Transcatheter Aortic Valve Replacement (TAVR)

The Centers for Medicare & Medicaid Services (CMS) will cover Transcatheter Aortic Valve Replacement (TAVR) for the treatment of symptomatic aortic valve stenosis through Coverage with Evidence Development (CED).

1. The procedure is furnished with a complete aortic valve and implantation system that has received FDA premarket approval (PMA) for that system's FDA approved indication.
2. The patient (preoperatively and postoperatively) is under the care of a heart team: a cohesive, multi-disciplinary, team of medical professionals. The heart team concept embodies collaboration and dedication across medical specialties to offer optimal patient-centered care. The heart team includes the following:
   1. Cardiac surgeon and an interventional cardiologist experienced in the care and treatment of aortic stenosis who have:
      1. independently examined the patient face-to-face, evaluated the patient’s suitability for surgical aortic valve replacement (SAVR), TAVR or medical or palliative therapy;
      2. documented and made available to the other heart team members the rationale for their clinical judgment.
   2. Providers from other physician groups as well as advanced patient practitioners, nurses, research personnel and administrators.
3. The heart team's interventional cardiologist(s) and cardiac surgeon(s) must jointly participate in the intra-operative technical aspects of TAVR.
4. TAVR must be furnished in a hospital with the appropriate infrastructure that includes but is not limited to:
   1. On-site heart valve surgery and interventional cardiology programs,
   2. Post-procedure intensive care facility with personnel experienced in managing patients who have undergone open-heart valve procedures,
   3. Appropriate volume requirements per the applicable qualifications below:

   There are two sets of qualifications; the first set outlined below is for hospital programs and heart teams without previous TAVR experience and the second set is for those with TAVR experience.

   Qualifications to begin a TAVR program for hospitals without TAVR experience:

   The hospital program must have the following:

   1. ≥ 50 open heart surgeries in the previous year prior to TAVR program initiation, and;
   2. ≥ 20 aortic valve related procedures in the 2 years prior to TAVR program initiation, and;
   3. ≥ 2 physicians with cardiac surgery privileges, and;
   4. ≥ 1 physician with interventional cardiology privileges, and;
   5. ≥ 300 percutaneous coronary interventions (PCIs) per year.

   Qualifications to begin a TAVR program for heart teams without TAVR experience:

   The heart team must include:

   1. Cardiovascular surgeon with:
      1. ≥ 100 career open heart surgeries of which ≥ 25 are aortic valve related; and,
   2. Interventional cardiologist with:
      1. Professional experience of ≥ 100 career structural heart disease procedures; or, ≥ 30 left-sided structural procedures per year; and,
      2. Device-specific training as required by the manufacturer.

   Qualifications for hospital programs with TAVR experience:

   The hospital program must maintain the following:

   1. ≥ 50 AVRs (TAVR or SAVR) per year including ≥ 20 TAVR procedures in the prior year ; or,
   2. ≥ 100 AVRs (TAVR or SAVR) every 2 years, including ≥ 40 TAVR procedures in the prior 2 years; and,
   3. ≥ 2 physicians with cardiac surgery privileges; and,
   4. ≥ 1 physician with interventional cardiology privileges, and
   5. ≥300 percutaneous coronary interventions (PCIs) per year;and,


5. The heart team and hospital are participating in a prospective, national, audited registry that: 1) consecutively enrolls TAVR patients; 2) accepts all manufactured devices; 3) follows the patient for at least one year; and, 4) complies with relevant regulations relating to protecting human research subjects, including 45 CFR Part 46 and 21 CFR Parts 50 & 56. 

   The following outcomes must be tracked by the registry; and the registry must be designed to permit identification and analysis of patient, practitioner and facility level variables that predict each of these outcomes:

   1. Stroke;
   2. All-cause mortality;
   3. Transient Ischemic Attacks (TIAs);
   4. Major vascular events;
   5. Acute kidney injury;
   6. Repeat aortic valve procedures;
   7. New permanent pacemaker implantation;
   8. Quality of Life (QoL).


6. The registry shall collect all data necessary and have a written executable analysis plan in place to address the following questions (to appropriately address some questions, Medicare claims or other outside data may be necessary). Specifically, for the CED question iv, this must be addressed through a composite metric. For the below CED questions (i-iv), the results must be reported publicly as described in CED criterion k.
   1. When performed outside a controlled clinical study, how do outcomes and adverse events compare to the pivotal clinical studies?
   2. What is the long term durability of the device?
   3. What are the long term outcomes and adverse events?
   4. What morbidity and procedure-related factors contribute to TAVR patients outcomes?

   Consistent with section 1142 of the Act, the Agency for Healthcare Research and Quality (AHRQ) supports clinical research studies that CMS determines meet the above-listed standards and address the above-listed research questions.

B. TAVR is covered for uses that are not expressly listed as an FDA-approved indication when performed within a clinical study that fulfills all of the following:

1. The heart team's interventional cardiologist(s) and cardiac surgeon(s) must jointly participate in the intra-operative technical aspects of TAVR.
2. As a fully-described, written part of its protocol, the clinical research study must critically evaluate not only each patient's quality of life pre- and post-TAVR (minimum of 1 year), but must also address at least one of the following questions:
   • What is the incidence of stroke?
   • What is the rate of all-cause mortality?
   • What is the incidence of new permanent pacemaker implantation?
   • What is the incidence of transient ischemic attacks (TIAs)?
   • What is the incidence of major vascular events?
   • What is the incidence of acute kidney injury?
   • What is the incidence of repeat aortic valve procedures?


3. The clinical study must adhere to the following standards of scientific integrity and relevance to the Medicare population:
   1. The principal purpose of the study is to test whether the item or service meaningfully improves health outcomes of affected beneficiaries who are represented by the enrolled subjects.
   2. The rationale for the study is well supported by available scientific and medical evidence.
   3. The study results are not anticipated to unjustifiably duplicate existing knowledge.
   4. The study design is methodologically appropriate and the anticipated number of enrolled subjects is sufficient to answer the research question(s) being asked in the National Coverage Determination.
   5. The study is sponsored by an organization or individual capable of completing it successfully.
   6. The research study is in compliance with all applicable Federal regulations concerning the protection of human subjects found in the Code of Federal Regulations (CFR) at 45 CFR Part 46. If a study is regulated by the Food and Drug Administration (FDA), it is also in compliance with 21 CFR Parts 50 and 56. In addition, to further enhance the protection of human subjects in studies conducted under CED, the study must provide and obtain meaningful informed consent from patients regarding the risks associated with the study items and /or services, and the use and eventual disposition of the collected data
   7. All aspects of the research study are conducted according to appropriate standards of scientific integrity.
   8. The study has a written protocol that clearly demonstrates adherence to the standards listed here as Medicare requirements.
   9. The study is not designed to exclusively test toxicity or disease pathophysiology in healthy individuals. Such studies may meet this requirement only if the disease or condition being studied is life threatening as defined in 21 CFR §312.81(a) and the patient has no other viable treatment options.
   10. The clinical research studies and registries are registered on the www.ClinicalTrials.gov website by the principal sponsor/investigator prior to the enrollment of the first study subject. Registries are also registered in the Agency for Healthcare Quality (AHRQ) Registry of Patient Registries (RoPR).
   11. The research study protocol specifies the method and timing of public release of all prespecified outcomes to be measured including release of outcomes if outcomes are negative or study is terminated early. The results must be made public within 12 months of the study’s primary completion date, which is the date the final subject had final data collection for the primary endpoint, even if the trial does not achieve its primary aim. The results must include number started/completed, summary results for primary and secondary outcome measures, statistical analyses, and adverse events. Final results must be reported in a publicly accessibly manner; either in a peer-reviewed scientific journal (in print or on-line), in an on-line publicly accessible registry dedicated to the dissemination of clinical trial information such as ClinicalTrials.gov, or in journals willing to publish in abbreviated format (e.g., for studies with negative or incomplete results).
   12. The study protocol must explicitly discuss beneficiary subpopulations affected by the item or service under investigation, particularly traditionally underrepresented groups in clinical studies, how the inclusion and exclusion criteria effect enrollment of these populations, and a plan for the retention and reporting of said populations on the trial. If the inclusion and exclusion criteria are expected to have a negative effect on the recruitment or retention of underrepresented populations, the protocol must discuss why these criteria are necessary.
   13. The study protocol explicitly discusses how the results are or are not expected to be generalizable to affected beneficiary subpopulations. Separate discussions in the protocol may be necessary for populations eligible for Medicare due to age, disability or Medicaid eligibility.

Consistent with section 1142 of the Act, the Agency for Healthcare Research and Quality (AHRQ) supports clinical research studies that meet the above-listed standards and address the above-listed research questions.

The principal investigator must submit the complete study protocol, identify the relevant CMS research question(s) that will be addressed, and cite the location of the detailed analysis plan for those questions in the protocol, plus provide a statement addressing how the study satisfies each of the standards of scientific integrity (a. through m. listed above), as well as the investigator's contact information, to the address below. The information will be reviewed, and approved studies will be identified on the CMS Website.

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

Email address for protocol submissions: clinicalstudynotification@cms.hhs.gov
Email subject line: "CED [NCD topic (i.e. TAVR)] [name of sponsor/primary investigator]"

See Appendix B for the NCD manual language.

TO:		Administrative File: CAG-00430R

FROM:		Tamara Syrek Jensen, JD
		Director, Coverage and Analysis Group

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

		Lori Ashby, MA
		Director, Division of Policy and Evidence Review

		Rosemarie Hakim, PhD
		Acting Director, Evidence Development Division

		Sarah Fulton, MHS 
		Technical Advisor

		Kimberly Long 
		Lead Analyst

		Steven A. Farmer, MD, PhD
		Senior Medical Advisor
		
		Joseph Dolph Hutter, MD, MA
		Lead Medical Officer


SUBJECT:	National Coverage Determination for Transcatheter Aortic Valve Replacement (TAVR) 

DATE:		June 21, 2019

I. Decision

The Centers for Medicare & Medicaid Services (CMS) will cover Transcatheter Aortic Valve Replacement (TAVR) for the treatment of symptomatic aortic valve stenosis through Coverage with Evidence Development (CED).

1. The procedure is furnished with a complete aortic valve and implantation system that has received FDA premarket approval (PMA) for that system's FDA approved indication.
2. The patient (preoperatively and postoperatively) is under the care of a heart team: a cohesive, multi-disciplinary, team of medical professionals. The heart team concept embodies collaboration and dedication across medical specialties to offer optimal patient-centered care. The heart team includes the following:
   1. Cardiac surgeon and an interventional cardiologist experienced in the care and treatment of aortic stenosis who have:
      1. independently examined the patient face-to-face, evaluated the patient’s suitability for surgical aortic valve replacement (SAVR), TAVR or medical or palliative therapy;
      2. documented and made available to the other heart team members the rationale for their clinical judgment.
   2. Providers from other physician groups as well as advanced patient practitioners, nurses, research personnel and administrators.
3. The heart team's interventional cardiologist(s) and cardiac surgeon(s) must jointly participate in the intra-operative technical aspects of TAVR.
4. TAVR must be furnished in a hospital with the appropriate infrastructure that includes but is not limited to:
   1. On-site heart valve surgery and interventional cardiology programs,
   2. Post-procedure intensive care facility with personnel experienced in managing patients who have undergone open-heart valve procedures,
   3. Appropriate volume requirements per the applicable qualifications below:

   There are two sets of qualifications; the first set outlined below is for hospital programs and heart teams without previous TAVR experience and the second set is for those with TAVR experience.

   Qualifications to begin a TAVR program for hospitals without TAVR experience:

   The hospital program must have the following:

   1. ≥ 50 open heart surgeries in the previous year prior to TAVR program initiation, and;
   2. ≥ 20 aortic valve related procedures in the 2 years prior to TAVR program initiation, and;
   3. ≥ 2 physicians with cardiac surgery privileges, and;
   4. ≥ 1 physician with interventional cardiology privileges, and;
   5. ≥ 300 percutaneous coronary interventions (PCIs) per year.

   Qualifications to begin a TAVR program for heart teams without TAVR experience:

   The heart team must include:

   1. Cardiovascular surgeon with:
      1. ≥ 100 career open heart surgeries of which ≥ 25 are aortic valve related; and,
   2. Interventional cardiologist with:
      1. Professional experience of ≥ 100 career structural heart disease procedures; or, ≥ 30 left-sided structural procedures per year; and,
      2. Device-specific training as required by the manufacturer.

   Qualifications for hospital programs with TAVR experience:

   The hospital program must maintain the following:

   1. ≥ 50 AVRs (TAVR or SAVR) per year including ≥ 20 TAVR procedures in the prior year ; or,
   2. ≥ 100 AVRs (TAVR or SAVR) every 2 years, including ≥ 40 TAVR procedures in the prior 2 years; and,
   3. ≥ 2 physicians with cardiac surgery privileges; and,
   4. ≥ 1 physician with interventional cardiology privileges, and
   5. ≥300 percutaneous coronary interventions (PCIs) per year;and,


5. The heart team and hospital are participating in a prospective, national, audited registry that: 1) consecutively enrolls TAVR patients; 2) accepts all manufactured devices; 3) follows the patient for at least one year; and, 4) complies with relevant regulations relating to protecting human research subjects, including 45 CFR Part 46 and 21 CFR Parts 50 & 56. 

   The following outcomes must be tracked by the registry; and the registry must be designed to permit identification and analysis of patient, practitioner and facility level variables that predict each of these outcomes:

   1. Stroke;
   2. All-cause mortality;
   3. Transient Ischemic Attacks (TIAs);
   4. Major vascular events;
   5. Acute kidney injury;
   6. Repeat aortic valve procedures;
   7. New permanent pacemaker implantation;
   8. Quality of Life (QoL).


6. The registry shall collect all data necessary and have a written executable analysis plan in place to address the following questions (to appropriately address some questions, Medicare claims or other outside data may be necessary). Specifically, for the CED question iv, this must be addressed through a composite metric. For the below CED questions (i-iv), the results must be reported publicly as described in CED criterion k.
   1. When performed outside a controlled clinical study, how do outcomes and adverse events compare to the pivotal clinical studies?
   2. What is the long term durability of the device?
   3. What are the long term outcomes and adverse events?
   4. What morbidity and procedure-related factors contribute to TAVR patients outcomes?

   Consistent with section 1142 of the Act, the Agency for Healthcare Research and Quality (AHRQ) supports clinical research studies that CMS determines meet the above-listed standards and address the above-listed research questions.

B. TAVR is covered for uses that are not expressly listed as an FDA-approved indication when performed within a clinical study that fulfills all of the following:

1. The heart team's interventional cardiologist(s) and cardiac surgeon(s) must jointly participate in the intra-operative technical aspects of TAVR.
2. As a fully-described, written part of its protocol, the clinical research study must critically evaluate not only each patient's quality of life pre- and post-TAVR (minimum of 1 year), but must also address at least one of the following questions:
   • What is the incidence of stroke?
   • What is the rate of all-cause mortality?
   • What is the incidence of new permanent pacemaker implantation?
   • What is the incidence of transient ischemic attacks (TIAs)?
   • What is the incidence of major vascular events?
   • What is the incidence of acute kidney injury?
   • What is the incidence of repeat aortic valve procedures?


3. The clinical study must adhere to the following standards of scientific integrity and relevance to the Medicare population:
   1. The principal purpose of the study is to test whether the item or service meaningfully improves health outcomes of affected beneficiaries who are represented by the enrolled subjects.
   2. The rationale for the study is well supported by available scientific and medical evidence.
   3. The study results are not anticipated to unjustifiably duplicate existing knowledge.
   4. The study design is methodologically appropriate and the anticipated number of enrolled subjects is sufficient to answer the research question(s) being asked in the National Coverage Determination.
   5. The study is sponsored by an organization or individual capable of completing it successfully.
   6. The research study is in compliance with all applicable Federal regulations concerning the protection of human subjects found in the Code of Federal Regulations (CFR) at 45 CFR Part 46. If a study is regulated by the Food and Drug Administration (FDA), it is also in compliance with 21 CFR Parts 50 and 56. In addition, to further enhance the protection of human subjects in studies conducted under CED, the study must provide and obtain meaningful informed consent from patients regarding the risks associated with the study items and /or services, and the use and eventual disposition of the collected data
   7. All aspects of the research study are conducted according to appropriate standards of scientific integrity.
   8. The study has a written protocol that clearly demonstrates adherence to the standards listed here as Medicare requirements.
   9. The study is not designed to exclusively test toxicity or disease pathophysiology in healthy individuals. Such studies may meet this requirement only if the disease or condition being studied is life threatening as defined in 21 CFR §312.81(a) and the patient has no other viable treatment options.
   10. The clinical research studies and registries are registered on the www.ClinicalTrials.gov website by the principal sponsor/investigator prior to the enrollment of the first study subject. Registries are also registered in the Agency for Healthcare Quality (AHRQ) Registry of Patient Registries (RoPR).
   11. The research study protocol specifies the method and timing of public release of all prespecified outcomes to be measured including release of outcomes if outcomes are negative or study is terminated early. The results must be made public within 12 months of the study’s primary completion date, which is the date the final subject had final data collection for the primary endpoint, even if the trial does not achieve its primary aim. The results must include number started/completed, summary results for primary and secondary outcome measures, statistical analyses, and adverse events. Final results must be reported in a publicly accessibly manner; either in a peer-reviewed scientific journal (in print or on-line), in an on-line publicly accessible registry dedicated to the dissemination of clinical trial information such as ClinicalTrials.gov, or in journals willing to publish in abbreviated format (e.g., for studies with negative or incomplete results).
   12. The study protocol must explicitly discuss beneficiary subpopulations affected by the item or service under investigation, particularly traditionally underrepresented groups in clinical studies, how the inclusion and exclusion criteria effect enrollment of these populations, and a plan for the retention and reporting of said populations on the trial. If the inclusion and exclusion criteria are expected to have a negative effect on the recruitment or retention of underrepresented populations, the protocol must discuss why these criteria are necessary.
   13. The study protocol explicitly discusses how the results are or are not expected to be generalizable to affected beneficiary subpopulations. Separate discussions in the protocol may be necessary for populations eligible for Medicare due to age, disability or Medicaid eligibility.

Consistent with section 1142 of the Act, the Agency for Healthcare Research and Quality (AHRQ) supports clinical research studies that meet the above-listed standards and address the above-listed research questions.

