, regarding the requirement “A contraindication to warfarin”, I assume that CMS wasn’t trying to encourage off-label use of the Watchman device (since patients in the Watchman clinical trials did receive 6 weeks of oral anticoagulation). But it would be useful if this statement was clarified to include relative contraindications. As CMS undoubtedly knows, the manufacturer of the Watchman device (Boston Scientific Inc.) is currently negotiating the design of ASAP-TOO with the FDA; this is a multicenter randomized controlled trial of truly contraindicated patients (who cannot take even the 6 weeks of oral anticoagulation) in which patients will be randomized to either Watchman implantation or antiplatelet agents.

Condition 6: I agree that it is entirely appropriate, indeed advantageous, to mandate the inclusion of all patients into a nationwide Registry. However, the idea of mandating a “control” arm of patients treated with oral anticoagulant therapy is both impractical and non-contributory. After all, how would the follow-up data be reliably gathered without an extremely costly clinical trial infrastructure? It would make much more sense to perform a propensity-matched analysis with one of the exiting databases (eg, patients treated with warfarin or NOACs in the various clinical trials or other ongoing prospective registries) in which we would have more confidence in the reliability of the data on patients taking oral anticoagulation.

From a clinical perspective, from a public health perspective, and from an individual patient perspective, there is a pressing clinical need for a therapeutic option for patients at risk for stroke but are poor candidates for oral anticoagulation. LAAC is one such FDA-approved option that our patients need as an option.

Thank you very much in advance for your consideration.

Sincerely,
Vivek Reddy, MD
Helmsley Trust Professor of Medicine
Icahn School of Medicine at Mount Sinai
New York, New York

Less

1] There needs to be clarification about how we are going to use CHADS2vasc and HAS BLED. Also the condition that warfarin is contraindicated does not make sense, since patients have to be on warfarin for at least 45 days post Watchman. A better way of stating the warfarin condition is that the patient is at high risk of bleeding based upon previous history (i.e. history of GI, GU, intra-articular, intra-ocular or cutaneous bleeding), risk of fall/trauma, or high HASBLED score. Indication for Watchman might be stated as follows: patients with a history of non-valvular atrial fibrillation requiring anticoagulation based upon clinical history of atrial fibrillation and CHADS2VASC score and are at high risk of bleeding based upon clinical history and/or HASBLED score. Since the PROTECT study included patients with CHADS2 of 1 and above, I think we should include those patients.

With regard to #6. "The registry must include contemporaneous patients followed on oral anticoagulant (OAC) therapy to serve as non-interventional controls." I have concerns regarding how these patients are going to be chosen. There will obviously be bias.

Less

Less

December 9, 2015

Ms. Tamara Syrek-Jensen, J.D.
Director, Coverage & Analysis Group, OCSQ
Centers for Medicare and Medicaid Services
7500 Security Boulevard, C1-14-15
Baltimore, Maryland 21244

Dear Ms. Syrek-Jensen:

The Association of Black Cardiologists (ABC) would like to thank the Centers for Medicare and Medicaid Services {CMS) for accepting our comment on the very important coverage determination focused on left atrial appendage closure for stroke prevention in the setting of atrial fibrillation. Founded in 1974, the ABC is a nonprofit organization with an international membership of 1,500 health professionals, lay members of the community (Community Health Advocates), corporate members, and institutional members. The ABC is dedicated to eliminating the disparities related to cardiovascular disease in all people of color. Today, the ABC's public and private partnerships continue to increase our impact in communities across the nation. We believe that good health is the cornerstone of progress. We are firm in our resolve to make exemplary health care accessible and affordable to all, dedicated to lowering the high rate of cardiovascular disease, including stroke, in minority populations and committed to advocacy and diversity. The ABC achieves its mission through community awareness, education and screening; continued professional education; scientific research and patient advocacy.

The ABC is in favor of technology to reduce the burden of stroke on the American public. African Americans suffer disproportionately from nearly all forms of cardiovascular disease, including stroke, and are nearly twice as likely as whites to die from stroke. People of African descent in America carry nearly double the toll of premature disability, morbidity and mortality due to stroke.ⁱ Based on Medicare beneficiary data from the National Minority Quality Forum, the incidence of atrial fibrillation in the African American community is estimated to be approximately 300,000. However, the incidence of stroke in African American patients with atrial fibrillation is almost double that seen in the general population (65 yrs age and older: 18.9% AA vs. 10.9% whites). In the 50-64 year old age group the incidence of African Americans with atrial fibrillation associated stroke is more than double that of whites. In fact, the Southeastern region of the United States, where the proportion of African Americans residing in the US is greatest, is euphemistically labeled the ‘stroke belt’. The high incidence of stroke among African Americans and Hispanics is due to ischemic strokes of all causes.ⁱⁱ The risk of stroke in the setting of atrial fibrillation increases almost 5-fold throughout all ages.ⁱⁱⁱ Prophylactic antithrombotic therapy dramatically reduces the incidence of stroke in patients with atrial fibrillation and has been incorporated into the current clinical guidelines for atrial fibrillation management.^iv,v Unfortunately, disparities in treatment of atrial fibrillation persist, leaving many patients at risk of stroke due to undertreatment, treatment noncompliance or treatment intolerance.^vi

Although the benefit of anticoagulation to prevent stroke in African Americans has been documented, there are still many unknowns remaining regarding the natural history of atrial fibrillation in African Americans. The prevalence of atrial fibrillation in African American and other minority communities continues to be debated.^vii African Americans are more likely to have intracranial bleeds with AF. Intracranial bleeds are four times more fatal than ischemic strokes.^viii When the higher intracranial hemorrhage rates in hypertensive patients are combined with the higher prevalence of hypertension in the Black community, the magnitude of benefit of longterm anticoagulation in African Americans with atrial fibrillation has not been established. Despite the development of novel oral anticoagulant agents, the safety and efficacy endpoints in the pivotal trials of these drugs was underpowered in the African American cohort.^ix

The Association of Black Cardiologists feels the establishment of mandatory registry with every attempt made to accurately reflect the US demographics and inclusion of minority populations would provide much needed data to better understand the impact of this novel technology on the American population. Atrial fibrillation is seen in approximately half the number of African Americans as in whites. African Americans with atrial fibrillation are also more likely to have bleeding complications (particularly intracranial hemorrhages) while taking oral anticoagulants despite having therapeutic INR’s. Thus the stakes are higher in African Americans patients with atrial fibrillation on oral anticoagulation and left atrial appendage closure could prove to be a better therapeutic approach. A post-market registry with a requirement that at least 10 percent of registry submissions be comprised of data from minority populations is a reasonable approach to this issue.

In summary, the Association of Black Cardiologists supports technologies, such as left atrial appendage occlusion, to reduce the impact of stroke on the American public. More evidence is needed to determine the role of this novel therapy in the management of atrial fibrillation in the African American and other minority communities. Appropriate reimbursement may improve access to this novel therapy and hence reduce treatment disparities observed in the management of chronic atrial fibrillation. The establishment of a post-market registry with a mandate to enroll a sufficient number of minority patients is strongly recommended as well.

Thank you for your consideration of these comments.

Sincerely,

Icilma V. Fergus, MD
President
Association of Black Cardiologists

Less

December 10, 2015

Tamara Syrek Jensen, JD
Director, Coverage & Analysis Group
Office of Clinical Standards and Quality
Centers for Medicare & Medicaid Services
Mailstop C1-09-06
7500 Security Blvd
Baltimore MD 21244

Dear Ms. Syrek Jensen:

On behalf of the Medical Device Manufacturers Association (MDMA), a national trade association representing the innovative sector of the medical device market, I am filing the following comments on the Proposed Medicare Coverage Decision Memorandum for Percutaneous Left Atrial Appendage (LAA) Closure Therapy (CAG-00445N). MDMA represents hundreds of medical device companies, and our mission is to ensure that patients have access to the latest advancements in medical technology, most of which are developed by small, research-driven medical device companies.

MDMA is providing comments and concerns regarding two of the proposed conditions of coverage for left atrial appendage closure (LAAC) therapy: Condition 2 and Condition 6. We will also respond to CMS’ request for comments on the use of coverage with evidence development (CED) in this decision.

MDMA recommends that CMS ensure appropriate patient access by aligning coverage for LAAC and other medical therapies with the available clinical evidence. We therefore strongly encourage CMS to eliminate the requirements for a high HAS-BLED score and a contraindication to warfarin, neither of which were requirements within the IDE clinical trials for the WATCHMAN Device. MDMA recommends that CMS instead align coverage with Food and Drug Administration (FDA) approved labeling for the device: for patients with non-valvular atrial fibrillation who, “Have an appropriate rationale to seek a non-pharmacologic alternative to warfarin, taking into account the safety and effectiveness of the device compared to warfarin.”

MDMA strongly recommends that CMS eliminate the requirement for the registry to include “contemporaneous patients followed on oral anticoagulant (OAC) therapy to serve as non-interventional controls.” This requirement directly contradicts CMS guidance for CED under Section 1862(a)(1)(E) which state that, “CED will lead to the production of evidence complementary to existing medical evidence.”

We presume that CMS would like to characterize the patient population receiving LAAC therapy, and evaluate performance as compared to other treatment options. However there are numerous logistical and methodological challenges associated with the proposed registry and, as proposed, the registry would not support a scientifically valid comparison.

First, there is no need to collect information on patients on OAC therapy. Extensive information is already available regarding the population of patients managed on OACs, making the collection of this additional information unnecessary. Moreover, it is difficult to include these patients in the registry in a way that would produce appropriate data for analysis. The OAC population is extremely large and dispersed, managed by family practitioners and general cardiologists, as opposed to LAAC device implanters. These patients and their physicians have no incentive to participate in a registry. Any prospective enrollment would be non-sequential and capture a subset of the population, introducing substantial selection bias that would make data comparisons problematic.

Second, and even more important, patients being well-managed on OACs are not a relevant comparator, as they are not appropriate or eligible for LAAC therapy. As a result, the proposed comparison will not address the clinical questions of interest to CMS. Because of the methodological and logistical challenges, the resulting analysis would not be scientifically sound and would lead to inaccurate conclusions that could negatively impact LAAC coverage.

MDMA generally supports CMS’s use of CED to provide beneficiary access to new medical technologies as additional information is collected. However the particular use of CED as proposed in this situation raises several significant concerns, with implications not only for the manufacturers of LAAC devices, but for the entire medical device industry.

Bringing a new therapy to market requires require significant investment of time and resources, and is undertaken in the context of ongoing and careful assessments regarding the likelihood of success. When market access requirements change unexpectedly, as is the case in this situation with the introduction of an OAC registry arm, it creates uncertainty in the industry. This uncertainty lessens the willingness of investors to take on the significant risks associated with funding novel medical technologies. For the small to mid-size device manufacturers represented by MDMA, this challenge can be insurmountable.

In the proposed NCD for LAAC, the requirement that LAAC therapy be prospectively evaluated against available pharmaceutical therapies represents an overwhelming and potentially prohibitive burden to the manufacturer. It is inappropriate for CMS to require the device manufacturer to support data procurement on pharmaceutical therapies that are already approved and covered, one of which has been in clinical use for over 50 years.

While we fully support CMS’ pursuit of real-world information regarding LAAC therapy, MDMA also urges consideration for the associated real-world execution challenges. In the case of the NCD for LAAC, failure to eliminate the OAC registry arm could have the unintended consequences of preventing this novel therapy option from becoming available in the US market, while also stifling future investment in other potentially life-saving medical technologies.

MDMA appreciates the opportunity to comment on the proposed coverage decision and looks forward to working with CMS on future coverage analyses.

Sincerely,
Mark B. Leahey
President & CEO, MDMA

Less

Dear Ms. Tamara Syrek-Jensen,J.D.

On behalf of the physicians and patients I care for at Methodist Hospital in San Antonio,Texas and South Texas, I would like to take this opportunity to comment on CMS's Proposed Coverage Decision for Percutaneous Left Atrial Appendage Closure using the Watchman device. I have been personally involved with this device for now for 1 year, having implanted my first Watchman in early 2015 once comerciallization was approved. I have implanted several devices with excellent outcomes.

I am in agreement with all 7 proposed conditions. That being said, I believe greater clarification of the "eligible patient population" is in order. I would like to share with you the perspective I have formed over my recent experience as a new operator with this device. There are several scenarios where I believe this device should be used. Broadly, all of these fall under what I would term "A CONTRAINDICATON TO LONG TERM THERAPY WITH ORAL ANTICOAGULATION".

The most important is the population of patients that have already had a significant adverse bleeding complication on an oral anticoagulant. These patients typically STOP the therapy leaving them UNPROTECTED. Understanding the catastrophic events that occur when let unprotected and having cared for these patient, understanding the strain families occur, and the long term effects on the medical cost wise on the health care system are like the most important reasons for approving this device. Data supports its effect and benefits for these patients.

Patients that have not yet had a documented adverse bleeding event, but at are extrermely high risk thereof. The HAS-BLED score is a reasonable tool to assess this, however there must also be an allowence for important clinical and social factors that increase bleed risk not embedded in the HAS-BLED score. Although these risk models a CHADS VASC score also allows for determination of the current risk. These issues also along with patient frailty play important considerations into long term care and potential poor outcomes.

Lastly COMPLIANCE issues are something I also believe should be taken into account to determine eligibility. If, after robust discussion with patients and families, it is determined a patient simply cannot be compliant with OAC therapy, or, if there are social issues that are of concern, these patients should be covered for LAA closure to reduce that risk of stroke. Occassionally lifestyle factors come into play here as well, and could be considered.

I appreciate the time to express my concerns and hopefully express my strong support for approval for this new technology.

Jorge Alvarez M.D.

Less

I am a cardiac electrophysiologist with experience in treating nonvalvular AF patients who are at risk for stroke. I also have performed numerous percutaneous left atrial appendage closure procedures in such patients.

As a physician that practices evidence based medicine, I cannot understand the rationale for basing coverage on patients being contraindicated to warfarin therapy. The Watchman trials (PROTECT AF, PREVAIL, etc.) have all included patients who are eligible for warfarin therapy and to my knowledge this decision would not be based on the current body of evidence.

Less

To Whom It May Concern:

Leadership within the Baptist Health South Florida health-system have been informed of the Centers for Medicare and Medicaid Services (CMS) decision to decline coverage of percutaneous left atrial appendage closure (LAAC) therapy—including the Watchman device. As a healthcare entity that is expanding our clinical services and endeavoring to continue a reputation of cutting-edge treatment options for patients in South Florida and beyond, Baptist Health has recently harmonized our cardiac and vascular service line under our division The Miami Cardiac and Vascular Institute. This branch of our health-system brings together a collection of skilled clinicians that support innovations in cardiovascular treatment and have the knowledge and facilities to offer patients a full suite of treatment options—among those the options of surgical intervention and device placement. The ability to provide LAAC therapy as an option, particularly the Watchman device, is considered paramount to our clinicians. During this open comment period, our facility would like to provide support in favor of CMS reconsidering the position on coverage of LAAC therapy in the Medicare population—one of the groups of patients that often benefit most from interventions such as this.

The Centers for Disease Control estimate that between 2.7 and 6.1 million people in the US have atrial fibrillation (1). The condition is estimated to hospitalize 750,000 people per year and results in approximately 130,000 deaths annually (1). It is projected that atrial fibrillation (hereafter A.Fib) costs the US healthcare system around $6 billion per year, and that A. Fib patients generally cost almost $9,000 more to manage from a healthcare standpoint annually than non-A.Fib patients (1). The majority of the long-term costs with these patients is in the treatment—the main management method for A.Fib has traditionally been pharmacotherapeutic. Patients with A.Fib generally require three classes of pharmacotherapy to decrease morbidity and mortality—medications that provide rate and rhythm control and blood-thinning therapy. The blood-thinning component to the patient’s regimen is particularly crucial, as around 5% of non-anticoagulated patients have an ischemic stroke and overall accounting for 15—20% of ischemic strokes that occur in the US (1,2). While the costs of the medications typically prescribed to achieve rate and rhythm control are generally not expensive, the blood-thinning effect—typically achieved by anticoagulants—can result in expense that make them costly for health-systems as well as patients. The risk of A.Fib increases with age (1), and with the large majority of elderly Americans using Medicare benefits for their medical needs, the fixed-income distinction of this demographic must be taken into account. Anticoagulation for U.S. A.Fib patients is typically achieved by use of warfarin or novel anticoagulants (hereafter NoAcs). Each class of medication brings different levels of indirect and direct expenses that must be considered when evaluating the full-body of interventions available to patients, as indicated in the table below:

Drug Class Avg. Prescription Cost (5) Laboratory Monitoring and Costs Additional Extended Costs Medicare Status Reversal Process and Avg Cost
Vitamin K Antagonist
Warfarin (Coumadin®/Jantoven®) $13/month x life of patient $6.19 $145.70 per INR testing (3) Laboratory technician:
$53,000/year (4)

Clinical pharmacist (if pharmacist-led clinic):
$103,000/year (4) Generic medication: consistent copayment

My name is Jacob Koruth and I practice electrophysiology at Mount Sinai Hospital in New York City. I am writing to support approval of the National Coverage Determination for the WATCHMAN left atrial appendage device. I am happy to see the proposal by CMS to provide coverage for the WATCHMAN procedure. This device represents a first of its kind alternative treatment for patients with non-valvular atrial fibrillation who are at high risk of thromboembolic events, but who are not good candidates for long term warfarin. At the present time, there is no national coverage and so local coverage is inconsistent across the country. I believe that a national coverage determination to cover this device according to the instructions for use approval would be very important to improve the outcome of patients and to provide uniform coverage across the country

Data from the PROTECT-AF and PREVAIL studies have clearly demonstrated the safety and efficacy of the WATCHMAN device.

My comments are surrounding several conditions in the proposal.

Condition 2 states that a patient will need to have a “contraindication to warfarin.” If this language is meant to suggest that a patient needs to have an absolute contraindication to warfarin, this is concerning. A “contraindication to warfarin” conflicts with the device IFU as well as the PROTECT-AF and PREVAIL studies, which require 45 days of warfarin post implant. This language would potentially eliminate a large pool of patients who will benefit from WATCHMAN therapy – including patients with a high fall risk, patients with difficult to control INRs and those that cannot have their INRs regularly monitored, or patients. These are some of the best candidates for WATCHMAN therapy and should not be excluded from coverage.

Thank you for the opportunity to submit these comments.

Less

To CMS

I am a practicing Electrophysiologist in San Antonio, TX. As a full time Electrophysiologist, a large percentage of my practice is devoted to taking care of patients with atrial fibrillation. One of the most catastrophic complications of atrial fibrillation is CVA. Patients with CHADS-VASC score >= 2 are deemed to be high risk for CVA and data suggests the anticoagulation is warranted. Many of these patients are taking Coumadin or one of the new NOAC agents (Pradaxa, Eliquis, Xarelto, or Savaysa). While a large percentage of patients tolerate the anticoagulants without complications, there is a subset of patients who either do poorly (due to bleeding issues, risk of falls, and non-compliance) or are contraindicated in taking anticoagulants. These patients have a high risk of CVA. The PREVAIL and PROTECT AF trials have shown that the Watchman left atrial occluder device has been shown to be non-inferior to warfarin for prevention of CVA and systemic embolism. I think the Watchman device adds a valuable tool in our arsenal to prevent CVA in patients with atrial fibrillation. I believe that if the device is used in the appropriate patient (CHADS-VASC score >= 2), it will be cost effective as well as life saving. I am somewhat puzzled by CMS stating that a person who is contraindicated to warfarin be eligible for the device. The FDA has mandated that the patient be placed on warfarin for 45 days post implant (which has been my practice). Lastly, HASBLED scores have not been a large part of clinical practice in deciding whether to anticoagulate patients with atrial fibrillation.

Less

As an electrophysiologist who treats thousands of patients with atrial fibrillation per year, the FDA approval of LAAC was extremely important and impactful for our practice as the only academic medical center in the state of SC. Patients clearly need an alternative to OAC, and LAAC is the best studied alternative, providing for better safety and outcomes compared to Warfarin therapy.

The CMC coverage decision is equally important, as it will dictate if many of our patients will have coverage for this procedure. The current wording of the decision is a bit unclear. A "contraindication to Warfarin therapy" needs further clarification, because many patients have a relative contraindication to long-term warfarin therapy and that is why they are referred for the device, but they require short term warfarin therapy around the time of implant per clinical trial data and FDA indications. Therefore, an absolute contraindication mandate would be impossible to comply with an preclude any Medicare patients from receiving this therapy. Additionally, the criteria for "high risk" CHADS2VASC and HASSLED scores should also be clarified. Any CHADS2VASC score of 2 or greater should be considered high risk as they are indicated for OAC. Also, any relative contraindication to long-term OAC should also supersede the patient's HASBLED score as they cannot tolerate long-term OAC safely. IF the patient does not have a relative contraindication to long-term Warfarin, then a HASBLED score of 3 or greater would be a reasonable criteria since that would make the patient relatively high risk for bleeding with OAC and LAAC could be considered. Thank you for your time and consideration.

Less

Less

Less

I appreciate the opportunity to comment on the proposed CMS coverage decision on the Watchman device. The safety and efficacy of this technology has been established and summarized in previous clinical trial data and has been outlined in previous comments.

As our population grows older, the numbers of frail, elderly patients with poor gait stability and atrial fibrillation continues to grow. Anticoagulation in this population is clearly a concern with regards to bleeding and specifically intracranial hemorrhage. Along the same lines, this patient population is at the highest risk of stroke from their atrial arrhythmia. This device offers an attractive option in these patients.
I would urge that CMS please take this into consideration. In addition, as a physician, at times I feel we have newer technologies available that would benefit our patients, but are unable to provide this technology due to lack of adequate coverage by CMS /payors. A collaborative working group between the FDA and CMS may be beneficial to design approval and then CMS coverage in a timely and appropriate manner.

Thanks for this opportunity.

Less

December 10, 2015

Ms. Tamara Syrek-Jensen
Director, Coverage & Analysis Group
Centers for Medicare & Medicaid Services
7500 Security Boulevard
Baltimore, MD 21244

RE: Proposed National Coverage Decision (NCD) Memorandum for Percutaneous Left Atrial Appendage Closure (LAAC) Therapy (CAG-00445N)

Dear Ms. Syrek-Jensen:

The American College of Cardiology (ACC), Heart Rhythm Society (HRS), and Society for Cardiovascular Angiography and Interventions (SCAI) are the non-profit professional associations representing the majority of practicing interventional cardiologists and electrophysiologists in the United States. These members have expertise and experience in the treatments and therapies for stroke prevention. ACC, HRS and SCAI appreciate the opportunity to submit joint comments on the proposed National Coverage Decision for percutaneous left atrial appendage closure (LAAC). This document represents the consensus of the three societies.

CMS’s proposal to cover LAAC for non-valvular atrial fibrillation (AF) patients with elevated risk of stroke performed by experienced operators at high quality facilities with data reported to a registry after a shared decision-making discussion is a significant and positive step for these patients. The societies support this framework for coverage. However, the societies also recommend revisions to certain details and requirements within that framework that will further ensure that the correct patients are receiving this therapy and that unnecessary administrative obstacles are not created.

• Specifically, the societies provide recommendation on appropriate contraindications to anticoagulation that would support coverage as described in Appendix A.
• With respect to institutional and operator requirements we recommend CMS rely on the peer- reviewed document published online today in the Journal of the American College of Cardiology, HeartRhythm, and Catheterization and Cardiovascular Interventions. The societies recommend CMS utilize the standards included in that document in place of the previously proposed requirements. These important changes are reflected in the attached redline version of the NCD.
• The societies question the scientific basis of the proposed requirement for data collection on contemporaneous patients managed with oral anticoagulant (OAC) therapy to serve as non- interventional controls and recommend it be removed.
• Currently, an evidence-based tool is not available for the anticipated patient population who may be treated with LAAC. As an alternative, the societies suggest an algorithm (Appendix B) as a decision-making tool to assist in determining the most appropriate treatment pathways for patients.

Comments pertaining to each of the proposed seven conditions for coverage are presented below. In addition, a redline version of suggested changes to the language proposed by CMS in the draft coverage policy is attached.

Available evidence aligns with the proposed condition that the device used to close the LAA should be approved by the Food and Drug Administration (FDA) for non-valvular AF. In comparison to non-valvular AF, insufficient evidence exists to support coverage of patients with valvular AF. The higher risk of left atrial thrombus outside the appendage in patients with rheumatic valvular disease suggests that patient population is not currently suitable for this therapy.

FDA approval for the Watchman, the only device currently with specific indications for this use, is for patients who have a high CHADS2 or CHA2DS2-VASc score. CMS also proposes to use a high CHADS2 or CHA2DS2-VASc score as a condition of coverage. The societies agree that a high score on this metric should be required for coverage, but suggest that it be further defined. Specifically, the policy should state a threshold of a CHADS2 score greater than or equal to 2 or a CHA2DS2-VASc score greater than or equal to 3.

The rationale for a patient to seek a non-pharmacologic alternative to warfarin will likely include the conditions captured in a HAS-BLED score. As such, the societies again suggest a numerical score be included. Specifically the policy should explicitly state that high bleeding risk be defined as a HAS-BLED score of greater than or equal to 3. While these thresholds for high stroke risk and high bleeding risk are to some degree arbitrary, sufficient expert opinion exists to support their use.

CMS also proposes to provide coverage to patients who have a contraindication to long-term warfarin therapy. The societies recommend CMS use the list of contraindications included in Appendix A as a guide to specific patient populations who could potentially fall into this category.

The draft policy, as written, requires patients to have a high CHADS2 or CHA2DS2-VASc score, high HAS- BLED score, AND contraindication to warfarin. Mandating that all three of these requirements be met would prevent patients who may not reach the HAS-BLED threshold but have long-term contraindications to anticoagulation from qualifying for coverage. Such patients would benefit from LAAC. More consistent with the FDA indications, the societies strongly recommend qualifying patients who have a high CHADS2 or CHA2DS2-VASc score AND EITHER a high HAS-BLED score OR a contraindication to warfarin. The redline attachment presents this solution.

As with other recent emerging technologies, the societies have worked for the appropriate qualifications for both, operators and institutions to optimize effectiveness and patient safety. That peer-reviewed document was published online today in the Journal of the American College of Cardiology, HeartRhythm, and catheterization and Cardiovascular Interventions. It is also attached with these comments. The societies strongly recommend CMS utilize the standards for both operators and institutions included in that document in place of the previously proposed requirements. These important changes regarding the appropriate criteria required to qualify operators and institutions are reflected in the attached redline version of the NCD.

CMS proposes that operators must undergo training “by the device manufacturer on the safe and effective use of the device prior to performing implant procedures” by a physician who already has experience implanting the device. In the draft policy CMS proposes that a new operator be supervised for a minimum of two cases and observed for a minimum of 2 cases. The proposal also includes a requirement that the training physician not be a representative of the manufacturer.

While the societies agree that proctors should be free of undue influence, proctors need to be compensated for their time and travel. The Societies recommend that the NCD conform to prior training requirements for procedures such as Transcatheter Aortic Valve Replacement (TAVR). Specifically, the Societies recommend that:

• Physicianproctors be selected on the basis of their experience, knowledge of the procedure and teaching skills;
• The Societies wholly endorse the concept that proctors be free of undue influence by manufacturers.; and,
• Quality proctors need to be compensated and we believe it would be counterproductive to have proctors who have no remuneration for their time and expertise and no relationship with the device manufacturer.

Revisions are proposed in the attached redline that are more consistent with previous NCDs such as those formulated for TAVR

As with other recent coverage with evidence development (CED) NCDs for cardiovascular therapies such as Transcatheter Aortic Valve Replacement, CMS proposes that physician teams must participate in a prospective national registry that consecutively enrolls all LAAC patients and tracks certain outcomes for at least five years.

The societies expect all operators and institutions to participate in an LAAC registry. The LAA Occlusion Dataset as part of the National Cardiovascular Data Registry (NCDR) LAA Occlusion Registry has already been developed. Previous meetings between FDA, CMS, the societies, as well as NCDR suggest that some mechanism of adjudication will be provided by the registry sponsors for strokes through the first two years after implantation. The existing design of the LAA Occlusion Dataset will allow tracking of all seven parameters specified in the Proposed Decision Memo, with the caveat that adjudication of strokes as currently proposed by the sponsor and the FDA is limited to two years after implantation, although site reporting will continue beyond that window. CMS’s proposal to review the qualifications of the registry is consistent with other recent CED policies.

In addition to collection of data on complications and outcomes of LAAC, CMS proposes participating registries must also “include contemporaneous patients followed on oral anticoagulant (OAC) therapy to serve as non-interventional controls.” For this population, the term “control” is inaccurate. The “non-interventional controls” would in fact represent a widely mismatched group that should not be comparable to patients who receive the device. The societies strongly object to this requirement and recommend its deletion.

• Such a requirement violates the standards of scientific integrity and relevance to the Medicare population referenced in the proposed NCD. While the desire to obtain information on a control group undergoing oral anticoagulation is understandable, that has been the subject of two well conducted trials that randomized patients, using a model and scientific rigor that could not remotely be replicated with the proposed control arm.
• Moreover, such a control arm would require that every institution performing LAAC set up an elaborate screening process, obtain investigational review board approval, and seek informed consent from patients and follow up from patients, most of whom may be seen entirely in an outpatient setting by a myriad of clinicians few of whom will have any involvement with LAAC.
• The “non-interventional controls” would in fact be largely mismatched with the patients who receive the device. The proposed control arm would constitute a research study with skewed selection and will be of limited if any benefit from a research or public health standpoint.

In sum, observational comparisons of patients who are treated with procedures versus those who are treated without interventions are highly prone to bias and other confounding factors which would severely limit any useful clinical information which could be obtained from such an undertaking in addition to all the other associated problems as described above.

Coverage of LAAC would also be contingent upon documentation of a formal shared decision-making interaction between the patient and physician using an “evidence-based decision tool” on anticoagulation. The societies support the role of shared decision-making but caution that the level of evidence base required for an “evidence-based decision tool” as used in other settings is not available for this patient population which is being proposed in the draft NCD proposal. As an alternative, the societies propose the use of a decision algorithm as outlined in appendix B as a guide to appropriate treatment pathways that is consistent with current clinical practice.

In benchmarking this draft NCD with similar CED LCDs for transcatheter cardiovascular therapies, it was noted that a section explicitly covering LAAC in clinical trials is not included in the LAAC draft NCD. It is not clear from the draft whether CMS intends to cover LAAC in clinical trials since no definitive statement is included other than that, “All CMS-approved clinical studies and registries must adhere to” a list of standards relevant for scientific integrity and relevance to the Medicare population. NCD 310.1 governing coverage of costs in clinical trials should apply, but the absence of clinical trial coverage is concerning. To remove uncertainty and confusion regarding CMS’s intent, the societies recommend the NCD be revised to expressly state that, “LAAC is covered within a clinical trial consistent with NCD 310.1 which provides coverage for the routine costs of qualifying clinical trials.” This revision is also presented in the redline attachment.

The societies look forward to working with you through the rest of the coverage process. If you have questions or need additional information regarding any of these comments, please contact James Vavricek at jvavricek@acc.org, Kim Moore at kmoore@hrsonline.org, or Dawn Gray at dhopkins@scai.org.

Sincerely,

/s/
James C. Blankenship, MD, MSc, FSCAI
SCAI President

/s/
John D. Day, MD, FHRS
HRS President

/s/
Kim Allan Williams, Sr., MD, FACC, FAHA, FASNC
ACC President

Enclosures

1. History of intracranial bleeding (intracerebral or subdural) where benefits of LAAC outweigh risks
2. History of spontaneous bleeding other than intracranial (e.g. retroperitoneal bleeding)
3. Documented poor compliance with anticoagulant therapy
4. Inability or significant difficulty with maintaining patients in therapeutic anticoagulation range
5. Intolerance of warfarin and NOACs
6. High risk of recurrent falls
7. Cognitive impairment
8. Severe renal failure
9. Occupation related high bleeding risk
10. Need for prolonged dual antiplatelet therapy
11. Increased bleeding risk not reflected by the HAS-BLED score (e.g. thrombocytopenia, cancer, or risk of tumor associated bleeding in case of systemic anticoagulation)
12. Other situations for which anticoagulation is inappropriate

Less

Dear Mrs. Syrek-Jensen, JD,

I am a practicing interventional cardiologist in Charleston, SC. I have been involved in left atrial appendage closure by the Lariat device, Amplatzer Cardiac Plug and the Watchman device. I am writing to support the NCD for the Watchman device.

Protect AF and Prevail along with CAP-1 and CAP-2 have demonstrated the efficacy of the Watchman for preventing stroke and systemic embolism with a reduction in intracranial hemorrhage in patients with nonvalvular atrial fibrillation. The 5-year follow up of the Protect AF trial showed an all cause mortality benefit. The safety of the device was also demonstrated with clear decrease in procedure related complications with device experience/procedure refinement. The ability of the Watchman to be safely implanted by new users was demonstrated by the low complication rates of new operators in the Prevail trial. These data support the Watchman as an alternative to Coumadin for patients and that it can be safely be offered across the country.

We must keep in mind that compliance with anticoagulation is an issue with high proportion of anticoagulation indicated patients not taking anticoagulation. Approximately 20% of stroke patients discharged on Coumadin are no longer taking anticoagulation at 3 months. With time this percentage increases. A Watchman device offers a solution that is long term.

