Columbia, Dartmouth, U of Washington, Yale
Aortic stenosis (AS) is the most common valvular heart disease requiring intervention in the United States, and the past decade has witnessed a profound, rapid evolution in treatment paradigms. Based on an unprecedented clinical trial effort, transcatheter aortic valve replacement (TAVR) has been approved and rapidly adopted for the treatment of intermediate and higher surgical risk patients with symptomatic severe AS. (Leon MB et al. NEJM 2016; Reardon MJ et al. NEJM 2017) Indeed, by 2017, the volume of transcatheter aortic valve replacement (TAVR) surpassed surgical aortic valve replacement (SAVR) in the United States. (Michaels J. Structural Heart Disease Summit, June 21, 2018) Recent clinical trials in low-risk surgical candidates have now provided further high quality evidence that TAVR is a reasonable option for the vast majority of patients with AS. (Mack MJ et al. NEJM 2019; Popma JJ et al. NEJM 2019) The new availability of evidence-based therapeutic alternatives for AS demands the incorporation of patient values and preferences into final decision making by an experienced heart team.
The extremely rapid change in treatment paradigms for AS presents a significant challenge for regulatory agencies and payers. We therefore appreciate the opportunity from CMS, in response to clinicians advocating for their patients (Drs. Pelikan, Robertson and Wright, CMS website), to reconsider the 2012 National Coverage Determination (NCD) for TAVR. Our recommendations for revisions to the new Proposed Decision Memo, along with rationale and data to support them, are humbly submitted below.
1. Current wording: “One cardiac surgeon has independently examined the patient face-to-face, evaluated the patient's suitability for surgical aortic valve replacement (SAVR), TAVR or medical or palliative therapy, and has documented the rationale for their clinical judgment, and the rationale is available to the heart team.”
Recommendation: “Both a cardiac surgeon experienced in surgical approaches and a cardiologist experienced in transcatheter approaches to aortic valve disease have independently examined the patient face-to-face, evaluated the patient's suitability for all options including surgical aortic valve replacement (SAVR), TAVR or medical or palliative therapy, and have documented the rationale for their clinical judgment, with the rationale available to the heart team.”
Although the randomized data on low surgical risk patients were not available at the time of the formulation of the proposed NCD, we believe that it is profoundly important to account for these landmark trials when proceeding with a coverage decision that is likely to have a lasting impact on clinical practice. Along with the existing evidence base, these trials unequivocally establish that TAVR is at least equivalent to SAVR with respect to important early clinical outcomes, including death, stroke, and rehospitalization. TAVR also offers a more rapid recovery and improvements in quality of life, particularly when performed from the transfemoral access route. It is therefore no longer appropriate that cardiac surgeons act as “gatekeepers” and unilaterally control patient access to this therapy. Indeed, it is of particular concern that approximately half of all SAVR centers in the United States currently do not offer TAVR, and it is unclear whether patients at these centers will have equal access to TAVR. We urge CMS to treat all patients with AS equally, regardless of the therapy for which they are initially referred, and mandate a true heart team approach with a shared decision making process for every patient.
Shared decision making is a process in which clinicians perform 3 distinct tasks: they 1) present all of the reasonable options available to patients, along with their risks and benefits; 2) listen to patients’ informed values and preferences and 3) use those preferences to deliberate with patients and families, and come to consensus on a decision. (Hess EP et al. Circ CV Qual and Outcomes 2014; Barry MJ, Annals of Int Med 2002) CMS has several precedents for mandating shared decision making in NCDs: lung cancer screening, left atrial appendage (LAA) closure and implantable cardioverter defibrillators (ICD). Evolution of how National Coverage Decisions can most effectively respond to stakeholder input regarding mandates for a shared decision making process led to a broader inclusion of team members, including advanced practice providers (as in the ICD NCD).