The principal investigator must submit the complete study protocol, identify the relevant CMS research question(s) that will be addressed, and cite the location of the detailed analysis plan for those questions in the protocol, plus provide a statement addressing how the study satisfies each of the standards of scientific integrity (a. through m. listed above), as well as the investigator's contact information, to the address below. The information will be reviewed, and approved studies will be identified on the CMS Website.

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

Email address for protocol submissions: clinicalstudynotification@cms.hhs.gov
Email subject line: "CED [NCD topic (i.e. TAVR)] [name of sponsor/primary investigator]"

See Appendix B for the NCD manual language.

II. Background

Throughout this document we use numerous acronyms, some of which are not defined as they are presented in direct quotations. Please find below a list of these acronyms and corresponding full terminology:

AATS – American Association for Thoracic Surgery
ACC – American College of Cardiology
ACCF – American College of Cardiology Foundation
AF – Atrial Fibrillation
AHA – American Heart Association
AKI – Acute Kidney Injury
AS – Aortic Stenosis
AVR – Aortic Valve Replacement
CAD – Coronary Artery Disease
CED – Coverage with Evidence Development
CMS – Centers for Medicare & Medicaid Services
COPD - Chronic Obstructive Pulmonary Disease
CI – Confidence Interval
CT – Computerized Tomography
CV – Cardiovascular
CVA – Cerebrovascular Accident
ECG – Electrocardiogram
EQ-5D - the EuroQol – 5D
FDA – Food and Drug Administration
HF – Heart Failure
HR – Hazard Ratio
IQR – Interquartile Range
ITT – Intention to Treat
KCCQ – Kansas City Cardiomyopathy Questionnaire
LV – Left Ventricle / Left Ventricular
LVEF – Left Ventricular Ejection Fraction
MI – Myocardial Infarction
NCA – National Coverage Analysis
NCD – National Coverage Determination
NIS – National Inpatient Sample
O:E – Observed to Expected Mortality Ratio
OR – Odds Ratio
PARTNER - Placement of AoRTic TraNscathetER Valve trial
PCI – Percutaneous Coronary Intervention
PDM – Proposed Decision Memorandum
PPI – Permanent Pacemaker Implantation
PROM - Predicted Risk of Mortality
PVL – Paravalvular Leakage
PVR– Paravalvular Regurgitation
QoL– Quality of Life
RCT – Randomized Controlled Trial
RR – Risk Ratio
SAVR - Surgical Aortic Valve Replacement
SCAI – Society for Cardiovascular Angiography and Interventions
SDM - Shared Decision Making
SF-12 - Medical Outcomes Study Short-Form-12
STS – Society of Thoracic Surgeons
TA – Technology Assessment
TAVR – Transcatheter Aortic Valve Replacement
TAVI – Transcatheter Aortic Valve Implantation
TVT - Transcatheter Valve Therapies Registry
TIA – Transient Ischemic Attack
US – United States
VARC - Valve Academic Research Consortium
VHD – Valvular Heart Disease

Aortic Stenosis
Aortic stenosis is a potentially serious condition that affects heart function by partially obstructing the blood flow from the heart to the aorta. Normally, the aortic valve has three small flaps, or leaflets, that open to allow blood to flow out of the heart and then close to prevent blood from flowing backwards into the heart again (Rajamannan, 2011). Aortic stenosis occurs if the valve opening narrows and cannot open all the way, restricting blood flow out of the heart (Mrsic, 2018).  Aortic stenosis is usually caused by degenerative calcification (thickening of the valve trileaflets and deposits of calcium that form nodules) or less commonly, rheumatic fever, a valvular infection leading to rheumatic heart disease (Ray, 2010).  As the ultimate consequence of calcific aortic disease, aortic stenosis begins with aortic sclerosis (abnormal hardening), leading to progressive valve obstruction with an ongoing process of valve remodeling and calcification, and then a gradual reduction in the mobility of the cusps of the aortic valve (Rajamannan, 2011). The risk factors for the development of degenerative calcific aortic stenosis, which are similar to those for the development of vascular atherosclerosis, include male gender, diabetes mellitus, systemic hypertension, cigarette smoking, elevated levels of low-density lipoprotein cholesterol and lowered levels of high-density lipoprotein cholesterol (Aronow, 2001).

Aortic stenosis is the most common valvular heart disease (VHD) in the developed world (Carabello, 2009) and the most prevalent form of cardiovascular disease in the Western world after hypertension and coronary artery disease (Maganti, 2010). Aortic stenosis is progressive and if left untreated carries a poor prognosis and short average course after symptom onset (Ross, 1968). Symptoms related to left ventricular failure include marked dyspnea (shortness of breath), orthopnea (shortness of breath while lying flat), nocturnal dyspnea (episodes of shortness of breath that occur at night), and pulmonary edema (excess fluid in the lungs) (Ross, 1968). On average, survival is two to three years after symptoms develop, with a high risk of sudden death (Bonow, 2008). Five-year mortality has been reported at 60% after a first hospitalization with a diagnosis of AS (Lung, 2014).

The estimated prevalence of moderate to severe aortic stenosis in ≥75 year old patients is 2.8% in the United States (US) (Nkomo, 2006). The proportion of individuals ≥75 years old in the US is predicted to increase to 10.7% in 2025 and 16.6% in 2050 (United States Census Bureau 2011). Based on these estimates, there will be approximately 0.8 million and 1.4 million patients with symptomatic severe aortic stenosis in 2025 and 2050 in the US, respectively (Osnabrugge, 2013). The American Heart Association (AHA) and American College of Cardiology (ACC) consider surgical or transcatheter AVR in patients with severe, symptomatic, and calcific aortic stenosis as the only effective treatment resulting in improved survival rates, reduced symptoms, and improved exercise capacity (Nishimura, 2014). The risk of operation, patient frailty, and comorbid conditions are considered when decisions are made with regard to proceeding with surgical versus transcatheter aortic valve replacement (AVR) (Nishimura, 2014).

Aortic Valve Replacement
For decades, the only available treatment for aortic stenosis was surgical aortic valve replacement (SAVR)(Bonow, 2006). It is a major operation that requires opening the chest and using a heart-lung bypass machine, but the risks associated with SAVR are far less than those of leaving severe aortic valve stenosis untreated (Bakaeen, 2010).  In this open-heart operation, the damaged valve is removed and replaced with a new artificial valve.

TAVR
Over the last decade TAVR has emerged as an alternative to SAVR (Arnold, 2015). TAVR treats aortic stenosis by displacing and functionally replacing the aortic valve with a bioprosthetic valve delivered on a catheter. In most TAVR cases, the proceduralist (an interventional cardiologist or cardiothoracic surgeon trained in TAVR) make a small opening in an artery near the groin to insert a catheter, a long tube, to deliver and implant the new valve. This procedure does not require a heart-lung bypass machine to support blood circulation. It is most often performed using a transfemoral approach, inserting the delivery catheter through the femoral artery (Grover, 2017). If transfemoral TAVR is not feasible, other arteries may be used as entry sites (e.g., the subclavian artery, the common carotid artery, or direct to the aorta). A transapical approach can also be used, where TAVR is performed using an incision in the chest; the new valve is inserted through the heart’s left ventricle (Smith, 2011).

Surgical Risk
Risk-adjustment models have been used to predict hospital mortality after surgery and to classify patients in published studies. For example, the Society of Thoracic Surgeons (STS) predicted risk of mortality (PROM) score (also referred to the as STS risk score) predicts mortality during the first 30 days after cardiac surgery, based on baseline patient characteristics. An STS risk score has been used in determining patient inclusion for TAVR trials. Brennan et al. (2017) reported a method of categorization for low-risk cases (STS PROM score < 4%), intermediate-risk cases (4% to 8%), and high-risk cases (> 8%).

III. History of Medicare Coverage

CMS issued an NCD on May 1, 2012 establishing the first CMS coverage policy for TAVR under Coverage with Evidence Development (CED). For TAVR procedures used to treat symptomatic aortic valve stenosis when furnished according to Food and Drug Administration (FDA)-approved indications, the NCD contains requirements including volume requirements for heart teams and hospitals as well as mandatory participation in a prospective, national, audited registry.

The NCD requires TAVR procedures for uses that are not expressly listed as an FDA-approved indication to be performed in clinical studies that meet requirements set forth in the NCD and are approved by CMS.

Since there is an existing NCD for TAVR, this review is a reconsideration of the current policy. The current policy is codified in section 20.32 of the Medicare National Coverage Determination Manual (Pub. 100-03). Section 20.32 of the NCD Manual is included in Appendix C.

A. Current Request

CMS received a complete, formal request to reconsider the TAVR NCD from Drs. Peter Pelikan and John Robertson with Providence Saint John's Health Center and Dr. Richard Wright with the Pacific Heart Institute. The request letter is available at https://www.cms.gov/Medicare/Coverage/DeterminationProcess/downloads/id293.pdf.

B. Benefit Category

For an item or service to be covered by the Medicare program, it must fall within one of the statutorily defined benefit categories outlined in the Social Security Act [§1812 (Scope of Part A); §1832 (Scope of Part B); §1861(s) (Definition of Medical and Other Health Services)].

TAVR qualifies as:

• Inpatient hospital services.
• Physicians’ services.

Note: This may not be an exhaustive list of all applicable Medicare benefit categories for this item or service.

IV. Timeline of Recent Activities

V. Food and Drug Administration (FDA) Status

On November 2, 2011 the FDA approved the first TAVR device for marketing in the United States. The Edwards’ SAPIEN Transcatheter Heart Valve (THV) was approved "for transfemoral delivery in patients with severe symptomatic native aortic valve stenosis who have been determined by a cardiac surgeon to be inoperable for open AVR and in whom existing co- morbidities would not preclude the expected benefit from correction of the aortic stenosis" (https://www.accessdata.fda.gov/scripts/cdrh/cfdocs/cfpma/pma.cfm?id=P100041). Since this first approval, devices have been approved for:

• Lower surgical risk groups, including high and intermediate;
• Alternate access sites, such as transapical and transaortic; and
• Valve-in-valve use for failed surgical bioprosthetic valves.

Table 1 below provides a timeline of TAVR device approvals to date.

Table 1

VI. General Methodological Principles

When making NCDs, CMS generally evaluates relevant clinical evidence to determine whether or not the evidence is of sufficient quality to support a finding that an item or service falling within a benefit category is reasonable and necessary for the diagnosis or treatment of illness or injury or to improve the functioning of a malformed body member. The critical appraisal of the evidence enables us to determine to what degree we are confident that: 1) the specific assessment questions can be answered conclusively; and 2) the intervention will improve health outcomes for beneficiaries. An improved health outcome is one of several considerations in determining whether an item or service is reasonable and necessary.

A detailed account of the methodological principles of study design that CMS uses to assess the relevant literature on a therapeutic or diagnostic item or service for specific conditions can be found in Appendix A.

Public comments sometimes cite published clinical evidence and give CMS useful information. Public comments that give information on unpublished evidence such as the results of individual practitioners or patients are less rigorous and therefore less useful for making a coverage determination. Public comments that contain personal health information will be redacted or will not be made available on the CMS website. CMS responds in detail to the public comments on a proposed NCD when issuing the final NCD.

VII. Evidence

For this reconsideration, we reviewed the published medical literature from 2012 to 2019 to determine reasonable and necessary indications for TAVR and whether the registry data collection questions have been answered. Additionally, we reviewed the published literature on TAVR to determine whether the CED questions have been answered. During our review, newer TAVR devices and different patient populations have been included in published studies, consensus statements, and guidelines. These devices and patient populations have similar considerations and have been included in our review, analysis, and decision. This section provides a summary of the evidence we considered during our review. The evidence focuses on overarching TAVR population risk factors and endpoints. It excludes research reports that focus on patient subgroups such as those concerning a specific disease or risk factor (such as studies of patients with obesity, diabetes, or kidney disease) or research reports that focus on a subset not related to a disease (such as studies that focus on a single manufacturer, a single provider, or a single geographic region).

Our evidence review focused on whether to continue data collection and CED for TAVR devices.

1. Evidence Questions

Our review and analysis of the evidence on the clinical utility of TAVR for Medicare beneficiaries with cardiac symptoms and severe aortic stenosis is guided by the following questions:

• Is the evidence sufficient to conclude that TAVR improves health outcomes for Medicare beneficiaries with cardiac symptoms and severe aortic stenosis who are not candidates for SAVR?
• Is the evidence sufficient to conclude that TAVR improves health outcomes for Medicare beneficiaries with cardiac symptoms and severe aortic stenosis who are candidates for SAVR, and are at either high or intermediate surgical risk?

If the answer to either or both of the questions above is positive, is the available evidence adequate to identify the characteristics of the patient, practitioner or facility that predict which beneficiaries are more likely to experience overall benefit or harm from TAVR?

CMS did not request an external technology assessment (TA) on this issue. The 2016 Ontario Health Technology Assessment made the following key points in their technology assessment that compared TAVR and SAVR.

Health Quality Ontario. Transcatheter Aortic Valve Implantation for Treatment of Aortic Valve Stenosis: A Health Technology Assessment. Ont Health Technol Assess Ser. 2016 Nov 1;16(19):1-94. eCollection 2016.

Five randomized controlled trials (RCTs) that evaluated the effectiveness and safety of TAVR compared with SAVR or balloon aortic valvuloplasty were published before September 2015. The trials included patient populations at different levels of surgical risk with the mean STS score for the TAVR group ranging from 2.9% to 11.8%. The authors concluded that "TAVR and surgery had similar rates of death, and both improved patients’ quality of life in the first year. TAVR was associated with higher risk of stroke, major vascular complications, leakage of blood around the valve (aortic regurgitation), and the need for a pacemaker. Surgical aortic valve replacement was associated with a higher risk of bleeding."

Two Cochrane reviews on TAVR (Thyregod, 2015; Vilela, 2015) were withdrawn. The Evidence-based Practice Centers (EPC) Program of the Agency for Healthcare Research and Quality (AHRQ) technology assessment report (Coeytaux, 2010; Williams, 2010) on percutaneous heart valve replacement was published in 2010 and utilized older data. Three cost-analysis studies on TAVR (Kularatna, 2016; Neyt, 2012; Van Brabandt, 2012) were identified. The TAVR technology assessment by the California Technology Assessment Forum (Tice, 2014) analyzed older data from 1945 to January 2012.

Literature Search Methods
CMS searched PubMed (MEDLINE and OVID) from January 2012 to July 2018. Search terms included combinations of: transcatheter aortic valve replacement, transcatheter aortic valve implantation, TAVR, TAVI, postoperative complications, adverse effects, adverse events, mortality, death, fatality, stroke, transient ischemic attack, major vascular events, acute kidney injury, myocardial infarction, bleeding complications, aortic insufficiency, atrial fibrillation, pacemaker, repeat aortic valve replacement, and quality of life. The search was limited to English language articles of studies involving human subjects.

We then restricted the studies to RCTs and meta-analyses, with the meta-analyses selecting RCTs and observational studies. We further restricted these studies to those whose scope in the analysis among surgically inoperable, and high-, intermediate-, and low-risk study populations examined the effect of TAVR on adverse effects after the procedure, durability of the TAVR device, or quality of life after TAVR. In order to reference other trials in our analysis, we reviewed the reference lists of the meta-analyses and RCTs to select the TAVR pivotal trials that would help to answer the questions about TAVR and adverse effects, durability, and quality of life and to identify evidence gaps.

For TAVR volume – mortality outcome studies, CMS searched PubMed from January 1, 2012 to June 12, 2018. Search terms included combinations of: transcatheter aortic valve replacement, transcatheter aortic valve implantation, TAVR, TAVI, hospitals, centers, institutions, facilities, high volume programs, low volume programs, cardiology service in hospital, operating rooms, patient care team, program development, program evaluation, health impact assessment, utilization, standards, mortality, fatality, and death. The inclusion criteria limited the search to English language articles. The exclusion criteria excluded studies not involving human subjects. Letters, commentaries, and editorials were excluded. We then restricted the studies to those whose scope included volume in the analysis as a predictor or confounder and mortality as the outcome. This evidence review primarily focuses on observational studies that assess the association between TAVR case volume and mortality to assess the volume requirements in the 2012 TAVR NCD.

For the TVT registry publications, CMS searched PubMed from January 1, 2012 to November 2, 2018. Search terms included combinations of: transcatheter aortic valve replacement, transcatheter aortic valve implantation, TAVR, TAVI, registry, registries, and TVT registry. The inclusion criteria limited the search to English language articles. The exclusion criteria excluded studies not involving human subjects. Letters, commentaries, and editorials were excluded.  We then restricted the studies to those whose scope included TAVR in the analysis as a predictor or confounder and the outcomes as mortality, stroke, permanent pacemaker insertion, acute kidney injury, major vascular complication, atrial fibrillation, major bleeding, valve durability reflected in aortic regurgitation and aortic valve reintervention, and quality of life. This evidence review primarily focuses on TVT registry studies that assesses trends in outcomes such as mortality and stroke among patients having TAVR to assess the extent to which the published literature addressed the data collection questions as stated in our 2012 NCD.

In answering these questions, we focus primarily on the major clinical trials as the foundation for the evidence base for TAVR. We then consult secondary analyses on the trial data, follow-up studies to assess stability of trial outcomes (both benefits and harms), to include such things as quality of life, and device durability; and TVT registry studies that assess if the trial outcomes for specific populations are generalizable to similar, non-trial patients who undergo TAVR in broader community practice.

Consistent with requirements for CED in our 2012 NCD, all of the studies discussed below (including trials appearing in Table 2) have reported on, in addition to measures for quality of life pre- and post-TAVR, rates or incidence of: stroke, transient ischemic attacks, all-cause mortality (death from any cause), major vascular events, acute kidney injury, and repeat aortic valve procedures, among other outcomes.

Benefits and Harms, Durability, and Quality of Life

Adams DH, Popma JJ, Reardon MJ, et al. Transcatheter aortic-valve replacement with a self-expanding prosthesis. N Engl J Med. 2014 May 8;370(19):1790-8.

The aim of the U.S. CoreValve High Risk Study was to assess the safety and effectiveness of TAVR with a self-expanding prosthesis as compared with surgical valve replacement in patients with severe aortic stenosis who were at increased risk of death during surgery.  From 2011 to 2012, 795 patients with severe aortic stenosis who were at increased surgical risk underwent randomized assignment to TAVR with the self-expanding transcatheter valve (TAVR group) or to SAVR (surgical group). In the intention-to-treat TAVR group, the mean age was 83.2+7.1 years and 46.4% were women. Based on the STS PROM score, the average predicted mortality at 30 days was 7.4%.