Protect AF and Prevail targeted patients with increased stroke risk and normal bleeding risk. In my practice I see a significant proportion of a fib patients that are at both increased stroke risk and increased bleeding risk. Although the studies did not particularly target this population, it seems appropriate that these patients would benefit from LAAC and avoidance of long-term anticoagulation.

There is no suitable alternative to the Watchman for LAAC. Surgical ligation is more invasive and has been demonstrated to be unreliable for LAA closure. There also is no prospective outcomes data for a surgical approach.

With regards to the seven conditions, it is unclear how condition #2 will be implemented. Elevated CHADS2 and CHA2DS2-VASc scores seem straight forward based on accepted cutoffs for anticoagulation. But should a high HAS-BLED score be one of the requirements for a Watchman procedure? The HAS-BLED score is devised from retrospective data to predict major bleeding defined as intracranial hemorrhage, hemoglobin drop of 2 g/L, hospitalization and or transfusion. This score has not been prospectively validated. Although clinically helpful concepts are gleaned from this score it is not all encompassing. There are numerous other clinical scenarios that clearly place a patient at increased bleeding risk such as prior intracranial hemorrhage or thrombocytopenia to name a couple. The term “a contraindication to Coumadin” should be clarified or replaced. “A contraindication to long term anticoagulation due comorbidities and patient characteristics” could be a more encompassing statement. Or, “clinical rationale for seeking an alternative to anticoagulation.” Simply stating “a contraindication to anticoagulation” for a group of patients that will need to anticoagulation for at least 45 days has adverse medicolwegal implications. Also, if the above two conditions are retained in some fashion, stating that a high HAS-BLED AND a Coumadin contraindication should probably be changed to HAS-BLED OR Coumadin contraindication.

I support a NCD for the Watchman device. A large proportion of my patients can benefit from this device.

Thanks,
Matthew O'Steen

Less

LAA closure has become an essential option for many patients in my practice at Vanderbilt. I have performed 31 Lariat LAA ligations, 15 Watchman LAA occluder implants, and referred 67 TT Atriclip device implants for patients with high stroke risks and high bleeding risks. I do not have an appreciation for what a patient who is "contra-indicated" to anti-coagulation looks like. Many of these patients had major bleeding while on OAC/NOAC, but were not actively bleeding to death and the time I performed these procedures.

In the absence of continuous bleeding, there can always be short term exposure to anti-coagulation. For example with Watchman implant and LARIAT require heparin infusion intra-operatively. Thus I feel the labeling of "contra-indication to warfarin" should be changed.

I personally feel having these cases in a formal registry is a good idea, but mandating this across the country is probably not feasible given the remote regions where this therapy may be delivered, with no resources dedicated to clinical registry research. I do agree that this should be carefully be made available to skilled operators with at least >100 transeptal cases or more, and in centers with collaborative relations between EP and Interventional cardiologists.

Over time Watchman has proven to be a viable alternative to anticoagulation, and the longer we observe, the greater the effect. Reducing bleeding, while delivering site specific stroke prevention therapy to the LAA will become first line therapy for AF patients in the near future.

Less

I do not agree with the CMS proposal regarding coverage of LAAC. I believe the evidence informing the use of LAAC is sufficient, and that LAAC does indeed treat an illness: left atrial stasis coagulopathy in atrial fibrillation, essentially the major cause of stroke in this chronic disease.

That the FDA and CMS reached disparate conclusions regarding the role for LAAC comes as little surprise to me. Warfarin has proven effective in preventing stroke, just not convenient, predictable, or in many cases safe enough. I will submit that warfarin, when used appropriately, followed frequently, in a stable patient with consistent metabolism, no interacting drugs, and with a touch of luck, can be quite effective in preventing stroke. It is terrible at preventing the most feared complication of hemorrhage(sic).

Even the NOAC's needed massive numbers of patients to prove noninferiority and superiority. LAAC has not the benefit of this type of analysis. Registries will tell the full story and we will all wait impatiently for these findings as they emerge over years.

Additionally, CMS has a standard to meet that may diverge from the role of providers and even medical device manufacturers, namely medical cost containment. While this is important, we must ensure that this standard does not deprive patients of what can be definitive, life-long therapies, such as LAAC. In fact, by decreasing risk of hemorrhage, equaling prevention of ischemic stroke, and decreasing medication and lab costs, LAAC could ultimately save healthcare dollars.

Finally, a clinical perspective. These are things I see nearly everyday:
(1) patients nervous about bleeding while on anticoagulants
(2) MD's nervous about the same
(3) frustration with the vagaries of warfarin use amd management

Things I see at least every month that implicate the LAA as THE major source of debilitating thromboemboli:
(1) strokes thought to be result of undiagnosed afib leading to loop recorder implants (if we are convinced enough of this connection to perform ILR procedure, are we so much less convinced of the culpability of the LAA as a source to justify LAAC?)
(2) LAA thrombi (twice this month in patients on warfarin being evaluated for Watchman!)

I have never been more convinced of the involvement of the LAA in afib related stroke as I am now. After seeing this anatomy in detail, I am amazed that warfarin works as well as it does. LAAC makes sense, seems safe when done with experienced hands, and is long-anticipated option that is finally here. This device was not simply actualized by engineers, it was beseeched by doctors. I hope that CMS allows us the latitude to choose LAAC in patients we deem reasonable, not just with pure contraindication to warfarin. Imagine: If you were 55 years with old with afib and facing decades of warfarin exposure, would you not want the option???

Less

Currently, I am Director of the Electrophysiology Laboratory for a community hospital in Northern New Jersey. I spend 100% of my clinical time in the office or electrophysiology laboratory. We have recently started our Watchman program; the program is a joint collaboration between our Electro-physiology and Structural Heart services.

I read with great interest the CMS’s proposed coverage decision for percutaneous left atria appendage (LAA) closure therapy (CAG-00445N). Based on my review, I feel that several proposed conditions for coverage merit further consideration.

Condition # 2:

1. A definition for a high HAS-BLED score should be provided. Furthermore, it should be acknowledged a high HAS-BLED score is sometimes not present in patients who are otherwise excellent candidates for this procedure. Examples include (a) patients who have previously bled on anticoagulation but have no other risk factors for bleeding and (b) patients who engage in high-risk professions or hobbies where continued anticoagulation adversely impacts their desired lifestyle.

2. Patients obviously cannot have an absolute contraindication to warfarin as FDA labeling currently mandates a minimum of 45 days of warfarin following device implantation. I am assuming that the intent is a relative or absolute contraindication to long-term use of warfarin; however, this should be explicitly clarified.

Condition # 6:

1. While the request for a registry is understandable, there are 2 major issues. First, a five-year follow-up in patients who receive the device is difficult given the costs inherent to such an endeavor. This becomes even more problematic if source data from other hospitals is necessary to complete the data elements that comprise the registry. Therefore, I would request that CMS consider mandating a shorter period of follow-up.

2. The second issue is the requirement of “contemporaneous patients followed on oral anticoagulation therapy to serve as non-interventional controls.” This is entirely impractical and of uncertain scientific utility. First, these are not the types of patients followed in the typical practice of an Interventional Cardiologist or Electrophysiologist. Thus, where would we even find these patients? Second, if these patients are being continued on anticoagulation, they are an entirely different cohort than those being referred for left atrial appendage closure. By design, these patients will do considerably better than Watchman patients. Thus, the registry would compare “apples to oranges”.

Less

CAG-00445N. As an implanter of this new and exciting technology in the field of atrial fibrillation, I wish to comment regarding the proposed NCD. There are countless number of patients who have atrial fibrillation in this country who are not anticoagulated, and suffer debilitating and devastating preventable strokes. I have seen first hand the removal of anticoagulation for falls, being elderly, GI bleeding, trauma, urinary hemorrhage, and fear resulting in preventable strokes which have left patients with significant morbidity. This is the first approved technology I have been able to offer my patients as a reasonable alternative to anticoagulation in the right clinical setting. In reading your NCD guidelines, I do not think this should be implanted in Warfarin ineligible patients, because that is outside of the FDA approved guidelines. I think you can word it to say Warfarin intolerant or significant bleeding or intolerance of Warfarin. A CHADS cut off is reasonable, but you must allow for physician discretion here. Suggestions, but not absolutes, as you know each and every patient is different and offers different clinical scenarios. Similarly, a HAS BLED score is unfair clinically. If my patient has a CNS hemorrhage on therapeutic Warfarin, his HAS BLED score is irrelevant because he will not be getting any anticoagulation from most clinicians. Regarding the CED portion, I believe a registry is important and patients should be followed for device related adverse effects and outcomes. However, I do not believe every patient should be forced to be enrolled into a trial because some will refuse and not be able to receive this valuable therapy. In addition, not every implanting institution will be offered these specific trials. Registry data, however, is important to collect, but finding appropriate non-implanted patients as controls is cumbersome for busy practitioners and full of bias. In addition, this is a procedure which requires a level of skill and expertise that should be compensated fairly. Not every Cardiologist can or will be doing these, like ablation or device implants. The level of training, and skill, and ability to perform this delicate procedure should allow for a generous compensation, especially since it has the ability to save a significant amount of money to the country if we can avoid large strokes.

All of the patients I have implanted have done very well to date, and most had no other options. Lets not limit this so much that it becomes unreasonable and difficult to offer to patients Thanks.

Less

To whom it may concern:

I am a physician practicing general cardiology and interventional cardiology with Heart Associates of Northern California, a private practice group in San Jose, California. My main hospital affiliation is with Good Samaritan Hospital in San Jose, CA. I have been in practice for more than 2 years and completed my interventional and structural heart training at Cedars-Sinai Medical Center in Los Angeles, CA. I have significant experience with the Watchman device through my training, as Cedars-Sinai was an important site that participated in some of the pivotal Watchman registries, including the Prevail Continued Access Protocol (CAP) registry.

Atrial fibrillation is a disease affecting millions of Americans and the incidence continues to increase as our patient population continues to live longer and longer. The traditional management of stroke risk reduction with anticoagulation (coumadin or novel oral anticoagulants) is difficult for patients, and often times underutilized due to perceived high bleeding risk in those elderly patients who are actually at the highest risk of stroke. The Watchman device provides an effective alternative to chronic anticoagulation to reduce the risk of stroke. There is no device that has been more heavily studied and scrutinized and the trial and registry data from the Protect AF and Prevail trials, along with the CAP registries show that the device is effective with a relatively low procedural risk that continues to improve as experience with the device increases.

The CMS proposal for Watchman coverage includes seven specific conditions that must be met. I disagree with Condition #2 which requires patients to have a contra-indication for anticoagulation. The FDA indication for device implantation actually requires that patient be eligible for anticoagulation, as the device requires a period of anticoagulation to allow adequate endothelialization of the device (usually about 6 weeks). Therefore, the CMS coverage proposal is in direct conflict with the approved FDA indication.
I have numerous patients with atrial fibrillation who are at high risk for bleeding but are terrified of having a stroke. These patients are forced to make a difficult decision with regards to stroke risk management. I have discussed Watchman with these patients which has provided them with a great sense of hope and security. They are willing to exchange the short-term procedural risk of Watchman implantation for the long-term bleeding risk of chronic anticoagulation. Given the overwhelming data showing noninferiority of the device against anticoagulation with coumadin, I feel it is not right, and even unethical, to withhold this technology from the patients that would derive the greatest benefit.

I ask you reconsider the proposed coverage conditions for the Watchman device.

Sincerely,

Amin Manuchehry, MD, FACC

Less

LAAC with the WATCHMAN device

Relating the above evidence to the proposed National Coverage Determination (NCD) for Left Atrial Appendage Closure (LAAC):

THE FOLLOWING ITEMS SHOULD BE RECONSIDERED for Condition#2:
*CHADS2 score is redundant and not needed; nor is it as effective as HAS-BLED in assessing the potential for bleeding.
*“HIgh” HAS-BLED score has no meaning. In the case of a CHADS2 score of 1, the HAS-BLED score must exceed 2 for the potential harm caused by OAC to offset its beneficial effect. It should contain wording that would encompass a “HAS-BLED =2 score is present”.
*The use of OAC should not be “contraindicated”, but other clinical conditions should be considered in quality of life and “an alternative to long term anticoagulation” is sought.

THE FOLLOWING ITEMS SHOULD BE RECONSIDERED for Condition #6:
*Multiple studies through thousands of patients have identified the role of OAC and its effects and outcomes.
*Candidates for OAC and LAAC are not the same, and would be comparing different groups, no matter the parameters that would be chosen.
*A contemporaneous registry would add nothing to evidence nor patient care.
*Consider creating a HAS-BLED form that needs to be completed at the time of LAAC (WATCHMAN), that could be placed in the WATCHMAN packaging.

Less

Less

We commend the Centers for Medicare and Medicaid Services (CMS) for their detailed consideration of coverage for percutaneous left atrial appendage closure (LAAC) therapy using an implanted device (CAG-00445N). However, we strongly disagree with the primary conclusion of that analysis, i.e. that such therapy “is not reasonable and necessary to diagnose or treat an illness or injury or to improve the functioning of a malformed body member and, therefore, is not covered under § 1862(a)(1)(A) of the Social Security Act.”

The illness in question is atrial fibrillation (AF). Warfarin, and novel oral anticoagulants (NOACs), are widely used for reduction of the risk of stroke and systemic embolism among patients with AF. Such therapy is guideline-driven and evidence-based, and it constitutes an integral part of the medical management of patients with AF. CMS currently covers these therapies for the treatment of AF.

LAAC therapy using the Watchman device, which is currently the only FDA-approved percutaneous LAAC therapy device, was found in a randomized trial to be superior to warfarin therapy for the reduction of stroke, systemic embolism, and cardiovascular or unexplained death among patients with AF.¹ Thus, this therapy has emerged as a potential alternative to oral anticoagulation for the treatment of AF, which is already covered by CMS as a reasonable and necessary treatment of AF. When used according to its FDA labeling, i.e. as an alternative to warfarin, the Watchman device is also reasonable and necessary for the treatment of AF, and it is superior to an already covered therapy. Thus, the primary conclusion of the present analysis is not evidence-based, and LAAC device therapy should be fully covered because it is reasonable and necessary to treat the illness of AF. We respectfully request that this line of reasoning be carefully considered, as it is sound.

However, we recognize that the proposal to cover LAAC therapy using an implanted device may ultimately be considered by you to be reasonable and necessary only for research under §1862(a)(1)(E) of the Social Security Act using the Coverage with Evidence Development (CED) paradigm. Although we would consider this an incorrect conclusion, we do wish to comment upon the present proposal for such coverage.

In particular, item #2 of the proposal, which addresses patient characteristics, requires revision. Most importantly, it lists “a contraindication to warfarin” as a necessary patient characteristic for use of LAAC therapy. However, this directly contradicts the FDA labeling of the Watchman device, which states that it is indicated only for patients who “Are deemed by their physicians to be suitable for warfarin”, and also contradicts the inclusion criteria for the pivotal trial which led to Watchman device approval.¹ The reason for this is that the device requires short-term post-procedure anticoagulation for safe use, as there is a risk of thromboembolism prior to device endothelialization.

We believe that this basic contradiction results from an intention on CMS’ part to limit use of the Watchman device to patients with a relative contraindication to long-term anticoagulation, for example due to bleeding complications. We consider this highly restrictive and non-evidence-based, and we request that it be removed. However, failing removal, it would of course be essential to change the wording to “a relative contraindication to long-term anticoagulation”, to allow for safe use and to avoid a contradiction of FDA labeling.

Finally, we request that the items in section #2 regarding bleeding risk and stroke risk be revised. LAAC device technology has not been shown to be more effective or safe for any subgroups of warfarin-indicated patients. Similarly, there are no available data suggesting that the Watchman device is more, or less, beneficial among particular HAS-BLED subgroups. Furthermore, the pivotal trial utilized a CHADS score of one or more as an inclusion criterion, and there were no requirements for particular bleeding risk profiles other than a requirement for eligibility for long-term anticoagulation.¹ Finally, the FDA labeling of the Watchman device has no such requirements other than the above-mentioned suitability for warfarin.

Therefore, based upon clinical trial data and FDA labeling, it is medically appropriate for clinicians to consider LAAC device technology for any patient in whom warfarin therapy is clinically appropriate per current guidelines, i.e. a CHADSVASC2 score greater than equal to one. The CHADS, CHADSVASC2, and HAS-BLED requirements in the current proposal should be removed pending availability of new relevant evidence to support such requirements.

Thank you for your consideration.

Respectfully,

The LAAC team at Hackensack University Medical Center:

1. Reddy VY et al. Percutaneous Left Atrial Appendage Closure vs. Warfarin for Atrial Fibrillation. JAMA 2014;312:1988-1998.

Less

Thank you for the opportunity to comment on the proposed coverage decision for left atrial appendage (LAA) closure. I am an Interventional Cardiologist specializing in structural heart disease and I currently serve as Director of the Cardiac Catheterization Laboratory at Scripps Green Hospital in La Jolla, California. My experience with percutaneous LAA closure includes participation in the PREVAIL and CAP2 studies, and the evaluation of patients for LAA closure in the current commercial environment. I have the following comments regarding the proposed decision:

Point 2.
- “The patient has…a contraindication to warfarin”. This wording is troublesome for the practicing clinician, as it does not reflect the current dataset supporting the safety and efficacy of LAA closure, would promote off-label treatment in patients in whom there is a paucity of data, and result in treated patients who are unable to receive the accepted and FDA-approved post-implant pharmacologic regimen. Only patients who were eligible for warfarin were included in the randomized clinical dataset that supports the safety and efficacy of the WATCHMAN device (Reddy VY et al, JAMA. 2014;312:1988-98; Holmes DR et al, J Am Coll Cardiol. 2014;64:1-12; Holmes DR et al, J Am Coll Cardiol. 2015;65:2614-23). Furthermore, the FDA indication for use for the WATCHMAN device specifies that the procedure should be performed only in patients who are "suitable for warfarin", and mandates the use of warfarin for 6 weeks post-implantation.

- “The patient has…a high HAS-BLED score”. While a high HAS-BLED score might be one element to consider in deciding on a treatment strategy for the AF patient, many patients with low HAS-BLED scores may be at substantial risk for major bleeding on oral anticoagulation: for example, a normotensive 68 year-old male with prior gastrointestinal bleeding. In specific, the C-statistic for the HAS-BLED score ranges from 0.50 to 0.66 (Lip et al, J Am Coll Cardiol. 2011;57:173-80), i.e., its ability to predict bleeding is only slightly better than chance. According to the 2014 American Heart Association/American College of Cardiology/Heart Rhythm Society Guidelines for the Management of Patients with Atrial Fibrillation, "HAS-BLED better discriminates risk than the HEMORR2HAGES or ATRIA scoring systems but all 3 scores had C-indexes < 0.70 in their receiver operating curves, indicating only modest performance and poor predictive accuracy", and do not recommend the use of the HAS-BLED score to risk stratify AF patients (January CT et al, J Am Coll Cardiol. 2014;64:2246-80). Importantly, a recently published pooled analysis of the PROTECT-AF and PREVAIL randomized trials demonstrated that the long-term bleeding reduction with WATCHMAN LAA closure was substantial and consistent across patients with both high and low HAS-BLED scores (Price MJ et al, JACC Cardiovasc Interv. 2015).

I suggest that rather than specifying “contraindicated” and “high HAS-BLED scores”, the coverage decision states that LAA closure should be reserved for “patients in whom the risk of long-term oral anticoagulation is felt to be unacceptable”, which would still be consistent with the FDA label and also identify patients in whom the overall benefit of LAA closure would be the greatest.

Point 6. “The registry must include contemporaneous patients followed on oral anticoagulant therapy to serve as non-interventional controls”. This is problematic for several reasons. First, any such comparison of LAA closure outcomes to contemporary “controls” suffers from substantial confounding, which would make the results uninterpretable; second, it puts an undo burden on implanting sites akin to a randomized trial; third, such a “control” arm was not required for other technologies approved under the CED paradigm, such as MitraClip and TAVR. Therefore, I suggest that such a "control" arm be removed from the proposed registry.

Thank you for your consideration.

Sincerely,

Less

RE: Comments on the Proposed Decision Memo for Percutaneous Left Atrial Appendage (LAA) Closure Therapy (CAG-00445N)

Dear Ms Syrek-Jensen

I want to respond to the Proposed Decision Memo for national coverage for Percutaneous Left Atrial Appendage Closure (LAAC) Therapy using the Watchman device. I serve as Service Line Director for the Heart Institute at Cedars Sinai Medical Center, Los Angeles, CA. Cedars Sinai Medical Center has been the lead enrolling center for the three major structural heart disease intervention trials in the US, resulting in the one of the largest experiences in transcatheter valve repair and replacement and percutaneous left atrial appendage closure for atrial fibrillation. These trials have made us well versed with the underlying disease, the procedural benefits for our patients, the interventional procedure and the limitations. While we are pleased that CMS(Centers of Medicare and Medicaid Services), has made a proposal of coverage of percutaneous Left atrial appendage(LAA) closure therapy, we have some concerns and comments about the criteria of coverage, physician training requirements, and requirements of the post market registry.

Comments on Condition 2 (Patient coverage criteria (e.g., high CHADS2 or CHA2DS2-VASc score and a high HAS-BLED score and contraindication to warfarin) :

Comments on CMS Guidance Related to Watchman Device for Left Atrial Appendage Closure

From: Brett Baker MD FACC
Clinical Cardiac Electrophysiologist and Cardiologist
Private Practice in Charleston, South Carolina since 1999
Multi-year experience with anticoagulation for the prevention of stroke and systemic embolism with oral anticoagulants including warfarin, dabigatran, rivaroxaban, and apixaban
Investigator: Amplatzer Cardiac Plug for occlusion of the left atrial appendage
> 10 year experience with ablation for the treatment of atrial fibrillation (includes radiofrequency and cryoablation)

Practice Setting:
Roper St. Francis Health System

December 10, 2015

Ms. Tamara Syrek Jensen, JD
Director, Coverage and Analysis Group
Centers for Medicare & Medicaid Services (CMS)
7500 Security Blvd., Mail Stop S3-02-01
Baltimore, MD 21244-1850

Re: Proposed Decision Memorandum for Percutaneous Left Atrial Appendage Closure (LAAC) Therapy (Administrative File: CAG #00445N)

Dear Ms. Syrek Jensen:

St. Jude Medical, Inc. (SJM) is driven by our vision and mission to transform the treatment of expensive epidemic diseases including atrial fibrillation, heart failure, stroke, coronary artery disease, congenital heart defects, Parkinson’s disease and chronic pain. We strive to achieve our goal by providing innovative solutions that improve care and reduce the economic burden of costly diseases on health care systems worldwide. We believe Left Atrial Appendage Closure (LAAC) therapy is an excellent reflection of the kind of innovation our company is committed to achieving, as we work to transform the treatment options available to patients and providers, alike.

We are writing to the Centers for Medicare & Medicaid Services (CMS) to provide comments on the Proposed Decision Memorandum for LAAC Therapy (Administrative File: CAG #00445N). The proposed decision memorandum states that percutaneous LAAC therapy will be covered for patients with non-valvular  atrial fibrillation only, when all of the seven coverage conditions are met.

SJM would like to comment on the following:

1. Coverage Condition 2 – Contraindication to warfarin
2. Coverage Condition 6 – Design of the LAAC CED registry
3. Coverage Condition 7 – Appropriate rationale to seek a non-pharmacologic alternative to anticoagulants
4. Coverage of LAAC therapy during IDE and FDA-approved clinical trials

1.       Coverage condition #2

2.       Coverage condition #6

3.        Coverage condition #7

4.     Coverage of LAAC therapy during IDE and FDA-approved clinical trials

In summary, we request CMS to consider our following recommendations:

1. Coverage condition #2
   1. Replace the requirement for “a contraindication to warfarin” with “intolerance to long-term warfarin use.”
2. Coverage condition #6
   1. Remove the requirement to enroll non-interventional control patients on OAC into the LAAC CED registry, and
   2. Leverage the data from the FDA-mandated post-approval studies to inform the CMS coverage decisions, where appropriate.
3. Coverage condition #7
   1. Adopt the proposed clinical conditions that impact the risk-benefit ratio of anticoagulation to be used in addition to the stroke and bleeding risk assessment by a combination of CHADS2 or CHA2DS2-VAS, and HAS-BLED as appropriate rationale to seek  a non-pharmacologic alternative to anticoagulants.
4. Coverage of LAAC therapy during IDE and other FDA-approved clinical trials that adhere to the CMS standards of scientific integrity and relevance to the Medicare population and, as such, are approved by CMS.

We appreciate the opportunity to comment on this proposed decision memorandum.

Should you have any questions or require additional information, please do not hesitate to contact me at (651) 756- 5717 or mkoullick@sjm.com.

Sincerely,

Maria Koullick, PhD
Senior Director of Healthcare Economics

Less

I’m an interventional cardiologist affiliated with Ohiohealth-Riverside Methodist Hospital in Columbus Ohio. We are a tertiary care institution who sees and treat complex patients with a multitude of comorbid conditions. We participated in the landmark trials upon which the FDA used the supporting evidence for the Watchman device approval. We treat patients on a daily basis at high risk for bleeding and whom we have seen had many complications from anticoagulation therapy used in the management of atrial fibrillation for the prevention of strokes. We therefore understand the importance for left atrial appendage occlusion with Watchman device. I understand the conditions for which CMS is proposing national coverage. However, I strongly believe, based on available scientific evidence, that this device should be approved for patients with high bleeding risk and not just patients who have a contraindication to take warfarin. Our patients who are at high risk for bleeding benefit tremendously from this new therapy. Based on my clinical experience, patients who have trouble with gastrointestinal bleeding and receive the Watchman device, come off Coumadin with a better quality of life from being off anticoagulation.

I would strongly encourage CMS to review the proposing condition to cover patients only with warfarin contraindication, as this will exclude patients who could absolutely benefit from this therapy and impact their quality of life in so many positive ways by eliminating bleeding risk off their life but at the same time receiving a therapy to prevent cardioembolic strokes from atrial fibrillation.

Less

December 9, 2015 (also submitted as PDF)

Tamara Syrek Jensen, JD
Director, Coverage & Analysis Group
Office of Clinical Standards and Quality
Centers for Medicare & Medicaid Services
Mailstop C1-09-06
7500 Security Blvd
Baltimore MD 21244

Dear Ms. Syrek Jensen:

The Advanced Medical Technology Association (AdvaMed) offers the following comments on the Centers for Medicare & Medicaid Services’ (CMS) Proposed Coverage Decision Memorandum for Percutaneous Left Atrial Appendage (LAA) Closure Therapy.¹ In the proposed decision memorandum, CMS proposes to limit Medicare coverage for LAA closure therapy to patients who are contraindicated for warfarin. CMS further proposes that patients be enrolled in a prospective national registry that also includes contemporaneous patients followed on oral anticoagulant (OAC) therapy to serve as non-interventional controls. CMS has specifically requested comments on its use of coverage with evidence development (CED) in this decision. AdvaMed’s comments below address these issues, as well as the use of CED.

AdvaMed’s member companies produce the life-saving and life-enhancing medical devices, diagnostic products and health information systems that are transforming health care through earlier disease detection, less invasive procedures and more effective treatments. AdvaMed members range from the largest to the smallest medical technology innovators and companies.

The proposed NCD contains several issues of concern, which we address in more detail in the comments below. In addition, CMS has specifically requested public comments on the Agency’s use of CED in this proposed coverage determination.

In the proposed decision, CMS states that coverage for LAAC therapy would be limited to “clinical studies” that meet particular conditions. AdvaMed is specifically concerned with conditions (2) and (6) as described in the decision memorandum.

Condition (2) describes requirements for patients for whom coverage would apply. CMS appears to propose to limit coverage of LAAC therapy under CED for an FDA-approved device (at this time the only FDA approved device is the Watchman LAAC Therapy) to patients who are contraindicated to warfarin. The proposed decision memorandum does not provide additional detail or explanation regarding CMS use of the term “contraindication.”

The FDA-labeled indication is for patients who (among other requirements) are deemed by their physicians to be suitable for warfarin, and who have an appropriate rationale for seeking a non-pharmacologic alternative to warfarin, taking into account the safety and effectiveness of the device compared with warfarin.

CMS’ language could be interpreted to mean that only patients who are not suitable for warfarin would be eligible for the limited coverage under CED described in this proposed decision. This is not only confusing, but seems to result in an unintended consequence. Under this reading of the proposed decision, patients who are not suitable for warfarin (and therefore not included in the FDA label) would be the only patients covered by the policy.

Furthermore, the proposed NCD is not clear as to whether CMS’ current coverage policy for IDE trials or other FDA-approved clinical trials would still apply. The final NCD should make clear that coverage for percutaneous LAAC therapy furnished to all other patients with non-valvular atrial fibrillation enrolled in an FDA-approved IDE trial is provided under the current IDE coverage policy.

Condition (6) of the NCD requires participation in a registry that enrolls LAAC patients as well as patients followed on oral anticoagulant therapy (OAC), to serve as a control arm. AdvaMed has multiple concerns with this proposal. In general, we do not believe that it is necessary or appropriate to include a control arm where CMS requires CED, and the Agency should carefully consider whether a control arm is necessary. CMS should collaborate with stakeholders and seek their input in the decision-making process. With respect to the requirement for a control arm in the proposed NCD, we have a number of specific comments:

First, CMS appears to be seeking a better understanding of the patient population receiving LAAC therapy, and wishes to compare the treatment group with patients on oral anticoagulant therapy. We do not agree that gathering additional data on patients who are well-managed on OAC is the appropriate comparison, as these patients would not be eligible for the Watchman device.

Furthermore, it is not clear that the OAC population is the group CMS is genuinely interested in learning more about, and collecting such data would not appear to answer the clinical questions that are of concern to CMS. A comparison between the patients receiving the Watchman device and patients on OAC would be extremely difficult, if not impossible, to interpret.

Finally, there are numerous operational and logistical issues associated with collecting these data, including very real issues associated with identifying and enrolling those control-arm patients in the registry. The potential number of OAC patients alone could be daunting. In addition, the physicians managing those patients will differ from those managing Watchman patients, and there is little incentive for physicians managing OAC patients, and for the patients themselves, to participate in the registry. For all of these reasons, it would be difficult to comply with the requirements of the registry as contemplated in the proposed decision memo.

AdvaMed has often advocated that CMS should work with stakeholders to identify the appropriate clinical questions to be addressed in order for the Agency to ultimately make a permanent coverage determination. We recommend that CMS meet with stakeholders to work through the clinical questions together to develop appropriate study endpoints that will meet CMS’ needs.

Additionally, with respect to Condition (6), CMS alternately describes this condition as a registry, and as a prospective, randomized, controlled clinical study. The language in the proposed coverage decision has caused confusion and should be clarified. We recommend that in addition to working with stakeholders to determine the appropriate comparator, if any, CMS should make clear whether the continued data collection requirement lies in a registry or a clinical trial.

In the proposed NCD, CMS specifically asks for public comment on the use of Coverage with Evidence Development in this decision. AdvaMed supports the use of sound evidence to inform medical practice, and has generally believed that CED may be a means to generate additional evidence in cases where the available evidence regarding a promising medical technology or procedure is not sufficient to support national coverage. In these circumstances, coverage contingent on the development of additional evidence can provide beneficiaries with access to innovative medical treatments when the alternative may be non-coverage.

CED allows CMS to provide coverage within the context of studies that conform to standards detailed in Medicare’s Clinical Trial Policy.² The research meeting these standards can include a range of research approaches.

AdvaMed has submitted comments to CMS regarding its CED guidance generally, and many of the concerns we have raised in that context are relevant with respect to the proposal for coverage of LAAC therapy under CED. AdvaMed strongly believes that Medicare’s coverage policies should encourage innovation, and that CED should expand, not restrict, patient access to new technologies and treatments. We also urge CMS to continue to find avenues in NCDs whereby the Agency may impose certain conditions for coverage (for example, specific provider or institutional requirements) without finding that the evidence is insufficient under §1862(a)(1)(A) of the Act and therefore placing the NCD under §1862(a)(1)(E).

We reiterate that when Medicare coverage is contingent on the collection of additional clinical or scientific evidence (beyond FDA requirements for safety and efficacy), CMS should:

1. collaborate with stakeholders to clearly identify the data collection objectives,
2. consider the minimum data necessary to achieve those objectives, and
3. clearly identify, with input from interested stakeholders, scientifically supported study endpoints and the duration of data collection in advance.

In considering the minimum data necessary, CMS should also consider whether additional data being developed through FDA-mandated post-market studies satisfies CMS’ evidentiary needs for coverage. In addition, coverage policies should be sensitive to the practical challenges associated with generating evidence, and should support patient participation in clinical trials.

If you have questions regarding these comments or if you require additional information, please contact me or Chandra Branham, JD, Vice President, Payment and Health Care Delivery Policy, at (202) 434-7219 or cbranham@AdvaMed.org.

Sincerely,

/s/
Donald May
Executive Vice President
Payment and Health Care Delivery Policy

Less

Less

Dear Ms. Syrek Jensen, J.D.,

As clinical electrophysiologists in a large cardiology practice in New Hampshire, we care for numerous patients with atrial fibrillation in our region. We strongly support CMS coverage for LAA occlusion which we believe will be an important tool in our management of stroke risk in patients with non-valvular atrial fibrillation.