The current TAVR NCD moves to the other end of the spectrum, requiring only a cardiac surgeon speak with the patient about the choices available to them, recording their interaction somewhere that a “heart team” can review. There is no evidence that cardiac surgeons alone have the interest or skillsets to lead patients and their families through the decision making process for the treatment of aortic stenosis, with equal consideration of all options as listed in the proposed NCD. In fact, the literature suggests that individual physicians continue to fall short when it comes to shared decision making, suggesting the importance of the heart team approach in this specific task. (Hsu, et al. Med Dec Making 2017; Taylor, et al. JAMA Surgery 2017; Coylewright et al. Circ CV Qual and Outcomes 2016). Therefore, we recommend using the precedent of the transcatheter mitral valve repair NCD, written in 2014, that calls for both a surgeon and a cardiologist experienced in valvular heart disease to meet with the patient and their family, and present all reasonable options that are available to the pati
2. Current wording: “CMS supports patient shared decision making in AVR but there is not a fully developed tool available at this time.”
Recommendation: “A formal shared decision making encounter must occur between the patient and a member of the heart team, with patient values and preferences documented in the medical record. Current best practices include use of an independent patient decision aid when available.”
The TAVR NCD, breaking in precedent from recent decisions including those covering other preference-sensitive decisions such as stroke prevention in atrial fibrillation (LAA closure NCD) and prevention of sudden cardiac death (ICD NCD), does not highlight a major role for shared decision making. In the proposed NCD background information, there is recognition that recent Expert Consensus Documents highlight the importance of this approach. (Otto CM, et al. JACC 2017; Bavaria JE, et al. JACC 2019; Nishimura RA et al. JACC 2019) However, CMS raises concern that there are not adequate tools available to support shared decision making. Indeed, there are at least 3 available patient decision aids publically available, developed using rigorous standards and with independent support from the American College of Cardiology and the Patient Centered Outcomes Research Institute (PCORI). These decision aids cover intermediate, high and prohibitive risk patients (see sharedcardiology.org, “tools”).
Notably, the shared decision making literature is clear that patient decision aids are helpful, but not necessary, to lead to a shared decision making process. This process is defined by the elicitation of informed patient values and preferences, and incorporation of these patient goals into final decision making. While decision aids are important, clinician skillsets in shared decision making are even more critical, and favorable clinician attitudes surpass both in the requirements for implementation of shared decision making. (Joseph-Williams, et al. 2017 BMJ) Clinician attitudes towards shared decision making are clearly shaped by health policy requirements, and thus the TAVR NCD has an opportunity to embed shared decision making within the clinical visit for aortic valve disease even as patient decision aids are being refined and made more readily available.
3. Current wording: “The heart team's interventional cardiologist(s) and cardiac surgeon(s) must jointly participate in the intra-operative technical aspects of TAVR.”
Recommendation: “The heart team's interventional cardiologist(s) and cardiac surgeon(s) must jointly participate in the intra-operative technical aspects of TAVR.” (no change)
In order to maintain the excellent outcomes seen in the randomized clinical trials and national registries, it is essential that the skillsets of the proceduralists are highly developed and maintained. Several analyses demonstrate the existence of learning curves and volume-outcome relationships in the setting of TAVR. (Carroll J, et al. JACC 2017; Vemulapalli et al. NEJM 2019) A recent publication by prominent cardiac surgeons that are highly skilled in TAVR implored their professional societies to move towards “an emergency overhaul” of training for cardiac surgeons to ensure the development of appropriate skills for transcatheter procedures. (Nguyen TC, et al. J Thorac Cardiovasc Surg in press 2019) We are in full support of the need for development of wire and catheter-based skills amongst cardiac surgeons for the intra-operative aspects of TAVR, as in interventional cardiology. (Raphael C, et al. Circ Cardiovasc Interv 2017) At present, we believe that maintenance of a team approach to the intra-operative technical aspects of TAVR is most likely to ensure reproducibility of the excellent safety profile demonstrated not only in trial settings, but through the national registries that serve us well.
In conclusion, we believe that the heart team approach remains critical for decision making regarding which options are reasonable to offer to patients with severe AS; for true engagement of patients and families in shared decision making about those options; and for appropriate partnering in the intra-operative technical aspects of TAVR, with continued evolution of training programs. Therefore, we recommend a heart team approach for all patients with severe AS considering aortic valve replacement.
Megan Coylewright, MD; Dartmouth-Hitchcock Medical Center
John K. Forrest, MD; Yale University School of Medicine
James M. McCabe MD; University of Washington Medical Center
Tamim M. Nazif, MD; Columbia University Irving Medical Center