For the results in the intention-to-treat analysis, the rate of death from any cause at 1 year was significantly lower in the TAVR group than in the surgical group (13.9% vs. 18.7%), with an absolute reduction in risk of 4.8 percentage points (upper boundary of the 95% confidence interval, −0.4; P<0.001 for noninferiority; P = 0.04 for superiority). The rates of any stroke were 4.9% in the TAVR group and 6.2% in the surgical group at 30 days (P = 0.46) and 8.8% and 12.6%, respectively, at 1 year (P = 0.10).

Major vascular complications at 30 days and 1 year and permanent pacemaker implantations at 30 days and at 1 year were significantly higher in the TAVR group than in the surgical group. Major bleeding at 30 days and 1 year, acute kidney injury at 30 days and 1 year, and new onset or worsening atrial fibrillation at 30 days and at 1 year were significantly more common in the surgical group than in the TAVR group.  The rates of paravalvular regurgitation were significantly higher in the TAVR group than in the surgical group at all time points after the procedure. The authors concluded that "in patients with severe aortic stenosis who are at increased surgical risk, TAVR with a self-expanding transcatheter aortic-valve bioprosthesis was associated with a significantly higher rate of survival at 1 year than surgical aortic-valve replacement."

Arora S, Misenheimer JA, Jones W, et al. Transcatheter versus surgical aortic valve replacement in intermediate risk patients: a meta-analysis. Cardiovascular Diagnosis and Therapy. 2016 Jun;6(3):241-9.

The aim of this study was to focus specifically on the population considered intermediate risk for valve replacement surgery. The Medline, EMBASE, Google Scholar, Web of Science and Cochrane databases were searched using standard methodology to search for clinical trials and observational studies including intermediate risk patients. One study was an RCT and five were observational studies consisting of one case control study and four cohort studies. Mean STS score ranged between 2.9% and 8% and the mean EuroSCORE ranged between 4.7% and 10.2%. The average age ranged from 78 to 81 years and the percent women ranged from 47% to 59%. For the TAVR group, the mean age was 80 years with a range from 78 to 81 years and the percent female was 54% with a range from 46% to 59%.

For the overall results, analysis of the TAVR and SAVR cohorts revealed no statistically significant differences in the outcomes of 30-day mortality (odds ratio [OR] 0.85, 95% confidence interval [CI]: 0.57-1.26, P = 0.41) or 1-year mortality (OR 0.96, 95% CI 0.75-1.23, P =0.74). No statistically significant difference was detected between TAVR versus SAVR at 30 days in regards to MI (OR 0.54, 95% CI 0.24-1.21, P = 0.14), 30-day stroke (OR 0.61, 95% CI 0.31-1.20, P = 0.15), or 30-day adverse neurological events (OR 0.63, 95% CI 0.35-1.14, P = 0.76). A statistically significant decrease in risk of post-procedural 30-day acute renal failure in the TAVR group (OR 0.51, 95% CI 0.27-0.99, P = 0.05) was observed, but so was a statistically significantly higher rate of pacemaker implantations for the TAVR group (OR 6.51, 95% CI 3.23 -13.12, P < 0.00001). The authors concluded "that in intermediate risk patients undergoing aortic valve replacement, the risk of mortality, neurological outcomes, and MI do not appear to be significantly different between TAVR and SAVR. However, there appears to be a significant reduction in risk of acute renal failure at the expense of an increased risk of requiring a permanent pacemaker in low and intermediate risk patients undergoing TAVR compared to SAVR."

Arora S, Strassle PD, Ramm CJ, et al. Transcatheter versus surgical aortic valve replacement in patients with lower surgical risk scores: A systematic review and meta-analysis of early outcomes.  Heart, Lung and Circulation. 2017 Aug;26(8):840-845.

The aim of this study was to examine the results on TAVR in lower risk surgical patients from outside of the United States.  The Medline, EMBASE, Google Scholar, Web of Science and Cochrane databases were searched using standard methodology through October, 2016 for studies reporting results comparing TAVR and SAVR. Four studies, including one randomized clinical trial and three propensity score-matched cohort studies met the inclusion criteria. The four studies were published between 2015 and 2016 utilizing data collected between 2008 and 2013. The STS Risk score was 3.0% and the EuroSCORE ranged between 6.3% and 9.9%. Mean age ranged between 78.3 years and 83.7 years, and percent male ranged between 39.0% and 58.5%.

For the overall results, compared to SAVR, TAVR had a non-statistically significant lower risk of 30-day mortality (risk ratio [RR] 0.67, 95% CI 0.41-1.10, P = 0.12) and 30-day stroke (RR 0.60, 95% CI 0.30-1.22, P =0.16). TAVR was associated with a statistically significantly lower risk of 30-day bleeding complications (RR 0.51, 95% CI 0.40-0.67) and a lower risk of 30-day acute kidney injury (RR 0.66, 95% CI 0.47-0.94). However, a statistically significantly higher risk of 30-day vascular complications (RR 11.72, 95% CI 3.75-36.64), 30-day moderate or severe paravalvular leak (RR 5.04, 95% CI 3.01-8.43), and 30-day permanent pacemaker implantations (RR 4.62, 95% CI 2.63-8.12) was noted for TAVR. The authors concluded that "among lower risk patients, TAVR and SAVR appear to be comparable in short term outcomes. Additional high quality studies among patients classified as low risk are needed to further explore the feasibility of TAVR in all surgical risk patients."

Arora S, Vaidya SR, Strassle PD, et al. Meta-analysis of transfemoral TAVR versus surgical aortic valve replacement. Catheterizations and Cardiovascular Interventions: Official Journal of the Society of Cardiac Angiography and Interventions. 2018 Mar 1;91(4):806-812.

The aim of this study was to compare the effect of transfemoral TAVR (TF-TAVR) on clinical outcomes, regardless of patient risk, when compared with SAVR to provide more information on the effect of the access route on patient complications.  The Medline (PubMed), EMBASE, Google Scholar, BIOSIS (Web of Science), and Cochrane Central Register of Controlled Trials (CENTRAL) databases were searched for all comparison studies between TAVR and SAVR and mortality as an outcome, irrespective of surgical risk, from database inception to April 15, 2017.

For the overall results, three studies were RCTs and four were observational cohort studies. Across the seven studies, the mean age ranged from 77.5 years to 84.1 years and the percent male ranged from 39.6% to 57.1%.  One study included low-risk patients, two with intermediate risk, one with intermediate / high risk, one with low / intermediate risk, one with high risk, and one study included all risk categories. Compared with SAVR, TF-TAVR had comparable 30-day mortality (RR 0.79, 95% CI 0.58-1.06; P = 0.12), 1-year mortality (RR 0.91, 95% CI 0.78-1.08, P = 0.28), 30-day stroke (RR 0.82, 95% CI 0.49-1.38, P = 0.46), 30-day transient ischemic attack (RR 1.94, 95% CI 0.46-8.22; P = 0.37), and 30-day risk of bleeding (RR 0.70, 95% CI 0.31-1.57, P = 0.39). But TF-TAVR was associated with higher incidences of 30-day vascular complications (RR 6.10, 95% CI 2.92-12.73, P < 0.00001) and 30-day pacemaker implantations (RR 3.29, 95% CI 1.41-7.65, P = 0.006). The authors concluded "TF-TAVR to be associated with comparable mortality, both at 30-day and 1-year as compared to SAVR. In concordance with previous studies, TF-TAVR was associated with statistically significantly lower 30-day risks of atrial fibrillation and renal failure, at a cost of a higher incidence of pacemaker implantations and vascular complications, when compared to SAVR. However, we noted TF-TAVR to have lower post-procedural risks of MI."

Baron SJ, Arnold SV, Reynolds MR, et al. Durability of quality of life benefits of transcatheter aortic valve replacement: Long-term results from the CoreValve US extreme risk trial. Am Heart J. 2017 Dec;194:39-48.

The aim of this study was to assess the durability of health status outcomes beyond one year of follow-up among patients with severe aortic stenosis at extreme surgical risk after TAVR. The CoreValve U.S. Extreme Risk Trial was a single-arm study that enrolled patients with severe symptomatic aortic stenosis, who were classified as being at extreme risk (i.e., 30-day mortality/morbidity was estimated at ≥ 50%) for traditional SAVR. In this trial, 639 patients with severe aortic stenosis at extreme surgical risk underwent TAVR between February 2011 and August 2012. In the iliofemoral extreme risk cohort, the mean age was 83.5 years and 47.8% were male.  The mean STS risk score was 10.4%.

For the overall results, after TAVR, there was substantial health status improvement in disease-specific and generic scales by 6–12 months. Overall, patients experienced significant health status improvement after TAVR. For both the KCCQ and SF-12, these differences generally peaked between 6 and 12 months after TAVR and were largely sustained through 3 years of follow-up for both the iliofemoral and non-iliofemoral cohorts. The authors concluded that "extreme risk patients with severe AS who were treated with TAVR using the self-expanding CoreValve experienced large improvements in both disease-specific and generic health status that were generally sustained at 24 and 36 months."

Carnero-Alcázar M, Maroto LC, Cobiella-Carnicer J, et al. Transcatheter versus surgical aortic valve replacement in moderate and high-risk patients: a meta-analysis. European Journal of Cardiothoracic Surgery. 2017 Apr 1;51(4):644-652.

The aim of this meta-analysis was to compare early and late outcomes of TAVR versus SAVR in patients with moderate or high risk for SAVR. The National Library of Medicine’s PubMed database, the Cochrane Central Register of clinical trials and the ISI Web of Science were searched to identify relevant clinical studies from January 2009 to June 2016. The meta-analysis included 5 clinical trials and 37 observational propensity score matching studies published between 2011 and 2016, enrolling 20,224 patients. The age and gender distributions were not reported.

For the overall results, the pooled analysis combining intermediate- and high-risk patients comparing TAVR to SAVR suggested no differences in early (30 days post-procedure or in-hospital) ( OR 1.11, 95% CI 0.89–1.39, P = 0.355) or late (follow-up > 12 months) mortality (RR 0.91, 95% CI 0.78–1.05, P = 0.194).  The sensitivity analysis by subgroup for intermediate-risk patients comparing TAVR to SAVR suggested no differences in early (30 days post-procedure or in-hospital) (OR 0.91, 95% CI 0.63–1.33, P = 0.637) or late (follow-up > 12 months) mortality (RR 0.82, 95% CI 0.65–1.03, P = 0.092). The analysis for intermediate-risk patients demonstrated no statistically significant difference in the risk of > 1 year stroke (RR 0.66, 95% CI 0.4–1.08, P = 0.1) among patients assigned to TAVR versus SAVR.  TAVR compared with SAVR had an increase in the incidence of pacemaker implantation for intermediate-risk patients, (OR 3.08, 95% CI 1.94–4.89, P < 0.001), as well as for high-risk patients (OR 1.86, 95% CI 1.29–2.68, P < 0.001). The authors concluded that "TAVR and SAVR have similar short and long-term all-cause mortality and risk of stroke among patients of moderate or high surgical risk. TAVR decreases the risk of major bleeding, acute kidney injury and improves hemodynamic performance compared with SAVR but increases the risk of vascular complications, the need for a pacemaker and residual aortic regurgitation."

Elmaraezy A, Ismail A, Abushouk AI, et al. Efficacy and safety of transcatheter aortic valve replacement in aortic stenosis patients at low to moderate surgical risk: a comprehensive meta-analysis. BMC Cardiovascular Disorders. 2017 Aug 24;17(1):234.

The aim of this study was to compare the safety and efficacy of TAVR to SAVR in low-to-moderate surgical risk patients with aortic stenosis. Five databases, PubMed, Scopus, Web of Science, Embase, and Cochrane Central Register of Controlled Trials, were searched. Eleven articles were included, of which four eligible studies were RCTs, while the remaining seven studies included five prospective cohort and two retrospective studies. The mean age ranged from 68.1 years to 83.3 years and the percent male ranged from 26.5% to 60.4%. Mean STS score ranged from 2.9% to 5.8% and mean EURO score ranged from 6.1 to 24.4.

For the overall results, at one-year of follow-up, the pooled-effect estimates showed no statistically significant difference between TAVR and SAVR groups in terms of all-cause mortality ( 1.02, 95% CI [0.83-1.26]), stroke (RR 0.83, 95% CI 0.56-1.21), and myocardial infarction (RR 0.82, 95% CI 0.57-1.19). The overall risk ratio did not favor either of the TAVR or SAVR groups in terms of in-hospital all-cause mortality (RR 1.11, 95% CI 0.63- 1.95, P = 0.72), 30-day all-cause mortality (RR 0.95, 95% CI 0.74-1.21, P = 0.66), 1-year all-cause mortality (RR 1.02, 95% CI 0.83-1.26, P = 0.86), or 2-year mortality (RR 0.91, 95% CI 0.76-1.08, P = 0.27). The risk ratio of 3-year mortality was reported only by the OBSERVENT study, which showed a statistically significantly higher risk of mortality in the TAVR group than the SAVR group (RR 1.63, 95% CI 1.21-2.19, P = 0.001).  The overall risk ratio did not favor either of the two groups in terms of stroke incidence within 30 days (RR 0.99, 95% CI 0.73-1.35, P = 0.94), 1 year (RR 0.83, 95% CI 0.561.21, P = 0.33), or 2 years (RR 0.88, 95% CI 0.63-1.23, P = 0.45) after the procedure. The OBSERVENT study reported a higher 3-year risk of stroke in the TAVR group (RR 2.54, 95% CI 1.36-4.74, P = 0.003), compared to the SAVR group. However, compared to SAVR, the risk of permanent pacemaker implantation was higher in the TAVR group at 30 days (RR 3.31, 95% CI 2.05-5.35), 1 year (RR 2.57, 95% CI 1.36-4.86), but not after 2 years (RR 1.57, 95% CI 0.91-2.70), probably due to the small number of included studies at the 2-year endpoint. The authors concluded that "due to the comparable mortality rates in SAVR and TAVR groups and the lower risk of life-threatening complications in the TAVR group, TAVR can be an acceptable alternative to SAVR in low-to-moderate risk patients with aortic stenosis."

Kapadia SR, Leon MB, Makkar RR, et al. 5-year outcomes of transcatheter aortic valve replacement compared with standard treatment for patients with inoperable aortic stenosis (PARTNER 1): a randomized controlled trial. Lancet. 2015 Jun 20;385(9986):2485-91.

The aim of this study was to present the prespecified final 5-year follow-up of patients deemed inoperable in the Placement of Aortic Transcatheter Valve (PARTNER-1) trial. Patients with severe symptomatic inoperable aortic stenosis were randomly assigned (1:1) to transfemoral TAVR or to standard treatment of medical management without AVR. For the 358 patients who were enrolled from May, 2007 through March, 2009, the mean age was 83 years, and 54% were female. The mean STS PROM score was 11.7%.

For the overall results, the risk of all-cause mortality at 5 years was 71.8% in the TAVR group versus 93.6% in the standard treatment group (HR 0.50, 95% CI 0.39–0.65, P < 0.0001). Risk of stroke at 5 years was 16.0% in the TAVR group versus 18.2% in the standard treatment group (HR 1.39, 95% CI 0.62–3.11,P = 0.555). The authors concluded that "TAVR is more beneficial than standard treatment for treatment of inoperable aortic stenosis."

Kapadia SR, Tuzcu EM, Makkar RR, et al. Long-term outcomes of inoperable patients with aortic stenosis randomly assigned to transcatheter aortic valve replacement or standard therapy. Circulation. 2014 Oct 21;130(17):1483-92.

The aim of this study was to report the 3-year or longer clinical and echocardiographic outcomes of inoperable patients randomly assigned to TAVR or standard therapy in the PARTNER-1) trial. In the PARTNER-1 cohort B study, 358 surgically inoperable patients with severe aortic stenosis were randomly assigned to TAVR or standard therapy between May, 2007 and March, 2009. The STS PROM score was high in both groups (mean [SD] STS score in TAVR and standard therapy groups: 11.2% [5.8] and 12.1% [6.1], respectively). For the TAVR group, the mean age was 83 years and 47.7% were male.

For the overall results, the 3-year mortality rate in the TAVR and standard therapy groups was 54.1% and 80.9%, respectively (HR 0.53; 95% CI 0.41–0.68, P < 0.001). Landmark analyses demonstrated that the differences in survival remained statistically significant after the first year of follow-up, and after the second year as well.  The incidence rate of stroke in the TAVR arm of 15.7% was significantly higher than the cumulative incidence rate of 5.5% observed at 3-year follow up in the standard therapy arm (HR 2.81, 95% CI 1.26–6.26, P = 0.012). The risk of new pacemaker implantation at 3-years follow up was similar between TAVR and standard therapy (P = 0.75).  The authors concluded that "TAVR in comparison with standard therapy results in better survival and functional status for patients with severe aortic stenosis who were inoperable, and the survival benefits increased during continued follow-up through 3 years."

Khan SU, Lone AN, Saleem MA, et al. Transcatheter vs surgical aortic-valve replacement in low- to intermediate-surgical-risk candidates: A meta-analysis and systematic review.  Clinical Cardiology. 2017 Nov;40(11):974-981.

The aim of this study was to discover whether TAVR can be as effective as SAVR in low- to intermediate-surgical-risk candidates. Four RCTs and eight prospective matched studies were selected using PubMed/MEDLINE, Embase, and Cochrane Central Register of Controlled Trials (inception: March 2017). Mean age of the total population ranged from 78 to 83 years, and mean logistic European System for Cardiac Operative Risk Evaluation (LES) was 13.3%.

For the overall results, among 9851 patients, analyses of RCTs showed that all-cause mortality was comparable with no statistically significant difference between TAVR and SAVR (short term ≤ 30 days, OR 1.19, 95% CI  0.86-1.64, P = 0.30; mid-term 1 year, OR 0.97, 95% CI 0.75-1.26, P = 0.84; and long term >1 year, OR 0.97, 95% CI 0.81-1.16, P = 0.76).  There was no difference in outcomes between both the TAVR and SAVR arms with regard to MI (≤ 30 days MI: RCTs, OR 0.66, 95% CI: 0.39-1.12, P = 0.13; matched studies, OR 0.51, 95% CI 0.22-1.20, P = 0.12; 1 year MI: RCTs, OR 0.91, 95% CI 0.60-1.36, P = 0.64; matched studies, OR 0.26, 95% CI 0.04-1.61, P = 0.15; > 1 year MI: OR 1.15, 95% CI 0.82-1.61, P = 0.42).  At short term ≤30 days, TAVR was associated with increased risk of ≤ 30 days vascular access complications (RCTs, OR 3.12, 95% CI 1.17-8.34, P = 0.02; matched studies, OR 9.49, 95% CI 1.62-55.62, P = 0.01) and ≤ 30 days permanent pacemaker implantation (RCTs, OR 4.86, 95% CI 1.37-17.23, P = 0.01; matched studies, OR 2.74, 95% CI 1.20-6.22, P = 0.02).  There was no difference in outcome in terms of ≤ 30 days major bleeding (RCTs, OR 0.47, 95% CI 0.10-2.27, P = 0.34; matched studies, OR 0.25, 95% CI 0.04-1.48, P = 0.13). The authors concluded that "in patients with symptomatic severe AS who carry low to intermediate surgical risk, SAVR and TAVR can provide similar mortality outcomes. Both interventions are associated with their own array of adverse events."