Despite the weight of evidence and a class I recommendation set forth by the 2014 AHA/ACC/HRS Task Force for the use of oral anticoagulants in patients at risk of thromboembolism, numerous studies have shown that oral anticoagulants are grossly underutilized [1,2]. There are numerous reasons for underutilization including patient acceptance, the actual or perceived risk of bleeding, and compliance to name a few. We are confident that the availability of an alternative to anticoagulants will enable us to protect more patients from thromboembolism than is possible with oral anticoagulants alone.

We would also like to mention that your proposal for coverage on the condition that patients have “a contraindication to warfarin” is both confusing and contradictory to the FDA labeling which requires patients to be “suitable for warfarin.” It is important to note that the standard procedure has been to provide warfarin anticoagulation for at least 45 days after Watchman implantation. Your proposal would invalidate this approach.

We urge the committee to establish coverage for LAA occlusion with the Watchman device. It is only with your help that we can offer the best available therapy individualized to each patient. Without coverage, we fear that our ability to offer this therapy as a viable alternative will be severely restricted.

Thank you for your consideration.

Connor J. Haugh, MD FACC
Jamie H. Kim, MD FACC

Less

Dear Ms. Syrek Jensen, J.D.,

As clinical electrophysiologists in a large cardiology practice in New Hampshire, we care for numerous patients with atrial fibrillation in our region. We strongly support CMS coverage for LAA occlusion which we believe will be an important tool in our management of stroke risk in patients with non-valvular atrial fibrillation.

Despite the weight of evidence and a class I recommendation set forth by the 2014 AHA/ACC/HRS Task Force for the use of oral anticoagulants in patients at risk of thromboembolism, numerous studies have shown that oral anticoagulants are grossly underutilized [1,2]. There are numerous reasons for underutilization including patient acceptance, the actual or perceived risk of bleeding, and compliance to name a few. We are confident that the availability of an alternative to anticoagulants will enable us to protect more patients from thromboembolism than is possible with oral anticoagulants alone.

We would also like to mention that your proposal for coverage on the condition that patients have “a contraindication to warfarin” is both confusing and contradictory to the FDA labeling which requires patients to be “suitable for warfarin.” It is important to note that the standard procedure has been to provide warfarin anticoagulation for at least 45 days after Watchman implantation. Your proposal would invalidate this approach.

We urge the committee to establish coverage for LAA occlusion with the Watchman device. It is only with your help that we can offer the best available therapy individualized to each patient. Without coverage, we fear that our ability to offer this therapy as a viable alternative will be severely restricted.

Thank you for your consideration.

Connor J. Haugh, MD FACC
Jamie H. Kim, MD FACC

Less

Dear Sir/ Madam

I read with concern your proposals for coverage determination/ decision for Left Atrial appendage closure using the Watchman device, an FDA approved device , for use in high risk patients with non valvular atrial fibrillation, who are seeking an alternative to long term oral anticoagulation.

As a clinical cardiac electrophysiologist in practice for 15 years , i have witnessed multiple misfortunes and unfavorable outcomes related to the use of oral anticoagulants for prevention of thromboembolic disease in this population. I have welcomed the availability of LAA closure options and have pursued adoption of Watchman at my institution with the manufacturer and have offered it to several of our patients with excellent procedural and short term outcome . I believe avoidance of OAC will translate in significant reduction of morbidity and healthcare cost savings in these patients.

I would like to make specific comments and recommendations to the esteemed panel in charge of the final decision on this matter:

To summarize, I think there is enough evidence to recommend approval and CMS coverage for Watchman LAA closure in high risk non valvular AF patients, at high risk for bleeding complications, at select implanting facilities, by select implanters. I believe there will be significant savings achieved in avoidance of the costly treatment of bleeding on anticoagulants.

Regards ! Ahmed F. Osman, MD FACC, FESC, FHRS

Less

I comment in my capacity of the Director Of Cardiac Clinical Electrophysiology at Heart and Vascular Institute at Henry Ford Hospital in Detroit, Michigan. As can be expected, I deal with patients with AF on a daily basis, including the managing of these patients anticoagulation when indicated. In that capacity and throughout twenty five years of practice, I have a fair experience in identifying risks and conditions of patients that have experienced absolute and relative contraindications to oral anticoagulants. I am currently a certified implanter of Watchman devices and have by now developed considerable experience in implanting and managing these patients.

As a summary statement and before commenting in the specific conditions, I want to emphasize that I favor the use of LAA occlusion devices ONLY on those patients that have demonstrated a clinically relevant intolerance to the use of anticoagulation.

Condition 1: Watchman should only be used in non-valvular Atrial Fibrillation patients. The only exception is the concomitant presence of non-rheumatic Mitral Regurgitation judged to be unrelated to AF

Condition 2:

• Appropriately indicated patients for anticoagulation based on Chads2Vasc scores with documented episodes of AF that are clinically significant, or asymptomatic episodes documented by devices considered to carry a significant risk for stroke as documented in prior studies (i.e. ASSERT, longer than six minutes)
• Relative Contraindication to Coumadin as documented by a prior bleed considered by the attending physician to be relevant and likely to recur, significant fall/trauma risk, or a high HAS- BLED score. Patient should be able to likely tolerate 45 days of Coumadin as indicated post implant.
• Absolute Contraindications to Coumadin, such as intracranial bleeds, in which Coumadin can’t be used for any period of time, should not be covered. A randomized trial to determine the efficacy and safety of Watchman implants in the absence of any anticoagulation, versus standard medical care (no anticoagulation) should be performed to allow for coverage at this time. Such a trial is being implemented now, and its results available before LAAC is offered without the 45 days of anticoagulation as used in the pivotal trial.

Condition 3-5: Agreed as described

Condition 6: A national prospective registry of LAAC is extremely important and I support fully the CMS initiative of a long term effectiveness and safety registry as described. It is less clear what if any, the control OAC registry is going to add to the available information from large randomized trials of OAC, unless the control registry will include only patients that, after being offered LAAC, refuse and continue OAC.

Condition 7: I fully agree with the CMS proposed strategy of developing a tool allowing documenting the discussion between the patient and clinician regarding the rationale risks and benefits of LAAC. Patients need to understand that the bulk of evidence is for OAC to be the standard of care, unless actual bleeding or high risks of bleeding are a consideration. The potential for abuse and the false premise that the data already justifies LAAC as an equivalent non inferior treatment to OAC should be discouraged by the standard use of a decision making tool and the existence of a detailed registry documenting the clinical scenario.

Less

Less

Dear Ms. Syrek-Jensen:

My comments relate to the issue of non-coverage for Watchman® device LAAC for stroke prevention in patients with non-valvular atrial fibrillation (AF). This is a very significant clinical problem and is getting more pronounced due to the rapidly aging American population.

Multitudes of patients have non-valvular AF with increased risk of ischemic stroke based on CHADS-Vasc risk stratification, but are not candidates for conventional oral anticoagulation (OAC) with either warfarin (VKA) or novel agents (NOACs) due to prior bleeding, multiple risk factors for bleeding or failure of OAC to prevent stroke. Although anticoagulants are effective in stroke prevention in this group, 40-50% of patients at risk are not treated. This is particularly a problem in the patient groups over 70 who have accumulated higher CHADS-Vasc scores but also more co-morbidities that prevent safe administration of OAC.

Many patients do well with OACs for years only to have life-threatening complications from the drugs develop after they have been on them for years or decades. Cerebral amyloid angiopathy (CAA) is a risk factor for intracranial hemorrhage (ICH), dementia, and cerebral micro-infarcts. CCA prevalence accelerates with age and accounts for 15% of all ICHs in patients over the age of 60 and up to 50% of all non-traumatic ICHs in patients over the age of 70; OACs only increase these risks. CCA increases in prevalence with age going from 5% in the 60s to 50% in the 90s. CCA is more common in women and interestingly, there appears to be a less beneficial preventative effect for warfarin use in elderly women for stroke prevention in the setting of AF. Detection of CCA as a risk factor for ICH (and therefore OAC complication-augmentation as the individual ages) is increasing as neurologists recognize and characterize this entity further. Thus, a non-pharmacologic management modality option to reduce the use of OACs in the rapidly growing healthy elderly population remains imperative.

The Watchman® LAAC device has produced similar outcomes to warfarin at 5-year follow-up and is markedly better in patients in whom anticoagulants either are contraindicated or are not administered. Recent data from the Europe was reported at the American Heart Association annual Scientific Sessions in November, 2015, (EWOLUTION registry); this study included 1,021 patients with AF, mean age 73 years, 62% ineligible or OAC, 40% women. 31% had had a prior stroke or TIA and 39% had either a prior major bleed or major risk factor for one; this group had a higher risk of stroke or bleeding than the PROTECT-AF and PREVAIL trials and the CAP registry. Implant success was 98% with a 30 day mortality rate of 0.8%. There appeared to be only 2 procedure related deaths. These data suggest with appropriate experience and learning curves, the safety of the procedure continues to drop.

Given the magnitude of stroke hazard and the large numbers of patients at risk, we need alternative approaches for stroke prevention. Selected high risk patients should have CMS-access to this FDA-approved technology. Variable payer access is the current state; the minority of US states have adopted payer models for the device. Thus, we are seeing unfair and inconsistent access to this technology for American patients, depending on which State they happen to reside in. This perplexes both providers and patients. Stroke results in increased disability, mortality, and healthcare costs. Stroke is among the top 3 reasons for institutionalization of patients to nursing homes.

By mail, I have enclosed to you 5 de-identified case scenarios from my own practice that emphasize the importance of having LAAC as an option for “select” patients facing enhanced stroke risk because of unacceptability of long-term OAC therapies, either due to life threatening bleeding episodes, OAC inefficacy, or OAC contraindication.

Thank you for your consideration.

Less

I am an Interventional Cardiologist at Ochsner Foundation Hospital in New Orleans, LA for past 21 years. I have 7 years experience with the Watchman device and have spent my entire career treating atrial fibrillation patients at risk for stroke and heart disease. For years the only therapy available was warfarin, but in the past decade, several alternative medications and devices have attempted to remedy the downfalls and inadequacies of warfarin therapy. Outpatient monitoring, hemorrhagic stroke risk, GI bleeding, indigestion, dizziness, latent INR and a host of other side effects have had both patients and physicians seeking other treatment for years. With the approval of Watchman in March 2015, I feel like I now actually have enough tools to give patients real choices in stroke reduction therapy. I applaud the FDA for their choice to approve Watchman and I also applaud CMS for their choice to offer proposed coverage. I sincerely thank you.

CMS coverage proves it's acceptance and need for the therapy with the decision to propose coverage, but my concern lies in some of the proposed language and it is my desire to address a patient scenario with you and suggest amending the proposed language to reflect the intended patient studied in trial and currently being implanted within the guidelines the FDA set forth in labeling. Unfortunately, intending use for a product and communicating alternative recommendations for payment is a disparity. In order for me to properly follow the labeling, I must put the patient on warfarin, but your Condition #2 prohibits that by asking for only contraindicated patients. My suggestion is a closer alignment to the Watchman implant protocol and subsequent modification to the proposed condition #2. In addition, condition #6 is not very clear to me, do you want my clinic to follow all orally anti-coagulated patients in a registry as a "control"? I think that the evidence and data has all been submitted for the science of all of these medications individually and should a control be required, CMS should use the multitudes of data studied prior to FDA approval and post-approval studies to evaluate control rather than burden physicians and institutions with the task of gathering more data. Left atrial appendage closure should be part of a registry and limited to only patients with intervention of the appendage. Please consider modification for the registry to include modified atrial appendages only to make this a registry with validity and specific science. Finally, HAS-BLED has not been the standard practice for stroke protection in the US. It has garnered great usage in neighboring countries, but this is not part of the FDA process of evaluating potential Watchman patients and the wording is unclear in your proposal for patient selection. If you are requesting this data be gathered in registry, I think it is appropriate, but asking for it to be required at certain level for reimbursement, I must ask where the science is located for that. The device wasn't studied this way and labeling is not reflective of such recommendations

Here is an example of an actual Watchman indicated and implanted patient. 62 year-old man with persistent atrial fibrillation, and a previous medical history of diabetes mellitus, coronary disease with large myocardial infarction requiring stenting of circumflex artery, and ischemic cardiomyopathy complicated by congestive heart failure. He has been compliant on warfarin for some time, but has been unable to maintain an INR between 2.0 and 3.0. His time in therapeutic range (TTR) has been approximately 35%, with the majority of his INR’s < 2.0. Despite attempts to remedy his INR with weekly INR clinic checks and hospital integrated pharmacy follow-up, the patient is not contra-indicated to warfarin. The patient has attempted rivaroxiban with subsequent GI bleeding and after an extended inpatient stay for transfusion and stabilization, he desires to remain on warfarin and not be at risk for GI bleeding with a warfarin alternative medication.

The current proposed indication would require this man to be on warfarin for his lifetime despite a high CHADs-VASC score of 2 and rising HAS-BLED score (2 currently) for the remainder of his lifetime due to the fact he can take warfarin and is not contraindicated. Furthermore, he would have to join a registry since he is on an oral anti-coagulant and await his fate. The question remains in my mind as to why the proposed coverage will not cover patient's currently approved per the FDA approved directions for use. Speaking from experience with TAVR, I agree that boundaries need to be in place to perform these advanced heart procedures safely and responsibly, but asking physicians to give warfarin to the patient's per the approval guidelines but deliver the device only to contraindicated patients is irresponsible and should be altered prior to your final decision. In addition, Left Atrial Appendage Closure registry data should be limited to patients that have undergone the therapy. If the organization desires a registry for stroke risk including anti-coagulants, please realize that the number of FTE's involved at our facility would be greater than 2.0 and reimbursement for such employees simply isn't built into the system as it exists today.

Less

To: Ms. Tamara Syrek-Jensen, J.D., Director of Coverage & Analysis Group

Dear Ms. Syrek-Jensen:

I have been involved with the WATCHMAN LAAC Therapy since the PREVAIL trial. Many patients have benefited from the device. Most have had relative contraindications to coumadin.
The HASBLED score isn't a full measure of who could benefit as it doesn't include relative contraindications such as compliance, other medical issues such as syncope, occupation, lifestyle, socioeconomic status, need for chemotherapy etc.
I believe it should be available for patient with a reasonable desire to seek alternative to standard anticoagulation (although it is reasonable to collect HASBLED in a registry).
I do understand need for national coverage with some restriction to ensure correct patient population for this complex procedure that is used for prevention of CVA in patients with limited alternatives.

Sincerely,

Mark Freher

Less

Less

Ms. Tamara Syrek-Jensen
Director, Coverage & Analysis Group
Centers for Medicare & Medicaid Services
7500 Security Boulevard
Baltimore, MD 21244

RE: National Coverage Analysis (NCA) for Percutaneous Left Atrial Appendage Closure (LAAC) Therapy (CAG-00445N)

Dear Ms. Syrek-Jensen:

The CMS national coverage decision on the Watchman device, if left unchanged, will restrict a valuable treatment option to my practice and our atrial fibrillation program. It is well known that the left atrial appendage is the origin of the large majority of thrombi in patients with nonvalvular atrial fibrillation. Excision of the left atrial appendage concomitant with surgery of the mitral valve is recommended in patients with mitral valve disease and atrial fibrillation who need surgical mitral valve repair or replacement. This recommendation is based on very small studies that demonstrate significant reduction in the stroke rate when the appendage is ligated.

The Watchman device has been thoroughly tested and has a clear mortality benefit when compared to warfarin on long term follow up of patients in the Protect AF study. Enrollment criteria would have excluded patients who were warfarin intolerant in both clinical trials, Protect AF and Prevail. This makes the national coverage decision perplexing. Patients need at least 6 weeks of systemic anticoagulation post implant. It is true that some patients are not candidates for systemic anticoagulation, but those patients would be excluded from Watchman as well.

Patients want choice. Systemic anticoagulation, despite recent advances in drugs and antidotes, places patients at risk of serious, sometimes fatal bleeding events. The Watchman clinical trials and long term follow up have demonstrated the utility of understanding the pathophysiology of a disease. Targeting the location of thrombus, excluding it from the vascular space and maintaining a patient on long term aspirin is superior to systemic anticoagulation over the long haul. Event rates beyond 4 years in the Watchman arm of Protect AF study would be 50% less than event rates in the warfarin arm if the implant procedural risk were adjusted to current day rates. The Protect AF CAP and Prevail both demonstrate 4% procedural implant risk. We have learned lessons regarding air emboli and the strokes associated with them.

It is time for CMS to make the right decision and not simply a short-term financial decision. Exclusion of the LAA with Watchman is superior to systemic anticoagulation with warfarin. The clinical trials have convinced the FDA and the major medical societies including the ACC and HRS. There is a need to offer patients a clear alternative to lifelong systemic anticoagulation and all of the hazards associated with this therapy. All patients with nonvalvular atrial fibrillation recommended to have lifelong systemic anticoagulation, those with CHADS-2 Vasc score of 2 or more, should have this choice. It is a decision between the patient and physician after careful discussion of the risks and benefits.

Please reconsider your national coverage decision.

Sincerely,

Brian Schuler, MD
Atrial Fibrillation Program Director
Wellspan Cardiology
York, PA

Less

The Watchman LAA closure device is a novel therapeutic option for patients that cannot be safely anticoagulated. As a clinician I see a lot of Medicare patients who unfortunately suffered strokes while being off anticoagulation. Its not an easy decision to take one off Warfarin and now there is another safe and effective option to prevent strokes with the Watchman device.

Less

Dear Ms. Tamara Syrek-Jensen,J.D.

I am an Electrophysiologist with Texas Cardiac Arrhythmia. I would like to express my concerns regarding the CMS's Proposed Coverage Decision for Percutaneous Left Atrial Appendage Closure using the Watchman device.

I think the descriptions regarding recommendations for the "eligible patient population" and indications need significant clarification.

I am unclear regarding the statements relating to indications which pre-specify the requirement for a high CHADS2 Vasc score and a high HAS-BLED score.

A high CHADS2 Vasc score should mean one that is high enough to have the patient indicated for anti-anit-coagulation or >=2. I am not clear that any greater hurdle would make sense as we are considering a therapy to replace anti-coagulation

With regards to HAS-BLED scoring I am unclear as to the indication nor scoring as this was not addressed in any of the studies to date. I would agree that one may use the HAS-BLED to ascertain patients that are at increased risk of bleeding ( who have not yet had a bleed). However, there are any number of other factors which may influence the risk of bleeding. These may include, frailty, risk of falling, compliance and other lifestyle considerations.

Finally, I am unclear as to the logic or motivation to suggest that device be restricted to patients who are contraindicated to OAC. These patients were not included in the studies to date as the study population required the ability to be randomized to anticoagulation. In fact a patient who is contraindicated for short term anticoagulation would not be a candidate for this device as Coumadin is required post implant.

Thank you for allowing me the opportunity to comment on the approval process. I hope you and you Committee come to a set of guidelines that will allow our patients to benefit from this new technology.

David Tschopp MD FACC FHRS
Texas Cardiac Arrhythmia

Less

There are several issues with the coverage decision that I believe would adversely affect patients who would benefit from LAAC (left atrial appendage closure).

I have specific concerns about coverage condition #2 using the phrase "contraindications to warfarin." There are many clinical scenarios where the patient may not have a contraindication to warfarin but has an intolerance to anticoagulation. The use of the word contraindication implies a significant criteria that would exclude many patients who would benefit. A more inclusive and appropriate language would be "intolerance to anticoagulation". Additionally in the coverage condition #2 there should be more specific language describing eligible regarding CHADSVASC score. This would ensure that patients who are at low risk (CHADSVASC 0 OR 1)are appropriately excluded. Those patients CHADSVASC 2 or greater however should absolutely be included. The expeditious approval for coverage is greatly appreciated as the delay in coverage will continue to pose patients at risk. For example I have many patients at the highest risk in my practice with CHADSVASC of 9 have a yearly stroke and thromboembolic risk of 17.4%. While these patients are awaiting coverage they are unfortunately accepting a month risk of 1.4%. While coverage considerations are important, it is critical to understand that there are risks to patients with delays in approval and coverage.

Left atrial appendage closure is a complex procedure involving a unique skill set of trans septal puncture and understanding three dimensional anatomy of the appendage. It utilizes multi modality imaging as well as the coordination with noninvasive imaging Cardiology experts, Anesthesia, and the operator. In comparison to ASD/VSD closure, Left atrial appendage closure is much more complex, with numerous additional steps and therefore reimbursement should reflect the additional complexity and skill set.

Less

Less

There is perhaps no medical device that has undergone as much scrutiny and been subject to as much rigorous trial data as the Watchman device.
Multiple trials involving thousands of patients studied over several years have proven the superiority of Watchman over anticoagulation. Subjecting a patient to decades of anticoagulation exposes him or her to much more risk and from the societal standpoint is much more expensive than a single device implant.
We strongly support a national coverage decision regarding Watchman but would like to point our that some of the proposed conditions in the coverage decision are completely unsupported by the available data.
In Condition # 2, the HAS-BLED score while useful as bleeding risk tool is not a good metric to assess which patients would benefit from Watchman, something the CHADs-Vasc score does a very good job with. Therefore, the HAS-BLED assessment should either be removed entirely or at the very least made an optional metric, and not a requirement.
Also, in Condition # 2, the fact that the coverage decision approved this device in patients that have a warfarin contraindication is in direct contradiction to the patients that were studied in the trials. Furthermore, since warfarin is strongly recommended for a short period post implant, patients with a contraindication to warfarin would never be able to have a Watchman!
Finally, regarding Condition # 6, we are in agreement that as in any device or procedure, there should be appropriate and close follow up in a registry setting. However, the need to have a comparison with an OAC population would not be very helpful at all. This question has already been studied in detail and answered conclusively with hundreds of thousands of data points, that Watchman is superior to OAC. Therefore the need to include this patient arm would not add to the science and also be difficult, expensive and time consuming for almost every institution.
Thank you for your consideration.

Less

Less

In our practice, we are selecting patients with CHADsVasc score of >=2. Patient must be able to tolerate short term anticoagulation but have a reasonable indication for being a poor long term anticoagulation patient. Patients with High HASBLED scores (>=3) or a history of bleeding complication on anticoagulation, or a fall risk.

Patients (with CHADSvasc >2) who do not do well on warfarin (labile INR), poor compliance, or low time therapeutic range are reasonable candidates, since they are at high risk of stroke.

Less

LAA closure with the WATCHMAN device is a major advance in the prevention of stroke in patients with atrial fibrillation. It has now been proven to reduce the risk of serious stroke, intracranial hemorrhage, and major bleeding in atrial fibrillation patients at risk for stroke. Even more importantly, a mortality benefit has been demonstrated in patients receiving the WATCHMAN device.

Coverage for WATCHMAN therapy is crucial for patients with an elevated risk for stroke, high bleeding risk, and a contraindication to warfarin. However, there are other populations of atrial fibrillation patients who would remain at risk for stroke and bleeding if excluded from coverage.

As an interventional cardiologist, a significant portion of my atrial fibrillation patients also have carotid, coronary, and peripheral arterial disease. Those treated with stents, particularly drug eluting stents, are committed to dual anti-platelet therapy. These patients often have an elevated risk of stroke and anti-coagulation is indicated. Therapy with dual anti-platelet therapy and warfarin would place them at significantly increased risk of major bleeding events. These patients would benefit greatly from WATCHMAN therapy.

Another set of patients who would benefit from WATCHMAN therapy are those with an elevated risk of stroke who have had difficulty maintaining their INR in the therapeutic range on warfarin. These patient's spend the majority of their time either sub-therapeutic and at risk for stroke or supra-therapeutic and at risk for bleeding.

Serving a large indigent population as well as an elderly population, our practice also has many patients who are challenged by great hardships in terms of obtaining the needed monitoring for warfarin. These are also the very patients who can not afford anticoagulation with the novel oral anticoagulation agents.

Finally, there is a sizeable population of atrial fibrillation patients at elevated risk of stroke who are at higher risk for bleeding complications from anticoagulation due to profession (ex. police officers and firefighters) or religious/personal beliefs that prohibit blood transfusion (Jehovah's Witnesses).

WATCHMAN therapy would be of great benefit to all these populations.

Less

Dear CMS

As a clinical electrophysiologist and newly trained Watchman implanter, I feel compelled to comment on your proposed LAAC Coverage proposal.

While I believe you have carefully reviewed the current literature on Watchman, I am concerned by the potential limits being placed on this therapy for potential "high" risk populations that may not meet your criteria.
Specifically, I am concerned by the requirement for a high CHADS2VASC score AND high HASSLED score AND contraindication for coumadin. In clinical practice there are many patients I care for that I feel are at increased risk for anticoagulation due to profession or hobbies that may be excluded from Watchman given your criteria.

Specifically I am concerned by patients I have that may have a modest CHADS2VASC scores but low HASBLED scores who ranch or farm or work in logging or other professions where they are concerned and may refuse anticoagulation due to potential risk. The same applies to my patients who downhill ski or hunt or fish or hike or bike. Yes, many of my medicare patients with Afib are still active and engage in these activities. They are often reluctant to consider long term anticoagulation. The same is true for patients who live remotely or who travel (i.e. truck drivers). While NOACs are a good alternative for some of these patients, drug cost is also a major limitation for many patients.
While I understand your intent to limit this therapy to the populations upon which this therapy was studied, I also believe you need to consider giving physicians some lattitude for discussing this therapy when there are legitimate concerns about increased risk of coumadin or NOACs.

I agree that this procedure should be performed by physicians well trained in transeptal technique ideally as a conjoint effort between electrophysiology, interventional cardiology and nonintervention cardiology (TEE) to obtain safe effective outcomes in our patients. I hope that medicare realizes that this approach is labor intensive and you will provide a fair reimbursement for this procedure which supports a team effort. I believe this will result in the best patient outcomes.

I also agree with the recommendation for a registry to document outcomes and ensure this is an appropriate safe procedure.
Thank you for considering my comments.

Less

I am an interventional cardiologist who specializes in structural heart disease. I have been an investigator with the Watchman device since 2005 and have participated in the pilot study, PROTECT-AF, CAP-1, PREVAIL, and CAP-2.

Overall I think that the CMS proposed NCD is excellent. However, I do have several comments that relate to conditions 2 and 6. First, I am unclear what CMS means by "contraindication to warfarin" since the clinical trials (and FDA approval) requred that patients be eligible for warfarin therapy and that patients be treated with warfarin for 45 days following implant. I think it would be clearer to change to "relative contraindication to warfarin". Secondly, I agree that high HAS-BLED score identifies patients at hihg risk of bleeding that could benefit from Watchman LAAC I thik that are situations that HAS-BLED does not necessarily cover. Two scenarios that I have recently seen are: 1) patient with comlex bifurcation stenting that will require life long ASA and clopidogrel but has had major bleeding on triple therapy (but no other factors that drive high HAS-BLED)and 2) eldery patient with frequent falls but no other significnat risk factors. I think that these types of patients could clearly benefit from Watchman LAAC. Thus, I think that HAS-BLED may be a sufficient but not a necessary criteria for LAAC. Lastly, I wholeheartedly agree with the importnace of a LAAC registry. However, I think that a required cohort of oral anticoagulation paients is overly burdensome and impractical. Additionally, I suspect that the patients in the LAAC group and OAC group would be significantly different and scientific comparisions would be implausible.

Thank you for allowing me to comment.

Less

As a practicing electrophysiologist for a large practice, for years have had droves of patients who suffer from atrial fibrillation and either have had a stroke or at high risk for a stroke based on their CHADs2-vasc score.

Most of these patients are able to tolerate oral anticogulation. However there are numerous patients who have had significant bleeding on oral anticoagulation ranging from frequent skin bleeding, epistaxis to life threatening GI bleeding requiring surgery and/or transfusion.

LAA closure technology with WATCHMAN has been an amazing therapy option to offer these patients who are not able to take long term anticoagulation. Placement of WATCHMAN does not obviously completely eradicate the risk of CVA but based on robust clinical data clearly reduces the risk of ischemic stroke.

Less

Less

Agree with current CMS proposed coverage decision but will kindly request consideration of the following.

1- Benefit of Watchman device extends to population of patients who have contraindication to OAC and to those who are high risk but not contraindicated to OAC. the term "contraindication to warfarin" is not well defined in the literature as there are absolute contraindications, strong contraindications and week contraindications. but even the week contraindications can bring about significant bleeding risk, such as for example patients who are committed to long term dual anti PLTs or high doses NSAIDs and being asked to take OAC as well (11% increase in bleeding risk)

2- Benefit of Watchman device extends to population of patients who have an established chronic pattern of non-compliance with medications even if they do not present contraindication to OAC. (Previous trials indicate that interruption of OAC carry significant stroke risk) NOACS have a discontinuation rate that ranges between 20-25%.

3- Benefit of Watchman device extends to population of patients who strongly reject taking OAC even if they verbalize agreement to taking the medicine. (these patients have high likelihood of discontinuing medication without informing their physicians and being left unprotected against increased stroke rate)

4- Registry with control arm of OAC: even if we place Warfarin aside the group of NOACS is a very heterogeneous group of drugs (different pharmacokinetics, pharmacodynamics, efficacy, bleeding risk etc) and with such wide spectrum of differences I am not certain we can have a scientifically sound control arm combining them all as one.

5- The utilization of HASBLED score as indicated or similar scores: these scores have only moderate performance and inadequate bleeding risk predictability

Less

Thank you for affording me the opportunity to provide comments regarding the recent proposed NCD by CMS for the Watchman procedure.

In my current role I serve as the administrative director for the Electrophysiology Service Line with the Sanger Heart & Vascular Institute/Carolinas HealthCare System. I have programmatic and operational responsibilities for heart rhythm care across the SHVI footprint. My background as a registered nurse, certified cardiac device specialist and electrophysiology expertise has prepared me to critically review the proposed conditions for the Watchman national coverage decision.

Sanger Heart & Vascular Institute is part of one of the largest public health systems in the country. We care for tens of thousands of individuals with atrial fibrillation across the system. Our facility served as a research site for PREVAIL study which investigated the safety of the Watchman device. Our implanting physicians have significant experience with the implantation of this left atrial appendage occlusion device as well as the pre and post procedure care of patients with the Watchman device. Our center continues to serve as an implant center for the FDA approved, commercially available Watchman devices.

Sanger Heart & Vascular Institute and Carolinas HealthCare System pride ourselves in offering many therapies to atrial fibrillation patients, including anticoagulation, atrial fibrillation ablation and left atrial appendage occlusion with the Watchman device. Our patients deserve the right to have access to these therapies. The CMS proposed national coverage decision will enable our patient’s access to this device but will need to cover the cost of providing this therapeutic solution to our patients at our not-for-profit facility.

I will leave it to the physician responders to comment on the bleeding risks and contraindication to anticoagulation therapy. Risk of stroke associated with atrial fibrillation and risk of bleeding while on oral anticoagulation are well documented. However, we have an obligation to treat those atrial fibrillation patients who are not candidates for anticoagulation, for whatever reason, to reduce the significant risk of stroke associated with atrial fibrillation.
I would like to address some of my concerns regarding the seven conditions for this coverage proposal:

• Condition 1: I am in agreement that coverage should include only the Watchman device. If separate clinical trials demonstrate similar benefit and risks for other technology, approval can be considered at that time.
• Condition 2: Defer to physician feedback but essentially should include both absolute and relative contraindications to anticoagulation. This determination should be at the physician’s discretion.
• Conditions 3, 4, 5: I fully support the requirement for strict physician training as well as infrastructure requirements to support patient safety.
• Condition 6: National registry development for the Watchman device should be strongly considered. Information aggregated in the registry, such as what has been done for TAVR, MitraClip and ICDs will provide valuable information.

However, the inclusion of a “control arm” consisting of patients treated with anticoagulation would cause undo expense and time drain on practices and hospitals. Furthermore, I believe establishing a control group of non-randomized oral anticoagulation therapy patients would not provide any valuable or meaningful data that would assist clinicians. Lastly, this ‘control arm’ would be quite impractical and difficult to validate.

• Condition 7: In my opinion, informed consent along with specific documentation of risk, benefit and alternative therapies for Watchman therapy is the appropriate mechanism for providing shared decision making for Watchman therapy.

The widely accepted standards of practice dictate that physicians should review the risks, benefits and alternatives to Watchman therapy with patients for informed consent to be given. This is the standard of care we apply to all procedures and therapies. I do not support a requirement of an additional documentation form. This method is burdensome and provides no additional value to patients.

The Watchman device signifies a revolutionary technology for stroke prevention in patients afflicted with atrial fibrillation. Watchman addresses an unmet need for the growing number of atrial fibrillation patients. I appreciate the opportunity to comment. I look forward to the final coverage decision from CMS.