Lazkani M, Singh N, Howe C, et al. An updated meta-analysis of TAVR in patients at intermediate risk for SAVR. Cardiovascular Revascularization Medicine:  including molecular interventions. 2018 Apr 20. pii: S1553-8389(18)30129-5.

The aim of this study was to assess the safety and efficacy of TAVR compared to SAVR in intermediate-risk patients. The study used articles that were searched in PubMed, EMBASE, Web of science, and the Cochrane Central Register of Controlled Trials databases that compared TAVR versus SAVR in patients at intermediate surgical risk, with a mean STS risk score of 3%–8% or a mean LES risk score of 10%–20%. Study designs included four RCTs and seven observational studies.

For the overall results, there were no statistically significant differences in all-cause and cardiac mortality at 30 days, 1- year and > 2-years of follow up. The study demographics showed the mean age in the TAVR and SAVR groups were 80.2 and 80.3 years, respectively. Study results indicated that the Forest plots showed no statistically significant differences in all-cause mortality including short-term mortality at 30-days (3.9% vs. 3.5%; Mantel Haenszel risk ratio (MH-RR) 1.05, 95%, CI 0.79–1.39, P = 0.74), medium term mortality at 1-year (11.1% vs. 10.6%; MH-RR 1.00, 95% CI 0.86–1.17, P = 0.97) and long term mortality at ≥ 2 year follow up (15% vs 15.4%; MH-RR 0.93, 95% CI 0.76–1.13, P = 0.45) between the TAVR and SAVR groups. No statistically significant difference was found in stroke between TAVR compared with SAVR at 30 days (MH-RR 0.81, 95% CI 0.62–1.05, P = 0.11), 1-year (MH-RR 0.90, 95% CI 0.72–1.13, P = 0.36) and ≥ 2 years follow up (MH-RR 1.02, 95% CI 0.83–1.27, P = 0.84). Vascular access complications (MH-RR 4.43, 95% CI 1.61–12.14, P = 0.004), and permanent pacemaker placement (MH-RR 2.81, 95% CI 1.43–5.52, P = 0.003) occurred at higher rates in the TAVR group compared to the SAVR group. At 30-days TAVR had statistically significantly higher rate of PVL irrespective of severity (MH-RR 5.05, 95% CI 3.06–8.31, P < 0.001). The authors concluded that "a meta-analysis such as this, provides confidence that in spite of criticisms of the individual studies, there is no statistical difference in all-cause mortality; cardiac mortality; stroke; myocardial infarction and major bleeding between SAVR and TAVR in the intermediate risk patient with severe aortic stenosis (AS)."

Leon MB, Smith CR, Mack M, et al.  Transcatheter aortic-valve implantation for aortic stenosis in patients who cannot undergo surgery. N Engl J Med 2010;363:1597-607.

The aim of the study was to report the outcomes with TAVR as compared with standard therapy among the patients in the PARTNER-1 trial who were not suitable candidates for surgery. From 2007 to 2009, 358 patients with severe aortic stenosis, whom surgeons considered not to be suitable candidates for surgery, were randomly assigned to standard therapy (including mainly balloon aortic valvuloplasty and a few SAVR and some medical therapy) or transfemoral transcatheter implantation of a balloon-expandable bovine pericardial valve. The overall patient population was at high risk, with a STS risk score of 11.6%. The mean age of the TAVR group was 83.1years and 45.8% were men.

For the results at 1 year, the rate of death from any cause was 30.7% with TAVR, as compared with 50.7% with standard therapy (HR with TAVR, 0.55; 95% CI 0.40-0.74, P < 0.001). At 30 days after randomization, the rate of death from any cause was 5.0% in the TAVR group as compared with 2.8% in the standard-therapy group (P = 0.41). Major strokes were observed to be not statistically non-significantly more frequently in the TAVR group compared to that in the standard therapy group at 30 days (5.0% vs. 1.1%, P = 0.06) and at 1 year (7.8% vs. 3.9%, P = 0.18). Major vascular complications were significantly higher in the TAVR group compared to the standard therapy group at 30 days (16.2% vs. 1.1%, P<0.001) and at 1 year (16.8% vs. 2.2%, P < 0.001). Major bleeding was significantly higher in the TAVR group than in the standard therapy group at 30 days (16.8% vs. 3.9%, P < 0.001) and at 1 year (22.3% vs. 11.2%, P = 0.007). The authors concluded that "in patients with severe aortic stenosis who were not suitable candidates for surgery, TAV(R), as compared with standard therapy, significantly reduced the rates of death from any cause, despite the higher incidence of major strokes and major vascular events."

Leon MB, Smith CR, Mack MJ, et al. Transcatheter or surgical aortic-valve replacement in intermediate-risk patients. N Engl J Med. 2016 Apr 28;374(17):1609-20.

The aim of this study was to evaluate TAVR and SAVR in the PARTNER-2  cohort A randomized trial, in which TAVR with a second-generation valve system was compared with conventional surgery in patients with severe aortic stenosis and intermediate-risk clinical profiles. The risk score guideline was an STS risk score of at least 4.0%; the upper limit applied by the case review committee was 8.0%. In the TAVR group, the mean age was 81.5 years and 54.2% were male.  The mean STS score was 5.8% in the TAVR and SAVR groups.

For the overall results, at 2 years, the rate of death from any cause was 16.7% after TAVR and 18.0% after surgery (P = 0.45), and the rate of disabling stroke was 6.2% after TAVR and 6.4% after surgery (P = 0.83).  Comparing TAVR to surgery, the risk of transient ischemic attack was similar at 30 days (P = 0.17), 1-year (P = 0.38), and 2-years (P = 0.09) of follow-up after the procedure. The need for new permanent pacemakers within 30 days after the procedure was similar in the TAVR group and the surgery group (8.5% and 6.9%, respectively, P = 0.17), as well as at 1-year (P = 0.43) and 2-years (P = 0.29) of follow-up. Repeat aortic-valve interventions was uncommon and similar in both the TAVR group and the surgery group (2-years rate of reintervention, 1.4% and 0.6%, respectively, P = 0.09; 1-year, P = 0.10; and 30 days, P = 0.05 of borderline insignificance, of follow up). The authors concluded that "in intermediate risk patients with severe symptomatic aortic stenosis, surgical and transcatheter valve replacement were similar with respect to the primary end point of death or disabling stroke for up to 2 years and resulted in a similar degree of lessening of cardiac symptoms."

Liu Z, Kidney E, Bem D, et al. Transcatheter aortic valve implantation for aortic stenosis in high surgical risk patients: A systematic review and meta-analysis.  PLoS One. 2018 May 10;13(5):e0196877.

The aim of this study was to assess the clinical effectiveness and safety defined as mortality and other important clinical outcomes up to 5 years post treatment of TAVR for patients with severe aortic stenosis for whom SAVR was not an option or presented a high risk of mortality. Electronic databases including the Cochrane Library (CDSR, DARE, HTA and CENTRAL), Centre for Reviews and Dissemination Databases (DARE, NHS EED and HTA), MEDLINE, MEDLINE in Process, EMBASE, ZETOC and PubMed were searched from January 2002 to August 2016. The mean age of patients enrolled in the three RCTs included in the analysis ranged from 83.1 to 84.5 years, and percent female ranged from 42.2% to 54.2%. In the TAVR group, the mean STS risk score ranged from 7.3% to 11.8%.

For the overall results, in surgically inoperable patients, there was no statistically significant difference in 30-day mortality between the TAVR and medical therapy (TAVR versus medical therapy: 2.6% versus 5.9%, P = .09). TAVI was superior to medical therapy for all-cause mortality at 1 year (HR 0.58, 95% CI 0.36−0.92, P = 0.02), 2 years (HR 0.50, 95% CI 0.39−0.65, P < 0.001), 3 years (HR 0.53, 95% CI 0.41-0.68, P < 0.001) and 5 years (HR 0.50, 95% CI 0.39−0.65, P < 0.001).  TAVR was superior to medical therapy in quality of life (QoL) at least for 1 year (the Kansas City Cardiomyopathy Questionnaire (KCCQ) summary score, the 12-Item Short Form Health Survey (SF-12) physical score and SF-12 mental health score). Including high-risk but surgically operable patients, TAVR showed no statistically significant differences from SAVR in all-cause mortality at two years (HR 1.03, 95% CI 0.82−1.29) and up to 5 years (HR 0.97, 95% CI 0.83−1.12, P = 0.63). The authors concluded "that all-cause mortality up to 5 years of follow-up did not differ significantly between TAVI and SAVR in patients surgically operable at a high risk, but favored TAVI over medical therapy in patients surgically inoperable. TAVI is a viable life-extending treatment option in these surgical high risk groups."

Mack MJ, Leon MB, Smith CR, et al. 5-year outcomes of transcatheter aortic valve replacement or surgical aortic valve replacement for high surgical risk patients with aortic stenosis (PARTNER 1): a randomised controlled trial. Lancet. 2015 Jun 20;385(9986):2477-84.

The aim of this study was to report on 5-year clinical and valve performance outcomes for high-risk patients in the PARTNER-1 trial comparing TAVR to SAVR. The trial randomly assigned high-risk patients with severe aortic stenosis to either SAVR or TAVR with a balloon-expandable bovine pericardial tissue valve. Overall, 699 patients were enrolled. Overall mean STS risk score was 11.7%.  Mean age was 84.1 years.

For the overall results, the study showed that death and stroke are much the same for each treatment at 5 years. At 5 years, risk of death from any cause was 67.8% in the TAVR group compared with 62.4% in the SAVR group (HR 1.04, 95% CI 0.86–1.24, P = 0.76).  The 5-year rate of stroke alone (P = 0.61) and the 5-year rate of transient ischemic attack alone (P = 0.30) was the same comparing SAVR to TAVR.  At 5 years of follow-up, the incidence of 5-year myocardial infarction (P = 0.15), endocarditis (P = 0.65), 5-year renal failure (P = 0.69), or need for 5-year new pacemaker (P = 0.64) were similar in the SAVR and TAVR groups; however, the incidence of 5-year vascular complications (P = 0.0002) was higher in patients in the TAVR group than those in the SAVR group, and the incidence of 5-year major bleeding complications (P = 0.003) was lower in the TAVR group than in the SAVR group. The authors concluded that "the final 5-year follow-up of high risk surgical patients shows equivalent outcomes after TAVR and SAVR. We detected no significant differences in all-cause mortality, cardiovascular mortality, stroke, or need for repeat hospital admission."

Mack M, Leon M, Thourani V, et al. Transcatheter aortic-valve replacement with a balloon-expandable valve in low-risk patients. N Engl J Med. 2019 March 16; DOI: 10.1056 /NEJMoa1814052.

The aim of this "PARTNER 3" study was to evaluate TAVR with transfemoral placement of a third-generation balloon-expandable valve versus SAVR in symptomatic patients with severe aortic stenosis and low surgical risk for death.

From 2016 to 2017, 1,000 low-risk patients from 71 sites were randomized 1:1, with the assigned procedure performed in 950 (496 in the TAVR group and 454 in the SAVR group). The primary outcome was a composite of death from any cause, stroke or rehospitalization in the as-treated population at 1 year. However, the protocol required clinical and echocardiographic follow-up for at least 10 years.  The investigators reported that "patients enrolled in this trial were younger (mean age, 73 years), included more men (69.3%), and had lower STS predicted risk of mortality scores (mean score, 1.9%) and fewer coexisting conditions than patients enrolled in previous randomized trials of TAVR."

In the as-treated analysis, the rate of the primary composite end point at 1 year was significantly lower in the TAVR group than in the SAVR group (8.5% vs. 15.1%; absolute difference, −6.6 percentage points; 95% CI −10.8 to −2.5, P < 0.001 for noninferiority; HR 0.54, 95% CI 0.37-0.79, P = 0.001 for superiority). The event rate for death from any cause was low in both groups and there was no significant difference between TAVR and SAVR in death from any cause at 1 year (HR 0.41, 95% CI 0.14-1.17). At 30 days, TAVR resulted in a lower rate of stroke than surgery (P = 0.02) and in lower rates of death or stroke (P = 0.01) and new-onset atrial fibrillation (P < 0.001). TAVR resulted in a shorter index hospitalization than surgery (P < 0.001) and in a lower risk of a poor treatment outcome (death or a low KCCQ score) at 30 days (P < 0.001). Life-threatening or major bleeding occurred less frequently with TAVR than with SAVR. There were no significant differences between TAVR and SAVR in major vascular complications, new permanent pacemaker insertions, or moderate or severe paravalvular regurgitation.

The investigators concluded that in patients with severe aortic stenosis and low surgical risk, the rate of the composite of death, stroke, or rehospitalization at 1 year was significantly lower with TAVR than with SAVR. They also concluded that "the most important limitation of this trial is that our current results reflect only 1-year outcomes and do not address the problem of long-term structural valve deterioration."

McNeely C, Zajarias A, Fohtung R, et al.  Racial Comparisons of the Outcomes of Transcatheter and Surgical Aortic Valve Implantation Using the Medicare Database. Am J Cardiol. 2018 Aug 1;122(3):440-445. doi: 10.1016/j.amjcard.2018.04.019.

The aim of this study was to assess the racial disparities in whites, blacks and Hispanics undergoing TAVR in comparison to SAVR.

The study used CMS claims data for patients who underwent TAVR or SAVR between November 2011 and December 2013.

No primary outcome was specified.  The study assessed numerous patient outcomes, but with a focus on risk-adjusted 30-day and 1-year mortality, 30-day readmissions, and discharge destination.

The analysis method used Kaplan–Meier survival curves to assess unadjusted survival rates, and log-rank tests to assess differences in the survival curves.  Multivariate logistic regression produced adjusted odds ratios for 30-day mortality and readmission by race/ethnicity; covariates included patient demographics, comorbidities, and valve type (for SAVR) or procedure approach (for TAVR); hospital was treated as a random effect to account for clustering.  Adjusted odds ratios for blacks and Hispanics were in comparison to whites. 

The study included a total of 113,051 (17,973 TAVR, 95,078 SAVR) Medicare fee-for-service beneficiaries ≥ 65 years of age, after exclusions for patients with < 1 year of Medicare Part A coverage, patients with multiple race codes, and races other than blacks and Hispanics (e.g., Asian, Native American; due to the low representation of other races). 

The investigators found, in terms of baseline patient demographics and characteristics, that the TAVR cohort was 3.9% black and 1.0% Hispanic; and the SAVR cohort, 4.8% black and 1.3% Hispanic (compared to 11% black and 8% Hispanic among all Medicare beneficiaries).  Thus, "minorities were underrepresented in both SAVR and TAVR relative to what would be predicted by population prevalence."  In both TAVR and SAVR cohorts, black patients were younger and more likely to be female than whites and Hispanics.  Blacks had the highest proportion of patients with a history of stroke, heart failure, and renal failure in both cohorts.

With respect to outcomes, after TAVR there were no significant racial differences in both unadjusted and risk-adjusted outcomes for: 30-day and 1-year mortality; 30-day and 6-month hospital readmissions; discharge destination including to home or nursing facility. 

After SAVR there were racial differences in unadjusted 30-day and 1-year mortality which disappeared after risk adjustment.  However, black patients had higher 30-day readmission rates compared to whites after SAVR (20.1% vs 25.2% vs 21.7% for whites, blacks, and Hispanics, respectively, P = 0.0001), which persisted after risk adjustment.

The investigators concluded that blacks had worse outcomes after SAVR compared with whites or Hispanics, but race did not impact mortality, readmission, or discharge to home after TAVR.  The authors opined that "A better understanding of the racial differences observed, particularly the factors that seem to mitigate the racial disparity in outcomes of TAVR may be important for targeted quality improvement."

Nielsen HH, Klaaborg KE, Nissen H, et al. A prospective, randomised trial of transapical transcatheter aortic valve implantation vs. surgical aortic valve replacement in operable elderly patients with aortic stenosis: the STACCATO trial. EuroIntervention. 2012 Jul 20;8(3):383-9.

The aim of this prospective RCT was to compare transapical TAVR) with a SAPIEN balloon-expandable valve to SAVR in operable elderly patients. The study was planned as an academic prospective multicenter clinical trial in the Nordic region with a 1:1 randomization of a total of 200 patients to transapical TAVR versus SAVR. Operable patients with isolated aortic valve stenosis and age ≥75 years were included. The primary endpoint was the composite of all-cause mortality, cerebral stroke, or renal failure requiring hemodialysis at 30 days.

After inclusion of 11 patients, there were three potentially severe adverse events in the transapical TAVR group (one case of left main occlusion, one case of aortic rupture and one case of up-stream valve embolization). The study was put on hold, and the Data Safety Monitoring Board (DSMB) contacted. On advice from the DSMB, the study was prematurely terminated after the inclusion of 70 patients because of too many adverse events and procedure-related complications in the transapical TAVR group. The last patient was included May 2011.

For the overall results, a total of 72 patients were randomized. Two patients were excluded after randomization: one patient declined participation, and the other unexpectedly met the exclusion criteria of impaired pulmonary function. In the transapical TAVR group the mean age was 80 years and 26.5% were men. The mean STS risk score was 3.1% in the transapical TAVR group and 3.4% in the SAVR group. The primary endpoint was met in five (14.7%) TAVR patients (two deaths, two strokes, and one case of renal failure requiring dialysis) versus one (2.8%) stroke in the SAVR group (P = 0.07). During the 3-month follow-up period in the TAVR group, there were two more deaths with another death occurring at day 38. Three patients received a permanent cardiac pacemaker. Nielsen and colleagues (2012) concluded that "the STACCATO (Surgical Aortic Valve Replacement [AVR] in Operable Elderly Patients With Aortic Stenosis) trial was prematurely terminated because of an overall excess of adverse events in transcatheter treated patients in comparison with patients receiving surgical aortic valve replacement."

Popma J, Deeb G, Yakubov S, et al. Transcatheter aortic-valve replacement with a self-expanding valve in low-risk patients. N Engl J Med. 2019 March 16; DOI: 10.1056/NEJMoa1816885.

This was a randomized non-inferiority trial evaluating TAVR with a self-expanding supra-annular valve (CoreValve, Evolut R, or Evolut PRO) versus SAVR in symptomatic patients with severe aortic stenosis and low surgical risk for death. The primary outcome was a composite of death from any cause or disabling stroke in the as-treated population (patients who underwent an attempted procedure) at 24 months.

From 2016 to 2018, 1,468 low-risk patients were randomized 1:1, with the assigned procedure performed in 1,403 (725 in the TAVR group and 678 in the SAVR group). All patients had a low STS risk score (mean of 1.9%); mean age was 74 years, and 34.9% were women.

In the as-treated analysis, 24-month follow-up was available for 72 patients in the TAVR group and 65 patients in the SAVR group; outcomes for patients who did not complete 24 months of follow-up were imputed based on the patient’s last known clinical status. The as-treated analysis demonstrated no significant difference between the TAVR and SAVR groups in the 24-month estimated incidence of the primary composite endpoint (5.3% vs. 6.7%; absolute difference, −1.4 percentage points; 95% Bayesian credible interval [CI] for difference, posterior probability of noninferiority > 0.999).  At 30 days, patients who underwent TAVR, as compared with SAVR, had a lower incidence of disabling stroke (0.5% vs. 1.7%), bleeding complications (2.4% vs. 7.5%), acute kidney injury (0.9% vs. 2.8%), and atrial fibrillation (7.7% vs. 35.4%) and a higher incidence of moderate or severe aortic regurgitation (3.5% vs. 0.5%) and pacemaker implantation (17.4% vs.6.1%).

The investigators concluded that in patients with severe aortic stenosis and low surgical risk, TAVR with a self-expanding supraannular valve was noninferior to SAVR for the composite outcome of death from any cause or disabling stroke at 24 months. They also concluded that "The most important limitation is that this prespecified interim analysis occurred when 850 patients had reached 12 months of follow-up, and complete 24-month follow-up of the entire cohort has not been reached. Definitive conclusions regarding the advantages and disadvantages of TAVR as compared with surgery await long-term clinical and echocardiographic follow-up, which is planned to continue through 10 years for all patients."

Reardon MJ, Van Mieghem NM, Popma JJ, et al. Surgical or transcatheter aortic-valve replacement in intermediate-risk patients. N Engl J Med. 2017 Apr 6;376(14):1321-1331.

The aim of the Surgical Replacement and Transcatheter Aortic Valve Implantation (SURTAVI) trial was to compare the safety and efficacy of TAVR with a self-expanding bioprosthesis to SAVR in patients at intermediate surgical risk. A total of 1,746 patients underwent randomization at 87 centers between 2012 and 2016. Mean age was 79.8 years, and all were at intermediate risk for surgery with a mean STS risk score of 4.5%. The percent male was 58% in the TAVR group.

For the overall results at 24 months, the rate of death from any cause was 11.4% in the TAVR group and 11.6% in the surgery group (95% credible interval [CI]for difference, −3.8 to 3.3%). The rate for aortic valve reintervention was similar for the two groups at 30 days (95% CI for difference, -0.1 to 1.4), but aortic valve reintervention occurred more frequently in the TAVR group at 1 year (95% CI for difference, 0.4 to 2.7) and at 2 years (95% CI for difference, 0.7 to 3.5). For quality of life, as measured by the KCCQ summary score, the TAVR group had a statistically significantly higher proportion of patients with improvement at 1 month than did the surgery group (95% CI difference, 10.0 to 15.1) but there was no difference at 12 months in the KCCQ (95% CI for difference, -2.2 to 2.9). The authors concluded that "in a comparison between TAVR and surgical replacement in patients with symptomatic, severe aortic stenosis at intermediate risk for surgery, TAVR was a statistically noninferior alternative to surgery with respect to death from any cause or disabling stroke at 24 months. However, each procedure had a different pattern of adverse events."

Reynolds MR, Magnuson EA, Wang K, et al. Health-related quality of life after transcatheter or surgical aortic valve replacement in high-risk patients with severe aortic stenosis: results from the PARTNER (Placement of AoRTic TraNscathetER Valve) Trial (Cohort A). J Am Coll Cardiol. 2012 Aug 7;60(6):548-58.

The aim of this study was to compare health status and patient-reported quality-of-life outcomes for patients with severe aortic stenosis and high surgical risk treated with either TAVR or SAVR) as part of the PARTNER-1 cohort A trial. The study evaluated the health status of 628 patients with severe, symptomatic aortic stenosis at high surgical risk who were randomized to either TAVR or SAVR in the PARTNER-1 trial. The overall mean age was 83 years and the percent male in the TAVR group ranged from 51% to 60.4%.  The mean STS score in the TAVR group was 11.8%.

For the overall results, the primary outcome, the KCCQ summary score, improved more rapidly with TAVR compared to SAVR, but was similar for the two groups at 6 and 12 months. For the overall population, TAVR resulted in more rapid improvement in the KCCQ than SAVR, with a statistically significant benefit at 1 month (mean adjusted difference, 5.5; 95% CI 1.2 to 9.8, P = 0.01) but no statistically significant difference at either 6 months (mean adjusted difference, - 2.6; 95% CI -6.7 to 1.6, P = 0.22) or 12 months (mean adjusted difference, -0.5; 95% CI -4.8 to 3.8, P = 0.82). The authors concluded that in the PARTNER-1 trial "in high-risk patients with severe aortic stenosis, health status improved substantially between baseline and 1 year after either TAVR or surgical AVR. TAVR via the transfemoral, but not the transapical route, was associated with a short-term advantage compared with surgery."

Siemieniuk RA, Agoritsas T, Manja V, et al. Transcatheter versus surgical aortic valve replacement in patients with severe aortic stenosis at low and intermediate risk: systematic review and meta-analysis. BMJ (clinical research ed.). 2016 Sep 28;354:i5130.

The aim of this study was to examine the effect of TAVR versus SAVR in patients with severe aortic stenosis at low and intermediate surgical risk. The data sources included Medline, Medline-in-process, Embase, and Cochrane CENTRAL from January 2012 to May 2016. Four RCTs published after 2012 with 3,179 patients and a median follow-up of 2 years were included.  The percent women ranged from 46% to 70% and the mean age ranged from 79 to 83 years. The STS risk score of patients ranged from 3.0% to 7.4%.

For the overall results, at the longest follow-up (median 2 years), 319 of the 1,578 (20%) patients undergoing TAVR and 340 of 1,550 (22 patients randomized to SAVR died (HR 0.86, 95% CI 0.74-1.01). The hazard for stroke was lower with TAVR but the confidence interval overlapped no effect (HR 0.81, 95% CI 0.3-1.01). New onset 2-year atrial fibrillation, including transient perioperative atrial fibrillation, was less common in patients randomized to TAVR (three studies, RR 0.43, 95% CI 0.35-0.52). TAVR increased the risk of 2-year aortic valve reintervention (RR 3.25, 95% CI 1.29-8.14), and 2-year permanent pacemaker insertion (RR 2.45, 95% CI 1.17-5.15). TAVR compared to SAVR might have little or no impact on 2-year health-related quality of life as measured by the KCCQ score (risk difference 3.5, 95% CI −1.9 to 8.9). The authors concluded that "many patients, particularly those who have a shorter life expectancy or place a lower value on the risk of long term valve degeneration, are likely to perceive net benefit with transfemoral TAVI versus SAVR."

Singh K, Carson K, Rashid MK, et al. Transcatheter aortic valve implantation in intermediate surgical risk patients with severe aortic stenosis: A systematic review and meta-analysis.  Heart Lung and Circulation. 2018 Feb;27(2):227-234.

The aim of this study was to perform a systematic review to evaluate the 30-day and 12-month mortality of TAVR compared to SAVR in intermediate-risk patients with severe aortic stenosis. The study used data that was based on a comprehensive search of four major databases (EMBASE, Ovid MEDLINE, PubMed, and Google Scholar) that was performed from their inception to April 29, 2016. Three randomized and five observational studies with propensity-matched data were included. Across the eight studies for those receiving TAVR, the average age ranged from 77 to 82 years and 27 to 62% were men.  For the TAVR group, the mean STS risk score ranged from 2.9% to 8%.

For the overall results, all-cause mortality at 30 days (P = 0.07) and 12 months (P = 0.34) was similar between the two groups. The 30-day all-cause mortality was lower in patients undergoing TAVR compared to SAVR, but this did not reach a statistically significant level (OR 0.76, 95% CI 0.57–1.02, P = 0.07).  There was no difference in 12-month all-cause mortality (OR 0.90, 95% CI 0.72–1.12 P = 0.34) between the two groups.  There was no statistically significant difference in the rate of stroke between the two groups (TAVR 4.1% vs. SAVR 4.8%, OR 0.86, 95% CI 0.62–1.20 P = 0.37). The rate of new pacemaker implantation was significantly higher in the TAVR group (11.6% versus 5.1%, OR 4.85, 95% CI 1.68–14.00, P < 0.00001). The authors concluded that "in the intermediate-risk patients, the 30-day and 12-month mortality are similar between TAVI and SAVR. Increased operator experience and improved device technology have led to a significant reduction in mortality in intermediate-risk patients undergoing TAVI."

The aim of this meta-analysis was to compare collective outcomes and adverse events of TAVR versus SAVR across the entire spectrum of surgical risk in patients with cardiac symptoms and severe aortic stenosis who were candidates for both procedures.  This updated a previous meta-analysis to include data from two recently published RCTs in low-risk patients.

The primary outcome of the meta-analysis was all-cause mortality up to 2 years.  Secondary outcomes included stroke, cardiovascular death, myocardial infarction, acute kidney injury, new-onset atrial fibrillation, major bleeding, major vascular complications, valve endocarditis, and permanent pacemaker implantation, up to 2-year follow-up.

The investigators performed a systematic literature search of Medline, Embase, and the Cochrane Library Central Register of Controlled Trials focusing on peer-reviewed publications of RCTs.  A total of 14 papers were found that reported on seven RCTs comparing TAVR to SAVR in high-, intermediate-, or low-risk patients.  This included a total of 8,020 patients randomly assigned to TAVR (4,014) or SAVR (4,006).  Mean STS scores ranged from 1.9% to 11.8%.  For the TAVR arm, the combined mean STS score was 9.4%, 5.1%, and 2.0% for high-, intermediate-, and low-surgical risk trials, respectively.  Mean age ranged from 73 to 85 years and the percent women ranged from 26% to 47%.

The investigators found that compared to SAVR, TAVR was associated with a statistically significant reduction of all-cause mortality (HR 0.88, 95% CI 0.78–0.99, P = 0.0300).  This effect was consistent across the entire spectrum of surgical risk (P-for-interaction = 0.410) and irrespective of type of transcatheter heart valve system (P-for-interaction = 0.674).  "Survival benefit was particularly evident in patients undergoing transfemoral TAVR, with a 17% relative reduction in the risk of all-cause mortality (HR 0.83, 95% CI 0.72–0.94); whereas there was no advantage of transthoracic TAVR over SAVR (HR 1.17, 95% CI 0.88–1.55) with a P-for-interaction = 0.032 for the two alternative routes of access."  TAVR also resulted in lower risk of stroke (HR 0.81, 95% CI 0.68–0.98, P = 0.028).  SAVR was associated with a statistically significant lower risk of major vascular complications and permanent pacemaker implantations compared to TAVR.

The investigators concluded that "in this meta-analysis of 7 landmark trials comparing TAVI with SAVR in patients with symptomatic, severe aortic stenosis, TAVI was associated with a reduction in all-cause mortality and stroke up to 2 years.  The mortality benefit of TAVI was observed consistently in patients at low, intermediate, and high procedural risk" and irrespective of FDA-approved valve type.  The authors stated that "additional studies of TAVI in younger, low-risk populations, and all-comers are underway.  Further research is required to investigate the long-term (> 5 year) valve durability, and to develop strategies for the optimal management of transcatheter and surgical bioprosthetic valve degeneration."

Smith CR, Leon MB, Mack MJ, et al. Transcatheter versus Surgical Aortic-Valve Replacement in High-Risk Patients. N Engl J Med 2011;364:2187–98.

The aim of the study was to describe results for the high-risk, operable patients in the PARTNER-1 trial who were randomly assigned to TAVR or SAVR. At 25 centers between 2007 and 2009, 699 high-risk patients with severe aortic stenosis were randomly assigned to undergo either TAVR with a balloon-expandable bovine pericardial valve or SAVR.  The mean age of the TAVR group was 84 years and 58% were males. The overall mean STS score was 11.8%.

For the results, the rate of death from any cause was 3.4% in the TAVR group and 6.5% in the SAVR group at 30 days (P = 0.07) and 24% and 27%, respectively, at 1 year (P = 0.44), a reduction of 2.6 percentage points in the TAVR group (95% CI, -9.3 to 4.1). For death from any cause, the HR was 0.95 (95% confidence interval, 0.73-1.23). The rates of major stroke were 3.8% in the TAVR group and 2.1% in the SAVR group at 30 days (P = 0.20) and 5.1% and 2.4%, respectively, at 1 year (P = 0.07). At 30 days, major vascular complications were significantly more frequent with TAVR (11.0% vs. 3.2%, P < 0.001); adverse events that were more frequent after SAVR included 30-day major bleeding (9.3% vs. 19.5%, P < 0.001) and 30-day new-onset atrial fibrillation (8.6% vs. 16.0%, P = 0.006). There was no statistically significant difference in new pacemaker implantation at 30 days (P = 0.89) or 1 year (P = 0.68). The authors concluded that "transcatheter and surgical procedures for aortic-valve replacement were associated with similar rates of survival at 1 year, although there were important differences in periprocedural risks."

Søndergaard L, Steinbrüchel DA, Ihlemann N, et al. Two-year outcomes in patients with severe aortic valve stenosis randomized to transcatheter versus surgical aortic valve replacement: The all-comers Nordic Aortic Valve Intervention Randomized Clinical Trial. Circ Cardiovasc Interv. 2016 Jun;9(6). pii: e003665.

The aim of this study was to evaluate 2-year clinical and echocardiographic outcomes among lower-risk patients who underwent TAVR or SAVR in the NOTION trial. A total of 288 patients from three centers in Denmark and Sweden were randomized to either TAVR (n=145) or SAVR (n=135) with follow-up planned for 5 years.  TAVR patients were slightly younger than the traditional TAVR patient (mean age, 79 years) and more commonly male (53%). The overall mean STS risk score was 3%.

For the overall results, there was no difference in all-cause mortality at 2 years between TAVR and SAVR (8.0% versus 9.8%, respectively; P=0.54) or cardiovascular mortality (6.5% versus 9.1%; P = 0.40). There was no difference in stroke at 2 years (P = 0.46) or transient ischemic attack at 2 years (P = 0.30) between TAVR and SAVR. There was a lower frequency of new-onset or worsening atrial fibrillation at 2 years in the TAVR group compared to SAVR (P < 0.001). There was a higher frequency of permanent pacemaker implantation at 2 years in the TAVR group compared to SAVR (P < 0.001). The authors concluded that " the NOTION trial was the first to randomize all-comers and lower-risk patients to TAVR or SAVR. The 2-year results presented here demonstrate the continuing safety and effectiveness of the TAVR procedure in these patients, but with continued differences in aortic regurgitation, pacemaker implantation, and atrial fibrillation."

Takagi H, Mitta S, Ando T. Long-term survival after transcatheter versus surgical aortic valve replacement for aortic stenosis: A meta-analysis of observational comparative studies with a propensity-score analysis. Catheter Cardiovasc Interv. 2018;00:1–12.

The aim of this study was to synthesize evidence regarding long-term survival after TAVR versus SAVR for severe aortic stenosis from real-world clinical practice by conducting a meta-analysis of observational studies with a propensity-score analysis and > 3-year follow-up. Databases including MEDLINE and EMBASE were searched through April 2017 using the Web-based search engines PubMed and OVID.  Fourteen observational comparative studies published between 2012 and 2016 enrolling a total of 4,197 patients were identified. For the TAVR groups, the age ranged from 78 to 85 years and the percent female ranged from 25% to 80%. For the TAVR groups across the 14 studies, the STS risk score ranged from 6.6% to 12.1%.

For the overall results, a pooled analysis of all 14 studies across all risk categories demonstrated a statistically significant 54% increase in > 3 year mortality with TAVR relative to SAVR (HR 1.54; 95% CI 1.31–1.81, P for effect < 0.00001, P for heterogeneity = 0.14).  As part of the sensitivity analyses, excluding a study that included only low-risk patients with a EuroSCORE II of < 4% yielded a mortality HR = 1.52 (95% CI 1.26-1.83) favoring SAVR. The authors concluded based on "a meta- analysis of 14 observational comparative studies with a propensity-score analysis including a total of > 4,000 patients, TAVI is associated with worse > 3-year overall survival than SAVR."

Tam DY, Vo TX, Wijeysundera HC, et al. Transcatheter vs surgical aortic valve replacement for aortic stenosis in low-intermediate risk Patients: A meta-analysis. The Canadian Journal of Cardiology. 2017 Sep;33(9):1171-1179.

The aim of this study was to determine differences in 30-day and late mortality in patients at low or intermediate surgical risk (STS risk score < 10%) treated with TAVR compared with SAVR. Medline and Embase were searched from 2010 to March 2017 for studies that compared TAVR with SAVR in the low and intermediate surgical risk populations, restricted to RCTs and matched observational studies. Four RCTs (n = 4,042) and 9 propensity score-matched observational studies (n = 4,192) were included in the meta-analysis (n = 8,234). Patients in the RCTs had mean STS risk scores ranging from 2.9% to 5.8% while mean LES scores ranged from 8.4% to 11.9%.  Patients in the observational studies had mean STS risk scores ranging from 4.6% to 7.2% while mean LES scores ranged from 4.2% to 24.4%).  No demographic data was presented.