Less

We are excited to have for the first time an FDA approved nonpharmacologic therapy for the reduction of the risk of cardioembolic events and stroke in patients with atrial fibrillation. This is a unique and valuable tool which can be implemented in patients who are at risk for bleeding. Strong clinical data suggests that the implantation of the percutaneous left atrial appendage occlusion device is far superior to no therapy at all and is currently the only option available for patients who cannot utilize anticoagulant therapy safely. Patients who have atrial fibrillation are at risk for potentially disabling stroke. The ability to utilize the left atrial appendage occlusion device in those patients who have demonstrated a bleeding complication or have been deemed to be too high risk for long-term anticoagulation is absolutely paramount. Any barriers that impedes the utilization of this important new technology Will undoubtedly result in the presence of cardioembolic events and stroke that could otherwise have been prevented. A stroke is one of the most devastating medical events that can occur to a patient. We fortunately have an effective therapy in place in the form of oral antithrombotic medications. Many patients cannot safely utilize these pharmacotherapies and this places them at risk when they do not have access to the left atrial appendage occlusion device. Percutaneous left atrial appendage closure provides a long awaited alternative option for patients where anticoagulation is not possible. Approval for the utilization of this exciting and effective new technology in those high risk patients who aren't able to be safely anticoagulated should be approved without additional barriers.

Less

I am an electrophysiologist at the University of Virginia, where I have been implanting Watchman devices starting with the PREVAIL CAP program. I treat patients with atrial fibrillation.

I have some concerns regarding the CMS National Coverage Decision. Specifically, the language stating "contraindication to warfarin" is troubling. While often patients can not tolerate warfarin in the long term due to bleeding, with careful vigilance, these patients can often tolerate short periods of warfarin. With the high rates of appendage closure at the 45 day TEE related to Watchman, the vast majority of our patients are able to stop warfarin at the 45 day mark. I think that a specific stipulation of "contraindicated" may severely limit the number of patients who may otherwise be able to safely get this device implanted.
Furthermore, I have concerns regarding the stipulation of "high" CHADS-VASc scores. American and European guidelines differ, and it would be helpful to have some clarity with regards to what is considered high risk.

Less

This letter is intended to respond to the CMS proposed national coverage decision for Percutaneous Left Atrial Appendage Closure (LAAC) Therapy. I am an electrophysiologist with Summa Health which is a large community hospital system in Akron Ohio. Our practice is hospital owned with a primary focus of delivering optimal healthcare for our patients with cardiac disease as well as development of research and education programs including our cardiology fellowship. Management of patients with atrial fibrillation is a core component of our practice, including heart rate and rhythm control with both medication and catheter based and surgical ablation, as well stroke risk reduction.

I am writing to address several specified conditions outlined in the proposed coverage decision. We participated in the initial clinical trial, Protect AF, evaluating the use of the Watchman Device as an alternative to Coumadin for stroke prevention. In addition, we have now successfully implanted five patients this year since the FDA approval. Although, relatively small, our experience has been positive and mirrors the thousands of patients studied clinical trials and registries over the past ten years. As outlined in the proposal however, the device is also clearly not necessary for patients doing well with oral anticoagulation (OAC) and low long term bleeding risk. Unfortunately, subpopulations of patients with atrial fibrillation remain at risk for adverse events despite warfarin or the novel oral anticoagulant (NOAC) medications. Specifically management of atrial fibrillation remains suboptimal in the following moderate to high risk patients (ie CHADS2VaSC = 2) in our practice:

1. Those with prior bleeding events or underlying conditions which result in contraindication for long term use of OAC
2. Those with increased risk for bleeding due to vocation (or avocation) over potentially many years of treatment with OAC (ie 45 year old manual laborer with high blood pressure and diabetes)
3. Patients with either difficult to control INRs on warfarin or difficulty affording NOAC treatment
4. Patients who would benefit from medications that increase risk of bleeding when used concurrently with OACs (ie NSAID use in patients with severe arthritis or those with vascular disease in need of dual antiplatelet therapy)

The FDA approval statement indicates the patient be deemed by their physician to be suitable for warfarin; and have an appropriate rational to seek a non-pharmacologic alternative to warfarin, taking into account the safety and effectiveness of the device compared to oral anticoagulation. Patients with clinical scenarios such as those outlined above would in my opinion have ‘appropriate rational to seek a non-pharmacologic alternative’. Patient selection should be more clearly defined in the proposal to identify this at risk population who would most benefit from the Watchman device.

Beyond patient selection, the proposal also inadequately addresses perioperative anticoagulation with warfarin. The protocol in both clinical trials, Protect AF and Prevail, required short term use of warfarin leading up to implant and then for minimum of six weeks after implant. Regarding Proposal condition 2: the “contraindication to warfarin” should be restated to allow peri-procedural anticoagulation as defined by trial protocols.

In summary, the proposal should more clearly identify the patients who would uniquely benefit from left atrial appendage occlusion. Specifically, the coverage decision should address those at moderate or high risk of stroke who also have an unacceptable increased risk of bleeding with OAC. As an example, I care for a 70 year old woman with a history of persistent atrial fibrillation with a CHADS2VASC score of 4 (4.8% annual risk of stroke). She has previously had two gastrointestinal bleeds, one with warfarin several years ago and the other with rivaroxaban this past year. She is felt to be at increased risk for rebleeding with long-term oral anticoagulation therapy, but acceptable risk if use is short term (consistent with current Watchman labeling). My patient would clearly benefit from alternative strategies to reduce stroke, and based on clinical trials would be a good candidate for the Watchman device. Implantation of this specific device has been extensively studied with over 2,000 patients enrolled in clinical trials and registries who have now been followed for up to ten years since initial enrollment in Protect-AF. As the experience grows, future implantation of the Watchman device should require active participation in a national registry to track outcomes similar to other recently developed device therapies such as transcutaneous aortic valve replacement (TAVR).

I appreciate the opportunity to comment. Please contact me questions or if I can be of additional assistance.

Tyler L Taigen, MD
Clinical Electrophysiology
Medical Director, Atrial Fibrillation Program
Chair, Clinical Quality Improvement Electrophysiology Section
Summa Cardiovascular Institute
95 Arch Street Suite 350
Akron, Ohio 44304

Less

Dear Ms. Syrek-Jensen:

I am Chair of the Division of Cardiovascular Diseases at Mayo Clinic Rochester, Minnesota, and a practicing structural interventional cardiologist. I have no conflicts of interest related to WATCHMAN or Boston Scientific.

As almost all comments on your website indicate, I would like to extend my strong support for a decision in favor of coverage for the WATCHMAN device. Atrial fibrillation is a massive public health problem and many patients could potentially benefit from left atrial appendage closure. The clinical trial data are robust, have been carefully scrutinized by the FDA on multiple occasions, and the device has been approved for sale. The ultimate decision as to whether or not a specific patient is a good candidate for left atrial appendage closure involves careful consideration of individual patient factors, the potential risks of performing the procedure, and the demonstrated long-term benefits. This is a decision that the patient and physician should be able to make without regard to worries over reimbursement or other extraneous concerns. Just the fact that almost all patients who receive the device are able to successfully discontinue chronic oral anticoagulation is of huge benefit to many patients. Avoiding a potentially dangerous drug, one with many interactions, is a boon for our elderly patients. Of course, there are other patients who may be better suited for anticoagulation; but again, this is a decision that should be made between the treating physician and patient.

In closing, I would like to reiterate my strong support for a national coverage decision for percutaneous left atrial appendage closure and rapid implementation nationwide of a consistent and fair reimbursement policy. Thank you very much and good luck in your deliberations.

Sincerely,
Charanjit S. Rihal, M.D., MBA

Less

Less

Percutaneous Left Atrial Appendage Closure Therapy represents a unique and valuable treatment option for atrial fibrillation patients who are stuck between a risk of stroke and a risk of bleeding. For these patients, there are already ample data to suggest that implantation of this device is a better option than providing no treatment at all, which has been the only alternative to date. To impose approval conditions or restrictions for this group of patients who have already demonstrated a bleeding complication or who are deemed too high risk for bleeding on long-term anticoagulation is to place a barrier that impedes care and adds unnecessary additional risk for a disabling stroke. For such patients, Percutaneous LAA Closure Therapy should be unconditionally approved and reimbursed appropriately.

For atrial fibrillation patients who are not in the very high bleeding risk category (i.e. those for whom anticoagulation remains a viable option, even if not ideal), Percutaneous LAA Closure is a potential alternative therapy with its own risks and benefits when compared to long-term anticoagulation. For patients who have the luxury of weighing these risks and benefits with their health care provider, and then making an informed decision regarding treatment modality, it would be of great importance to gather additional data regarding clinical outcomes for patients who opt for LAA Closure vs those who choose to take a long-term anticoagulant. The financial burden and manpower hours of data collection, however, should not be unilaterally imposed on health systems, especially in the current health care financial climate of "do more with less."

A stroke is one of the most devastating medical events that can occur to a person, robbing them of their independence, their identity, and their dignity, while leaving them alive. We are blessed to have had an effective treatment option to reduce this risk in patients with atrial fibrillation - oral anticoagulant medications. Percutaneous LAA Closure Therapy represents a long-awaited option in patients for whom anticoagulation is not possible - full approval for these at-risk patients should be granted without additional conditions so they have the ability to take action, without any barriers, to reduce their risk for such a grievous complication from atrial fibrillation.

Less

Less

I am an experienced Interventional Cardilogist and have been implanting the Watchman device since it was FDA approved in May of 2015. I am confused about how to qualify a patient based on the proposed NCD. The proposal covers patients who have contra-indication to oral anitcoagulants.

The Watchman device was approved for coumadin eligible patients. The NCD seems at odds with the IFU, please clarify. I have two patient scenarios that I would like clarified. I saw a patient with chronic A Fib who is on dagibitran. He is a fisherman and got a fish hook stuck in his thumb. He had to spend 12 hours in an ER getting the bleeding to stop. He wants to be off OAC. Is he eligible under the NCD?

A second patient is an active 70 yr old man with chronic A Fib on coumadin with no contra-indication. He is an acitve downhill skier and is afraid that if he has a major fall he may have a major bleeding event. He wants to be off OAC's so he can continue a sport he loves. Both patients have valid individual reasons for wanting to be off OAC's. Are they eligible in the current language?

Less

We commend the CMS decision to allow coverage for Left Atrial Appendage Occlusion (LAAO) with an implanted device under Coverage with Evidence Development for patients with non-valvular atrial fibrillation (NVAF). However, we would like to comment on two conditions of coverage in the proposed National Coverage Determination.

Condition 2, bullet 3: The patient has a contraindication to warfarin.

Currently the Watchman™ Left Atrial Appendage Closure Device is the only FDA-approved device for LAA occlusion. The FDA-approved indications for this device include patients with NVAF who are deemed by their physicians to be suitable for warfarin and have an appropriate rationale to seek a non-pharmacologic alternative to warfarin¹. Therefore we do not agree with the proposed NCD requirement to limit coverage to patients with a contraindication to warfarin. The NCD coverage requirements for the patient should align with the FDA-approved indications for the device, and with the referenced clinical studies which offered LAA occlusion as an alternative to warfarin therapy. Additionally, short-term warfarin therapy is required post LAAO. Therefore, we recommend updating this verbiage in the NCD to read, “The patient is at high risk for long-term warfarin therapy”. The patient’s physician would determine anticoagulation risk based on contraindications outlined in the drug prescribing information for warfarin sodium (Coumadin) ² and consideration of the patient’s occupation or lifestyle that may place them at high risk of major bleeding secondary to trauma.

Condition 5: Training should also include a physician who already has experience implanting the device, but who is not a representative of the manufacturer, as well as a minimum of two supervised and two observed cases. A physician “who is not a representative of the manufacturer” may be interpreted in several ways. We understand the physician would not be employed by the manufacturer, but would the physician be contracted by the manufacturer to provide the training and case oversight; or not be under contract but approved/certified by the manufacturer to provide the training and case oversight?

Clarification of this statement will help determine the physician population available for this training. Supervision of cases will require the physician be privileged and credentialed at the facility providing the service. Again we commend CMS for allowing coverage of this important new technology. Thank you for the opportunity to comment on this proposed NCD.

Sincerely,

Ravi Chari

Less

Less

To the Director of Coverage and Analysis Group,

I am a practicing cardiac electrophysiolgist whose primary patient population suffers from atrial fibrillation. It is the most common cardiac arrhythmia and afflicts patients in many ways. The most worrisome, however is the potential for stroke. Risk of stroke is estimated to be 5X greater with atrial fibrillation. After many stagnant years of medical research (warfarin, as the sole option), patients now have access to new oral anticoagulants and in the near future percutaneous left atrial appendage closure. After reviewing the CMS’s Proposed Coverage Decision for Percutaneous Left Atrial Appendage Closure (LAAC) Therapy, I have a few comments.

Overall, I am in agreement with the CMS’s 7 proposed conditions and believe that WATCHMAN technology should be considered in a carefully selected patient population. There are a few conditions that require further clarification. Condition 2 proposes implant of the WATCHMAN in patients with contraindication to warfarin. This proposal is contrary to the inclusion/exclusion criteria for the first trial WATCHMAN trial, PROTECT AF. In clinical practice, I believe patients with relative contraindication have the most to gain from WATCHMAN. One of the patients referred to me for watchman is a pleasant 68 year old female with prior history of TIA, Hypertention and diabetes who was not on anticoagulants due to severe orthostatic hypotension and recurrent falls. This patient with CHADS-VASC score of 6 has over 9% annual risk of CVA. Although she does not have a contraindication to warfarin, she was felt to be at high risk due to falls and was not offered systemic anticoagulation. She is a candidate for short-term anticoagulation and has a tremendous advantage in stroke reduction with WATCHMAN.
In atrial fibrillation patients, treatment with anticoagulation exposes patients to bleeding. Although several Risk-Scores have been used and validated none captures the risk of bleeding adequately. One of my patients, 66 yom with CHADS-VASC score of 3 is an avid soccer player. He has a low HAS-BLED score, however, is at a potentially life threatening bleeding risk with systemic anticoagulation.

As it pertains to condition 2, it is my opinion that the benefits of watchman are cumulative and should include patients with high risk of CVA (CHADSVASC >2) who have a reason to seek alternative treatment to systemic anticoagulation.

I believe strongly in careful monitoring of new therapies and having accurate data. I am excited to hear about the national registry proposed in condition 6. The practical implications of the current proposed registry however are significant. I work for a private practice group with patients coming from several different hospital networks. A registry of all AF patients, who are on oral anticoagulation would be difficult if not impossible to maintain by one provider or hospital system. There is a significant overlap of patients across different healthcare systems and data collected on all patients by one hospital that offers WATCHMAN therapy would be inaccurate. Fortunately, with the recent publication of new oral anticoagulant trials we have large sets of data on risks of CVA and bleeding complications. WATCHMAN registry should include inhospital efficacy (successful implantation) and complications, so the therapy can be offered at highly qualified centers to assure patient safety.

Overall, I am excited for my atrial fibrillation patients who will have access to a new therapeutic option for stroke reduction.

Less

To whom it may concern:

I am writing in response to the CMS position on percutaneous left atrial appendage closure therapy. I am a cardiac electrophysiologist who has been in practice for over 10 years at both a major academic medical center (Duke University ~10 years) and more recently one of the busiest cardiac electrophysiology practices in the world (Texas Cardiac Arrhythmia). My practice is dedicated to the treatment of complex arrhythmias particularly with catheter ablation and the majority of my patients suffer from atrial fibrillation. I have been performing catheter ablation for atrial fibrillation for over 12 years and presently perform approximately 320 ablations for atrial fibrillation annually.

I participated in the preliminary studies to show safety and efficacy of the WATCHMAN LAA occlusion device and continue to perform these procedures at my practice locations in Austin, Texas and Raleigh, North Carolina. I am pleased to see that CMS has issued a statement on percutaneous LAA closure therapy as this device satisfies a previously unmet need in a large number of patients suffering from atrial fibrillation.

I feel that the coverage criteria that CMS has come up with is a good foundation, but some of the wording could result in many patients who would otherwise benefit from this technology being denied based on payors interpretation of these criteria. For example, the lack of a precise definition for "high CHADS2 or CHA2DS2-VASc score" could result in patients being denied simply based on an individual interpretation of what that definition is. There were distinct enrollment criteria for patients who participated in the trials leading up to this decision and there are also published criteria to suggest which patients require anticoagulation based on CHADS2 and CHA2DS2-VASc scores. It would only seem appropriate to follow along with those distinct criteria (CHADS2 >/= 1 or CHA2DS2-VASc >/= 2). It is also left up to interpretation what a "contraindication to warfarin" is and more specific language should be used to define this to include, but not limited to: prior GI bleeding on warfarin, history of falls, history of spontaneous intracranial hemorrhage, allergic reaction to warfarin, inability to maintain a therapeutic INR, lack of access to a health care provider to manage warfarin, intolerance due to side effects such as GI upset, difficulty with medical compliance, etc.

CMS has also decided to incorporate HAS-BLED scores into these criteria, again, left to interpretation and not used nearly as routinely for making decisions requiring anticoagulation. If CMS elects to leave this criteria, a more specific definition of a “high” HAS-BLED score must be given.

The therapy of left atrial appendage occlusion/exclusion is the only existing alternative to warfarin and NOACs and has been proven in randomized trials to be safe and efficacious. There are over 3 million people in the U.S. with atrial fibrillation and the majority of that population carries a significant risk for stroke. It is imperative that as physicians who care for these patients that we make sure that ALL patients who need this therapy have access to it.

Less

Less

There is an ever increasing population of aging Americans with atrial fibrillation who have gait or balance disorders that put them at risk of life threatening bleeds from minor injuries simply because they are on anticoagulants. Additionally, there are many patients with a history of repeated GI or GU bleeding who cannot be placed on anticoagulation.

Less

In reference to the proposed coverage decision for the Watchman device, I would like to make a few comments regarding my clinical experience and practice.

I am an Interventional Cardiologist at Deborah Heart and Lung Center in New Jersey, and am part of the heart team at our institution. We have used the Watchman device in appropriately indicated patients for the prevention of left atrial appendage thrombus with good procedural success and clinical safety.

Although I agree with most of the proposed 7 conditions for coverage, condition 2 addressing coverage criteria needs attention. I agree that a patient with a “high” CHADS2 or CHADS-VASc score would benefit from therapy, however, any patient with a score > than 1 is at risk for thrombo-embolism with subsequent TIA/CVA. The term “high” is a vague term, and should reflect the fact that if a patient is at risk of stroke (CHADS>1), which is high enough to recommend chronic anticoagulation, he may be a potential candidate for LAA closure. I also agree that the HAS-BLED score is a useful predictor of future bleeding risk, but in the same light, a score of 3 reflects a 6% risk of a bleeding event (considered high risk). Therefore, a patient may be at high risk for stroke and bleeding without having dramatically elevated CHADS2 and HAS-BLED scores. These factors should be considered when evaluating potential candidates. Additionally, strictly limiting the Watchman device to patients with a contraindication to Warfarin may limit its benefit in certain patient populations. For example, patients with labile INR’s who are either sub-therapeutic (with risk of CVA), or supra-therapeutic (with risk of hemorrhage), may be ideal candidates to come off of chronic anticoagulation with LAA closure.

Furthermore, the PROTECT AF and PREVAIL studies showed that the implantation procedure can be done safely in the right hands, with a low risk of complications. Ultimately an extended discussion between the clinician and patient should be held to review the risks and benefits of implantation.

I agree that a Heart team approach comprised of Interventional cardiologists, EP specialists, and General cardiologists continue to work together in evaluating potential candidates for implantation, as we do in our institution. I believe this technology offers our patients a reasonable, safe option to chronic anticoagulation with Warfarin, and would ask that you consider my recommendations.

Thank you.

Vincent Varghese, DO, FACC, FSCAI
Director, Interventional Cardiology Fellowship Program
Deborah Heart and Lung Center

Less

I am writing on behalf of physicians and patients. As an electrophysiologist, there is a significant need alternatives to anticoagulation in order to treat patients with non-valvular AF at high risk for stroke. While novel anticoagulants have helped offer an option in patients with difficulty dosing warfarin or tolerating long-term anticoagulation, discontinuation rates for these drugs remain significantly high. Prior to the availability of this technology, patients who have to choose between high risk of bleeding or high risk of CVA. LAA closure with the WATCHMAN device is a third option that has been demonstrated in randomized controlled trials to decrease risk of CVA similar to that of warfarin without long-term risk of bleeding once off anticoagulion required post implant. Further, the data demonstrate less significant CVA severity as hemorrhagic CVA tends to be more significant then embolic. This is a significant medical advance for patients and a needed option.

Currently, the proposed NCD has been written in a way that limits implants to in whom warfarin is contraindicated. While admirable, as these patients have little other option, this ignores both all randomized trial data, FDA approval indications, and the FDA DFU. In fact the FDA SSED clearly summarizes data which has always included patients who are warfarin eligble. Therefore, the current NCD would ignore all data for the WATCHMAN device which was studied as an alternative to anticoagulation in patients who were warfarin eligible. Warfarin is required for 45 days post implant based on PROTECT AF and PREVAIL.

I would ask that CMS reconsider such a limitation and align a NCD with the available randomized control data, FDA approval indications, and patient needs.

Less

I am a cardiac electrophysiologist at VCU in Richmond VA. I have been treating atrial fibrillation for 20 years and have been involved in the Lariat procedure and more recently implanting the Watchman device. Many patient have difficulty taking oral anticoagulants. I have seen several major bleeds and hemorrhagic strokes. In patient who are difficult to anticoagulate I have seen strokes from subtherapeutic INR. Patient with renal failure may not be able to take NOACs.

Our center has years of experience with transeptal puncture for atrial fibrillation ablation and this is key for the Watchman procedure. We have had excellent results so far and have been very impressed with the device. The implant is not physically hard on our patients, In patient at high risk of bleeding on warfarin or in patients with inability to keep a therapeutic INR I believe the Watchman is an excellent choice to decrease stroke risk.

I agree patients should have significant stroke risk (CHA2DS2-VASc score)to benefit from the device. However the HAS-BLED score can be low and the patient still at risk of serious bleeding if they have wild INR swings or have frequent falls

I agree the procedure should only be done in established centers with extensive experience in transeptal punctures. The procedure should only be performed (as with all procedures) after extensive discussion with the patient concerning all risk and benefits

The Watchman is a wonderful addition to the treatments for stroke prevention in atrial fibrillation,

R K Shepard, MD

Less

To whom it may concern

We were excited that the LAA occlusion device was finally FDA approved a few months ago. There are many patients who are in desperate need for alternative to stroke prevention. In fact, we had a long list of patients who were not on blood thinners for various clinical reasons awaiting this approval.

At Spectrum health, we have implanted 16 devices thus far successfully. When assessing your proposed criteria, I have two reservations.

This device is approved for patients in need for an alternative to blood thinners. Under condition two, you are demanding a high HAS Bled score. While I fully agree that a high CHADS-Vasc score is needed, and that there should be a clinical concern not to use Warfarin long term, I disagree with the wording of contraindication to warfarin. As you well know, use of Warfarin is needed short term to receive a Watchman device. Also, adopting the HAS-Bled score as an absolute measure of risk of bleeding is also short sited. I am the co-director of the anticoagulation clinic at Spectrum, and manage the coumadin of 3000 patients. I do not use the HAS-Bled score clinically due to it's limitations. We have implanted 16 patients successfully since the Watchman was FDA approved last spring. Four of the patients we implanted would not be approved under your proposed criteria. Allow me to present them briefly so you can see that the Watchman made perfect clinical sense.

Two of the implants were in female patients ages 60 and 55 with a brain AVM, and brain aneurysm respectively. The brain aneurysm was coiled but continued to have a large residual aneurysm. Both of these patients did not have a bleeding complication yet. They both had HAS_BLED scores of zero. They both had CHADS-Vasc scores of 2 with PAF. They were not anticoagulated for fear of a catastrophic brain bleed, and fortunately both underwent successful Watchman implants tolerating the risk of Coumadin for a few weeks. Another clinical scenario was a patient who had two falls with large thigh hematomas disabling him for weeks at a time. His bleeding was not life threatening but prohibited his doctors from placing him on blood thinners. His HAS-Bled score was 1, which would disqualify him from the Watchman under your proposed criteria. My final example is of a patient with recurrent bleeding from his ostomy site despite trying Coumadin, Eliquis and Xarelto. His bleeding is not life threatening, but was again quite disabling.

My second reservation is for your call to include patients who do not receive a Watchman but stay with OAC in a national registry. This will be impractical for most centers and add significant cost. Currently, the majority of my consults for Watchman consideration do not receive one. However, they usually are referred to our tertiary center from other institutions often from other cities. We will not be able to keep track of all those patients and include them in a registry. It will add significant cost to our health care system. This was already done in the PROTECT_AF trial where patients were randomized.

I hope you find these comments useful. I urge you to adopt the language of the FDA indication for coverage. An educated discussion between the patient and his clinician should drive the decision on needing such therapy. On behalf of my patients awaiting coverage for this much needed technology, thank you for your time.

Less

We have read with interest the proposed CMS proposed decision memo on left atrial appendage closure. While we appreciate CMS’s concern regarding misuse of the technology and CMS’s role in safe guarding patients from potential harms of new technology I wish to comment on two issues.

Condition 2: Lists that the three conditions: high risk of stroke and high risk of bleeding and contraindication to warfarin have to be met for LAA occlusion to be reasonable. However this scenario would exclude patients who have a high CHADSVASC score and have had major intracranial/GI bleeding on Warfarin. Such patients will not have a high HAS BLED scores and as such would only satisfy two of the conditions. These patient would have a high risk of re-bleeding and would clearly benefit from LAA occlusion that would decrease the risk of ischemic stroke and obviating the risk of warfarin.
As such I propose to change condition 2 to High CHADSVASC in addition to a high HASBLED or contraindication to long term warfarin use.

Condition 6: I agree that a LAA occlusion registry is warranted to track outcomes. Mandating a contemporaneous registry of patients on ORAL anticoagulants is beyond the scope of this type of follow-up and will pose a large burden on hospitals. Prospective Long term follow-up of such patients is very laborious and expensive. In due time research will cover this subject in retrospective analyses which are much more feasible.

Less

The contraindication to oral anticoagulation requirement is concerning for me as a practioner.
Risk stratification using the HASBLED score (2 or higher) appears more reasonable:

Clinical Scenario:
82 year old female with long standing persistent atrial fibrillation. HTN, DM and prior history of stroke who is currently taking aspirin therapy (not oral anticoagulation due to frequent falls). She has a high CHADSVASC score, HASBLED score of 3. We are concerned about her falls risk and her family is concerned about intracranial hemorrhage on oral anticoagulation. With current NCD I am not allowed to offer her left atrial appendage occlusion.

Clinical Scenario 2:

68 year old man with persistent atrial fibrillation, hypertension and diabetes mellitus. He refuses to take oral anticoagulation despite multiple physicians advising him to take oral anticoagulation. His mother died of intracranial hemorrhage. Under current NCD I would not be able to offer him the watchman device.

Less

Less

I am a cardiac electrophysiologist who has been in active practice for 15 years. My practice focus is atrial fibrillation and, as a result, I care for thousands of patients with this condition. To date, I have implanted 3 Watchman devices and have been impressed with the device performance and its ability to completely occlude the LAA.

Regarding the proposed CMS guidelines for approval, I have the following concerns:

1- The wording currently states that qualifying patients have "high" stroke and bleeding risks. This is unacceptably vague. I favor more definitive wording which is structured so as to facilitate physician discretion. Specifically, I would propose elimination of the HAS-BLED criteria (allowing inclusion of patients who have high risk lifestyles that preclude Coumadin use) and CHADS-VASc >= 2 (patients that require anticoagulation).

2- Regarding the control arm in the registry, it should be noted that most Watchman implantation locations are referral centers. As such, warfarin monitoring may be performed by outside physicians. This will make compliance with the control arm of the registry overly burdensome and it should therefore be eliminated.

Thank you for considering my comments.

Less

Paul A. Friedman, M.D.
Cardiovascular Diseases
507-255-2446 Fax 507-255-2550
friedman.paul@mayo.edu

December 7, 2015

Director, Coverage and Analysis Group
Re: LAAC Therapy CED
Centers for Medicare & Medicaid Services (CMS)
7500 Security Blvd., Mail Stop S3-02-01
Baltimore, MD 21244-1850

Dear Director:

Thank you for the opportunity to comment on the proposed decision for percutaneous left atrial appendage closure (CAG-00445N). I have reviewed the seven conditions detailed and would like to propose some comments below.

1. I agree with the requirement that the patient have a high CHADS2 score or CHADS-VASc score, as these patients are at risk for thromboembolism and thus may stand to benefit and have been demonstrated to benefit in clinical trials from implantation of a WATCHMAN device in the left atrial appendage. I do have a deep concern about mandating an absolute contraindication to warfarin. In all of the published clinical trials to date, patients have been treated with warfarin for 45 days following WATCHMAN implantation. Promoting a practice that involves implanting the device in patients who are unable to receive warfarin, while an interesting area for further investigation, would eliminate the portion of the population that indeed has been found to benefit. Additionally, while I favor the use of a registry (detailed below) and that that registry should include recording of the HAS-BLED score, mandating a specific HAS-BLED score for use of the device for reimbursement does not appear warranted at this time in my opinion based on available evidence. This score has not been used to create the data that ultimately included improved long-term mortality benefit, and non-inferiority, and thus while recording it as important registratry information is logical, using it for patient selection seems premature.

2. I agree that the procedure should be furnished in a hospital that meets the institutional requirements as outlined in your coverage memo. To perform the procedure safely, the availability of a high degree of expertise and supporting personnel does indeed seem prudent.

3. I furthermore agree with the requirements regarding the qualifications of the physicians to ensure that the procedure be performed safely, and additionally agree with the suggested requirements for training.

4. The concept of enrolling the patients in a registry is an excellent one, which would add additional information to our current database to help insure this technology is used effectively. However, the requirement for contemporaneous patients followed on oral anticoagulation requires careful consideration. It is quite likely that these patients would have important different clinical characteristics than those undergoing closure, and careful consideration would be required to determine whether the non-closed cohort is similar to the WATCHMAN cohort to permit meaningful comparison.

5. The concept of requiring formal shared decision interaction between patient and provider using an evidence-based decision tool and anticoagulation is consistent with our current practice is an excellent concept to promote high standards of care, and I fully support.

I appreciate the opportunity to provide comments.

Sincerely,

/s/
Paul A. Friedman, MD, FHRS
Vice-Chair, Cardiovascular Medicine
Vice-Chair for Academic Affairs and Faculty Development
Medical Director, Remote Monitoring
PAF/brs

Less

December 7, 2015

Ms. Tamara Syrek Jensen, J.D.
Director, Coverage and Analysis Group
Re: LAAC Therapy CED
Centers for Medicare and Medicaid Services (CMS)
7500 Security Boulevard, Mailstop S3-02-0l
Baltimore, MD 21244-1850

Dear Ms. Syrek Jensen:

I am an interventional cardiologist in active practice at Mayo Clinic, Rochester, Minnesota. I have been involved in structural cardiac intervention as well as interventional cardiology for 35 years. I am writing concerning the proposed decision memo for percutaneous left atrial appendage (LAA) closure therapy (CAG-00445N) decision summary. I would like to write regarding two aspects of this summary with which I disagree.

Sincerely,

Less

Regarding Proposed Decision Memo for Percutaneous Left Atrial Appendage (LAA) Closure Therapy (CAG-00445N)

Recognizing that there is clinical judgement and patient choice that is incorporated into the decision to do a procedure on any inidividual patient, I think it is important to word the indication for the procedure as accurately as possible.

First, # 2 states that there should be a high CHADS2 or CHA2DS2-VASc score. I think that a CHADS2 or CHADS2-VASc score that is high enough to justify anticoagulation for stroke prevention is also high enough to consider implantation of a Watchman device in patients who also have a relative contraindication to anticoagulation. In the Protect Trial and in the current ACC/AHA guidelines anticoagulation is actually recommended for patients with a fairly low CHADS2 or CHA2DS2-VASc score.

Second, the patient's clinical history of bleeding is much more important than a high HAS-BLED score. The HAS-BLED score is an algorithm to predict the future risk of serious bleeding. However, there are patients with low HAS-BLED scores that have a history of serious intracranial bleeding or recurrent gastrointestinal bleeding who are appropriate candidates for the Watchman device. A high HAS-BLED score should not be a requirement for implantation, if the patient has a history of serious bleeding, a bleeding disorder, or an indication for dual antiplatelet therapy.

Third, #2 also states there should be a contraindication to warfarin. This statement is misleading in that the FDA approved the device with a recommendation for 45 days of anticoagulation with Warfarin after implantation of the device. It is patients with a strong contraindication to chronic anticoagulation therapy but not an absolute contraindication to anticoagulation who are selected for implantation of the Watchman device in the United States. As you know, aspirin and clopidogrel are perscribed after implantation of the Watchman device in Europe. There is a large registry which suggests that dual antiplatelet therapy after implantation of a Watchman device is safe. I suspect that in the United States, patients with contraindications to chronic anticoagulation and in the near future those with an absolute contraindication to anticoagulation will be good candidates for the device.

Finally, I am in favor of registries after approval of a new technology. However, I would request that the registry be kept relatively simple. It should be designed to track procedural complications and a very few well defined clinical endpoints. You can't repeat the Protect Trial.

Thanks for your consideration of these ideas

James Hopkins MD

Less

Thank you for the opportunity to comment on the proposed coverage decision for percutaneous left atrial appendage (LAA) closure therapy. I am a practicing interventional cardiologist with a special interest in structural heart disease. I received my MD degree from the University of Pennsylvania, my internal medicine training from the Brigham and Woman's Hospital, and my cardiology and interventional cardiology training from the Massachusetts General Hospital.