For the overall results, there was no difference in 30-day / in-hospital mortality between TAVR and SAVR (3.2% vs 3.1%, pooled RR 1.02, 95% CI 0.80-1.30, P = 0.99) or mortality at a median of 1.5-year follow-up (incident RR 1.01, 95% CI 0.90-1.15, P = 0.83). There was a statistically significant decrease in periprocedural stroke in the TAVR group (3.0%) compared with the SAVR group (3.9%) (pooled RR 0.76, 95% CI 0.60-0.97, P = 0.03). In the TAVR group, there was a statistically significant reduction in periprocedural atrial fibrillation (11.2% vs 35.2%; pooled RR 0.31, 95% CI 0.27-0.36, P < 0.00001).  In the TAVR group, there was an increased risk of periprocedural permanent pacemaker insertion (15.6% vs 4.9%; pooled RR 3.57, 95% CI 2.25-5.68, P < 0.00001). There was less periprocedural myocardial infarction in the TAVR group (0.8% vs 1.3%; pooled RR 0.64, 95% CI 0.41-1.00, P = 0.05). The authors concluded that "although there was no difference in 30-day and late mortality, the rate of complications differed between TAVR and SAVR in the low-intermediate surgical risk population."

Villablanca PA, Mathew V, Thourani VH, et al. A meta-analysis and meta-regression of long-term outcomes of transcatheter versus surgical aortic valve replacement for severe aortic stenosis. International Journal of Cardiology. 2016 Dec 15;225:234-243.

The aim of this study was to determine the long-term (≥1 year follow-up) safety and efficacy of TAVR compared SAVR in patients with severe symptomatic aortic stenosis who are at high and intermediate operative risk. A computerized literature search of PubMed, CENTRAL, EMBASE, the Cochrane Central Register of Controlled Trials, ClinicalTrials.gov, Google Scholar databases, and the scientific session abstracts in Circulation, Journal of the American College of Cardiology, European Heart Journal and American Journal of Cardiology was conducted from January 1, 2000, to April 10, 2016. Four RCTs and 46 observational studies satisfied inclusion criteria. For the 50 publications, publication year ranged from 2008 to 2016. Across the TAVR groups, the mean age ranged from 70 years to 91 years. Among the TAVR groups, STS score ranged from 2.9 to 11.8.

For the overall results, sensitivity analysis showed no differences in high risk (RR, 1.16; 95% CI 0.87–1.53; P = 0.32) and intermediate risk (RR, 1.15; 95% CI 0.83–1.60; P = 0.40) between both approaches in long-term (≥1 year) all-cause mortality. Sensitivity analysis of 30-day mortality showed no differences in high-risk (RR, 1.02; 95% CI 0.76–1.36; P = 0.91) and intermediate-risk (RR, 0.65; 95% CI 0.38–1.09; P = 0.10) patients between both TAVR and SAVR approaches. Sensitivity analysis showed that stroke risk was significantly lower in high-risk patients undergoing TAVR (RR, 0.79; 95% CI 0.66–0.95; P = 0.01) compared to SAVR. Sensitivity analysis showed that for atrial fibrillation the lower events with TAVR were observed in both high-risk (RR, 0.38; 95% CI 0.26–0.55; P = < 0.001) and intermediate-risk patients (RR, 0.39; 95% CI 0.25–0.62; P < 0.001). No difference in pacemaker implantation risk was observed in intermediate-risk patients (RR, 1.68; 95% CI 0.94–3.00; P = 0.08). The authors concluded that their meta-analysis showed "that TAVR is as effective as SAVR in high-risk patients with aortic stenosis for the endpoint of long-term survival; each intervention confers its own significant complications. There is early evidence that TAVR may be superior to SAVR in intermediate-risk patients."

Wang Y, Zhou Y, Zhang L, et al.  Midterm outcome of transcatheter versus surgical aortic valve replacement in low to intermediate risk patients: A meta-analysis of randomized controlled trials.  Journal of Cardiology. 2018 Jun;71(6):534-539.

The aim of this study was to assess the midterm outcome comparing TAVR and SAVR for the treatment of patients with severe aortic stenosis at low to intermediate surgical risk. PubMed, EBSCO, and Cochrane CENTRAL (Cochrane Central Registry of controlled trials) were systematically searched for RCTs that reported the clinical outcomes of TAVR versus SAVR in patients at low to intermediate surgical risk with at least 2 years of follow-up. From 2000 to 2017, 4 clinical studies comprising 4355 patients were identified.  The RCTs were published between 2016 and 2017.  Across the 4 TAVR groups, age ranged from 79.2 to 81.5 years and percent male ranged from 53.8% to 57.9%. STS score for the TAVR groups ranged from 2.9% to 5.8%.  Mean Logistic EuroSCORE I ranged from 8.4% to 11.9%.

For the overall results, at 2-year follow-up, TAVR was associated with similar rate of 2-year death from any cause (RR 0.86; 95% CI 0.67–1.10; P = 0.22), cardiovascular death (RR 0.88; 95% CI: 0.73–1.06), 2-year stroke (RR 0.90; 95% CI: 0.73–1.10; p = 0.31), or 2-year myocardial infarction (RR 0.99; 95% CI: 0.70–1.39; p = 0.93) between the two groups. TAVR reduced 2-year new atrial fibrillation (RR 0.46; 95% CI: 0.33–0.64; p < 0.0001).  However, 2-year PPM implantation (RR 3.01; 95% CI: 1.04–8.72; p = 0.04) and 2-year aortic-valve re-intervention (RR 3.22; 95% CI: 1.64–6.29; p = 0.0006) were more common in the TAVR group than the SAVR group. The authors concluded that "in patients with severe AS at low to intermediate surgical risk, compared with SAVR at midterm follow-up, TAVR has similar rate of mortality, myocardial infarction, and stroke, lower incidence of life-threatening bleedings, acute kidney injury, and new-onset atrial fibrillation, but increased incidence of permanent pacemaker implantation."

The aim of this study was to evaluate sex-specific differences in outcomes after SAVR or TAVR in high-risk patients with severe aortic stenosis.  While past trials demonstrated similar survival after SAVR or TAVR, sex-specific outcomes remained unknown.

This study was a secondary, subgroup analysis of data prospectively collected in the PARTNER trial (Smith 2011).  PARTNER randomized 699 (57% male, 43% female) high-risk patients with severe aortic stenosis to either SAVR or TAVR with a balloon-expandable valve (see the summary of Smith 2011 in this Evidence section for trial details).  The current study compared baseline characteristics, and procedural (30-day or in-hospital), and 2-year outcomes among males and females in that trial.

The current secondary analysis used Kaplan-Meier time-to-event methods to generate survival curves.  Log-rank tests compared these curves, and Cox proportional hazards models tested for interactions between sex and treatment type.

The investigators found significant gender differences in baseline characteristics.  "Despite higher STS predicted risk of mortality (PROM) scores (11.9 vs. 11.6; p= 0.05), female patients had lower prevalence of coronary artery disease (64.4% vs. 83.7%), prior coronary artery bypass graft surgery (19.8% vs. 61.2%), peripheral vascular disease (36.4% vs. 46.9%), diabetes mellitus (35.6% vs. 45.6%), and elevated creatinine (11.7% vs. 23.9%).  There were differences in outcomes as well.  "Among female patients, procedural mortality trended lower with TAVR versus SAVR (6.8% vs. 13.1%; p= 0.07) and was maintained throughout follow-up (hazard ratio [HR]: 0.67; 95% confidence interval [CI]: 0.44 to 1.00; p= 0.049), driven by the transfemoral arm (HR: 0.55; 95% CI: 0.32 to 0.93; p= 0.02).  Among male patients, although procedural mortality was lower with TAVR (6% vs. 12.1%; p= 0.03), there was no overall survival benefit (HR: 1.15; 95% CI: 0.82 to 1.61; p= 0.42)."  The investigators could not determine in this hypothesis-generating subgroup analysis whether these differences in baseline characteristics caused the differences seen in health outcomes between males and females.

The investigators concluded that "despite higher incidences of vascular complications and strokes, women had better late mortality with TAVR than with SAVR.  That was especially true in the transfemoral arm and suggests that for high-risk female patients, TAVR is a better option than surgery."

The authors cautioned that the study was hypothesis-generating only, and opined that "a randomized, controlled trial specifically in female patients is necessary to properly study differences in mortality between treatment modalities."

Witberg G, Lador A, Yahav D, et al. Transcatheter versus surgical aortic valve replacement in patients at low surgical risk: A meta-analysis of randomized trials and propensity score matched observational studies. Catheterizations and Cardiovascular Interventions:  official journal of the Society of Cardiac Angiography and Interventions. 2018 Feb 1;00:1–9.

The aim of this study was to conduct a systematic review and meta-analysis on the relative risks and benefits of TAVR versus SAVR in patients who are at low surgical risk for AVR.   The authors searched Medline, Embase, and Cochrane CENTRAL from January 1, 2005, up to March 31, 2017.

For the overall results, six studies, two RCTs and four propensity score matching observational studies, totaling 3,484 patients were included. For the six studies, the mean age ranged from 78 to 82 + 4.4 years, and 27% to 59% were male. Follow-up ranged from 3 months to 3 years, with a median of 2 years. The average LES score was 6.5% and average STS risk score was 3.0%.  Short-term mortality, defined as in-hospital or 30-day mortality, was similar with either TAVR or SAVR for five studies with 3,102 patients (2.2% for TAVR and 2.6% for SAVR, OR 0.89, 95% CI 0.56–1.41, P = 0). TAVR was associated with increased risk for intermediate-term mortality (4 studies, 1,804 patients, OR 1.45, 95% CI 1.11–1.89, P = 0.006). TAVR was associated with reduced risk for bleeding and renal failure (acute kidney injury or AKI). TAVR was associated with an increased risk for pacemaker implantation (TAVR rate 15.3%, SAVR rate 3.1%, OR 5.59, 95% CI 4.07-7.67, P < 0.00001). The authors concluded "in patients who are at low surgical risk, TAVR seems to be associated with increased mortality risk. Until more data in this population is available, SAVR should remain the treatment of choice for these patients."

Ad N, Holmes SD, Shuman DJ, et al. The effect of initiation of a transcatheter aortic valve replacement program in the treatment of severe aortic stenosis.  Seminars in Thoracic and Cardiovascular Surgery. 2016 Summer;28(2):353-360.

The aim of this study was to assess the effect of a TAVR program and heart team concept on their approach to severe isolated symptomatic aortic stenosis with regard to surgical practice, patient selection, perioperative outcomes, 1-year survival, and AVR volume. The study population included patients having isolated SAVR between January 2008 and August 2011.  When the program began, the pre-TAVR group (n = 282, 42 months), were compared with those who had isolated SAVR after the TAVR program began until February 2015, the post-TAVR group (n = 344, SAVR and n = 126, TAVR, 42 months).

For the overall results, operative mortality for isolated SAVR was similar in pre-TAVR and post-TAVR (2.1% vs 1.8%, P = 0.798). The study demographics for patients who had isolated TAVR (n = 126) showed a mean age of 83 years and 47% were women. The analysis showed that for all isolated AVR, the observed/expected (O/E) ratio was 0.91 pre-TAVR and 0.82 post-TAVR (n = 470), including O/E = 0.79 for patients who had TAVR. Limitations cited by the authors included that "the results reflect the experience of a single center with a well- established TAVR program and experienced surgeons; for this reason, they may not be generalizable to all centers." The authors concluded that "no changes were found in proportion of isolated surgical AVR cases or patient risk and outcomes after introduction of TAVR program and Heart Team."

Alli OO, Booker JD, Lennon RJ, et al. Transcatheter aortic valve implantation: assessing the learning curve. JACC Cardiovascular Interventions. 2012 Jan;5(1):72-9.

The aim of this study was to assess the learning curve of the physicians and team involved in TAVR via the transfemoral route at a single institution involved in the PARTNER trial. The study used data that was a retrospective analysis of the first 44 consecutive patients who underwent TAVR as part of the PARTNER-1 trial at one institution between November 2008 and May 2011. The analysis methods included the patients divided into tertiles based on case sequence, with approximately 15 patients in each group based on sequence number. Process measures, such as procedure times, radiation exposure, and contrast administration were chosen as markers for increased procedural proficiency.

For the overall results, the 30-day mortality for the entire cohort was 11%. The study demographics showed that the median age of the patients was 83 years and 50% were male. Study results indicated statistically significant decreases across the three tertiles in contrast volume (median: 180 to 160 to 130 ml, P = 0.003), valvuloplasty to valve deployment time (12.0 to 11.6 to 7.0 min, P < 0.001) and fluoroscopy times, from 26.1 to 17.2 and 14.3 min occurred from tertiles 1 to 3 (P < 0.001).  Limitations cited by the authors included that this study was "a retrospective single-center analysis and is subject to the limitations of such analyses. The sample size is small and may be a limitation in the interpretation of the data."  The authors concluded "experience accumulated over 44 transfemoral aortic valve implantations led to significant decreases in procedural times, radiation, and contrast volumes. Our data show increasing proficiency with evidence of plateau after the first 30 cases." Further, "the current data reveal a plateau of the proficiency curve with less variation in means and suggest that even in experienced centers a learning curve of at least 20 aortic valvuloplasties and at least 25 to 30 TAVI procedures will be needed for procedural proficiency."

Alli O, Rihal CS, Suri RM, et al. Learning curves for transfemoral transcatheter aortic valve replacement in the PARTNER-I trial: Technical performance. Catheter Cardiovasc Interv. 2016 Jan 1;87(1):154-62.

The aim of this study was to assess technical performance learning curves of teams performing transfemoral -TAVR, the accumulation and dissemination of new knowledge from experience, and clinical or technological refinement of the procedure. The study population in the PARTNER-I trial included 1,521 patients undergoing transfemoral TAVR from April 2007 to February 2012. For the learning curve analysis, the technical performance metrics selected were procedure time, fluoroscopy time, and contrast volume. Study demographics showed mean age of 84 years, 44% were female, and 93% were white.

For the overall results, as patient sequence number increased, average procedure time decreased from 154 to 85 minutes (P < 0.0001), and fluoroscopy time from 28 to 20 minutes (P < 0.0001). Procedure time plateau was dynamic during the course of the trial, averaging 25 cases (range 21–52) by its end. The distribution of minimum (asymptotic) procedure time revealed an average of 83 minutes, ranging from 52 to 140 minutes, and the average number of cases needed to reach the asymptote at 17 of the 26 institutions was 36, ranging from 21 to 52. Institutions entering the PARTNER-I trial earlier reached minimum procedure times after about 50 cases compared with 25 for institutions entering the trial later—a shorter learning curve. Limitations cited by the authors included "it is difficult to quantify the learning curve per institution, and more so for individual operators." The authors concluded that quantifiable "technical performance learning curves exist for TF-TAVR; procedural efficiency increased with experience, with concomitant decreases in radiation and contrast media exposure. The number of cases needed to achieve efficiency decreased progressively, with optimal procedural performance reached after approximately 25 cases for late-entering institutions."

Attias D, Maillet JM, Copie X, et al. Prevalence, clinical characteristics and outcomes of high-risk patients treated for severe aortic stenosis prior to and after transcatheter aortic valve implantation availability. European Journal of Cardiothoracic Surgery. 2015 May;47(5):e206-12.

The aim of this study was to compare the prevalence, characteristics and outcomes of high-risk patients treated prior to and after the availability of TAVR.  The retrospective study included all consecutive patients treated for native severe aortic stenosis in a high-volume surgical center. Patients who underwent SAVR or TAVR were identified from two national prospective registries: the EPICARD® database and the French Aortic National CoreValve® and Edwards (FRANCE2) registry, respectively. The study population included 879 consecutive patients treated 2 years before (‘pre-TAVR era’ from January 2008 to December 2009) and after (‘modern era’ from January 2010 to December 2011) the availability of TAVR in the institution.

For the overall results, 367 patients were treated by SAVR in the pre-TAVR era and 512 patients were treated in the modern era: 404 by SAVR and 108 by TAVI.  The study demographics showed that among the 879 consecutive patients, 460 were men and 419 women, with a mean age of 75 years. Study results indicated that the all-cause 30-day mortality rate was similar during both eras: 22% in the pre-TAVR era versus 13.8% in the modern era, P = 0.46. The overall 1-year survival was not different for high-risk patients treated in the pre-TAVI era or in the modern era (61% versus 68%, P = 0.52). Limitations cited by the authors included that "the two cohorts were compared without matching, making it a descriptive two-cohort study." This study was a single-center retrospective study conducted in a high volume center. The authors concluded that "the 1-year survival was similar for high risk patients treated before and during the modern era, by SAVR or TAVI."

Badheka AO, Patel NJ, Panaich SS, et al. Effect of hospital volume on outcomes of transcatheter aortic valve implantation. American Journal of Cardiology. 2015 Aug 15;116(4):587-94.

The aim of this study was to determine predictors of TAVR outcomes, such as mortality, with a specific focus on the effect of hospital volume. The study used data that was a cross-sectional study using study cohort data that was derived from the National Inpatient Sample (NIS) database of 2012, a subset of the Healthcare Cost and Utilization Project (HCUP) sponsored by the AHRQ. The NIS is an all-payer inpatient database and is a stratified 20% sample of discharges from US community hospitals. Annual hospital TAVR volume was divided into quartiles with the following cutoffs: first (< 5 TAVRs/year), second (6 to 10 TAVRs/year), third (11 to 20 TAVRs/year), and fourth quartile (>20 TAVRs/year). The study population included 1,481 (weighted n = 7,405) TAVR procedures performed in the U.S. during the study period.

Study results indicated that overall in-hospital mortality rate was 5.1%. Study demographics showed mean age of 82 years for the overall cohort; 49% were women, and 79% were white. Patients aged <60 years were excluded. The analysis showed that in-hospital crude mortality rates decreased with increasing hospital TAVI volume with a rate of 6.4% for lowest volume hospitals (first quartile), 5.9% (second quartile), 5.2% (third quartile), and 2.8% for the highest volume TAVR hospitals (fourth quartile) (P <0.001). The association between hospital volume quartile and the primary outcome persisted even after adjusting for potential confounding factors.

Compared to patients treated in the lowest quartile of hospital volume, adjusted odds ratios of mortality for the patients treated in second, third, and fourth quartiles of hospital volume were 0.92 (95% CI 0.70 - 1.21, P = 0.550), 0.80 (95% CI 0.60 - 1.06, P = 0.114), and 0.38 (95% CI .27 - 0.54, P <  0.001), respectively.  Limitations cited by the authors included the lack of long-term follow-up data. The authors concluded that "the highest volume hospitals had significantly better outcomes after TAVI." Further, "the mortality benefit in our study was significant only in the highest volume quartile lending support for hospital volume thresholds for quality control." However, "the volume cutoffs used in the study are applicable to NIS only, which represents a stratified 20% sample of US community hospital discharges and cannot be used to define volume cutoffs in clinical practice."

Clarke S, Wilson ML, Terhaar M.  Using clinical decision support and dashboard technology to improve Heart Team efficiency and accuracy in a Transcatheter Aortic Valve Implantation (TAVI) Program. Stud Health Technol Inform. 2016;225:98-102.

The aim of this study was to describe a clinical decision support system (CDSS) designed to assist heart team experts in treatment selection decisions. In describing the dashboard technology, an innovative feature was its ability to utilize algorithms to consolidate data and provide clinically useful information to inform the treatment decision. The team implemented a CDSS so it would integrate into the existing clinical workflows of the TAVR heart team.

The authors concluded that "computer algorithms and rule-based alerts can provide CDS to clinicians, and this prototype is aimed to improve team efficiencies, accurate assessment of the clinical information to ultimately promote improved decision-making."

de Biasi AR, Paul S, Nasar A, et al. National analysis of short-term outcomes and volume-outcome relationships for transcatheter aortic valve replacement in the era of commercialization.  Cardiology. 2016;133(1):58-68.

The aim of this study was to describe the short-term in-hospital outcomes for TAVR performed in the era during which they were commercialized and to characterize what effects the hospital experience, measured by annual procedural volumes, may have had on short-term outcomes of mortality and morbidity. The study used data from the 2012 NIS from the HCUP with sponsoring from the AHRQ. The cross sectional study consisted of patients ≥ 18 years of age who underwent TAVR as their principal procedure upon admission to an NIS-participating hospital in 2012. Hospitals were categorized into groups defined by the CMS 2012 NCD for TAVR. The lowest-volume group was defined as hospitals that performed <20 TAVRs in 2012, the next group included hospitals performing the minimum CMS requirement up to double the CMS-mandated volume (i.e., 20–39 TAVRs), the third encompassed centers performing 2–3 times the CMS requirement (i.e., 40–59 TAVRs) and the highest-volume group was set as those hospitals performing at least triple the CMS-mandated number of procedures (i.e., ≥ 60 TAVRs).

Study results indicated the overall short-term in-hospital mortality was 5.0% (n = 380).  Results of the study demographics showed 7,635 patients aged ≥ 18 years received TAVR during the one-year study period. The median age was 83 years and 51% of the patients were male while 84% were white.  The mean age was not shown.  Mortality following TAVR at hospitals performing at least the minimum but no more than double the number of procedures required by CMS (i.e., 20–39 TAVRs) was nearly twice the mortality observed for the highest-volume (i.e., > 60 TAVRs) centers (7.0% vs. 3.6%, respectively, P = 0.023). Annual hospital TAVR volume was slightly protective against mortality when treated as a continuous variable in univariable regression (OR 0.99, 95% CI 1.00–1.00, P=0.028) but was not predictive upon multivariable analysis. The adjusted multivariable analysis showed that annual hospital TAVR volume as a continuous variable did not predict short-term in-hospital mortality (OR 1.00, 95% CI 0.99– 1.00, P = 0.111). Limitations cited by the authors included that ICD-9-CM codes are not rigorously defined. The authors concluded that "while unadjusted data suggested a possible association between hospital TAVR volumes and short-term mortality, no such volume-outcome relationships emerged upon more rigorous multivariable regression analyses."

D'Onofrio A, Salizzoni S, Agrifoglio M, et al. Medium term outcomes of transapical aortic valve implantation: results from the Italian Registry of Trans-Apical Aortic Valve Implantation.  Annals of Thoracic Surgery. 2013 Sep;96(3):830-5.

The aim of the multicenter prospective study was to assess early and medium term clinical outcomes of patients undergoing transapical TAVR. From April 2008 through June 2012, the study population included a total of 774 patients enrolled in the Italian Registry of Trans-Apical Aortic Valve Implantation (I-TA) which included 21 centers. Outcomes were analyzed according to the impact of the learning curve comparing the overall survival of the first 50% of patients versus second 50% of patients for each center. The impact of case-volume on survival was analyzed by procedural volume, i.e., high-volume versus low-volume centers, by comparing survival of centers with more than 27 cases versus centers with less than 27 cases, with 27 cases as the cutoff value as this was the median number of cases performed in the participating centers.

For the overall results, thirty-day mortality was 9.9% (77 patients). The study demographics showed a mean age of 81 years and 58% were women. Study results indicated that 1-, 2-, and 3-year survival was 82%, 76% and 68%, respectively. The VARC (Valve Academic Research Consortium) 30-day mortality was significantly higher among the first 50% patients (12%) of each center when compared with the second 50% (7.9%, P = 0.04). But they found similar overall 3-year survival of the first 50% patients (67%) versus the second 50% patients of each center (69%, P = 0.64). Conversely, 30-day VARC mortality of low-volume centers was 12% whereas in high-volume centers it was 9%, this difference was not statistically significant (P = 0.22). The multivariate analysis identified as independent predictors of 30-day VARC mortality several variables including learning curve, second 50% (OR 0.57, 95% CI: 0.34 - 0.94, P = 0.02). A limitation cited by the authors was that the study population was not a homogeneous distribution of patients among the different centers, a common problem with multicenter registries with the results reflecting the real world nature of the study.

The authors concluded that "transapical transcatheter aortic valve implantation (TAVI) provides good early and medium term (up to 3 years) clinical and hemodynamic results".  They "observed that patients who received TA-TAVI during the first half of the experience at each center had a significantly higher 30-day VARC mortality when compared with patients operated on during the following period. Nevertheless, survival at follow-up was similar, reflecting once again the importance of comorbidities. The learning curve is therefore crucial for patient selection and procedure performance (valve sizing, access, positioning, postdilation), and at the multivariate analysis it was identified as an independent predictor of 30-day mortality. However, procedural volume does not seem to have a significant impact on outcomes because 30-day mortality was similar between low-volume centers and high-volume centers."  In conclusion, the authors "did not observe significant differences of survival at follow-up related to the learning curve."

Henn MC, Percival T, Zajarias A, et al. Learning alternative access approaches for transcatheter aortic valve replacement: Implications for new transcatheter aortic valve replacement centers. The Annal of Thoracic Surgery. 2017 May;103(5):1399-1405.

The aim of this retrospective study was to evaluate the learning curve for TAVR approaches and compare perioperative and 1-year outcomes. From January 2008 to December 2014, the study population included 400 patients who underwent TAVR (transfemoral [TF], n = 179; transapical [TA], n = 120; and transaortic [TAo], n = 101). Learning curves were constructed using metrics of contrast utilization, procedural, and fluoroscopy times. Patients within each access approach were sequentially numbered by the order in which they underwent TAVR, and was used as the x-axis variable of experience. To further evaluate the technical learning curve, each access group was divided into two groups: cases completed before proficiency, labeled "early"; and those completed after proficiency, labeled "late".

For the overall results, no statistically significant differences in 30-day or 1-year mortality were seen before or after proficiency was reached for any approach. The study demographics showed mean age across the different groups to vary from 77 to 84 years. Percent female ranged from 34% to 74% across the six groups. When comparing Kaplan-Meier 1-year survival curves for all three access approaches before and after proficiency, there were no statistically significant differences (P = 0.098, 0.333, and 0.658). Overall 30-day mortality regardless of access approach was not statistically significantly different before and after proficiency was reached (5 of 150 [2%] versus 12 of 250 [5%], P = 0.612). When evaluating the Kaplan-Meier 1-year survival curves of all three TAVR approaches combined, there were no differences between survival before proficiency and after proficiency (P = 0.198). Study results indicated that depending on the metric, learning curves for all three routes differed slightly but all demonstrated proficiency and approached their asymptote between the 25th and 50th case. There were no statistically significant differences in procedural times. When comparing the first 50 cases to subsequent cases within each access approach group, the TA and TF approaches demonstrated statistically significant improvements in contrast use (69 mL versus 50 mL, P = 0.002, and 104.8 mL versus 77.0 mL, P = 0.007, respectively). All three access approaches had decreased fluoroscopy times, although they were not statistically significant. Limitations cited by the authors included their study having a relatively small sample size and it being a retrospective, single-institution study.

The authors concluded that "the learning curves for TA and TAo are distinct but technical proficiency begins to develop by 25 cases and becomes complete by 50 cases for both approaches." "However, when comparing the outcomes of the cases before proficiency with those after proficiency was reached, there were no significant differences in outcomes, including 30-day mortality (and) 1-year mortality." The authors concluded that "the technical learning curve for all three access approaches had no effect, however, on outcomes without risk adjustment."

Jensen HA, Condado JF, Devireddy C, et al. Minimalist transcatheter aortic valve replacement: The new standard for surgeons and cardiologists using transfemoral access?  The Journal of Thoracic and Cardiovascular Surgery. 2015 Oct;150(4):833-9.

The aim of this retrospective study was to evaluate a minimalist approach for TAVR (MA-TAVR) cohort with specific characterization between early, midterm, and recent experience and whether an institutional learning curve influenced results. The study population included retrospectively reviewed 151 consecutive patients who underwent MA-TAVR with surgeons and interventionists equally as primary operator at Emory University between May 2012 and July 2014. Patient characteristics and early outcomes were compared using VARC 2 definitions with patients divided according to chronological operation date into three groups: group 1 consisting of the first 50 patients, group 2 included patients 51 to 100, and group 3 included patients 101 to 151.

For the overall results, in-hospital mortality and morbidity were similar among all three patient groups. The study demographics showed median age for all patients was 84 years and was similar among groups. The majority of patients were men (56%) and 85% were white. Study results indicated that the rate of the composite adverse outcomes was similar throughout the experience, and also when group 1 and group 3 were directly compared (95% CI, 0.31-2.53, P = .825). The overall 30-day mortality was 2% and was similar across groups 1, 2, and 3 (2%, 0%, and 4%, respectively, P = 0.369). The authors concluded "in a high-volume TAVR center, transition to MA-TAVR is feasible with acceptable outcomes and a diminutive procedural learning curve." Further, "clinical outcomes were similar throughout the experience." The authors showed "that in a high volume TAVR site no significant learning curve is apparent when the minimalist protocol is implemented."

Kim LK, Minutello RM, Feldman DN, et al. Association between transcatheter aortic valve implantation volume and outcomes in the United States.  The American Journal of Cardiology. 2015 Dec 15;116(12):1910-5.

The aim of this study was to analyze in-hospital outcomes after TAVR stratified according to hospital volumes. The study used data that was from the AHRQ HCUP NIS files from 2012. Hospitals performing transfemoral (TF)-TAVI and transapical (TA)-TAVI were stratified into high- volume and low-volume centers by volume of procedures performed, using the median number of TF-TAVI (20 cases) and TA-TAVI (10 cases) cases as the cutoff for the entire unmatched 2012 NIS cohort of patients. For the results, the study population included a total of 7,660 patients who underwent TAVI in 256 hospitals in 2012. The study demographics across the four groups of high versus low volume and TF versus TA- TAVI showed the mean age ranging from 76 years to 82 years. The percent female ranged from 47% to 59% and the percent white ranged from 79% to 86% across the four groups.

Study results indicated in the TF-TAVI cohort, there was a higher incidence of death in the group of patients undergoing procedures in the low- volume centers versus high-volume centers (6.5% versus 4.5%, respectively, P = 0.02). After adjustment for other potential predictors of outcome, multivariate logistic regression analyses demonstrated that low TF-TAVI volume status was an independent predictor of death (adjusted OR 1.55, 95% CI, 1.09-2.21, P = 0.02). In the TA-TAVI cohort, the unadjusted rates of death were significantly higher in low-volume hospitals versus high-volume hospitals (8.5% versus 4.1%, respectively, P = 0.002). Low-volume status remained an independent predictor of death after multivariable adjustment (adjusted OR 3.08; 95% CI, 1.69 – 5.65, P < 0.001). Limitations cited by the authors included that "assuming worse outcomes at low TAVI volume centers solely based on retrospective studies may be misleading especially because some of these low-volume centers may be in their initial stage of establishing a TAVI program. (Their) report is based on predominantly early generation technology, and therefore, it represents early experiences in the United States." The authors concluded that "low volume was associated with worse postprocedural outcomes, including postprocedural mortality, for both TF-TAVI and TA-TAVI." Further, "centers with lower volume of TAVI had more frequent adverse events compared with higher volume centers."

Landes U, Barsheshet A, Finkelstein A, et al. Temporal trends in transcatheter aortic valve implantation, 2008-2014: patient characteristics, procedural issues, and clinical outcome. Clinical Cardiology. 2017 Feb;40(2):82-88.

The aim of this study was to evaluate temporal trends in a large multicenter TAVR registry. The study population (n = 1,285) included patients who underwent TAVR between January 2008 and December 2014 at 3 high-volume Israeli tertiary medical centers. Patients were divided into 5 time quintiles according to their procedural date (Q1: 2008–2010, 260 patients; Q2: 2011, 251 patients; Q3: 2012, 266 patients; Q4: 2013, 261 patients; and Q5: 2014, 248 patients). Outcomes were analyzed and reported according to VARC-2. Study demographics showed a mean age of 82 years, and 57% were female.

For the overall results, Kaplan-Meier survival curves showed gradual decrease in cumulative mortality risk across procedure year (unadjusted P = 0.031). There was no difference in in-hospital all-cause mortality across the four years of the study (P = 0.583). By multivariate analysis, there was no statistically significant 1-year mortality decrease (HR per one calendar-year increment: 0.96, 95% CI: 0.83-1.10, P = 0.576). Limitations cited by the authors included the study "is a retrospective study, which carries the concern of unmeasured confounding variables and/or possible missing reported outcomes." The authors concluded that in-hospital mortality was small and no temporal trends were identified. They "found a significant temporal trend in survival, with improved long-term survival as the procedure calendar year advanced. There was no significant difference between the cohorts in short-term mortality rate, and the long-term survival variance lost its significance once we adjusted for age, multiple comorbidities, and STS score."

Minha S, Waksman R, Satler LP, et al. Learning curves for transfemoral transcatheter aortic valve replacement in the PARTNER-I trial: Success and safety. Catheter Cardiovasc Interv. 2016 Jan 1;87(1):165-75.

The aim of this study was to investigate whether outcomes of TF TAVR improved with experience, to identify the number of cases needed to maximize device success and minimize adverse events after TF-TAVR, and determine if adverse events were linked to the technical performance learning curve.  From April 2007 to February 2012, the study population included 1521 patients who underwent TF-TAVR in the PARTNER-I trial at 26 sites. Outcomes learning curves were defined as number of cases needed to reach a plateau for device success, adverse events, and post-procedure length of stay. The contribution of the procedure time learning curve to 30-day major adverse events was identified. Study demographics showed a mean age of 84 years; and 44% were female and 93% were white.

For the overall results, 80% device success was achieved after 22 cases; major vascular complications fell below 5% after 70 cases and major bleeding below 10% after 25 cases. It took an average of 28 cases to achieve a consistent low risk of 30-day major adverse events, but institutions entering in the middle of the trial achieved it after about 26. The risk of composite major adverse events (stroke, mortality, major bleeding, and major vascular complications) within 30 days fell from nearly 50% to ̴ 33% by case 45 (P = 0.0008).  The most statistically significant correlate of 30-day major adverse events was procedure time (P < 0.0001). However, this association was related to patient and unmeasured variables, not the procedure time learning curve (P = 0.6). Minimum (asymptotic) probability of a 30-day major adverse event was achieved between 24 and 32 cases across institutions, averaging 28 cases. Institutions entering the trial later reached the institution-specific asymptote for 30-day major adverse events after ̴ 26 cases.  A limitation cited by the authors included that "data collection, along with the collaborative interventional-surgical nature of the PARTNER-I trial, did not allow for assessment of individual operator experience."  The authors concluded that "without risk adjustment an outcomes learning curve appeared to exist for TF-TAVR. A consistent low risk of adverse events was achieved after ̴ 26 cases by end of trial. However, risk factors for adverse outcomes were strongly related to patient characteristics and procedure time that changed over the course of the trial. Once these factors were accounted for, outcomes were not adversely affected by the technical performance learning curve."

The aim of this study was to assess the association of hospital SAVR and combined SAVR and TAVR volumes with patient outcomes of TAVR procedures performed within 1 year, 2 years, and for the entire period after initiation of TAVR programs.  This study used data from the Medicare Provider and Analysis Review and Master Beneficiary Summary Files for patients who underwent TAVR between January October 2011 and December 2015.  The primary outcome was 30-day mortality.  Secondary outcomes included a composite of 30-day mortality or stroke, 30-day hospital readmission, and 1-year and 2-year mortality.

Hospital SAVR volume was initially dichotomized into high (≥97 per year) and low (<97 per year) categories based on the median of the entire cohort.  Then, to evaluate for the synergic effect of SAVR volume and accumulated TAVR performance (i.e., the impact of SAVR volume on TAVR performance), and to assess the association of SAVR volume in combination with hospital TAVR volume with patient outcomes, 4 hospital categories were constructed: (1) low SAVR and low TAVR, (2) high SAVR and low TAVR, (3) low SAVR and high TAVR, and (4) high SAVR and high TAVR.

The associations between SAVR volume, SAVR and TAVR volumes, and risks of death, death or stroke, and readmissions within 30 days were determined using standard hierarchical logistic regression models.  Adjusted analysis included covariates for patient demographics, comorbidities, procedure characteristics, and hospital factors.  Sensitivity analysis was performed for transfemoral TAVR alone, and to test other definitions for high TAVR and SAVR volumes, including "the Leapfrog Group–recommended SAVR volume (120 per year, with estimated 70% being Medicare recipients), and the 2018 [Consensus] recommendation for TAVR volume (50 per year or 100 in the prior 2 years)."  

The study included a total of 60,538 TAVR procedures performed at 438 hospitals, after exclusion of patients who were discharged in December 2015 (to ensure follow-up for all patients).  Important patient demographics included mean age of 82 ± 8 years, 52% male, and 93% white.

The investigators found that hospitals with high SAVR volume (mean annual volume, ≥97 per year) were more likely to adopt TAVR early and had a higher growth in TAVR volumes over time (P < .001).  After adjustment, high hospital SAVR volume alone was not associated with better patient outcomes after TAVR.  "When hospital TAVR and SAVR volumes were jointly analyzed, patients treated in hospitals with high TAVR volume had lower 30-day mortality after TAVR (high TAVR and low SAVR vs low TAVR and low SAVR: odds ratio [OR], 0.85; 95% CI, 0.72-0.99; high TAVR and high SAVR vs low TAVR and high SAVR: OR, 0.81; 95% CI, 0.69-0.95), the effect of which was more pronounced when hospitals also had high SAVR volume.  Patients treated in hospitals with high SAVR volume and high TAVR volume had the lowest 30-day mortality (vs hospitals with low SAVR volume and TAVR volume: OR, 0.77; 95%CI, 0.66-0.89)."