While the majority of the conditions for coverage are reasonable, I believe that some need to be revised.

Firstly, the statement that Watchman patients must have a contraindication to warfarin directly contradicts the FDA mandate that such patients be suitable for warfarin therapy. Indeed, all the patients enrolled in the randomized PROTECT AF and PREVAIL trials were specifically required NOT to have a contraindication to warfarin.

Secondly, the stipulation that Watchman patients have a high HAS-BLED score has no scientific basis. None of the clinical trials assessing the Watchman device had a HAS-BLED score requirement. While the HAS-BLED score has some usefulness in predicting near-term risk of bleeding complications in patients receiving oral anti-coagulation therapy, there is no evidence to support its utility in discriminating the risks:benefits of LAA closure. The primary driver of the net clinical benefit of the Watchman device over oral anti-coagulation therapy stems from the continuous, lifelong risk of bleeding inherent in chronic anti-coagulation strategies. As such, it is younger patients, who often have lower HAS-BLED scores and lower CHADS / CHADS-VASC scores, who have potentially the most to gain from LAA closure. Without LAA closure, such younger patients will be subject to the risks and costs of many years of oral anti-coagulation therapy. Thus these are the very patients in whom the Watchman device will likely have the most clinical benefit and cost effectiveness.

I support the decision to require a national registry for LAA closure therapies. However, I do not understand how CMS can require a control arm of long-term oral anti-coagulation patients in order to reimburse an FDA-approved therapy that has been thoroughly studied in multi-center, randomized, controlled trials. Who will pay for the enrollment and data collection for this control arm? I believe this issue needs further clarification and reconsideration.

Thank you in advance for thoughtfully reviewing theses comments.

Less

Less

There are several very concerning issues regarding the CMS NCD coverage decision for LAA closure. As I am sure that the decision makers are aware, the contraindication to warfarin therapy requirement directly contradicts the trials and requirements of LAA devices. Warfarin is used for the first 4-6 weeks with each of the devices until the physician is able to ensure that the closure was adequate. It would make more sense to state that the patient should have a relative contraindication to anticoagulation therapy or a clinical preference to not be on full anticoagulation.

The concept that the registry must include a oral anticoagulant arm to serve as non-interventional controls is poorly conceived. These patients are presumably not going to be randomized and therefore have inherent selection bias. It is also difficult to understand where these patients are going to come from, as consideration for appendage occlusion usually means that the patient already has significant comorbidities and appendage occlusion may be their only true mechanism of stroke prevention. Are we being asked to deny a proven therapy to a subset of patients just to follow them and see how they do? This was the purpose of the clinical trials leading to the device's approval in the first place. Perhaps it would be more valuable to include screening failures in the registry, rather than deny therapy to patients.

Thank you for your consideration.

Less

Several elements of the proposed coverage rule for left atrial appendage occlusion might be modified.

The second condition stipulates a high HAS-BLED score. This score is not clinically validated and thus it is difficult to justify this requirement. Furthermore the other 2 conditions, a high CHADS2 score in conjunction AND a contraindication to warfarin encompass the entire target population intended in the decision. Some patients who have a bleeding episode without a high bleeding score would be excluded from consideration for LA a device therapy as the decision was currently written. Thus, there is no negative impact to deleting the HASBLED requirement, and an important group of patients will not be excluded from having the option for left atrial appendage occlusion. An example would be a 75-year-old diabetic patient (with thus a CHADS score of at least 2) who has no other significant medical history suffers a a spontaneous bleeding episode.

The sixth condition stipulates a oral anticoagulant registry. The stated purpose is to serve as a control group. Oral anticoagulant tolerant patients will clearly be fundamentally different than device candidate patients. It is difficult to understand what “control” purpose would be served. This represents spoor science. Furthermore the outcomes from oral anticoagulant tolerant patients are well documented in all of the approval trials for the NOAC agents as are outcomes from warfarin. The well-established data from these multiple trials could easily serve the "control group" function without adding a remarkably burdensome additional requirement to the use of this therapy.

Less

I am an interventional cardiologist in a practice that treats more 3000 thousand patients with atrial fibrillation. There stroke risk is known to be more than 5 times without anticoagulation. I educate every single patient on anticoagulation but still 1/4 patients will not be on it for a variety of reasons which include 1.) cost 2.) bleeding 3.) inconvenience. The cost of bleeding and disabling stroke to the health care system is difficult to measure but is likely similar to heart failure costs. In my opinion, this device should not be considered " not reasonable and necessary to treat an illness or improve function and therefore not covered under the social security act". I find this terribly disappointing for my patients. The stroke and bleeding risk in these patients are real and there is a device (which is being used in Europe regularly in appropriate patients) that can help them. Please reconsider you decision. Atrial fibrillation patients will thank you.

Less

As an interventional cardiologist and Medical Director of the Cardiac Catheterization Laboratory at Abington Hospital (of Abington Jefferson Health), I applaud CMS for its proposal to cover LAAC using the WATCHMAN Device. We began implanting WATCHMAN in August, 2015, and our experience has been quite favorable so far. The patients have been carefully selected and the outcomes have all been positive. However, having reviewed the language of CMS's proposed coverage decision, I am concerned about 2 of the conditions which must be met: condition 2 and condition 6, under the current proposal

For condition 2 to be met, a patient must have a "contraindication to warfarin." While all of the patients I have treated to date meet this condition, by virtue of having had prior bleeding complications while taking oral anticoagulation, the word "contraindication" is too restrictive. I have a patient (a Medicare subscriber) who only recently recovered from a massive thigh hematoma while taking warfarin which developed while simply bending over; during the period of warfarin interruption he unfortunately suffered a stroke. This gentleman is now back on warfarin - which is not necessarily "contraindicated" at this juncture - but he has an OBVIOUS reason to seek an alternative to long term oral anticoagulation. My point is that the FDA-labeled population is a broader population than the "warfarin-contraindicated" population mentioned in the CMS proposal - and includes other patients of mine such as the skier who interrupts anticoagulation during winter months, and the horse-shoer whose livelihood depends on an alternative to long term anticoagulation.

Condition 6 is also problematic. While I fully support participation in a national registry - and at my center, we are certainly equipped to participate in such a registry as we have for our TAVR program - collecting data on a control arm of patients on OAC seems senseless. How would the control group be identified? Would the control patients be selected from an EHR database? Would only newly initated warfarin-treated patients be included? The control group would predictably be unmatched in terms of any relevant baseline characteristics. I would evision that even attempting to control for these variables would render any conclusion reached unscientific, and therefore it would be a waste of time and resources to require this control group be part of the registry analysis.

The other conditions mentioned in the CMS document are critical in my opinion. Sufficient physician and ancillary staff training - and at my institution a formal failure mode analysis - were prerequisites for WATCHMAN program initiation. Documentation of a shared decision making discussion between patient and clinician is already occurring, and indubitably important on multiple levels.

Thank you very much for your consideration of my opinions.

Less

I am somewhat troubled when considering clinical application of the NCD for left atrial appendage closure (LAAC. First, is the requirement for high HAS-BLED Score. While this score has been validated and compares favorably with other stratification tools for bleeding, it has not been validated in all pertinent cohorts yet it is to be used for all cohorts when determining whether LAAC. There are many where the HAS-BLED score does not define the totality of short and long term bleeding risk. In significant cohorts, the risk for bleeding is "not high" as assessed by a predictive model but there still is substantial risk for stroke, such as those who require dual anti platelet therapy for coronary stent placement in addition to OAC or those with more moderate HAS-BLED Scores but where the problem is ongoing bleeding,

Predictive scores were never designed to determine unilaterally determine whether OAC (for stroke prevention) yet these seem to be in the NCD to determine whether a procedure for stroke prevention can be performed without any clinical judgement as to the the equipoise between bleeding and risk for stroke and whether these scores even apply to the individual patient. In fact the coverage decision notes for OAC: "...it is difficult to determine with precision the absolute stroke risk in patients with NVAF, determining when the benefit from reduced stroke risk outweighs the harm of increased bleeding is likewise difficult. In these circumstances, patient preferences and physician judgment become especially important.” so it is unclear why that isn't the case for LAAC offered for stroke prevention when the risk of bleeding is deemed to be too high for OAC. I suggest this requirement be removed completely in lieu of formal documentation of the reasons why the clinician believes bleeding risk precludes OAC but stroke risk remains substantial.

Similarly, requiring a contraindication for warfarin does not allow patients with relative contraindications to warfarin to have access to this therapy and instead must be subjected either to an increased risk of bleeding or inadequate protection against stroke. Again, patients on dual anti platelet therapy who may require this in the long term, those with moderate renal failure (Creat < 2.3), social barriers to maintenance of OAC in therapeutic range and numerous other factors. I would suggest inclusion of acceptable relative contraindications to OAC be inserted rather than absolute contraindications only.

Thank you for the opportunity to comment.

Less

Currently I am Associate Chief of Cardiology at a tertiary hospital center. I spend 40% of my time in the EP lab performing complex procedures. The Watchman device can be delivered easily and safely in the vast majority of patients with non-valvular AF. The safety of this device is supported by >10 years of experience at European Centers and numerous clinical studies. Regarding CMS's proposed coverage decision several comments are in order, specifically regarding Conditions 2 and 6:

Condition 2: Specifically, requiring patients to have a high HAS-BLED score and a contraindication to Warfarin will eliminate a huge patient population that will potentially benefit from the Watchman device:

-A contraindication to warfarin conflicts with the device IFU as well as the protocol that resulted in the FDA device approval, which requires 45 days of warfarin post implant and up to 6 months of warfarin if a complete seal of the LAA is not seen on TEE at 45 days.
-There are numerous clinical scenarios where warfarin is not contraindicated by strict definition, but where it is to the patient's benefit not to be on warfarin:
-patients with difficult to control INRs
-patient who are non-compliant with medications
-patients, who because of age, mobility, transportation or access issues cannot have their INR properly monitored
-patients who are non-responders to warfarin: recent results from pharmacogenomic testing confirm that >30% of patients treated with warfarin are non-responders to the drug (Fawcett et al; Mercy Cardiovascular Institute Experience, Canton Ohio; Poster presentation and abstract)
-patients who are at a high fall risk
-A Low HAS-BLED does not identify all patients who are at a higher risk of bleeding:
-patients with CAD s/p percutaneous intervention with DES placement requiring ASA and a platelet inhibitor: "triple therapy" with an anticoagulant clearly increases their bleeding risk, yet there are no clear guidelines as to how to manage these patients.
-patients who are chronically on NSAIDS or steroids placing them at increased risk for major GI bleeding even in the absence of anticoagulant therapy
-patients

Less

Hello.

I am a cardiac electrophysiologist that treats hundreds of patient with atrial fibrillation every year. I have watched eagerly the development of percutaneous left atrial appendage closure therapy and was involved in publishing one of the initial series of patients treated with the Lariat suture exclusion of LAA system. After reviewing the totality of the literature on Watchman, I am convinced it is an excellent alternative to anticoagulation in patients with atrial fibrillation and both an increased risk of stroke and bleeding. I agree entirely with the stated FDA indications which largely match inclusion criteria of PROTECT AF which showed non-inferiority to warfarin for stroke risk reduction and a significant decrease in hemorrhagic stroke, disabling strokes, and CV or unexplained death (at the 5 year report).

I agree with most of the 7 proposed coverage conditions from CMS. However, I feel that the requirement for a contraindication to warfarin inappropriately restricts this needed therapy from a significant group of patients that need it. I would advocate Watchman be used only in patients with a clear and strong rationale to prefer non-pharmacologic alternative to warfarin for stroke risk reduction. Of course, people with an absolute contraindication to warfarin like intracarebral hemorrhages and recurrent life threatening bleeds would be the best candidates and ones to gain the most benefit from this. But I have many patients with relative contraindications or other strong reasons for preferring Watchman over oral anticoagulation, and I believe that patients and their physicians should be able to choose the best therapy for those patients. For example, [PHI Redacted] has atrial fibrillation. He has hyperention and is 68 years old, so his CHADS-VASc score is 2, giving him approximately a 2% risk of stroke per year off anticoagulation. He has had multiple nose bleeds while on warfarin. These are not life threatening or a clear contraindication to warfarin, but the fear of not knowing when the next one will happen and the disruption they cause to his life when they happen have become major issues for him. He has left business meetings and had to interrupt vacations, etc. to go to urgent care centers, etc. For many patients with atrial fibrillation, living with anticoagulation and their fear of stroke are their biggest concerns, and they alter their lives significantly because of that fear. Patients frequently come to my office on anticoagulation with hematuria, easy bruising and hematomas that degrade their already weak and frail skin. And most of them live in some degree of fear of what happens if they fall, or if their blood pressure becomes too high, etc. Also, I have many patients in their 50s or 60s that still have very active lifestyles. For example, I have a patient who is avid cyclist and still races at a competitive level nationally. He has a CHADS-VASc score of 3, but refuses to take anticoagulation because he occasionally gets in wrecks while he cycles. Based on the FDA indications, I could have a conversation with all these patients and discuss the risks/benefits of Watchman vs long term anticoagulation or just living with their stroke risk. Based on the proposed CMS conditions for coverage, I may not be able to offer them Watchman. I am not sure what qualifies clearly as a “contraindication” to warfarin. Does one GI bleed or first episode of hematuria qualify? Would [PHI Redacted], with is epistaxis, qualify? I certainly wouldn’t consider epistaxis or riding bicycles a real contraindication for warfarin, but I am certain that all the patients I referenced above would strongly prefer Watchman over anticoagulation, and I suspect that since all of them have an expected life span of greater than 5 years, that in the long run, they would be better off with Watchman than with warfarin or other oral anticoagulants. The bike rider and other patients that are at high risk of stroke that just refuse to take anticoagulation due to fear of bleeding, difficulty complying or affording anticoagulation or frequent monitoring would certainly be better off with Watchman than they would with no anticoagulation, which is what they are on now. Again, those patients may not have “contraindications” to warfarin, but they all have a clear rational for avoiding it. Ultimately, I think CMS is putting patients and physicians in a very tough spot forcing us to make a black/white determination of what is a contraindication or not to warfarin. There are many people living in a grey area that deserve the right to hear all the pros/cons of all options and then make an educated decision that makes sense to them. Additionally, it seems contrary to customary evidence based medicine principals and prior policy decisions to study and show noninferiority/superiority for a therapy and then only authorize its use in a subpopulation. Finally, I feel extraordinarily uncomfortable dealing with the contradictory FDA labeling that states Watchman is only indication in patients that ARE warfarin candidates, but CMS insisting it only be used in patients that ARE NOT warfarin candidates. Physicians are stuck in a situation where we are either using a product off-label or at risk of non-payment or medicare fraud. I hope that CMS removes the condition that warfarin be contraindicated and replace it with language that matches the FDA’s indications.

My second comment is regarding the registry requirements. I agree with CMS that this new and invasive procedure needs additional real-world data to help confirm or adjust our understanding of how to safely and effectively use this technology. I fully support the use of a registry at all implanting centers to track which patients are receiving the therapy, acute complications, and long term outcomes. However, I believe the requirement to have ongoing monitoring and tracking of patients that do NOT get the procedure as a control arm is unnecessary and overly taxing on implanting physicians and systems. We already have plenty of data on the expected outcomes of patients with atrial fibrillation on warfarin, novel agents, and off anticoagulation. All we really need is data on what happens to patients that receive the Watchman. Also, I am confident that additional studies will be preformed both by Boston Scientific and larger centers that are well done randomized trials or large series comparing Watchman to the various oral anticoagualants. We will see plenty of data on these subjects in the coming years. Asking every implanting site to essentially perform prospective studies tracking data on both implanted patients and similar patients that do not get Watchman will require considerable, time, effort, and expense, and some implanting centers will not be able to keep up, or will do a poor job with the data, which potentially contaminates the data and clouds our understanding. I want a prospective, national registry of Watchman implantations, but am strongly opposed to including patients that do not get Watchman in the registry.

Thank you for your ongoing work on the Watchman coverage determination and for your willingness to read about some of my concerns and suggestions.

Sincerely,
Stephen May, MD, FHRS

Less

Less

Dear Ms. Tamara Syrek-Jensen,J.D.

On behalf of the physicians and patients we serve at the Saint Luke's Mid America Heart Institute in Kansas City, I would like to take this opportunity to comment on CMS's Proposed Coverage Decision for Percutaneous Left Atrial Appendage Closure using the Watchman device. I have been personally involved with this device for now over 10 years, having implanted my first Watchman in 2005 as part of the Protect AF trial. I have implaned over 80 devices as part of the trials and registries upon which the FDA recently approved the device. Since commercialization, we have performed this procedure in 31 patients with excellent outcomes.

I am in agreement, in principal, with all 7 proposed conditions. That said, I believe greater clarification of the "eligible patient population" is in order. I would like to share with you the perspective I have formed over my decade of experience with this device. There are 4 distinct scenarios where I believe this device should be used. Broadly, all of these fall under what I would term "A CONTRAINDICATON TO LONG TERM THERAPY WITH ORAL ANTICOAGULATION".

1) The most important is the population of patients that have already had a significant adverse bleeding complication on an oral anticoagulant. These patients typically STOP the therapy leaving them UNPROTECTED. LAA closure with the Watchman device provides a non-pharmacologic therapy that reduces their risk of reccurent stroke by at least 70% (based on imputed placebo analytics shared at FDA panel #3), and is similar in effectivenedd to OAC .

2)Patients that have not yet had a documented adverse bleeding event, but at are extreemly high risk therof. The HAS-BLED score is a reasonable tool to assess this, however there must also be an allowence for important clinical and social factors that increase bleed risk not embedded in the HAS-BLED score. Although these risk models are useful and the best we have, a C-statistic of 0.6 makes them little better than the flip of a coin. Careful clinical assessment, and appropriate non-HAS-BLED bleeding risk factors documeted in the medical record should be able to also support coverage in some patients.

3)Frailty is another big one I have used. Formal frailty indicies exist for TAVR. This is important to consider because virtually every one of these patients is also UNPROTECTED because physicians and patients are unwilling to take the bleeding risk. Risk of falls is included in this category as well. One patient I recently cared for lives in a very remote area on the Missouri/Iowa border and was fearful of of falling (again) on the ice this winter. She has an unsteady gait (mild Parkinson's) an has had documented falls in the past; once including hip fracture two years ago.

4) Lastly COMPLIANCE issues are something I also believe should be taken into account to determine eligibility. If, after robust discussion with patients and families, it is determined a patient simply cannot be compliant with OAC therapy, or, if there are social issues that are of concern, these patients should be covered for LAA closure to reduce that risk of stroke. Occassionally lifestyle factors come into play here as well, and could be considered.

All of these examples, in my view, are conraindications to long-term OAC therapy. Currently, patients with ABSOLUTE contrindications to even short-term OAC therapy are not candidates for this therapy, as this group has not yet been studied in the US. I hope to be personally involved with the proposed ASAP-TOO trial that will study this population in a randomized fashion.

A final comment about my 10 year experience in now over 100 patients. Collectively, they have done extrordinarily well without even one large stroke I would have attributed to AF. We have seen a few small TIA's that, frankly, are most c/w a "vascualar" event (as most of these patients have significant diffuse vascular disease as well). I can't emphasize enough the dramatic improvement in QUALITY OF LIFE these patients express every time I see them.

Thank-you for lettting me share my experiences and opinions with you. This is an absolutely fantastic new therapy that in my opinion significantly improves the live of many AF patients and absolutely reduces the risk of stroke in those we know would otherwise be protected. I look forward to CMS finalizing and appropriately clarifying an "eligible" patient population that will be adequately covered so that we can improve the care to our ever growing AF population in the US.

Sincerly,
Ken Huber

Less

I am cardiac electrophysiologist at a tertiary care university hospital. I care for a wide range of patients with atrial fibrillation and our practice is a referral destination for difficult to manage patients with atrial fibrillation. Often these patients are those that have challenges to chronic anticoagulation despite a high risk of stroke based on prior stroke history or risk factors based on the CHADS VASC score. Their issues with anticoagulation often are not reflected in the HASBLED score. An example would be a pt who developed a spontaneous intracranial bleed whether on therapeutic Coumadin or no anticoagulation. This individual may have a HAS BLED score of only 1 or 2…but few would be comfortable offering or re-offering chronic anticoagulation for further stroke risk reduction. This individual, if he/she also had a high CHADS VASC score or prior cardio-embolic stroke would be the type of individual to greatly benefit from LAAC. With such an example I caution the use of absolute HAS BLED or CHADSVASC numbers, as many pts with little options at present may not fit such rigid definitions, but still benefit from LAAC as described in the Watchman Trials.

All in all I am in support of the 7 proposed criteria. The only percutaneous option covered is Watchman due to this device being the only one extensively studied in controlled trials. I am in favor of documenting the HAS BLED and CHADS VASC and also documenting the specific rationale to why the individual was offered LAAC or is a candidate of LAAC (condition2) . Clearly there needs to informed consent and discussion regarding options to LAAC (condition 7). I am in favor in the conditions 3,4 and 5 to ensure a safe and collaborative environment for a LAAC program. And the registry is paramount in gaining more data on benefit and risks and outcomes of LAAC.

Less

As a busy, board-certified electrophysiologist practicing in Philadelphia, PA who treats many atrial fibrillation patients, I anxiously await the CMS decision on LAA Closure Therapy. As you are well aware, atrial fibrillation is a major public health problem. It remains the largest cause of preventable stroke in the United States. While it is well-understood by physicians that anticoagulation can reduce this risk, we know many patients are not getting the therapy they need. I see this in my daily practice in the hospital and in the office. Patients who clearly would benefit from anticoagulation are not receiving the therapy because of true and perceived risks of chronic anticoagulation. Specifically patients who have suffered bleeding, falls, or are at risk of falls are often not placed, or are taken off of, anticoagulation because of the potential for serious adverse events. Unfortunately, these are often the patients with the highest CHA2DS2Vasc scores and therefore those who could extract the most benefit from stroke reduction strategies. While studies have suggested that the benefits of anticoagulation outweigh the potential risks, even in sicker patients, on an individual basis physicians often make the decision to avoid anticoagulation because of their motivation to “first, do no harm.” These decisions unfortunately can lead to a high rate of embolic strokes that are preventable and can be devastating to patients and their families. LAA Closure therapy offers us an alternative to provide protection in these situations. In my initial experience with the Watchman LAA Closure device, thus far, I have been able to implant the device successfully in several patients with elevated CHA2DS2Vasc scores. My first patient was an elderly gentleman with persistent atrial fibrillation and a CHA2DS2Vasc score of 4 who developed a spontaneous and significant retroperitoneal bleed that led to a lengthy hospitalization and rehabilitation stay. After he was stabilized, he was placed back on warfarin but the concern for a repeat bleeding event was high. Our team was successfully able to implant the Watchman and he ultimately has been able to stop his chronic anticoagulation without event, and is doing quite well. I think this example is important because this is a patient who was tolerating anticoagulation but clearly had proven himself to be a high risk patient for chronic therapy, and had a strong reason to seek an alternative option. This example is similar to other patients I have treated with the device.

Regarding the proposed CMS criteria, I would recommend the following:

Condition 2: I would not support using the device only in those with a “contraindication to warfarin.” I state this because, as in the case of my patient, he was tolerating warfarin after his first bleed but clearly it was uncertain whether he would continue to tolerate anticoagulation. Rather than require a “contraindication to warfarin,” I would propose that a careful discussion between the physician and patient is necessary to determine suitability of long-term anticoagulation. If it is deemed that the patient is not suitable for long-term anticoagulation, then LAA Closure Therapy should be considered as a possible alternative.

Lastly, I want to state that I work in an urban hospital that provides care to an underserved minority population. While there are significant attempts to provide appropriate outpatient care and outreach to our patients, a number of socioeconomic factors often prevent our highest risk patients from receiving adequate care. The requirements and expense of warfarin monitoring and use of the newer anticoagulant agents are often prohibitive for our patients. This significant healthcare disparity could, at least partially, be addressed in some circumstances by use of LAA Closure Therapy.

I thank you for your consideration of this new technology. I urge your approval of this therapy for the benefit of our patients suffering from atrial fibrillation.

Less

I'm an electrophysiologist with Dupage Medical Group in the West Suburbs of chicago and solely practice electrophysiology with 50% of my practice suffering from atrial fibrillation.

I've reviewed CMS's 7 proposed criteria for WATCHMAN coverage and agree that this new technology needs to be rolled out in a very deliberate way, that encourages only appropriate patient selection criteria. I appreciate the efforts to include a registry and believe further study should be performed over the course of the clinical life of the product to better understand its risks and benefits a la TAVR.

Please do not hesitate to contact me with ?'s

Less

Less

Colleagues at CMS,

As a practicing cardiologist and electrophysiologist, I am excited to have a national coverage decision for percutaneous left atrial appendage closure (LAAC) proposed. This is a revolutionary technology that provides an option for our increasing burden of atrial fibrillation patients, many of which who are contraindicated from long-term oral anticoagulation. I have reviewed the 55 page document and in particular focused on the 7 conditions for which LAAC therapy has been proposed. I can appreciate the difficulty and complexity when arriving to such a determination.

I think it is very well written. I had some comments that I think would make it easier to implement and execute. With respect to the second condition, the statement indicates a contraindication to warfarin. I think this could be better defined by discussing it as a contraindication to long-term oral anticoagulant therapy. This is for two specific reasons. One is the patient cannot have an absolute contraindication to warfarin because under the practice guideline and trial in the United States, the patient had to be able to take warfarin therapy for a minimum of 6 weeks. In addition since this therapy was initially studied, newer oral anticoagulants have been developed and for many of the patients, a very similar contraindication exists. Specifically, our patients with recurrent severe gastrointestinal bleeding would be an example of a group that would be a contraindication to all long-term anticoagulants but can typically be managed in a short window due to the intermittent but potentially challenging nature of this complication.

I also have concerns the burden of a new registry for patient's in a control arm. I do not see the added value. There are numerous studies available with current anticoagulants. The risk of stroke in patients on warfarin, dabigatran, rivaroxaban, apixaban, and edoxaban are well recorded in studies with thousands of patients. I wholeheartedly support an accurate and ongoing registry LAAC patients demonstrate short and long-term efficacy of the device as well as short and long-term potential complications.

Thank you again for preparing the document and having it available for our review.

Respectfully,

John Warren Holshouser, MD

Less

Less

Less

To the CMS Officer,

I am writing this response to your proposed decision that “left atrial appendage closure is not reasonable and necessary to treat an illness or improve function and therefore not covered under the social security act.”

I am a Board Certified Cardiologist and the Physician in Chief for Cardiovascular Services at Southcoast Health System in Fall River. We are the largest Health Care System in Southeastern Massachusetts, East Bay and Rhode Island. Per recent AHA/NCDR data we accounted for the second largest atrial fibrillation inpatient discharges in the state of Massachusetts in 2013-2014. We also serve about 70% Elderly Medicare and/or Medicaid patient population.

I strongly believe that restriction of Watchman device and low reimbursement will limit this valuable therapy to a variety of patients who may need an alternative to long term anticoagulation and were in fact tested in clinical trials done which led to the approval of this device therapy by FDA in March 2015. Specifically, the elderly at high risk of stroke but intolerant of anticoagulation; at high risk of falls, etc; as well as another important group of patients who should have a choice- active young and middle aged atrial fibrillation patients with lifestyle preference to remain active: bicycling, playing soccer, skiing, etc. and avoid the restrictions imposed by systemic anticioagulation.

Hence, I implore CMS to reconsider their recent decision memo for left atrial appendage closure device to a broader patient population as studied in clinical trials. This will provide patients a viable alternative to long-term anticoagulation in prevention of stroke secondary to atrial fibrillation.

Sincerely,
Margaret Ferrell, MD
Cardiovascular Care Center
Southcoast Health System
Fall River, MA

Less

I appreciate the opportunity to comment on the proposed decision by CMS regrading coverage for the Left Atrial appendage Occlusion device (WATCHMAN). I am a cardiac electrophysiologist in Sarasota Florida and my practice is entirely devoted to management of cardiac arrhythmias and a large part of it involves treating patients with atrial fibrillation. As cardiologists and electrophysiologists we were gratified to finally see the FDA approval for the Watchman device come through earlier this year.

The landmark PROTECT -AF and PREVAIL trials have clearly demonstrated the efficacy of these devices in patients with atrial fibrillation who are at increased risk of stroke.

It is commendable that CMS has proposed a coverage for this new technology and set certain conditions for its use as any new therapy or device should be subjected to. However I do have certain reservations regarding these "conditions'. Primarily it involves condition 2. As observed by many of my colleagues the "contra-indication to warfarin" is a bit confusing as one of the inclusion criteria for both the sides for WATCHMAN device was the ability to tolerate a short course of anticoagulation with coumadin (upto 45 days).Indeed if they have absolute contra-indication to coumadin they may need to be considered for alternative therapies like the left atrial clip performed by surgeons.We do recognize that this therapy would be best suited to patients who cannot tolerate long term anticoagulation. It would be desirable to bring some clarity to this and a statement which mentions a relative contraindication to anticoagulation or similar language would create less confusion to our patients and referring physicians. I am also not sure whether we could use HAS-BLED score in making these decisions as it is not widely use in clinical practice and was not part of the inclusion criteria for studies using the WATCHMAN device.

I strongly support the physician training and hospital requirement as detailed in conditions 3,4,5.

In terms of the registry (condition 6) it would be useful to obtain data on all our patients;however there are some practical difficulties in enrolling patients in the control arm on long term anticoagulation who would want to participate in a long term registry.Moreover the risk factor profile in terms of the CHADS-Vasc score and bleeding risks may be quite different and it may not be a "true" control arm.

Thank you once again for giving me the opportunity to comment and hope this will inform your final decision regarding the coverage.

Dilip J Mathew MD FACC FHRS

Less

I am a practicing interventional cardiologist. I see many patients with non-valvular afib. I have a few comments about the proposed 7 conditions of the CED, specifically related to condition #2 and 6.

Condition 2: In my real world clinical practice, I would find it difficult to completely view "warfarin contraindication" solely based on the HAS-BLED score. There are conceivable patients with a significant history of medical noncompliance, fall-risk or at- risk occupations for injury-related trauma/bleeding. These patients should also be included in the LAA closure treatment if they are at significant risks of a CVA.

Condition 6: I'm not clear why an OAC control group is necessary in the registry, since a registry is unlikely to produce similar patient cohorts for a good comparison. On the other hand, we already have good clinical data on OAC controls from the pivotal randomized clinical trials. I find this proposed OAC controls in the registry a redundant step and potentially add confusion rather than clarity to this particular clinical scenario.

Less

Less

I am a long-time practicing electrophysiologist whose primary focus is on atrial fibrillations. Over the years, I have cared for thousands of patients with AF. I have also cared for many patients who have suffered thromboembolic complications from this increasingly common problem. Strokes from AF tend to be devastating. I view my primary job is to assess a patients risk for stroke and do everything I can to minimize this risk. In the era prior to the novel oral anticoagulants, this was a very difficult discussion. With these new agents, we clearly have better options for our patients. However, there are a number of patients who cannot safely take long term anticoagulation with warfarin or a NOAC. Until the recent approval of the Watchman device, our options for these patients were terrible... referral for LAA surgical removal (which is a major and and potentially morbid procedure without any hard data endpoints) or telling patients to live with the risk. The latter is what occurred the vast majority of the time.

With FDA approval of Watchman, we finally have another data-supported option for these patients. Imaging living with the fact that you have an elevated CHADS2-VASc score denoting say an 8% annual risk of stroke. However, you are on long-term dual anti-platelet therapy for recurrent CAD and in-stent restenosis. You are placed on an anticoagulant and have recurrent epistaxis, smoldering GI bleeds, etc. Your stroke risk is diminished by ~65%, but you cannot continue on the OAC. You are forced to stop the drug and just wait for a stroke. Watchman affords this patient a safe and tangible means of dramatically dropping the risk of stroke. For this patient, this changes their life (both physically and pshychologically).

My concern is that the coverage recommendations would likely make it impossible for this patient or many, many others to receive this important therapy. The document is too limiting in its reference to life-threatening bleeds. Further, its listing anticoagulation as contraindicated goes against current FDA labeling. Rather, a patient currently needs to be able to tolerate short-term anticoagulation for the device. Finally, a registry of treated patients is an excellent means of tracking real world outcomes. However, mandating that non-treated patients are followed is both onerous, costly and logistically unfeasible in the real world.

For all of these reasons and more, I respectfully request you to rethink the currently proposed coverage agreement.

Less

Less

Less

December 3, 2015

Ms. Tamara Syrek-Jensen, JD
Director, Coverage and Analysis Group

Dear Ms. Syrek-Jensen,

This communication is in response to the request for comments to the proposed NCD on the Watchman™ device for left atrial appendage occlusion. The FDA language at the time of approval was excellent - "The Watchman device is indicated to reduce the risk of thromboembolism from the left atrial appendage in patients with nonvalvular atrial fibrillation and increased risk for stroke and systemic embolism based upon CHADS2 or CHADS2- V ASC scores who are deemed by their physicians to be suitable for warfarin and have an appropriate rational to seek a nonpharmacologic alternative to warfarin taken into account the safety and efficacy of the device compared to warfarin."

This is the language that was intended to mirror the results of the randomized clinical trials and registries upon which the device was approved. That data identified in patients treated with the device that there was a marked survival advantage in device treated patients compared with chronic warfarin, that the device was virtually identical to long-term warfarin in preventing all cause stroke or systemic embolization, there was a dramatic reduction in hemorrhagic stroke with the device compared with warfarin, and more than 90-95% of patients at follow up could be off of warfarin and still have the benefit of the reduction of risk.

In the randomized trials, all patients had to be eligible for warfarin. In the device treated patients, warfarin was administered typically for six weeks to facilitate healing; after six weeks, > 90-95% of patients were able to be taken off of warfarin.

I believe that the seven conditions of the CMS proposed NCO have some significant problems. I will refer to each condition in the following order:

1 - FDA approval is now in place. There is a strong patient need for such a device because many patients at risk for stroke (in some series 40% of patients) are not treated with anticoagulants whether that be either warfarin or a NOAC. In addition, up to 25% of patients treated with either warfarin or a NOAC discontinue that treatment after approximately two years and so they are therefore unprotected in terms of stroke prevention. That data has been confirmed in many articles and many registries.

2 - I think it is important that patients being considered for a Watchman device must have non-valvular atrial fibrillation and be at increased risk for stroke based upon assessment of the CHADS2-VASC scores. However, the proposed CMS language that says that a patient must have a contraindication to warfarin is extremely problematic. It is important to remember that warfarin was mandated in all past clinical protocols of Watchman studies because of prior animal and human data. As well as all collected randomized data of the device presented in all three past FDA panels includes the use of a short course (six weeks) of warfarin post procedure. This again is in direct conflict of the statement concerning "contraindication to warfarin" in the proposed CMS language. Following the six weeks, then warfarin can be discontinued.

The important piece of information was outlined in the FDA IFU; namely patients who are potential candidates are deemed by the physicians to be suitable for warfarin but have an appropriate rationale to seek a nonpharmacological alternative to warfarin taking into account the safety and efficacy of the device compared to warfarin. That language is extremely important because it allows physicians to provide the information to patients and then in classic decision making decide on what is the appropriate care for that individual patient. There are patients in whom by virtue of associated conditions such as frailty or frequent falls or history of bleeding in whom the physician as well as the patient do not think that long-term anticoagulation whether that be with warfarin or a NOAC is in the patient's best interest. I think that these decisions should be best made by the attending cardiologist and the patient in concert. -An arbitrary score like the HAS-BLED score is just that. There will be some patients who definitely have had recent bleeding. There are other patients who are elderly and are noncompliant to medications. In these patients, a device such as the Watchman device should be considered a very reasonable approach. It needs to be documented in the medical record but it is still a very reasonable approach.

3 through 5 - The discussion language about the institutional requirements in #3 as well as the physician qualifications and experience and that training should be accomplished prior to performing implant procedures is excellent. I think it is very appropriate that the patient should be enrolled in a prospective national registry that consecutively enrolls left atrial appendage occlusion patients and tracks their outcome at the patient data level. I think the specific pieces of information to be collected identified in the proposed CMS regulation are good. However, I think operator specific complications are somewhat problematic because in some circumstances there may be two operators involved with cases - an electrophysiologist and an interventional cardiologist. I think the most important thing is the participation in a prospective national registry which should be required for reimbursement. That is in my mind similar to the TVT Registry which has been a very important source of information on the science and outcomes of the procedure.

6 -A major problem with the CMS proposal is the comparator control grdup of patients treated with chronic oral anticoagulants. That information is not available. Many patients who are treated with oral anticoagulants never see anyone who provides information on left atrial appendage occlusion. They are not seen by physicians or hospitals experienced with the procedure. No one has data on these types of consecutive patients. If indeed you define a comparative group as patients who are treated with long-term anticoagulants, that is not the group of patients that would be candidates for this device. Those would be patients in whom the physician thinks that long-term anticoagulations is perfectly suitable for the patient. So I think the science of a comparator group is suboptimal at best and may be very misleading. It will be a different comparison group; it will be different physicians involved. It will have different follow up requirements and it will not only be unscientific but also very impractical. At the present time, those institutions performing Watchman procedures may not see patients on long-term anticoagulants. It is important to remember that the Watchman device is not felt to be a replacement for all patients with non-valvular atrial fibrillation rather than anticoagulation. In addition to that in the comparator group, all of the information on anticoagulants, whether that be warfarin or NOAC, identifies that perhaps 25% of patients at the end of two to three years discontinue that drug or group of drugs leaving them at high risk for stroke. It is the comparator group that poses the most problems. It should be eliminated in my opinion.

7 - In terms of the formal decision making interaction, I think that that is good medical care. At the present time, there is no evidence-based decision tool. However I think it is very appropriate in the medical record to include the discussion of the risk and the benefits of both anticoagulation as well as the device. That is good medical practice. That conversation should be documented in the medical record.

I would hope that these comments would be of some help to you in this process. Thank you for your consideration.

Sincerely,

David R. Holmes, Jr., M.D.
Scripps Professor in Cardiovascular Medicine
Professor, Mayo Clinic College of Medicine
Consultant, Cardiovascular Medicine
DRH:keh

Less

Dear CMS:

I am writing this comment to convey my deepest disappointment with your coverage decision regarding Watchman, LAA occluder. Orange County Heart Institute participated in the Protect AF trial and CAP, and CAP 2 registries. We have had significant experience with the treatment and based on the feedback from our patients, it has been one of the most rewarding treatments that we offer our patients with atrial fibrillation.

Managing the increasing number of patients with atrial fibrillation is very demanding and challenging. We know how well anticoagulation with warfarin and the newer oral anticoagulants reduces stroke risk and how it reduces mortality. Yet, due to the challenges of taking anticoagulation with warfarin, a significant number of patients still refuse to take it. The newer oral anticoagulants provide an alternative but many of my patients cannot afford to pay for it. Furthermore, there is still a high discontinuation rate for these newer agents and studies show that it is as high as 25%.

There is clearly a need and demand for an alternative treatment for stroke prevention in patients with atrial fibrillation. The patients who have received Watchman in my practice are the most satisfied patients. They are protected from stroke, bleeding, and their quality of life is significantly improved over taking anticoagulants. Quality of life is something that seems to get overlooked in the clinical studies and it is so important for patients.

I certainly hope that CMS changes their position on coverage for Watchman. And it should not only be for patients who are contraindicated for anticoagulation. Watchman patients all require a period of anticoagulation with warfarin during the post procedure period. This is contradictory to the FDA approval and to manner in which the clinical trials were conducted.

As an electrophysiologist who takes care of so many patients with atrial fibrillation, I see this treatment as essential in the care of my patients and my patients have been patiently waiting for it. Now that it has finally been approved by the FDA, it is very disappointing to read CMS's coverage statement.

I really hope that you reconsider your coverage decision, and that you get it right. For patients' sake.

Sincerely,

Jay H. Lee, MD, FACC

Less

Subject: Public comment on percutaneous left atrial appendage closure using the Watchman percutaneous device.

Please consider this memo in response to CMS Public comment request for the Watchman percutaneous left atrial appendage closure therapy.

I am a practicing, board-certified Cardiologist at the Minneapolis Heart Institute in Minneapolis Minnesota and have been in Cardiology practice for more than 30 years. I care for all types of cardiovascular patients, including those with both valvular and non-valvular atrial fibrillation. Atrial fibrillation constitutes a very large part of my clinical practice.

I have a prior conflict of interest, as an inventor of the Watchman device, and as a co-founder of Atritech, the startup company responsible for Watchman, subsequently acquired by Boston Scientific Corporation. Although I previously received payments from the acquisition transaction, all payments have now ended. I will not receive additional payment regardless of any CMS decision to cover or not cover this device.

The complexity and problems of clinical Cardiology caring for patients in nonvalvular Atrial Fibrillation were a key factor driving us to develop a device-based, alternative therapy to pharmacologic anticoagulation.

Many patients, especially the elderly, have a “perfect storm” of atrial fibrillation, high stroke risk, difficult INR control, and high recurrent fall risk. A typical patient may be in their 70s or 80s, and have a CHA2DS2-VASC score of 4 or greater (very high stroke risk, >6% annually). Annual risk of major bleeds is roughly 2% on warfarin anticoagulation therapy. These patients are at very high stroke risk yet simultaneously at major bleeding risk when exposed to long-term oral anticoagulation. The Watchman option in this patient group is associated with excellent, comparable efficacy yet substantially lower long-term risk.

Elderly patients are unable or unwilling to take anticoagulation in proper dosing for many years. Warfarin is a difficult drug to manage, especially in the elderly, evidenced by real-life INR values being outside the 2.0-3.0 range in 60% of all patients taking warfarin. These patients are thus either at risk of a bleed (high INR) or a stroke (low INR). Novel anticoagulants as warfarin alternatives are quite expensive, averaging $4100/year for medication alone at current discount prices.

Combination of data from the Prevail and Protect-AF trials show a stroke risk of 1.75/100 patient-years using the Watchman compared to 1.87/100 patient-years with warfarin (p=ns). Importantly however, Cardiac mortality however is reduced by roughly 50% using Watchman compared to warfarin, hazard ratio 0.48.

Proper clinical management of anticoagulation is difficult in many patients, but is especially difficult in the elderly, who are precisely those at highest risk of nonvalvular atrial fibrillation related stroke.

Thank you for requesting comments on Watchman left atrial appendage closure. It is important to recognize that this technology is the 1st clinically approved device, which is being compared against a very old and well established pharmaceutical therapy. It is my firm belief that while Watchman is a clear advance in clinical atrial fibrillation management, further refinements of the device technology and methods will improve the current data, a scenario that is made very difficult without adequate reimbursement.

Less

Dear Sirs,

I am currently an Interventional Cardiologist practicing in Southern New Jersey . My main practice is at Cooper University Hospital and Rowan Medical School- I'm a very involved in most interventional procedures especially the structural program including TAVR, mitral clip, and LAA closure using the Watchman device and the Lariat suture system.I have had the opportunity to become an early adopter of the WATCHMAN technology since FDA approval. Please allow me this opportunity to respond to the proposed CMS Coverage decision.

Currently in our practice, we have thousands of patients with a diagnosis of atrial fibrillation. Off these patients approximately 40% are on aspirin alone, or in combination with plavix for CVA risk reduction. Of these patients, those with CHADS2vasc scores of 2 or higher comprise the vast majority(85+%) and in these patients the number of patients on anti-coagulation is approximately 70%. The main determinants for withholding anti-coagulation are prior bleeding and concerns for patient frailty/risk of bleeding. Of those on blood thinners, a significant number (~10%) have had at least one major or multiple minor bleeding episodes in the past, yet remain on anticoagulants for fear of stroke.
My concerns about the proposed coverage decision revolve around where the line will be drawn in regards to bleeding risk and what does "contraindication to warfarin" mean. In my opinion, if someone has had a bleeding event on one anticoagulant, their risk of future bleeding is higher and they should be offered an alternative therapy such as WATCHMAN. The words contraindicated are nebulous to me in this regard, as some physicians might accept higher risk bleeding for their patient rather than expose them to the risk of stroke. The truly "contraindicated" patient is a rarity, and a separate problem.A contraindication to warfarin conflicts with the device IFU as well as the protocol that resulted in the FDA device approval, which requires 45 days of warfarin post implant and up to 6 months of warfarin if a complete seal of the LAA is not seen on TEE at 45 days. There are numerous clinical scenarios where warfarin is not contraindicated by strict definition, but where it is to the patient's benefit not to be on warfarin (e.g. non compliant patients, high risk occupation, difficult to control INR with recurrent bleed...)

The safety of this device is supported by a long European experience and numerous clinical studies. Having personally performed numerous Lariat procedures, I can testify that Watchman is clearly a technically much easier and safer one with way more outcome data. My concern also is that by placing severe restrictions on the use of Watchman might push physicians to start using the more complex and risky Lariat procedure.

Finally, although it is reasonable to conduct a registry of patients receiving the Watchman device as suggested in condition 6, collecting data on long term oral anticoagulants as a control arm would be at best of very limited utility, and at worst highly misleading. Unlike randomized controlled trials (which have already been done comparing Watchman with anti-coagulation), this type of registry would be plagued by selection bias and would not allow a scientific comparison .

I hope that CMS will understand and realize the true potential benefit of approving coverage of Watchman for a much broader spectrum of patients, including enhancing quality of life, improving long term outcomes and actually lowering the cost to the health care budget by reducing morbid events in large subsets of patients that would not be covered by more restrictive conditions.

Less

Thank you for the opportunity to comment on the Proposed Decision Memo for Percutaneous Left Atrial Appendage (LAA) Closure Therapy (CAG-00445N).

I am an interventional cardiologist practicing at a large, tertiary care academic medical center in Colorado. I have been in practice for over 6 years and have a focused area of expertise in structural and valvular heart disease and interventions. Specifically, I have extensive experience in various novel percutaneous structural heart device therapies, including TAVR and transcatheter mitral valve repair (TMVR) with the MitraClip system. In addition, our site actively participates in various patient registries including NCDR/CathPCI and TVT. In fact, several physicians at my site are executive members of these registry steering committees, including the registry proposed in the CMS coverage decision for LAA closure therapy.

While I have limited experience with the WATCHMAN device specifically, my facility will begin WATCHMAN implants in appropriate patients in December 2015. On that note and given my background of experience with novel structural heart disease devices, I have several specific concerns regarding the proposed CMS coverage decision for LAA closure therapy:

A. POINT #2

The wording of a “high CHADS2 or CHA2DS2-VASc score” should be modified to reflect the current ACC/AHA guidelines for anticoagulation (CHA2DS2-VASc score = 2). Furthermore, the wording should arguably focus specifically on CHA2DS2-VASc scores given the inherent limitations of the CHADS2 score in identifying potentially higher risk patients for thromboembolic events.

On that note, I believe that the term “elevated” would be a better choice over “high” as the latter potentially encourages physicians to only consider LAA closure therapy in patients with the highest (e.g., maximum CHA2DS2-VASc score of 6-9) overall risk for thromboembolism whereas a meta-analysis data of available studies show that most patients treated with WATCHMAN in clinical trials had CHA2DS2-VASc scores = 4 (Meta-Analysis: Holmes et al, J Am Coll Cardiol 2015;65:2614–23).

On a similar note, the wording of “high” in relation to HAS-BLED scores should likewise be modified to “elevated” since data shows that many patients who have had life-threatening bleeding events with oral anticoagulant (OAC) therapy do not have markedly elevated HAS-BLED scores. In support of this is that a meta-analysis of available studies shows that most patients treated with WATCHMAN in clinical trials had HAS-BLED scores of 1-2 (Meta-Analysis: Holmes et al, J Am Coll Cardiol 2015;65:2614–23).

Finally, the phrase that only patients who are “contraindicated to warfarin” should be changed to more accurately reflect the wording outlined in the current FDA approval document which is to treat patients who are “eligible for warfarin, but have reasons to seek another long-term therapeutic option.” Alternatively, the wording could be changed to “relative contraindication to long-term therapy with warfarin.” This change of wording is critical for two important reasons. First, patients undergoing WATCHMAN device implantation are currently mandated to be treated a minimum of 45 days of warfarin and aspirin therapy post implantation as guided by treatment utilized in the clinical trials (PROTECT-AF and PREVAIL, specifically). Thus, patients must be able to safely tolerate at least a short course of warfarin, and fundamentally cannot be “contraindicated” to such therapy. Second, the clinical trials conducted which ultimately led to the FDA approval of WATCHMAN did not study patients who were “contraindicated” to warfarin, so the currently proposed wording would directly encourage treatment of patients for which is there a fundamental lack of data to support safety and efficacy.

B. POINT #6

The proposed decision memo stating that the proposed national registry enrolling and following patients for at least five years undergoing LAAC therapy must “include contemporaneous patients followed on oral anticoagulant (OAC) therapy to serve as non-interventional controls” is unreasonable. First and foremost, similar novel device therapies (i.e., TAVR and TMVR with MitraClip) do not have mandated concurrent “control groups” per the TVT registry. Second, the inherent variability of choice in OAC therapy (warfarin versus alternative oral anticoagulants) for patients with non-valvular AF will very likely fail to provide any meaningful clinical data for comparison with LAAC therapy long-term. Finally, the level of patient selection and data collection on the sheer volume of patients that would need to be included would place undo hardship and unreasonable expectations of site personnel in order to maintain the level of excellence that would be required for such a data metric to be deemed both reliable and believable.

Thank you for your consideration of my comments.

Sincerely,

Michael S. Kim, MD, FACC, FSCAI
Assistant Professor, Cardiology
Director, Structural Heart Disease Program
University of Colorado Denver
University of Colorado Hospital
Aurora, Colorado
michael.kim@ucdenver.edu

Less

To Whom It May Concern:

My name is Dr. Robert Sommer and I am a practicing interventional cardiologist at New York Presbyterian-Hospital (Columbia University Medical Center). I am the leader of the multispecialty advisory group, which has been assembled here at the hospital, for the management of patients with atrial fibrillation. We are charged with optimizing both their medical management, to minimize the risks of stroke and bleeding, as well as their quality of life, including the oversight of the Left Atrial Appendage Closure program.

We have been involved with the Watchman device since early in the Clinical Trials, participating in the original PROTECT AF Study, the follow-up PREVAIL Trial and the subsequent CAP2 registry. We have a dedicated group of physicians who have been credentialed to perform this procedure. In experienced hands, the procedure has been extraordinarily safe here at Columbia, as has also been documented in the National Trials and in more than ten years of European Experience. The 5 year data from the PROTECT AF Study which indicates a statistically significant reduction in all cause mortality (35%), cardiovascular mortality (60%), all cause stroke (40%) and hemorrhagic stroke (85%), as well as the FDA’s approval of the device in March, has led both patients and physicians to seek out this therapy. All are anxiously awaiting a National Coverage Determination for this predominantly Medicare population.

We would like to comment on your recent Proposed Coverage Decision for the closure of the left atrial appendage. I will go through each of the proposed conditions of coverage below.

Less

I applaud CMS’s proposal to cover Watchman LAAC therapy. I am a very busy Interventional Cardiologists in Pennsylvania and I spend most of my time in the cath lab managing patients with various vascular issues that suffer from bleeding events of various manifestations. These patients are a significant financial burden to the hospital and the rest of the Medicare system. Major bleeding events can cost in excess of 23,000 dollars and Intra-cranial Hemorrhage can cost 50,000. Cost is one issue that Watchman can address with these issues, but consider patients who are intolerant to these medications, or have multiple events over the course of their lives. Finances aside, the quality of life these patients experience cannot be discounted and medication compliance has always been a major issue with my patient population.

My caution to CMS is to ensure patients with multiple clinical concerns are accounted for and the reimbursement decision has the levity to allow multiple clinical factors to be considered.

I also have concerns with condition 2: Requiring patients to have a high HAS-BLED score and a contraindication to Warfarin will eliminate a huge patient population that will potentially benefit from the Watchman device. A contraindication to warfarin seems to be in direct conflicts with the device’s FDA approved indication, which requires 45 days of warfarin post implant and up to 6 months of warfarin if a complete seal of the LAA is not seen on TEE at 45 days.

There are several scenarios where warfarin is not contraindicated by strict definition, but where it is to the patient's benefit not to be on warfarin: Liable INRs, non-compliance, age, limited mobility, non responder to warfarin (40% patients are above or below the therapeutic range).

The addition of the HAS-BLED score can also be a concern since it does not identify all of the patients who have a high risk of bleeding events. Patients with DES stents requiring Aspirin and Plavix are at an increased risk of traditional triple therapy protocols with no clear guidelines to date. I also see several patients that take NSAIDS that have a dramatic increase in bleeding events.

Condition 6 may also prove to be burdensome. It is a reasonable approach to conduct a registry of patients receiving the Watchman device, but I see very limited clinical benefit to collecting long-term OAC data in the proposed control arm. This data is currently well published and could be misleading.

I want to take this opportunity to again thank the FDA and CMS for their positive approval and coverage decision. I am hopeful that the board will appreciate that doctors need the ability to assess the patient and make the very best clinical decision. Making to narrow a scope will limit my ability to treat appropriately treat the patients individual clinical picture.

Less

Less

1) Currently I am the Medical Director of the Structural Heart Disease Program in the Section of Cardiology at Christiana Care Health Systems which is the largest health care system in Delaware.

2) I spend nearly 100% of my time in the cardiac catheterization laboratory or hybrid operating room and perform high volume coronary, peripheral and structural heart interventions and seeing complex patients.

3) I have placed multiple Watchman devices and have concluded that, in properly trained hands, this device can be delivered easily and safely in the vast majority of patients with non-valvular AF. The safety of this device is supported by >10 years of experience at European Centers and numerous clinical studies.

Regarding CMS's proposed coverage decision several comments are in order, specifically regarding Conditions 2 and 6:

Condition 2: Specifically, requiring patients to have a high HAS-BLED score and a contraindication to Warfarin will eliminate a huge patient population that will potentially benefit from the Watchman device:

It is also important to point out that the addition of NOACS has not had any major influence in the way non-valular A-fib patients are treated: NOACS have not resulted in an increased rate of anticoagulation in high risk non-valvular A-fib patients (Jani et al, Uptake of Novel Oral Anticoagulants in Patients with Non-valvular and Valvular Atrial Fibrillation: Results from the NCDR-Pinnacle Registry; ACC 2014) Discontinuation of anticoagulation therapy due to bleeding is no different with the NOACS than it is with warfarin (Connally et al; NEJM 2009;361: 1139-1151; Patel et al; NEJM 2011;365:883-891; Granger et al; NEJM 2011; 365:981-992) Also, CMS should consider the one time cost of device implantation, which is likely to be orders of magnitude less than a lifetime of pharmacotherapy, laboratory testing, major morbidity and mortality from bleeding events, as well as morbidity and mortality from embolic stroke in patients at risk from bleeding on anticoagulation, or who are unprotected from stroke by inadequate anticoagulation.

Condition 6: Although it is reasonable to conduct a registry of patients receiving the Watchman device, collecting data on long term oral anticoagulants as a control arm would be at best of very limited utility, and at worst highly misleading. Unlike randomized controlled trials (which have already been done comparing Watchman with anti-coagulation), this type of registry would be plagued by selection bias and would not allow comparison of "apples to apples". Patients on oral anticoagulation are likely to be at lower bleeding risk AND lower stroke risk than patients receiving the Watchman device. Data demonstrates that the higher the CHADS2 score, the less likely the patient is to be anti coagulated (Piccini et al; Pharmacotherapy in Medicare Beneficiaries with Atrial Fibrillation; Heart Rhythm 2012;9:1403-1408)

These comments are respectfully submitted in the hope that CMS will understand and realize the true potential benefit of approving coverage of Watchman for a much broader spectrum of patients, including enhancing quality of life, improving long term outcomes and actually lowering the cost to the health care budget by reducing morbid events in large subsets of patients that would not be covered by more restrictive conditions.

Less

1. It seems incomprehensible that all of the randomized data with this device upon which FDA approval was based was obtained in patients who by protocol were able to take anticoagulation with warfarin and yet the NCD proposes only to allow patients with a contraindication to coumadin. This is not the patient population that was studied and this criterion appears to represent a direct contradiction to the FDA approval document which states that it is approved for use in patients who are able to take coumadin.

2. The NCD states it will be approved in patients with a high CHA2DS2VASC score but does not define what that score is. The FDA document indicates a score of 2 or higher which is also the criterion for entry in the second randmized trial of the device (PREVAIL). The NCD should define a threshold for a high CHA2DS2VASC score so that we do not have different policies around the country in different MAC territories.

3. Similarly, if high HAS-BLED score is to be a criterion, then that should be defined precisely and implemented uniformly across the country. On the other hand I would reitereate that this CMS/NCD criterion does not reflect the patient population studied in the randomized trials nor the indication in the FDA document.

4. I strenuously object to the conept of mandating that implantation sites obtain concurrent comparative data from patients remaining on oral anticoagulation. Regardless of what such data might show relative to patients receiving the watchman device it would be relatively meaningless compared to prospective randomized data since it would be subject to many forms of selection bias. CMS often quotes the original founding legislation from the 1960's for what they can and cannot do or pay for. I do not believe that anything in that original mandate indicates that CMS should require healthcare providers to perform clinical research at their own expense. Who is going to decide which of our thousands of patients with atrial fibrillation on anticoagulation are included in such a registry? Who is going to obtain their consent and collect their data? Why would any of those patients who have no relationship to the watchman device agree to have that data collected?

Less

The majority of my practice involves taking care of patients with atrial fibrillation.

Oral anticoagulation is the mainstay of stroke risk reduction in patients at high risk of stroke in the setting of AF
Any patient with a CHA2DS-VASc score of 2 or more is at risk and needs OAC or other means of stroke risk reduction.
A substantial number of patients are unable to take OAC or remain unprotected, either due to a contraindication to or due to a failure of OACs respectively.
We are quite pleased to have an alternative of left atrial appendage closure to OAC for patients who remain unprotected.
While the CMS’ proposed coverage includes a large chunk of those patients who could not take OAC there are notable omissions to the coverage, which leave a large number of our clinical patients unprotected as they cannot avail this latest clinical advance.
I am giving an example of a few patients described below.

1. 47 yr old AA male with persistent AF, cardiomyopathy with EF 35%, and HTN. Was started on warfarin. Presented to ER 6 months later with embolic CVA. INR was therapeutic at 2.7 . TEE 3 months later shoed new thrombus in LAA. Switched to newer oral anticoagulant, had persistent thrombus 3 months later. After over a year of OAC his LAA is finally cleared of thrombus, but he has failed two OACs and clearly remains at risk for recurrent CVA at this young age. This patient would benefit from LAAC but under current CMS proposal, he no longer qualifies for implant as he is taking an OAC.

2. 75 yr old diabetic female, with AF, ESRD on hemodialysis, HTN (frequently high > 165 mm Hg), CAD and recurrent CVA maintained on Warfarin and ASA. Patient has to frequently stop and start Warfarin due to bleeding from his hemodialysis access site
CHA2DS2-VASc score is 8. HASBLED score is 6 (annualized risk of bleeding 5- 19%)
Although this patient is taking Coumadin, the benefit of stroke risk reduction is uncertain in patients with ESRD on hemodialysis ,and the risks of bleeding are much higher as projected by this HASBLED score.
Under current proposed CMS coverage this patient will not qualify for Watchman although he will clearly benefit from it.

3. 75 yr old male with HTN, paroxysmal AF, and prior h/o TIA.
CHA2DS2-VASc is 5
He has bladder carcinoma and has undergone therapy and is currently in remission. Requires frequent urological procedures, after which he has significant hematuria (not requiring hospitalization or transfusions) which prompts discontinuation of Warfarin for months, during which periods he remains unprotected for embolic CVA.

There are several other scenarios where patients would clearly benefit from LAAC such as patients who are already on dual antiplatelet therapy and adding an OAC would be detrimental. There are also patients who have a h/o prior intracranial bleed but are currently tolerating OAC but are at ongoing risk of further bleeding if they remain on OAC for the long term or CVA if they stop the OAC.

I hope the CMS will have a look at these real life scenarios and include high risk patients who have a CHA2DS2-VASc score of 2 or more and a HASBLED score of 3 or more for assessment of LAAC as these are the patients most likely to benefit from this important advance in the management of AF.

Reagrds

Sarfraz Durrani MD

Less

Condition 1: Agree

Condition 2:
-Simplify to state a CHADSVASC2 score of >= 2.
-Contraindication to anticoagulation should be labeled as a relative contraindication to warfarin at the physician's discretion with benefits felt of the device felt to at least equal or outweigh its risks.

Condition 3-5: Agree.

Condition 6: Disagree with registry control arm of anticoagulant Rx. Patient's will not be randomized and comparison due to biases will not be meaningful. Also, the process would lead to a tremendous amount of work for little or no gain.

Condition 7: Agree

Less

Dear CMS panel members,

I am writing to appeal your proposed decision: left atrial appendage closure is not reasonable and necessary to treat an illness or improve function and therefore not covered under the social security act. I am quite surprised that despite FDA approval for the Watchman device based on extensive clinical trials data showing safety and efficacy of the device, the CMS has put forward the proposed decision which counters the FDA regulartory/approval process. I have several patients in my practice currently who are not good candidates for systemic anticoagulation for atrial fibrillation, I have no choice but to leave them at their calculated risk for stroke and I have no way to prevent their strokes. Infant, I have already had patients who suffered devastating/disabling stroke because systemic anticoagulation was contraindicated. I have multiple patients who had life threatening GI bleed and have no options to reduce their risk of stroke, UNLESS the Watchman device receives CMS decision for reasonable and necessary treatment strategy for patients at high risk for systemic anticoagulation. I am puzzled by the fact that this device is available outside of the USA for a long time, has undergone significant regulatory process by the FDA and subsequently approved but still there is a significant hurdle by CMS in getting this treatment strategy to the US residents. Given prolonged duration of multiple clinical trials and lengthy FDA approval process, countless US patients have suffered for not being treated in a timely manner by this device. At national meetings for cardiology, the physicians from outside of the USA repeatedly comment on the disparity in treatment options for US residents vs Europeans or Canadians, with the US lagging far behind some of the other developed countries in the world. I have no conflict of interest with this device or with Boston Scientific, I am wholeheartedly requesting you to reconsider your decision for the sake of my patients. I am a TAVR operator and have seen life changing impact of TAVR on high risk/inoperable patients. I am eagerly awaiting to see similar impact from the Watchman device on patients who will have the risk of stroke reduced without hemorrhaging from being on longterm systemic anticoagulation.

Less

To the CMS Officer,

I am writing this response to your proposed decision that “left atrial appendage closure is not reasonable and necessary to treat an illness or improve function and therefore not covered under the social security act.”

I am a Board Certified Interventional Cardiologist and practice at Southcoast Health System in Fall River. We are the largest Health Care System in Southeastern Massachusetts, East Bay and Rhode Island. Per recent AHA/NCDR data we accounted for the second largest atrial fibrillation inpatient discharges in the state of Massachusetts in 2013-2014. We also serve about 70% Elderly Medicare and/or Medicaid patient population. Hence, I believe I am in a position to give you a good perspective on these “real world patients” with atrial fibrillation outside an ideal clinical trial setting.

As you surely know, there are between 3-6 million Americans suffering from atrial fibrillation in the United States with about 800,000 new patients diagnosed every year. Atrial fibrillation accounts for about 20% of all strokes and patients with atrial fibrillation are 5 times more likely to experience a stroke. Hence, stroke prevention is as important as symptom control in the treatment of atrial fibrillation. While coumadin does decrease the risk of stroke in patients with atrial fibrillation, in the best of settings the time in “therapeutic range” with coumadin is only about 73-75%. This is due to many reasons including effects of diet on INR levels, patient compliance due to recurrent blood tests, especially in the elderly Medicare population where transportation can be quite a challenge. Novel anticoagulants are a recent alternative but the perceived risk of fatal bleeding due irreversible anticoagulation and the higher cost is a major deterrent. Also, the rate of non-compliance to Novel Anticoagulants and Coumadin is similar based on recent PINNACLE registry study.

Elderly people are at a higher risk of falls and bleeding per HAS-BLED score and these are the same patients also at a higher risk of stroke based on CHADS2 and CHADSVASC score. Unfortunately, this patient population is frequently under-anticoagulated. Many studies are also establishing the long term cost benefit of Watchman device in patients over the course of 5-8 years compared to being on long term anticoagulation. Patients also have a major restriction in their lifestyle and are unable to ski, bike, travel and hike outside due to the risk of bleeding. This can be a major blow to quality of life for younger active patients with atrial fibrillation. PROTECT-AF trial also demonstrated a mortality benefit over 4-year follow up in patients receiving Watchman device vs. long term coumadin. Hospitalization for anticoagulation related bleeding and testing and transfusions performed comes at a great cost to the health system.

I have several patients both young and old who have benefitted from the use of this device and now consider it an important part of how we treat and protect our pateints from stroke. Clearly, these patients need alternative to anticoagulation.

Hence, restriction of Watchman device and low reimbursement will limit this valuable therapy to a variety of patients who may need an alternative to long term anticoagulation and were in fact tested in clinical trials done which led to the approval of this device therapy by FDA in March 2015.

Hence, I implore CMS to reconsider their recent decision memo for left atrial appendage closure device to a broader patient population as studied in clinical trials. This will provide patients a viable alternative to long-term anticoagulation in prevention of stroke secondary to atrial fibrillation.

Sincerely,

Mitchel Sklar, MD FACC FSCAI
Cardiovascular Care Center
Southcoast Health System
Fall River, MA
Clinical Assistant Professor of Medicine
Warren Alpert Sch of Medicine
Brown University
Providence RI

Less

Hello,

I am an electrophysiologist practicing at deborah Heart and Lung Center which is a tertiary cardiovascular center. I have a large population of patients with atrial fibrillation who would potentially benefit from WATCHMAN device. With regards to the recently released proposed coverage decision to cover percutaneous LAAC therapy by CMS, I had a few concerns that I would like to express here:

1) The use of the phrase, "contraindicated to receive coumadin" when describing patients who would benefit from WATCHMAN is both scientifically incorrect and practically unhelpful. In all of the trials of the WATCHMAN device, those patients who had contra-indications to Warfarin were excluded and therefore it would not be scientifically correct to require warfarin contraindication when we generalize the results of the trials to the clinical practice. Of course, the main reason for patients to seek an alternative therapy to anticuagulation is high risk of complications including bleeding from their use but this does not and should not be considered a contra-indication. For example, an elderly person who had previous falls may in the physician's clinical judgement be of high enough risk for intracranial bleeding to justify using WATCCHMAN instead of coumadin but strictly speaking that patient does not have a contra-indication to warfarin. I suggest using a language that reflects the balance between the risks and benefits and allows for physicians to use their clinical judgement in addition to objective measures such as CHADS and HAS BLED scores.

2) HAS BLED score does not reflect some of the other risks of bleeding complications such as intra-cranial bleeding due to falls or presence of dual anti-platelet therapy. These situations should be considered in the final decision policy.

3) In some younger patients who require anticoagulation, they may require working in professions such as construction or professional sports that puts them at higher risk of bleeding complications when in fact their HASBELD score might be low. There should be some provision in the guideline to allow for inclusion of these patients as well.

Regards,

Less

I have been practcing academic Interventional cardiology for over 20 years, during which i was part of the development phase for several medical technologies, the last several years were particulary exciting with the advances in structural heart disease, namely the TAVR procedure and the watchman procedure.

The Watchman procedure has stood significant scientific scrutiny , rigorous clinical trials and proved to be aplicable to daily clinical practice. while i agree that the procedure shouild be performed in selected institutions with the clinical expertise and available infrastructure, i don't agree with limiting the indication to the patients with 'anticoagulants contraindication' because this is not supported with the trials evidence and will eliminate significant proportion of the population that will benefit from this procedure.

Our clinical experience with the procedure has been major success on the efficacy side and the safety aspect. for this and many other reasons and on behave of many patients i would like to plea with CMS to widen the scope of coverage that i am sure it will prove to be profitable for every one over the long run.

Less

I agree with most all the requirements set forth in the proposed policy. I have 2 concerns:
1. The CHADS scoring system is no longer recommended for clinical use. I understand it was used in the clinical trials of the Watchman, but this was prior to CHADSVasc supplanting the CHADS scoring system. I think providers should use the CHADSVasc system and at their discretion, select appropriate candidates. My practice is to consider it's use only for CHADSVasc score >/= to 3.
2. The clinical trials of the Watchman all stipulated that patients be eligible for 5 years of warfarin therapy, as this was a randomized trial. To change the indications to warfarin-ineligible patients negates the positive effects the device demonstrated as an alternative to warfarin. I believe the device should be available to selected warfarin eligible patients, as the benefit was demonstrated in the trials. Thank you.

Less

To the CMS officer,

I am writing this response to your proposed decision that “left atrial appendage closure is not reasonable and necessary to treat an illness or improve function and therefore not covered under the social security act.”

I am a Board Certified Interventional cardiologist and practice at Southcoast Health System in Fall River. We are the largest Health Care System in Southeastern Massachusetts, East Bay and Rhode Island. Per recent AHA/NCDR data we accounted for the second largest atrial fibrillation inpatient discharges in the state of Massachusetts in 2013-2014. We also serve about 70% Elderly Medicare and/or Medicaid patient population. Hence, I believe I am in a position to give you a good perspective on these “real world patients” with atrial fibrillation outside an ideal clinical trial setting.

As you surely know, there are between 3-6 million Americans suffering from atrial fibrillation in the United States with about 800,000 new patients diagnosed every year. Atrial fibrillation accounts for about 20% of all strokes and patients with atrial fibrillation are 5 times more likely to experience a stroke. Hence, stroke prevention is as important as symptom control in the treatment of atrial fibrillation. While coumadin does decrease the risk of stroke in patients with atrial fibrillation, in the best of settings the time in “therapeutic range” with coumadin is only about 73-75%. This is due to many reasons including effects of diet on INR levels, patient compliance due to recurrent blood tests, especially in the elderly Medicare population where transportation can be quite a challenge. Novel anticoagulants are a recent alternative but the perceived risk of fatal bleeding due irreversible anticoagulation and the higher cost is a major deterrent. Also, the rate of non-compliance to Novel Anticoagulants and Coumadin is similar based on recent PINNACLE registry study.

Elderly people are at a higher risk of falls and bleeding per HAS-BLED score and these are the same patients also at a higher risk of stroke based on CHADS2 and CHADSVASC score. Unfortunately, this patient population is frequently under-anticoagulated. Many studies are also establishing the long term cost benefit of Watchman device in patients over the course of 5-8 years compared to being on long term anticoagulation. Patients also have a major restriction in their lifestyle and are unable to ski, bike, travel and hike outside due to the risk of bleeding. This can be a major blow to quality of life for younger active patients with atrial fibrillation. PROTECT-AF trial also demonstrated a mortality benefit over 4-year follow up in patients receiving Watchman device vs. long term coumadin. Hospitalization for anticoagulation related bleeding and testing and transfusions performed comes at a great cost to the health system.

Hence, restriction of Watchman device and low reimbursement will limit this valuable therapy to a variety of patients who may need an alternative to long term anticoagulation and were in fact tested in clinical trials done which led to the approval of this device therapy by FDA in March 2015.

Hence, I implore CMS to reconsider their recent decision memo for left atrial appendage closure device to a broader patient population as studied in clinical trials. This will provide patients a viable alternative to long-term anticoagulation in prevention of stroke secondary to atrial fibrillation.

Sincerely
Adam J. Saltzman, MD
Cardiovascular Care Center
Southcoast Health System
Fall River, MA

Less

Less

Less

Less

Thank you for this opportunity to discuss the draft National Coverage Determination for the Watchman left atrial appendage closure device. I would like to make a few comments about the device and highlight our group's experience.

I am an electrophysiologist in a private practice group. I have extensive experience managing patients with non-valvular atrial fibrillation, including performing hundreds of afib ablations. We have been performing Watchman implants since June 2015.

The Watchman device was approved by the FDA for left atrial appendage closure in the spring of 2015 after extensive investigation in 2 randomized, controlled trials (PROTECT-AF and PREVAIL) as well as 2 registries (CAP and CAP 2). In brief, those studies found the Watchman device to be non-inferior to warfarin in stroke prevention in patients with non-valvular afib. 5-year follow up data from the PROTECT AF trial was recently reported (Reddy et al, JAMA 312:1988, 2014), demonstrating the device to be superior to warfarin for the primary endpoint of stroke/systemic embolism/cardiovascular death. Finally, and most importantly, not only was there a statistically-significant 60% relative reduction in cardiovascular mortality, but there was also a 34% relative reduction (5.7% absolute reduction) in all-cause mortality (Reddy et al, JAMA 312:1988, 2014). Most recently, a meta-analysis of the aforementioned trials showed a significant reduction in hemorrhagic stroke; cardiovascular or unexplained death; and major, non-procedural major bleeding as well as a trend toward signifcant reduction in all-cause mortality with a mean follow up of 2.69 years (Holmes et al. J Am Coll Cardiol 2015;65:2614–23). Thus, there is extensive data supporting the efficacy of the Watchman device.

Since FDA approval of the Watchman device, my group has developed a team approach (consisting of 2 EPs, 1 interventional cardiologist, 1 echo cardiologist, and afib coordinator) in treating patients at risk for stroke (high CHADS2-VAS score) but with a relative contraindication for long-term anticoagulation. We created screening tools to ensure appropriate patient selection. We also developed a registry to track procedural data and post-implant follow up. We have now successfully implanted 25 Watchman devices with zero complications. 20 patients have successfully discontinued warfarin after their 45-day TEE, per protocol.

In talking with the patients who have discontinued warfarin, it is very striking how liberated they feel to be off anticoagulation. A lot of people, including myself, may not realize how much patients restrict themselves, either consciously or subconsciously, while on anticoagulation. They are now able to live without fear of having a major bleeding event but also remain at low risk for a stroke.

Finally, please consider the clinical need of our AF patients at risk for stroke. There remains a void in treating patients who are not candidates for long-term anticoagulation, most commonly due to a high bleeding risk. Even in the clinical trials for the novel anticoagulants (dabigatran, rivaroxaban, apixaban, and edoxaban), between 20% to 33% of the patients had to stop the NOAC by 2 years. Data also shows that these patients who are not deemed candidates for long-term anticoagulation are also the patients at highest risk for strokes. The Watchman device may be a potential therapy to fill that void.

So while oral anticoagulation will and should remain first-line therapy for our AF patients at risk for stroke, we still require an option for those patients who are not good candidates for long-term oral anticoagulation. Furthermore, I am convinced that the clinical and psychological effects of the Watchman device implantation and discontinuation of anticoagulation should be carefully taken into consideration.

I sincerely appreciate the opportunity to make these comments and hope that you consider them moving forward with this draft NCD.

Please do not hesitate to contact me directly at 901-201-1551 if you have any further questions or concerns.

Less

I appreciate the opportunity to comment on the CMS national coverage decision (NCD) for the Watchman device. I am an echocardiographer who has been in clinical practice for 19 years, and I help guide structural heart therapies including left atrial appendage occlusion with the Watchman device. I was a co-investigator in the PREVAIL study, which investigated the safety of the Watchman device. I continue to provide TEE guidance for implantation of the device as an FDA approved commercial therapy. I have substantial training and experience in clinical trials. I now lead the Division of Cardiology for one of the largest not for profit heath care systems in the US. I regularly care for patients with atrial fibrillation and patients at risk for stroke.

Our structural heart team applauds the recent proposed NCD by CMS to provide coverage for the Watchman procedure. Evidence provided through the landmark PROTECT AF and PREVAIL studies have clearly demonstrated the safety and efficacy of the Watchman device for stroke prevention as an alternative to oral anticoagulation therapy in patients with paroxysmal or chronic atrial fibrillation at risk for stroke. The FDA label has appropriately approved this therapy for patients at increased risk for bleeding with oral anticoagulation therapy, which would augment the benefit of Watchman over oral AC therapy in this patient subset. Based on the Pinnacle AF registry, as many as a third of patients with AF and an indication for oral AC do not actually take oral AC. Many of these patients at risk for stroke cannot take oral AC and have an unmet need. We believe this is a compelling public health problem, as these patients are left with significant stroke risk. Some of these patients go on to suffer strokes with all of the morbidity and mortality associated. We believe that many of these patients could genuinely benefit from Watchman therapy.

Seven conditions were proposed for coverage. My thoughts:

Condition 1. I agree that coverage should only include the Watchman device. Any other technology should only be approved upon completion of appropriate clinical trials that show clinical benefit.

Condition 2. The “contraindication to warfarin” language is vague and can be interpreted as an absolute contraindication to warfarin therapy. This would not be appropriate. The device has been should to be safe and effective as an alternative to warfarin therapy and was studied with a 6 week warfarin post implant treatment bridge. This is how the device is currently used and is mutually exclusive with an absolute contraindication to warfarin therapy. The device has not been studied without the six week post implant warfarin bridge. The FDA label indicates approval for patients with a relative contraindication to long term oral anticoagulation therapy not an absolute contraindication. I strongly suggest that the final CMS approval be for patients with a relative or absolute contraindication to oral AC therapy. The determination as to long-term oral anticoagulation therapy should be at the discretion of the treating physician; patient choice alone due to perceived bleeding risk concerns is not enough and should not be used in isolation for coverage. As long as some reasonable rational is given for increased bleeding risk concern, coverage should be provided.

Conditions 3,4, 5: I support strict physician training and infrastructure requirements to ensure patient safety.

Condition 6. I support the establishment of a national registry. This approach has become an invaluable source of information for other structural heart procedures such as TAVR and MitraClip therapy. However, I am very concerned that the registry include a “control” arm of patients receiving oral AC, as it would be non-randomized rendering any results of uncertain meaning.

Condition 7. I agree with the patient-centric approach of shared decision making for Watchman therapy. Physicians should review the risks benefits and alternatives to Watchman therapy with patients for informed consent to be given. This is the standard of care with all procedures and therapies. Documenting this in the pre-procedural note should suffice; additional paperwork is not necessary.

Thank you for the opportunity to comment on the NCD for the Watchman device. I believe it is a transformative technology that – if used appropriately – can improve the lives of many patients who suffer from atrial fibrillation.

Sincerely,

Geoffrey A Rose, MD FACC FASE
Chief of Cardiology
Sanger Heart & Vascular Institute
Professor of Medicine
Carolinas HealthCare System

Less

I appreciate the opportunity to comment on the CMS proposed decision regarding coverage for Left Atrial Appendage occlusion devices. Overall, the decision seems well thought out, and I especially agree with the need for a registry to track the safety and efficacy of these devices as they are used outside of clinical trials. I do however have concerns about some aspects of the coverage decision as outlined below:

Condition #2 lists a number of patient inclusion criteria, starting with a “high” CHADS2 or CHADS2-VASc score. This should be clarified; what exactly is a “high” score? Based on current guidelines from the ACC/HRS/AHA, anticoagulation should be considered for scores as low as 1. If these patients are felt to be high enough risk for embolic CVA to warrant anticoagulation, then it would seem reasonable to consider them for left atrial appendage occlusion as well. I would suggest either clearly defining what a “high” score is, or rephrasing it to something like “high enough CHADS2 score to clinically warrant use of oral anticoagulation therapy”.

Condition #2 also lists use of the HAS-BLED score. While it has been used in many clinical trials, in my experience, this is not a scoring system that is widely used in clinical practice (or well validated). By focusing primarily on risk of major bleeding requiring transfusion, it excludes patients who may have other very common and valid reasons for not tolerating oral anticoagulation. As examples, these would include, severe ecchymoses and bruising from anticoagulation, frequent nosebleeds (or other bleeding) not requiring transfusion, or difficulty maintaining a therapeutic INR (the HAS-BLED score tries to account for risk of bleeding from supra-therapeutic INRs, but not the increased risk of CVA in patients who are frequently sub-therapeutic despite best efforts). I would suggest broadening this criteria to be more in line with the FDA indication of “having an appropriate indication for a non-drug alternative” rather then restricting this to a high HAS-BLED score alone.

Finally, condition #2 states that the patient must have a “contraindication to warfarin”. This is problematic, as it directly contradicts the FDA approval for the Watchman which states that the device should only be used in patients who “are suitable for warfarin” primarily due to the use of warfarin for at least 45 days after implant in most of the clinical trials for this device. In addition, with the widespread adaptation of the novel oral anticoagulants, there are now many patients who can not tolerate warfarin but can tolerate one of the newer agents. I would suggest rephrasing this to “a contraindication to long term use of anticoagulation” which would both allow for short term anticoagulation use around the time of the procedure as needed, and also reflect the fact that there are multiple anticoagulation strategies that are tolerated differently by different patients.

Condition #6 requires participation in a national registry, which in itself, is an excellent idea. My concern involves the requirement for a control arm of patients on oral anticoagulation. From a scientific standpoint, this is obviously appealing, but will be nearly impossible to do from a practical standpoint since it will be very difficult to find patients on normal clinically indicated anticoagulation therapy who will be willing to participate in a long term registry. Speaking as a busy clinician who sees lots of patients with AF, I do not think there would be any realistic way to accomplish this goal of a meaningful registry “control arm”. In addition, the scientific utility would be limited since the patients suitable for left atrial appendage closure will by definition have different risk factors both for stroke and bleeding from those on oral anticoagulation making comparisons difficult. Randomized clinical trials with control arms demonstrating safety and efficacy will obviously be required by the FDA for any approval process, but I do not see a need for this to be part of a CMS required registry.

Thank you again for taking up this very important subject of left atrial appendage closure. From a clinical standpoint, there is an urgent need for a method to prevent cardioembolic stroke in atrial fibrillation patients who (for whatever reason) are not good candidates for long term oral anticoagulation.

Ian Woollett, MD, FACC, FHRS

Less

I am an academic Cardiologist at Columbia University/New York-Presbyterian Hospital in New York, and am writing regarding proposed National Coverage Decision for Percutaneous, Transcatheter Left Atrial Appendage Closure (LAAC) Therapy with the WATCHMAN device. I serve as the Director of the Cardiac Catheterization Laboratories at Columbia/New York-Presbyterian Hospital. Please note that I do receive institutional research support from Boston Scientific Corporation (the maker of this device), but this is NOT related to studies of this particular device. In addition to participating in clinical research and educational efforts in interventional cardiology, I treat a wide range of patients with cardiovascular disease, including many patients with atrial fibrillation.

First and foremost, as you know, atrial fibrillation is an increasing problem for an aging patient population, and anticoagulation is frequently underprescribed even in high-risk patients. This is partially due to educational gaps among treating physicians, but more frequently is related to patient preference and the considerable side effects of systemic anticoagulation with warfarin and other anticoagulants. In my own practice I have seen many patients who are at significant stroke risk due to atrial fibrillation but yet are either reluctant to take (or have had side-effects from taking) oral anticoagulation. Many of these patients may be able to take oral anticoagulation for a short time, but as a long-term strategy for stroke prevention, anticoagulation is highly sub-optimal for many patients.

It is in this context that the availability of left atrial appendage closure devices is so critically important. The ability to safely lower the risk of ischemic stroke (while at the same time reducing the risk of bleeding complications) would greatly enhance our ability to treat patients according both to their risk profile as well as to their preferences. Notably, all of the other stroke prevention therapies at our current disposal (save open surgical atrial appendage ligation) all systematically INCREASE bleeding complications concomitantly. Bleeding is an important problem in the atrial fibrillation population, and has been associated with many adverse events (and overall cost to the healthcare system).

My frank and honest assessment of the initial proposed national coverage decision is that it goes too far in stringently restricting the usage of this device for only those patients in whom oral anticoagulation is contraindicated. This is especially relevant given that oral anticoagulants are in fact typically prescribed for a short term after WATCHMAN implantation. Additionally, the approval of this device conditional upon enrollment in registries of both the device itself as well as a parallel medically treated registry seems too onerous at this time, especially given the abundant data that already exists in this regard.

Once again, as stated in my earlier letter regarding the initial assessment of the WATCHMAN data, I do hope that when considering the totality of the evidence in favor of the use of the WATCHMAN device, you recognize that this device will undoubtedly serve as an important addition to the armamentarium of treatments for a growing public health problem.

Less

Utilizing the HAS-BLED score inadequately captures the population of patients benefiting from LAA closure therapy. For example the frail patient who frequently falls, but has yet not had a major bleed from such a fall, is an example. Furthermore, the triple anti-thrombotic therapy patient may be not be at high-risk by HAS-BLED but clinically considered to be at high risk of a hemorrhagic complication. Finally, HAS-BLED does not perform well in predicting intracranial hemorrhage.

Requiring a post market surveillance study of those treated with an LAA device is appropriate and necessary; however, requiring data on a control group that is not treated with a device is impractical. This requirement will result in unnecessary effort, will ultimately fail, and won't shed any valid conclusions with respect to this therapy.

Less

Tolerance of oral anticoagulation exists along a spectrum. Most atrial fibrillation (AF) patients are well and most appropriately managed with chronic oral anticoagulation (OAC). However, there is a tremendous unmet need to prevent strokes and major bleeding complications among patients in whom the risk of chronic anticoagulation associated bleeding is either prohibitive or outweighs the benefits of this therapy. Alignment between the FDA indication and Medicare coverage for patients with reasons to avoid chronic OAC would convey a cohesive message of reserving Watchman as second line therapy with careful considerations of the risks and benefits of each therapy.

FDA approval of the Watchman device and Medicare reimbursement to facilitate its use are pivotal milestones for further research of the role of LAA closure for AF associated stroke prevention. In addition to monitoring the overall safety and efficacy of the Watchman device, a Medicare mandated registry is anticipated to address a large number of unanswered questions such as optimal adjunctive pharmacology and refined patient selection according to clinical variables. While inclusion in a national registry is welcomed, the proposed requirement for a contemporaneous OAC comparator group is a major distraction from more important therapy development questions, and is inappropriate given the proposed second line indication for Watchman. Such comparative outcomes research is inherently flawed and difficult to construct. Many high stroke risk AF patients who represent an unacceptable risk for chronic oral anticoagulation are currently waiting for the Watchman device. A robust registry without a contemporaneous OAC comparison requirement will allow deserving patients to receive the Watchman device while furthering this field in the best interest of patients and research.

Less

Dear Sirs,

I am an Electrophysiologist practicing in the New Jersey suburbs of Philadelphia. Because of my affiliation with the Cooper University Hospital and Rowan Medical School, I have had the opportunity to become an early adopter of the WATCHMAN technology since FDA approval. To date we have implanted approximately 25 patients, without complication. Please allow me this opportunity to respond to the proposed CMS Coverage decision.

Currently there are over 7000 unique patients in my practice with a diagnosis of atrial fibrillation. Off these patients approximately 40% are on aspirin alone, or in combination with plavix for CVA risk reduction. Of these patients, those with CHADS2vasc scores of 2 or higher comprise the vast majority(85+%) and in these patients the number of patients on anticoagulation is approximately 70%. The main determinants for witholding anticoagulation are prior bleeding and concerns for patient frailty/risk of bleeding. Of those on blood thinners, a significant number (~10%) have had at least one major or multiple minor bleeding episodes in the past, yet remain on anticoagulants for fear of stroke.

My concerns about the proprosed coverage decision revolve around where the line will be drawn in regards to bleeding risk and what does "contraindication to warfarin" mean. In my opinion, if someone has had a bleeding event on one anticoagulant, their risk of future bleeding is higher and they should be offered an alternative therapy such as WATCHMAN. The words contraindicated are nebulous to me in this regard, as some physicians might accept higher risk bleeding for their patient rather than expose them to the risk of stroke. The truly "contraindicated" patient is a rarity, and a separate problem. For example, I saw a patient last week who had atrial fibrillation and a recent major spontaneous intracranial hemorrhage on coumadin requiring craniotomy. Per her neurosurgeon, she is contraindicated to any anticoagulant. Meanwhile Boston Scientific's recommendation for WATCHMAN implant requires 6 weeks of anticoagulation post procedure followed by 6 months of dual antiplatelet therapy. This is a patient who IS "contraindicated" to warfarin in whom I might not be able to, or be willing to implant a WATCHMAN device.

My recommendation is that the final coverage decision be worded to offer physicians the option to implant a patient at appropriate CVA risk (CHADS2vasc> or = 2) with a WATHCMAN device if they are deemed at increased risk for bleeding. This could be demonstrated by a HASBLED> or = 2 which implies a moderate increase in bleeding risk. Warfarin or other novel anticoagulants should be considered acceptable for short term management periprocedurally. I also think that some thought should be given to bleeding risk outside of the realm of the HASBLED score. Patients in whom their choice of careers is accompanied with increased risk of physical injury, or those with inherent bleeding risk due to hematological issues or in need of frequent interventional procedures. Without leaving the door widely open, some language should be provided to offer physicians with options for these patients.

Let me close by saying that I recognize the difficulty of adopting new policy, both to assure appropriate use and prevent excessive medical care costs with novel therapies. I ask you to give me the framework within which I can offer patients in need and at risk, another choice for long term CVA risk reduction. To be too strict with coverage will leave some patients at risk, either for bleeding on anticoagulants, or stroke on inferior protection strategies such as aspirin.

Less

I am an interventional cardiologist at Baylor University Medical Center in Dallas TX.
I agree with CMS coverage conditions point 1,3-7.

I don't think point 2 with regard to contraindication to warfarin is consistent with prior watcham study data. In fact, by study design patients should be treated with warfarin for 45 days post device implantation until adequate seal is confirmed by repeat TEE. In fact if an adequate seal is not obtained—pts should remain on warfarin.

While I agree patients who are warfarin/anticoagualtion intolerate are ideal patients for LAA closure therapy, the requirement of having an absolute contraindication to warfarin is not consistent with prior study data.

Less

1) Currently I am Chair of Interventional Cardiology at a single specialty heart hospital.

2) I spend nearly 100% of my time in the cardiac catheterization laboratory or hybrid operating room and perform high volume coronary, peripheral and structural heart interventions.

3) I have placed multiple Watchman devices and have concluded that, in properly trained hands, this device can be delivered easily and safely in the vast majority of patients with non-valvular AF. The safety of this device is supported by >10 years of experience at European Centers and numerous clinical studies.

Regarding CMS's proposed coverage decision several comments are in order, specifically regarding Conditions 2 and 6:

Condition 2: Specifically, requiring patients to have a high HAS-BLED score and a contraindication to Warfarin will eliminate a huge patient population that will potentially benefit from the Watchman device:

It is also important to point out that the addition of NOACS has not had any major influence in the way non-valular A-fib patients are treated: NOACS have not resulted in an increased rate of anticoagulation in high risk non-valvular A-fib patients (Jani et al, Uptake of Novel Oral Anticoagulants in Patients with Non-valvular and Valvular Atrial Fibrillation: Results from the NCDR-Pinnacle Registry; ACC 2014) Discontinuation of anticoagulation therapy due to bleeding is no different with the NOACS than it is with warfarin (Connally et al; NEJM 2009;361: 1139-1151; Patel et al; NEJM 2011;365:883-891; Granger et al; NEJM 2011; 365:981-992) Also, CMS should consider the one time cost of device implantation, which is likely to be orders of magnitude less than a lifetime of pharmacotherapy, laboratory testing, major morbidity and mortality from bleeding events, as well as morbidity and mortality from embolic stroke in patients at risk from bleeding on anticoagulation, or who are unprotected from stroke by inadequate anticoagulation.

Condition 6: Although it is reasonable to conduct a registry of patients receiving the Watchman device, collecting data on long term oral anticoagulants as a control arm would be at best of very limited utility, and at worst highly misleading. Unlike randomized controlled trials (which have already been done comparing Watchman with anti-coagulation), this type of registry would be plagued by selection bias and would not allow comparison of "apples to apples". Patients on oral anticoagulation are likely to be at lower bleeding risk AND lower stroke risk than patients receiving the Watchman device. Data demonstrates that the higher the CHADS2 score, the less likely the patient is to be anti coagulated (Piccini et al; Pharmacotherapy in Medicare Beneficiaries with Atrial Fibrillation; Heart Rhythm 2012;9:1403-1408)

These comments are respectfully submitted in the hope that CMS will understand and realize the true potential benefit of approving coverage of Watchman for a much broader spectrum of patients, including enhancing quality of life, improving long term outcomes and actually lowering the cost to the health care budget by reducing morbid events in large subsets of patients that would not be covered by more restrictive conditions.

Less

To CMS

I appreciate the opportunity to comment. I am an interventional cardiologist in practice for 13 years and perform structural heart therapies including left atrial appendage occlusion with the Watchman device. I was an investigator in the PREVAIL study which investigated the safety of the Watchman device and continue to implant the device as an FDA approved commercial therapy. I serve as our Director of Clinical Research for the Sanger Heart and Vascular Institute and have substantial training and experience in clinical trials. I work for the cardiovascular service line for one of the largest not for profit heath care systems in the US. I regularly care for patients with atrial fibrillation and patients at risk for stroke.

I applaud the recent proposed national coverage decision by CMS to provide coverage for the Watchman procedure. Evidence provided through the landmark PROTECT AS and PREVAIL studies have clearly demonstrated the safety and efficacy of the Watchman device for stroke prevention as an alternative to oral anticoagulation therapy in patients with paroxysmal or chronic atrial fibrillation at risk for stroke. The FDA label has appropriately approved this therapy for patients at increased risk for bleeding with oral anticoagulation therapy which would augment the benefit of Watchman over oral AC therapy in this patient subset. Based on the Pinnacle AF registry as many as a third of patients with AF and an indication for oral AC do not take oral AC. While the absence of oral AC therapy does not indicate that all these patients have a valid contraindication for oral AC clearly some of these patients at risk for stroke cannot take oral AC and have an unmet need. This is a compelling public health problem. In my own practice I see such patients frequently. While most patients in my experience can safely be treated with oral AC therapy as many as 20% of patients at risk for stroke have bleeding risk that make long term oral AC unfavorable. These patients are left with significant stroke risk. Some of these patients go on to suffer strokes with all of the morbidity and mortality associated. Some of these patients could genuinely benefit from Watchman therapy.

Seven conditions were proposed for coverage. I have concerns about some of these conditions as written. I will address them below.

Condition 1. I agree that coverage should only include Watchman and other technology should be approved contingent on completion of appropriate clinical trials that show a similar benefit to risk profile.

Condition 2. The “contraindication to warfarin” language is vague and could be interpreted as an absolute contraindication to warfarin therapy. This would be inappropriate. The device has been should to be safe and effective as an alternative to warfarin therapy and was studied with a six week warfarin post implant treatment bridge. This is how the device is currently used and is mutually exclusive with an absolute contraindication to warfarin therapy. The device has not been studied without the six week post implant warfarin bridge. The FDA label indicates approval for patients with a relative contraindication to long term oral anticoagulation therapy not an absolute contraindication. I strongly suggest that the final CMS approval be for patients with a relative or absolute contraindication to oral AC therapy. Furthermore this contraindication to long term oral anticoagulation therapy should be at the discretion of the treating physician. I agree that patient choice alone due to perceived bleeding risk concerns should not be enough. Alternatively HAS-BLED score alone should not be the determinant as this risk score does not fully take into account legitimate bleeding risks such as fall risk and previous bleeding. In the absence of fall risk or previous bleeding HAS-BLED can be helpful in identifying patients at risk for bleeding on oral AC who might benefit from Watchman therapy. A HAS-BLED score of greater than or equal to 2 is associated with increased bleeding risk with even a HAS-BLED risk of 2 associated with nearly a 2% per year risk of major bleeding. That said I believe the physician is the best judge of bleeding risk in an individual patient as such estimates are complex and often not objectively quantifiable in individual patients. The HAS-BLED score is not widely used clinically and should not be used in isolation for coverage. As long as some reasonable rational is given for increased bleeding risk concern coverage should be provided. In terms of stroke risk criteria a CHADS VASC score of greater than or equal to 2 should be used as this is the FDA label and these patients would otherwise be considered candidates for oral AC. Setting a higher limit would be arbitrary and illogical.

Conditions 3,4,5. I completely support strict physician training and infrastructure requirements to ensure patient safety.

Condition 6. I agree with the establishment of a national registry and this has become a useful source of information for other structural heart procedures such as TAVR and MitraClip therapy. I am very concerned and perplexed over suggestions that the registry include a “control” arm of patients receiving oral AC. Such a “control arm” would not be randomized and therefore would provide no useful information beyond any generic registry of oral AC therapy. It could not be used to compare outcomes with Watchman treated patients as it would be terminally confounded based on the registry design. In addition to providing no useful information it would also be impractical to implement. Inputting data on literally thousands of patients uninvolved in Watchman therapy would be prohibitively expensive and time consuming. Finally compliance with such a “control arm” would be impossible to enforce as CMS would be unable to verify that all eligible patients were entered. The proposed “control arm” to the registry is absolutely unworkable.

Condition 7. I agree with shared decision making for Watchman therapy. Physicians should review the risks benefits and alternatives to Watchman therapy with patients for informed consent to be given. This is the standard of care with all procedures and therapies. I do not favor a requirement of formal paperwork documentation. This provides only additional burden without material benefit. We do not need more administrative paperwork and such paperwork cannot prevent bad actors from doing bad things. Simple documentation of a discussion of risks benefits and alternatives in consultation notes and procedure informed consent should suffice.

The Watchman device represents a breakthrough technology for stroke prevention in AF and provides for an unmet need in a significant number of patients. Thank you for the opportunity to comment and I look forward to the final coverage decision language.

Less

To: Ms. Tamara Syrek-Jensen, JD
Director of Coverage & Analysis Group
From Duane C. Berkompas, MD, FACC
Chief Division of Cardiology Spectrum Health
Grand Rapids, MI

Dear Ms. Syrek-Jensen

I am sending this comment to ask for your support on the proposed national coverage decision to cover percutaneous LAAC therapy with the Watchman Device.

As you know, based on several studies, most importantly the PROTECT AF and PREVAIL clinical trials, the Watchman LAAC device was approved for implant by the FDA early this year. Our hospital was chosen as one of the early sites in Michigan to provide this new and exciting therapy for patients. Our institution has a very successful Structural Heart program and are one of the leading centers for TAVR implants in the US. In addition, we have a very large Cardiology Electrophysiology program. There has been outstanding collaboration across specialties to make these programs successful.

We began implanting the Watchman LAAC device in June of this year after developing our team of implanting physicians and support staff. We have been constrained in our volume mainly due to the lack of insurance payers to authorize and approve this procedure. In spite of these limitations, we have successfully implanted 13 devices and have 3 more cases next week. We have 20 patients who are trying to patiently wait for the CMS proposal to cover the Watchman LAAC procedure.

Our typical patient so far has significant risk for stroke due to their history of atrial fibrillation with CHADS2Vasc scores ranging from 3 to 7. All have of course been able to tolerate being on Coumadin, but have had a major clinically significant bleeding episode on chronic anticoagulation, with HAS-BLED scores ranging from 2 to 6 as well. All our patients have been very eager to have the implant performed with the hopes of being able to be on aspirin only to protect them from stroke. As part of our consent process we spend a great deal of time reviewing the risks and benefits of the procedure compared to chronic anticoagulation and provide them with websites about the procedure and information brochures.

In review of the seven conditions CMS is proposing in order to cover percutaneous LAAC therapy, I believe all conditions are appropriate. I do believe that for a LAA procedure to be approved, there should be rigorous trials attesting to outcomes of stroke prevention compared to chronic anti-coagulation as we have with the Watchman Device.

Related to the requirement for a high HAS-BLED score, we believe there is a need to consider patients for this procedure that may not have a high HAS-BLED score. This includes patients in a high risk for bleeding lifestyle or occupation. Specifically, we have a 60 year old gentleman who competes in mountain bike racing on a national level. He does not want to give this up and has been willing to accept the risks of his racing while being on chronic anticoagulation, but is hoping to be approved for the Watchman procedure so that he can take aspirin only and not have to worry about a biking accident with significant trauma while being on Coumadin or a NOAC agent. We have other patients who suffer from atrial fibrillation at a young age and are looking at the possible need for chronic anticoagulation for up to 30 years, every year with a 2-3% risk for major bleeding event. Some of these patients would definitely benefit from having protection against stroke provided by the Watchman implant while being able to maintain a very active lifestyle without the risks of major bleeding caused by chronic Coumadin therapy.

Finally, we also advocate for specific physician and institution requirements to be met to be allowed to perform this procedure. The implant procedure has significant risks, and we know from other high risk procedures that higher volume leads to lower rates of procedural complications. I believe the team of an Interventional Cardiologist, Electrophysiologist and TEE specialist provide for excellence in outcome and safety of the procedure. In our short experience we have use CT angiography of the chest to further define a patient's LAA anatomy, as well as Intra-cardiac ECHO to assist with the procedure. Without the experience and expertise of our team to decide when additional tests and imaging were needed, we would not have been able to perform this procedure in at least 3 of our patients. Collaboration across specialty expertise, as we have seen in the TAVR arena, has proven to be needed and indeed a necessary component for the safety of this procedure.

Finally, we also approve the proposal to have a national registry to collect data on outcomes on patients treated with this technology. Despite the volumes of data obtained in the multiple Watchman studies, we can learn more through data collection in a registry to make sure our current device implant and follow up care is accomplishing that which we are attempting to provide - freedom from risk for stroke in afib patients and freedom from the bleeding events associated with chronic anticoagulation. However, any comparison against the NOAC's should be done in a proper fashion. Trying to compare patients treated with NOAC's to patients treated with the Watchman Device outside of a randomized clinical trial would of course have questionable results given the large potential for bias and statistical errors.

On behalf of the patients we are responsible for and the opportunity to provide new and exciting, and life altering procedures and therapies, I thank you for your consideration of the comments provided in this comment.

Less

I am Chairman of Cardiovascular Diseases and director of the Insitute for Heart and Vascular Health. Thus I look at this procedure and device from an operator perspective as an interventional cardiologist and from an administrative perspective.

At Einstein we reat an innercity populationthat is older than the average patient in the US and at a significant lower socio-economic status and in an under served area fo primary care services. This combination provides a difficult population to treat.

This device has provided opportunities for patients with increased risk of bleeding and in certain situations of long term noncompliance. This device has significant potential to address the health disparities that we face.

The following are comments on the recently released coverage decision by CMS regarding left atrial appendage closure (LAAC).

Comments will be directed to aspects of the seven proposed conditions for coverage.

Condition 1 addresses the WATCHMAN device. This is reasonable as it is the device that has the most evidence supporting its use.

When reviewing Cindition 2 regarding coverage criteria, there needs to be clarification on "contraindication to warfarin." The use of the device in patients with high risk of beeding on warfarin is appropriate. The FDA indications recommends warfarin for a period of time after the procedure. There should be a strong effort to allow coverage of all on-label indications. There should vbe an effort to not exclude a subset of patients who could benefit from the device per labeled indications.

Regarding Conditions 3, 4, and 5 physician training and hospital infrastructure requirements for performing the procedure should be consistent with clinical trials and the FDA guided commercialization experience to support safety and efficacy of the device.

Condition 6 addresses the proposal of a registry. All iperators involved in the device use agree with an ongoing registry. The inclusion of a control arm of patients on OAC could be problematic and would probably not include "all-comers." It would only likely contain those who prove to not be a candidate for the device. This may not be a representative selection of patients and has the potential to confuse analysis. Other options should be considered to find a method to look at patients on OAC.

Condition 7 addresses shared decision making. All operators support this concept. It might be reasonable to look at a heart team decision, also as in TAVR.

Less

I am currently the director of the cardiac cath lab at the Washington Hospital Center and participated in the WATCHMAN trials.

I have reviewed the proposal for reimbursement guidelines and find most recommendations for reimbursement both rational and reasonable. The only major ones that need to be contested include:

1. The selection of patients that have a high HAS-BLED score and are coumadin ineligible. Neither of these were evaluated in the trials. In fact, coumadin is required in the IFU once the device in implanted for the first 45 days.

2. Although a national coumadin national registry is an interesting proposition, it is outside the scope of this device application and would create excessive economic and resource burdens to both providers and institutions.

Less

In regards to the recently approved LAA Watchman closure procedure(CAG-00445N)and the CMS proposed coverage decision:

I am an interventional cardiologist and fully trained structural heart interventionalist, currenly working at the Cleveland Clinic Florida, were I serve as Section Head of Structural Heart Disease.

I have a special interest in Stroke prevention therapies such as Watchman LAA closure procedure. I am actively involved with it commercial use at JFK Medical Center in Florida, and have been actively involved with educating the physicians and patients on appropriate patient selection based on RCT and FDA approved indications.

Although, I endorse the proposed coverage decision. I have concerns about some of the conditions that have been outlined.

A) Language regarding "contraindication to warfarin" should be revised. Current, approved indications mandate eligibility for warfarin (at least short term); as supported by the results of the RCTs. The use of coumadin is indicated up to a least 45 days after the procedure.

B)The wording "high” CHADS2 or CHA2DS2-VASc scores should be revised, to reflect a specific number to avoid confusion and facilitate proper patient selection. RCT and current guidelines encourage use of OAC therapy in patients with CHADSVASC score > or equal to 2. It is not clear to me what "high" implies. This may lead to improper patient selection.

C) Although the HASBLED score is useful to quantify potential bleeding complications. The current risk model was never implemented for patient selection in the RCTs. Furthermore, the model has not yet been incorporated and supported by the current guidelines recommendations. In my practice, although the HASBLED score has theorical relevance, other important clinical considerations are ignored by the model that should be considered when making this important clinical decision; Including yet not limited to fall risk, dietary habits, medication compliance, physical level of activites.

Finally, I applaud CMS effort to provide coverage for this novel therapy which I have personally seen to truly affect the lives of patients at exposed risk of lifethreatening strokes.

Thank you for the consideration of my comments.
Robert J. Cubeddu

Less

Less

Thank for the opportunity to comment on the Proposed Decision Memo for Percutaneous Left Atrial Appendage (LAA) Closure Therapy (CAG-00445N)

I am an interventional cardiologist at a large tertiary care academic medical center in central California and have been in practice for > 10 years. My focus area of expertise is structural and valvular heart disease.

My clinical experience with the Watchman device includes participation in the pivotal PROTECT AF trial, the CAP registry, and as a current commercial implant site for the Watchman device since FDA approval in March 2015.

The patients that I have treated with the Watchman device to date have derived benefit from avoiding the bleeding complications of long term oral anticoagulation as demonstrated in multiple clinical trials and registries (PROTECT AF Trial Holmes et al., Lancet 2009; 374: 534–42; Meta-Analysis: Homes et al, J Am Coll Cardiol 2015;65:2614–23).

It is also important to note that patients derive significant physical and mental quality of life benefit from avoiding long term oral anticoagulation (Alli et al. J Am Coll Cardiol 2013;61:1790–8).

With this background of experience, I have two specific concerns regarding the proposed CMS coverage decision:

Point 2

The wording that patients have “high” CHADS2 or CHA2DS2-VASc scores should be revised to mimic the current guidelines for anticoagulation.

Current ACC/AHA and European Society guidelines state that any patient with a CHADS2 greater than or equal to 2, or CHA2DS2-VASc score greater than or equal to 1 by current should be considered for oral anticoagulation (OAC).

I believe the term “elevated” would be a better choice than “high” since patients considered for Watchman should mimic those patients currently considered for OAC under current guidelines.

Since the maximum CHADS2 score is 6, and the maximum CHA2DS2-VASc score is 9, we do not want to encourage physicians to only treat those patients with the highest scores who are at therefore at the highest overall risk for CVA. A meta-analysis of available studies shows that most patients treated with Watchman in clinical trials had CHA2DS2-VASc scores of 3-4 (Meta-Analysis: Homes et al, J Am Coll Cardiol 2015;65:2614–23).

Similarly, the phrase “high” HAS-BLED score should be changed to “elevated” HAS-BLED score since many patients who have had life-threatening bleeding with OAC do not have markedly elevated HAS-BLED scores.

For instance, a patient with prior life-threatening intracranial hemorrhage while on warfarin may have a HAS BLED score of only 1 or 2. The same applies to a patient with recurrent significant urologic or gastrointestinal bleeding. A meta-analysis of available studies shows that most patients treated with Watchman in clinical trials had HAS-BLED scores of 1-2 (Meta-Analysis: Homes et al, J Am Coll Cardiol 2015;65:2614–23).

Finally the phrase the only patients who are “contraindicated to warfarin” should be changed to reflect the wording of the current FDA approval which is to treat “atrial fibrillation patients who are eligible for warfarin, but have reasons to seek another long-term therapeutic option.”

This change in wording is important for two reasons. First, Watchman device implantation currently mandates 6 weeks of warfarin and aspirin therapy immediately following implantation as utilized in the clinical trials. Therefore, patients must be able to tolerate at least a short course of warfarin and cannot be “contradindicated” to any warfarin therapy. Second, clinical trials for Watchman have not studied patients “contraindicated” to warfarin so the proposed wording would encourage treatment of patients for which there is a lack of data to support efficacy.

Point 6

The proposed decision memo states that:

“The registry must include contemporaneous patients followed on oral anticoagulant (OAC) therapy to serve as non-interventional controls.”

This request is unreasonable for several reasons. First, neither current TAVR or MitraClip registries require a concurrent “control group.” Second, this requirement is not feasible nor will it be useful to answer any meaningful clinical question given the great heterogeneity of patients treated with OAC therapy. Third, this level of patient selection and data collection would place undo hardship on site clinical personnel and place demands more akin to a clinical trial. Finally, clinical follow-up in the registry > 5 years is an unreasonable demand on site personnel.

Thank you for the consideration of my comments.

Sincerely,
Jason H. Rogers MD, FACC, FSCAI
Professor, Cardiovascular Medicine
Director, Interventional Cardiology
University of California, Davis Medical Center
Sacramento, California
jhrogers@ucdavis.edu

Less

Kenneth A. Ellenbogen, M.D.
kenneth.ellenbogen@vcuhealth.org

Percutaneous Left Atrial Appendage (LAA) Closure Therapy (CAG-00445N)

To: Ms. Tamara Syrek-Jensen, J.D., Director of Coverage & Analysis Group

Dear Ms. Syrek-Jensen:
I am currently Chief of Cardiology at the VCU School of Medicine in Richmond, VA. We have a large multi-subspecialty Cardiology Practice at an urban medical center and an affiliated VA Medical Center. I am writing to comment on the recent CMS Proposed Coverage Decision for Percutaneous Left Atrial Appendage Closer Therapy (LAAC) with the WATCHMANTM Device. Our primary hospital is a large hospital with a bed capacity of almost 800 beds. We are frequently at or close to 100% capacity and see a large variety of patients who are economically disadvantaged and have limited financial resources. A large percentage of our patients are minority and indigent patients, who often cannot get care elsewhere.
In addition to my clinical responsibilities, I am on the editorial board of multiple medical journals. I am aware of through my professional association with multiple cardiology journals including the American Heart Journal, Journal of Cardiovascular Electrophysiology, Heart Rhythm, Circulation: Arrhythmia and Electrophysiology and PACE. I have already begun implanting this device in my patients.
I am fully cognizant of the fact that this device has been widely available at more than 50 countries in the world and has been implanted in Europe for over 5 years. Over 10,000 devices have already been implanted. The data is hard to argue with, and it is very clear to me that this device is the only device that has been extensively studied for patients with atrial fibrillation and has been proven in a large randomized double blind study to be superior to warfarin in long term cardiovascular outcomes and associated with a lower mortality than warfarin.
The CMS proposed decision is not supported by medical literature. It is contrary to the many papers published on this subject and is contrary to the European guidelines. It should be recommended for patients with a high stroke risk and elevated bleeding risk or labile INRs or recurrent risk of falls. The current guidelines worsen the already existing disparities in health care where wealthy patients are more likely to receive the WATCHMAN device or be placed on expensive NOACs (novel oral anticoagulants). I am surprised and discouraged by the fact that the current recommendations will especially effect poorer patients, patients with less educational and financial resources, and patients who have less access to tertiary providers. The description of patients having a “contraindication to warfarin” is unnecessarily vague and poorly defined. For example, a 78 year old female with a CHADS-VASc score of 3 with 6 INRs ranging from 1.5, 2.5, 2.9, 1.3, 4.2, 2.1 who has a time in therapeutic range (TTR) of 50% would in my mind be an excellent candidate for this device just as a 72 year old women with hypertension who has spinal stenosis and frequent falls would also be an excellent candidate for the WATCHMANTM.
I do, however, strongly support some sort of national registry but am totally confused and baffled by what value and what need there is to require a control arm. The purpose of the registry is to determine how the safety of device usage compares to that during the many clinical trials. It would be useful for determining if there are any long term problems with this device. However, it should be a subset of patients who will be receiving the device and would be an undue burden to physicians, hospitals, nurses and our limited financial resources to include all patients receiving the device.

Sincerely,

Kenneth A. Ellenbogen, M.D., FACC, FAHA, FHRS
Chairman, Division of Cardiology, VCU School of Medicine
Kontos Professor of Cardiology
VCU School of Medicine
Richmond, VA

Less

I strongly support a CMS coverage decision for the Watchman Left atrial appendage occlusion device. I believe the device should be made available to patients with atrial fibrillation unrelated to valvular heart disease who are indicated for long term anticoagulation but where long term anticoagulation is contraindicated because of a bleeding risk or a fall risk. This means patients with atrial fibrillation and a CHADS2 or CHADS2VASc score > or = to 1

Anticoagulation related bleeding may be overt, but the (frequent) cases of occult blood loss should also be covered in my opinion. The HASBLED score may be useful to assess bleeding risk in some patients who have yet to experience overt bleeding on anticoagulants, but stipulating the use of HASBLED should be avoided because this score was validated in patients taking warfarin - and very few patients are initiated on warfarin currently.

Fall risk can only be assessed in a well documented discussion between the patient and physician.

CMS should consider also covering patients who may be able to take an anticoagulant but whose occupation may make long term anticoagulation inadvisable. Examples include jobs involving hard physical activity such as farm work, carpentry and other building trades. Again this would need to be carefully documented in a physician patient discussion.

I strongly support a national registry of indications and complications related to implantation. I also support rigorous training standards for implanting physicians, and certain basic standards for implanting hospitals. For example cardiac surgery should in my opinion be available of site and at short notice in any implanting center.

Less

I would like to make a few comments regarding the draft National Coverage determination for the Watchman left atrial appendage closure device.

I have extensive experience with atrial fibrillation (having performed over 2000 ablations) as well as managing patients at risk for embolic stroke. I have performed left atrial ligation using the lariat device for over 4 years now. More recently, since june of this year, I started implanting the watchman left atrial closure device.

The Watchman device has secured FDA approval after a number of clinical trials and registries have confirmed its safety and efficacy as an alternative to oral anticoaculation. This device is also approved in Europe. It has been extensively studied in rigorous FDA clinical trials – two randomized studies comparing to Warfarin (PROTECT AF and PREVAIL) and two non-randomized FDA registries (CAP and CAP-2). Furthermore, it was also reported last year that not only did Watchman remain noninferior to warfarin for the primary endpoint even after a mean follow-up of 2.3 years, but there was also a statistically-significant ~60% decrease in cardiovascular mortality in the Watchman patients (Reddy et al, Circulation 127:720, 2013). Long-term (5-year) follow up data from the PROTECT AF trial was also recently reported (Reddy et al, JAMA 312:1988, 2014) and the device was found to be superior to warfarin for the primary endpoint of stroke/systemic embolism/cardiovascular death. Finally, and most importantly, not only was there a statistically-significant 60% relative reduction in cardiovascular mortality, but there was also a 34% relative reduction (5.7% absolute reduction) in all-cause mortality (Reddy et al, JAMA 312:1988, 2014).

I have now had the chance to see firsthand the clinical applicability of this device in treating my patients at risk for stroke, and who have relative contraintications for anticoagulation long term – such as patients at an increased risk of bleeding based to the HAS-BLED score.

Our group has implanted the watchman device in 25 patients since we started our program a few months ago. We have taken this very seriously, creating our own screening tools, our own registry and have carefully followed these patients. A few things that are worth highlighting from our watchman program:

• We are a private practice group, working out of 2 hospitals. Performing watchman implants in one of them.
• We have created a group with 2 Eps, 1 structural heart disease specialists, 1 TEE specialist, an afib coordinator and the research infrastructure we already had in place.
• 25 implants have been performed, no complications.
• 20 patients already underwent their 45 day TEE and have been able to discontinue Coumadin per protocol given the results
• All of our patients had CHADS2Vasc of 3 or greater and increased HASBLED scores

Our results have been similar to the registries, showing the feasibility and safety of this procedure when done outside the clinical trial realm, as long as by experienced physicians and in a team approach similar to what is being done for valve clinics/TAVR.

What has been truly amazing to me is the patient experience. I can honestly say that few times as a physician have I seen patients as happy as we can now witness when anticoagulants are discontinued. We have seen significant improvements in quality of life that are very easy to measure. I truly believe that the lack of constant fear from complications secondary to chronic anticoagulation has such a profound impact on our patients’ lives. We have now so many examples, and this has really gotten me to become an advocate for this technology.

Finally, please consider the clinical need of our AF patients at risk for stroke. Historically only a small percentage of patients with afib were adequately anticoagulated with warfarin and a large number of patients stop it. While the advent of the NOACs (Dabigatran, Rivaroxaban, Apixaban, and now Edoxaban) has certainly increased the options available, a large number of patients remain untreated. In fact, even in the NOAC clinical trials, between 20% to 33% of the patients had to stop the NOAC by 2 years. Data also shows that typically patients at the highest risk of stroke are the ones who are more likely to stop or have it stopped by their physicians for fear of complications. So while oral anticoagulation will and should remain first-line therapy for our AF patients at risk for stroke, we still require an option for those patients who are not good candidates for long-term oral anticoagulation. Furthermore, I am convinced that the clinical and psychological effects of the watchman device implantation and discontinuation of anticoagulation should be carefully taken into consideration. I sincerely appreciate the opportunity to make these comments and hope that you consider them moving forward with this proposed NCD. Please feel free to contact me directly at (205) 790-7776 with any questions.
Sincerely

Less

Public response RE: Proposed decision memo for percutaneous left atrial appendage closure therapy (CAG-00445N)

Dear CMS:

I am writing this response to your proposed decision that “left atrial appendage closure is not reasonable and necessary to treat an illness or improve function and therefore not covered under the social security act.”

I am a board certified cardiac electrophysiologist and Fellow of the Heart Rhythm Society. My practice is exclusively electrophysiology and 80% of my patients are seeing me for treatment of atrial fibrillation. I practice in a geographically isolated region in Michigan’s Upper Peninsula. While my practice is in a relatively rural setting (Marquette, Michigan), I see over 1,000 AF patients annually. Many of my patients drive 100 miles or more to see me and drive 30 miles or more to reach their closest critical access hospital where they can have an INR checked to monitor their warfarin therapy or treat their warfarin related bleeding episode. I have no financial ties to industry. I do not own stock in Boston Scientific or any other companies developing left atrial appendage occlusions products. I do not get speaker’s fees from any company. Boston Scientific does not support any research I’m involved in. I do hope to some day implant Watchman devices for my patients. I have done 12 implants during fellowship a decade ago. I realize that the assigned wRVU and therefore physician compensation will likely be so low that it will not be worth my time, radiation exposure and liability risk. Similarly I fully expect that the hospital reimbursement will probably be less than the cost of the device, making Watchman a money losing procedure for the hospital that employs me. To be clear, the only horse I have in this race is my patient’s interest, health and welfare.

As I’m sure you are keenly aware, up to 1/5th of all strokes are caused by atrial fibrillation. Atrial fibrillation patients are 5 times more likely to experience a stroke. Strokes due to atrial fibrillation are larger strokes and more likely to be fatal or disabling, and therefore more costly to society and the health care system. These are facts that every one of my new patients referred for AF leave the office understanding. I tell them that warfarin reduces the risk of stroke by up to 68% and reduces death by 25%. I tell them that the novel oral anticoagulants are comparably effective in preventing stroke with lower risk of intracranial hemorrhage and they’re easier to take, but perhaps more costly. I tell them that aspirin does very little to reduce the risk of stroke and in 3 of 5 studies didn’t show a statistically significant reduction in strokes. I take my obligation to educate my AF patients about stroke risk and it’s prevention very seriously. Despite that 1 in 4 of those at significant risk leave the office without agreeing to take oral anticoagulant therapy. Of the 3 in 4 who start an anticoagulant one of those will not be on it when they return for their follow-up visit. Some stop because of labile INRs on warfarin leading to unacceptably frequent blood draws and monitoring, some because of bleeding events on warfarin. Others stop because of the perceived risk of bleeding from the novel oral anticoagulants as they’re terrified by the onslaught of television and radio commercials from unscrupulous lawyers. Many stop because they have to choose between paying the $300 co-pay for their Eliquis or feeding their family and the risk of starving seems greater than the few percent per year risk of stroke. Some are forced to stop because of major bleeding events on the novel oral anticoagulants though honestly that’s quite uncommon. Studies of “real world” data suggest that half of those with AF and an increased risk of stroke are not on anticoagulant therapy and my practice experience mimics those numbers. Perhaps that’s in part because those at greatest risk of stroke (high CHADS2 scores) are also those at greatest risk of bleeding complications from oral anticoagulant therapy.

Being a sub-investigator in the Protect-AF study as a fellow at William Beaumont Hospital years ago I’ve followed the Watchman device closely and have been eagerly awaiting it’s availability to my patients. It seems as if the FDA did all it possibly could to suppress the development of the technology. Three advisory panels, additional randomized studies to evaluate safety, following patients out years after implant. An astute statistician recognizes that eventually all Kaplan-Meier survival curves come back together as everyone will eventually die, but the data from the Protect-AF study are compelling. Similar protection from ischemic stroke, less hemorrhagic stroke and intracranial hemorrhage, less bleeding and lower cardiovascular mortality in those randomized to the Watchman device. The longer the patients are followed the better the data look and the greater the statistical significance of those findings. For CMS to conclude that Watchman does not offer any advantage in the CMS patient population is simply preposterous. There are many people in my practice that would be helped by the availability of the Watchman device. While those who would derive greatest benefit, those who cannot take long term anticoagulant therapy, were labeled contraindicated by the FDA and not eligible, those who choose not to take anticoagulant therapy (many for good reason) could derive substantial protection from the Watchman device. Perhaps the CMS panel members have never taken warfarin or perhaps they are not subject to the same coverage, or more aptly, lack thereof for novel oral anticoagulants. CMS’s conclusion also appears counter to that of the FDA and this inconsistency across governmental agencies may be interpreted by some as another example of the incompetency of our governmental bureaucracy.

On behalf of my patients I am requesting that CMS reconsider it’s proposed coverage decision that percutaneous left atrial appendage occlusion does not offer benefit in the treatment of atrial fibrillation patients. Admittedly the technology will result in significant expenditures, and I suspect that is why my patients have been blocked access to this technology for a decade longer than those in Europe. This cost may be offset, by the savings from not treating bleeding events in those on warfarin, not paying $5,000+ yearly for each patient on novel oral anticoagulants and not paying for long term care for those who refuse anticoagulant therapy and are disabled by stroke.

Sincerely,

Rudolph F. Evonich, MD, FACC, FHRS

Less

Less

Less

The Watchman has been heavily anticipated among the medical community because it does offer a new hope to provide patients an effective tool at reducing stroke risks and cardiovascular mortality due to atrial fibrillation. Warfarin and NOACs have been a troublesome issue for numerous patients in our population due to bleeding, labile INRs, disruption due to surgery needs, difficulty with transportation to have routine bloodworm, and most importantly costs with respect to NOACs. We have seen many patients now opt for no therapy instead unfortunately we have seen some complications from this as well. I commend CMS on providing a framework for NCD for the Watchman device, however I would not include strict restrictions on having HAS-BLED scores included in the criteria. Furthermore, one of the main restrictions to having a greater exposure of TAVR to the general cardiology population has been the restriction to access. Not all patients from our community have the availability to travel to academic centers. We have been disappointed in this because we have been able to provide high level of care to our patients and have the necessary infrastructure to make both TAVR and Watchman devices successful. Yet, the limited access has disrupted this and I fear similar restrictions will hinder our ability to provide this therapy to a significant number of patients. I agree on the proposals to have necessary criteria in place for site availability and physical training along with an adequate statement on what constitutes procedural competency for EP and Interventional cardiology when it comes to transeptal punctures.

Less

I am writing to endorse approval of National Coverage Determination for the Watchman left atrial appendage occluder device (FDA approved). This device represents a significant advance for treatment of patients with non-valvular atrial fibrillation who are at high risk of serious bleeding.

While I applaud CMS for appreciating the need for this device, I have significant concerns regarding some of the language in the Proposed Decision Memo for Percutaneous Left Atrial Appendage (LAA) Closure Therapy, dated November 10, 2015.

Use of the HAS-BLED score should not be mandatory for device payment coverage. This score is useful clinically, but like all scores, has inherent limitations. There needs to be room for the expert judgment of an experienced clinician.

Additionally, the term “a contraindication to warfarin” is ambiguous. I recommend again that clinical judgement of the treating physician prevail, such that while a patient may be a able to tolerate warfarin, there may be situations where it is reasonable to seek a non-drug alternative.

Thank you for the opportunity to submit these comments.

Less

1. Currently the coverage proposal states that patients must have a 'contraindication to warfarin'. This should read, 'have a compelling to reason to avoid anticoagulation.' One could argue that even patients who have bleeding risks or episodes could still be on warfarin. The patients who benefited from LAAC in PROTECT AF did not have a contraindication to warfarin. Or it should read "Has a high HasBLED score..." OR "a contraindication to warfarin' not AND.

2. The TSP requirement is a catch-22 for some skilled operators. If the operator hasn't met the baseline 25/10 requirement, then the operator must perform the TSP with a physician proctor present and scrubbed into the case.

Less

I am a busy Electrophysiologist practicing in Delaware at Christianacare Hospital. I treat thousands of patients with atrial fibrillation. While I believe the vast majority of patients with atrial fibrillation can be appropriately managed with anticoagulation therapy for stroke prevention, the Percutaneous Left Atrial Appendage Closure device is a critical option for subgroups of these patients. After reviewing CMS proposal for coverage of this therapy I have the following comments:

1. Condition 2: uses language "contraindication to warfarin". This would seem to contradict the language used by the FDA during approval of the device which stipulates the this therapy would be appropriate for high risk patients seeking an alternative to long-term anticoagulation therapy. Furthermore, since warfarin use is required for at least 45 days after the procedure, I believe this language is confusing. Would suggest changing to "patients deemed to be at high risk for long-term warfarin use"

2. Condition 2: "high CHADS2, CHA2DS2-VASc score"
Needs clarification. Most of the clinical trials evaluating anticoagulant therapy and the clinical trials evaluating the Watchman LAA Closure device defined a high score as 2 or greater. A CHA2DS2-VASc score of 2 is associated with a 2.2% yearly risk of stroke or 5 year cumulative risk of 11% which most would agree is quite high.

3. Condition 2: uses language "high HASBLED score". Needs clarification. Again, HASBLED score of 2 equates with 1.88% yearly risk of major bleed or 5 year cumulative risk of 9.4% which most would agree is quite high.

4. Condition 2: requirement of high HASBLED score may eliminate patients with history of major bleeds on anticoagulation or needing antiocoagulation. For example a 62 year old female with congestive heart failure, diabetes and permanent nonvalvular atrial fibrillation develops life-threatening GI bleed while on warfarin. Her CHA2DS2-VASc score is 3 (3.2% yearly risk of stroke) but her HASBLED score is only 1. Clearly this patient should be offered access to Percutaneouos Left Atrial Appendage Closure therapy. Condition 2 language should be changed to "HASBLED score of 2 or more or history of major bleeding"

5. Condition 2: Should consider history of recurrent falls as risk factor for significant bleeds. This is a very common scenario for patients. I just saw a 66 year old male with permanent non-valvular atrial fibrillation,diabetes, prior thromboembolic stroke, and prior MI (CHA2DS2-VASc score= 5). He has gait instability with multiple recent falls resulting in hip fracture and many ER visits. He is obviously at high risk of bleeding on anticoagulation and should be considered candidate for Percutaneous Left Atrial Appendage Closure therapy.

Thank you

Brian H. Sarter, MD, FACC, FHRS

Less

I am writing to endorse the approval of National Coverage Determination for an FDA approved implantable device for percutaneous left atrial appendage closure for those patients with non-valvular atrial fibrillation for the purpose of reduction of cardioembolic risk of ischemic stroke. While I applaud CMS for appreciating the obvious need for a device of this nature, and the thoroughness of the evidentiary data, I have significant concern with some language within the Proposed Decision Memo for Percutaneous Left Atrial Appendage (LAA) Closure Therapy, dated November 10, 2015.

Within my group, administrative surveillance identified 6,182 patients with atrial fibrillation, with an average age of 81 years. The mean CHA2DS2-VASc in this group was 3.9. Active surveillance of a 10% sample identified 22.6% with significant history of bleeding (13.4% intracranial bleeding, 56.4% gastrointestinal bleeding, 38.9% anemia requiring transfusion). While the HAS-BLED score may be considered to be useful for identification at those with elevated risk for complication, it is obvious that the same limitations with HEMOR2RHAGES apply towards HAS-BLED; notably the broad overlap with CHADS2, and the weighted balance towards antiplatelet therapy within the model and significantly less, representation of those with bleeding complications as a results of systemic anticoagulation. While propensity towards bleeding is important, once the patient has a history of disabling intracranial bleed, or life-changing gastrointestinal bleed, as examples, the HAS-BLED score takes on a different clinical meaning, relative to the bedside clinical judgement. Within our cohort, of those with a prior history of intracranial bleeding, gastrointestinal bleed or anemia requiring transfusion, the use of systemic anticoagulation reduced to 63.6%, despite a CHA2DS2-VASc of 4.1 within this group. The difference in HAS-BLED score for those in whom bedside clinical judgement felt continued systemic anticoagulation was or was not reasonable and safe to continue were no different for those patients with a history of prior bleed. This proprietary data is available upon request. I recommend instead of requisite high HAS-BLED score, there be afforded clinical judgement, such that while the patient is recommended for systemic anticoagulation, and suitable for warfarin, that there exists a reason to seek a non-drug alternative.

Although my group does not have the volume of younger adults as seen in other groups, I spent 15 years in the military treating very young and very active adults with atrial fibrillation. While those young adults with lone atrial fibrillation may not have a high CHA2DS2-VASc score, certainly there are those patients with risk factors in which a recommendation for systemic anticoagulation is warranted. In the same fashion that biologic versus mechanical valve replacement is in many ways dependent on the risk of need for systemic anticoagulation in the young adult; even in the absence of an elevated HAS-BLED score, the 1-year predictive nature of such a propensity model has little application in a patient in whom a 10-year or greater requirement for systemic anticoagulation. Again, I recommend instead of requisite high HAS-BLED score, there be afforded clinical judgement, such that while the patient is recommended for systemic anticoagulation, and suitable for warfarin, that there exists a reason to seek a non-drug alternative.

As is obvious to the authors of the initial draft, the ambiguous nature of the phrase “a contraindication to warfarin” should be further refined. Whether the decision is made to state that such a contraindication is relative, to avoid direct conflict with approved FDA Directions for Use, or discarding such statement is imaginable. I recommend instead of contraindication to warfarin, there be afforded clinical judgement, such that while the patient is recommended for systemic anticoagulation, and suitable for warfarin, that there exists a reason to seek a non-drug alternative.

Thank you for the opportunity to comment on this technology, so boldly recognized by both the FDA and CMS, as a paradigm shift in the reduction of non-valvular atrial fibrillation associated cardioembolic risk reduction. If clarification is warranted, please contact me directly at 941-917-4250.

Less

I am EP physician, practicing in Phoenixville Hospital, PA.
I am excited and fully supporting CMS proposal for national coverage of the Watchman device.
Watchman device provide unique stroke prevention therapy for patients who can not take warfarin.
At the present time, private and HMO type insurances are denying this therapy for very straightforward indications: pt with history of stroke and brain bleeding, was denied. Patient with stroke 1 week after severe GI bleeding, was denied.
National coverage is essential to get access to this therapy for most of the patients who are needed Watchman device.
My suggestions for second condition to replace "high" CHA2DS2VASc score to more specific score.
Patients with CHA2DS2VASc score 2 or greater.
"High" score will be available for interpretations and some insurances might take advantage from this open statement.
Also, I recommend to use CHA2DS2VASc score rather than CHADS2 score which is currently recommended by most societies.

Less