The investigators concluded that "assessing hospital SAVR volume alone is not adequate and potentially misleading given the tendency for these hospitals to accumulate TAVR volume more quickly.  It was within hospitals with high SAVR volume that the association of accumulated TAVR volume with better outcomes became very strong."  Hospital SAVR volume alone was not associated with better TAVR outcomes, but hospitals with high SAVR volume were more likely to be fast adopters of TAVR.  And accumulating high volumes of TAVR in turn was associated with lower mortality after TAVR.  Hospitals that achieved the best TAVR outcomes were those with high volumes of both SAVR and TAVR.  The authors opined that, "With TAVR becoming more mature and being performed among patients with lower risk of complications, the association between hospital SAVR and TAVR volume and patient outcomes may change" and requires continual reassessment.

Patel HJ, Herbert MA, Paone G, et al. The midterm impact of transcatheter aortic valve replacement on surgical aortic valve replacement in Michigan. The Annals of Thoracic Surgery. 2016 Sep;102(3):728-734.

The aim of this study was to characterize the early to midterm, of up to four years, impact of TAVR dissemination on SAVR volume, patient profiles, and outcomes in the state of Michigan. The study used data obtained after SAVR (n = 15,288) and TAVR (n = 1,783) using the Michigan Society of Thoracic and Cardiovascular Surgeons Quality Collaborative between January 2006 and June 2015. During this period, the study population included 17 of 33 adult cardiac hospitals in the state of Michigan that developed TAVR programs.

For the overall results, the rates of 30-day mortality (pre-TAVR era, 3.9% vs post-TAVR era, 2.7%, P < 0.001) were lower in hospitals initiating TAVR programs. The study demographics showed that the mean age of the entire cohort was 71 years and 63% were men. For those receiving TAVR at TAVR hospitals, 93.3% were white. Non-TAVR hospitals did not display changes in mortality for either the entire or the high-risk SAVR cohorts after initiation of TAVR in Michigan.  A limitation cited by the authors included "the relatively long period of this study (9.5 years). Not only could there have been improvements in overall perioperative management but also the development of our quality collaborative may have instituted process and structure changes that positively impacted early outcomes independent of the effects of TAVR during this period in Michigan." The authors concluded that "the O/E (Observed / Expected) ratios for TAVR, however, do suggest improving results over the period, likely reflecting the learning curve associated with this procedure." In addition, "although multiple early-outcome measures consistently appeared to improve for TAVR, the results were mixed for non-TAVR hospitals."

Russo MJ, McCabe JM, Thourani VH, Case Volume and Outcomes After TAVR With Balloon-Expandable Prostheses: Insights From the TVT Registry. J Am Coll Cardiol 2019;73:427–40.

The aim of this study was to determine if: (1) after the initial learning curve, a volume-outcome relationship for balloon-expandable TAVR persisted; and (2) learning curves and volume-outcome relationships differed across different device generations.  

This study used data from the STS/ACC TVT registry for patients with who underwent TAVR with balloon-expandable (Sapien) devices between November 2011 and January 2017. 

Primary outcome measures were 30-day all-cause mortality, stroke, and major vascular complications.  The analysis method established case experience for each hospital by assigning chronological case-sequence numbers, which were then stratified into quartiles.  Kaplan-Meier methods were used to estimate 30-day mortality and stroke within each quartile.  A Cox proportional hazards frailty model tested the association between the 30-day outcomes and case sequence quartile groups while adjusting for STS PROM score and site random effect.  Log-rank methods were used to identify a learning curve termination (LCT), defined as the point after which there was no longer a relationship between increasing case experience and improved outcomes.  Cox proportional hazards frailty models then tested for associations between an outcome and case-sequence quartiles beyond the learning curve.  Similar analyses were performed for the most recent-generation balloon-expandable valve, S3, including for a subgroup of S3 valves used at hospitals with no prior experience with early-generation balloon-expandable valves (Sapien or Sapien XT).

The study included a total of 61,949 TAVR procedures performed at 450 hospitals (Sapien, n = 18,192; Sapien XT, n = 15,530; S3, n = 28,227).  Patients/procedures excluded were those that involved valve-in-valve procedures, emergent cases, and patients with primary aortic insufficiency or bicuspid valves. Important patient demographics included mean age ranging from 81 years and 46% male (in the 4^th case-sequence quartile) to 82 years and 50% male (in the 1st case-sequence quartile).

The investigators found that there was a stepwise decrease in the combined endpoint of 30-day mortality or stroke from Sapien to SXT to S3 (P < 0.001).  A learning-curve termination was identified as case 200.  After this, there was no significant association between 30-day mortality (P = 0.83) or stroke (P = 0.45) and increasing hospital implant frequency, but there was a significant association between major vascular complications and hospital implant frequency (P = 0.01).  In the S3-only analysis, the investigators did not detect a learning curve or volume-outcome relationship with respect to risk-adjusted mortality or stroke, including at hospitals with no prior experience with balloon-expandable valves.

The investigators concluded that when analyzing all generations of balloon-expandable valves, there was a learning curve termination at 200 cases.  After this learning curve, no volume-outcome relationship was evident.  In the analysis of the current-generation balloon-expandable valve (S3) only, no learning curve or volume-outcome relationship was detected.  The authors opined that "these findings support that good outcomes are not merely a function of quantity, but are influenced by a constellation of factors, including technological advancements, best practices, collaborative knowledge programs, and organizational culture."

Salemi A, Sedrakyan A, Mao J, et al. Individual Operator Experience and Outcomes in TAVR. JACC: Cardiovascular Interventions. Vol. 12(1), January 2019 DOI: 10.1016/j.jcin.2018.10.030.

The aim of this study was to assess the impact of individual operator experience on transfemoral TAVR outcomes, given the lack of published analyses of the volume-outcome relationship at the operator (as opposed to the hospital) level.  This study used data from the New York State Department of Health Statewide Planning and Research Cooperative System (SPARCS; an all–age group, all-payer database) for patients who underwent transfemoral TAVR between January 2012 and December 2016.  The primary outcome was a composite of in-hospital mortality, stroke, or acute MI.  Secondary outcomes were the individual components of this primary composite outcome.

The analysis method evaluated physician volume during the prior year, first as a categorical variable (low, 1 to 23 cases; medium, 24 to 79; high, ≥ 80), then as a continuous variable.  Hierarchical and restrictive cubic spline regression methods were used to assess the impact of individual operator volume on the risk-adjusted primary and secondary outcomes.  Adjusted analysis included covariates for patient demographics, comorbidities, cardiac procedures, procedure year, hospital volume, and the first year in which each physician began performing TAVR.  Sensitivity analysis excluded the initial 10, 20, and 30 TAVR cases "to assess whether the exclusion of initial cases would mitigate the impact of the initial learning curve and change the volume-outcome relationship."  

Of a total of 8,771 transfemoral TAVR procedures conducted by 207 operators in New York, 5,916 were included in the study, after exclusion of procedures that were elective, or for which a licensed physician operator could not be identified.  Important patient demographics included mean age of 83 years, 50% male, and 85% white.

The investigators found that patients undergoing TAVR performed by high-volume physicians (≥ 80/year) had a significantly improved risk-adjusted composite outcome of death, stroke, or acute MI (OR 0.59, 95% CI 0.37 to 0.93) compared with those treated by low-volume physicians (<24/year).  Being treated by operators who performed 200 procedures during the prior year was associated with significantly lower risks for post-procedural stroke (OR 0.41, 95% CI 0.17 - 0.97) and composite events (OR 0.45, 95% CI 0.26 to 0.78).  In sensitivity analysis, "excluding the first 10 procedures yielded similar results to the main analysis. When excluding the first 20 or first 30 procedures, the association between operator volume and improvement in patient outcomes became more linear."

The investigators concluded that there was a statistically significant and clinically important inverse relationship between increasing individual operator volume and a risk-adjusted composite outcome of in-hospital mortality, stroke, or MI, with the relationship being driven primarily by stroke.  Furthermore, "a clear and quantifiable learning curve on the operator level of at least 20 procedures exists, after which a less significant albeit continuous improvement in outcomes is demonstrated."  Regarding the requirement that heart teams (centers) must complete "at least 20 such procedures annually or 40 procedures biannually," the authors opined that while "these numbers appear to be generally consistent with our study results in general terms, perhaps these guidelines are more appropriately directed toward specific operators."  This also highlights "the importance of proctoring and partnerships with experienced TAVR practitioners during the early adoption period."

Suri RM, Minha S, Alli O, et al. Learning curves for transapical transcatheter aortic valve replacement in the PARTNER-I trial: Technical performance, success, and safety. J Thorac Cardiovasc Surg. 2016 Sep;152(3):773-780.e14.

The aim of this study was to evaluate the rate at which technical performance improved, assessed change in occurrence of adverse events in relation to technical performance, and determined whether adverse events after TA-TAVR were linked to acquiring technical performance efficiency, the learning curve. From April 2007 to February 2012, the study population included 1100 patients that underwent TA-TAVR in the PARTNER-I trial at 24 sites. The technical performance measures which were assessed for learning curves were selected to illustrate technical efficiency and the influence on the learning curve of accumulating external experience: procedure time, fluoroscopy time, contrast volume used, and number of postdeployment dilatations. Important study demographics included average age of 85 years, 52% female and 95% white.

For the overall results, mean procedure time decreased from 131 to 116 minutes within 30 cases (P = .06) and remained constant at about 117 minutes thereafter. The authors were unable to demonstrate that higher institutional volume, assessed as lower interval between sequential cases, was associated with procedure time after accounting for sequence number and trial entry date (P = .5). Study results indicated that within 30 days, 354 patients experienced a major adverse event (stroke in 29, death in 96), with possibly decreased complications over time (P  ̴ .08).

Intraprocedural adverse events fell from 31% to 25% by 15 cases, but with wide confidence limits. Occurrence of a composite event (adverse event or death) within 30 days fell from 51% initially to 29% by case 30, then rose slightly to 36% by case 90. Although longer procedure time was associated with more adverse events (P < .0001), these events were associated with change in patient risk profile, not the technical performance learning curve (P = .8). Limitations cited by the authors included they "were unable to account for variations in team members and whether individual operator experience was more important than a team’s accrued experience." The authors concluded that "the learning curve for TA-TAVR was 30 to 45 procedures performed, and technical efficiency was achieved without compromising patient safety." Further, "following the introduction of TA-TAVR across PARTNER-I institutions, procedure time, fluoroscopy time, and volume of contrast medium sharply decreased as patient sequence number increased, indicating a short technical performance learning curve from the perspective of number of cases."

Vaquerizo B, Bleiziffer S, Wottke M, et al. Impact of transcatheter aortic valve implantation on surgical aortic valve. International Journal of Cardiology. 2017 Sep 15;243:145-149.

The aim of this study was to investigate the impact of increasing TAVR volumes on SAVR volumes and to assess the evolution in baseline demographics and its impact on 30-day clinical outcomes across TAVR and SAVR patients.  From June 2007 through September 2015, this German single-center observational study included 3543 consecutive patients with severe aortic stenosis who underwent TAVR (n= 1407) or SAVR (n= 2136) in a single center and were subcategorized into nine cohorts defined by procedure year.  The study demographics showed that the mean age was 74 years and 42% were female.

For the overall results, the crude all-comers 30-day mortality for TAVR improved from 11% in 2007 to 3% in 2015 (P < 0.001). The overall 30- day mortality was similar between TAVR and SAVR after adjusting for the independent predictors of mortality (adjusted OR 0.758, 95% CI 0.504-1.139), P = 0.2). A limitations cited by the authors included "the single-center experience and retrospective nature" of the study.  The authors concluded that there was "a remarkable improvement in the crude 30-day mortality rates" over the nine enrollment periods (procedure years) for the TAVR cohort but not for SAVR. In addition, "overall 30-day mortality was similar between TAVR and SAVR after adjusting for baseline characteristics."

The aim of this study was to evaluate the association between hospital volume of TAVR procedures and patient health outcomes.  Specific research questions included: Is there a significant volume-outcome relationship, and does this relationship persist at "steady state," after any "learning curve" for TAVR operators and "start-up" period for hospitals ends (as measured at two time points, 6 months and 12 months of TAVR experience at each hospital)?  Do patient and hospital characteristics differ according to hospital procedural volume?

This study used data from the STS/ACC TVT registry for high-risk and intermediate-risk patients with symptomatic severe aortic stenosis who underwent TAVR with the latest devices, between January 2015 and December 2017.  The primary outcome was risk-adjusted mortality at 30 days.  Secondary outcomes included a 30-day composite of complications (stroke, moderate or severe paravalvular leak, major vascular access-site complications, major bleeding, or acute kidney injury) and outcomes for each component of this composite.

The primary analysis assessed the association between hospital procedural volume as a continuous variable and risk-adjusted 30-day mortality in patients who underwent a transfemoral TAVR approach.  (Similar analysis was performed for the more complicated non-transfemoral patients.)  Generalized linear mixed models assessed hospital TAVR volume–outcome relationships, using a three-level (patients, operators, and hospitals) hierarchical structure.  Hospitals were categorized into quartiles based on annualized TAVR procedure volume, and outcomes were compared across quartiles.  Adjusted analysis used covariates from a standard TVT Registry in-hospital risk model, along with individual operator case number to account for an operator "learning curve" when assessing the volume-outcome relationship at the hospital level.  Sensitivity analysis accounted for a hospital "start-up" period at two time points, excluding the first 6 months and then the first 12 months of procedures at each hospital. 

The study included a total of 96,256 transfemoral procedures performed at 554 hospitals by 2,935 operators.  Important patient demographics included, for the overall study, median age of 82 years, 53% male, and 90% white.

The investigators found a statistically significant but small absolute difference in adjusted 30-day mortality between low- and high-volume hospitals.  Mortality was higher and more variable at hospitals in the lowest-volume quartile (3.19%; 95% CI 2.78 to 3.67) than at hospitals in the highest-volume quartile (2.66%; 95% CI, 2.48 to 2.85; OR 1.21; P = 0.02).  A small but statistically significant difference persisted after exclusion of procedures done in a hospital’s first 12 months – with the goal of measuring a steady-state volume-outcome relationship.  Additional findings included: patients with higher risk scores (more complex patients) were treated at higher-volume hospitals; "a higher percentage of patients treated at hospitals in the lowest-volume quartile than in the highest-volume quartile were black or Hispanic (12.1% vs. 7.8%)."  However, the majority of blacks and Hispanics were treated at hospitals in the two highest-volume quartiles.

The investigators concluded that "An inverse volume–mortality association was observed for transfemoral TAVR procedures from 2015 through 2017.  Mortality at 30 days was higher and more variable at hospitals with a low procedural volume than at hospitals with a high procedural volume."

Wassef AWA, Alnasser S, Rodes-Cabau J, et al. Institutional experience and outcomes of transcatheter aortic valve replacement: Results from an international multicentre registry.  International Journal of Cardiology. 2017 Oct 15;245:222-227.

The aim of this study was to investigate the relationship between institutional experience and procedural and clinical TAVR outcomes. The study population included all consecutive patients who underwent TAVR at eight international sites in North America, South America and Europe since the initiation of the respective center's TAVR program. The study used data that was from 1953 patients undergoing TAVR which were grouped into chronological volume quantiles (Q) to assess temporal changes on procedural and clinical outcomes that comprised of the first 62 cases for Q1, 63–133 for Q2, 134 to 242 for Q3 and 243 to 476 for Q4.

For the overall results, 30-day all-cause mortality was significantly reduced in Q4 compared to Q1 (8.3% vs 3.7%, P=0.011). The study demographics showed that the mean age of patients was 81 years and 991 (51%) were female. Study results indicated that TAVR in Q4 was independently associated with lower mortality (OR 0.36, 95% CI 0.19–0.70, P = 0.002). A limitation cited by the authors was that the "study examined procedural experience of centres' heart team performing TAVR procedures and data on individual operator's experience and volume was not captured."  The authors concluded that "greater institutional experience with TAVR procedures improves device success and clinical outcomes. An experience of >243 cases is independently associated with lower mortality. These findings have important implications for defining minimum volume criteria for institutions and training standards for TAVR procedure."

The 27 studies reviewed below all used the TVT registry prominently in their analyses. These studies were not specifically designed to target a particular question for the registry as identified by CMS in 2012; nor were protocols for these studies submitted to or approved by CMS. However, we believe the research questions for each of these studies reviewed below are related to one or more of the questions for the registry identified in the 2012 NCD; and collectively, these 27 studies are related to all of the registry questions in the 2012 NCD.  We are aware that there are numerous other published studies that may be related to one or more of these questions for the registry in the 2012 NCD, or aspects of them. Discussion of the purpose, composition, function, and strengths of the registry appear in the analysis section of this decision. Our 2012 NCD required that the registry provide all data necessary to address the following questions (2012 NCD):

• When performed outside a controlled clinical study, how do outcomes and adverse events compare to the pivotal clinical studies?
• How do outcomes and adverse events in subpopulations compare to patients in the pivotal clinical studies?
• What is the long term (≥ 5 year) durability of the device?
• What are the long term (≥ 5 year) outcomes and adverse events?
• How do the demographics of registry patients compare to the pivotal studies?

Below are summaries of the 27 published papers that are based on registry data and are relevant to this NCD reconsideration.

Abramowitz Y, Vemulapalli S, Chakravarty T, et al. Clinical Impact of Diabetes Mellitus on Outcomes After Transcatheter Aortic Valve Replacement: Insights from the STS/ACC TVT Registry. Circ Interv. 2017;10(11).

The aim of this study was to assess the magnitude of risk and the incremental influence of diabetes status on short- and long-term mortality and morbidity associated with TAVR.  Previous publications were "limited by small sample size and contradictory results."

This study used data from the STS/ACC TVT registry for patients with and without diabetes mellitus (DM) who underwent TAVR from November 2011 through September 2015, and were linked to Medicare claims (for 30-day and 1-year outcomes).

Primary outcomes included post-TAVR mortality (in-hospital, 30-day, and 1-year mortality), as well as stroke, rehospitalization because of heart failure, new dialysis, and MI at 30 days and 1 year. The in-hospital outcomes were collected from the TVT Registry.

The analysis method used logistic regression models to assess unadjusted and adjusted associations between DM and in-hospital mortality. Covariates in the multivariate model were derived from a standard, validated TAVR in-hospital mortality risk model. Cumulative incidence methods were used to compare 30-day and 1-year outcomes between patients with and without DM; death was treated as a competing risk for non-fatal outcomes. Cox proportional hazards models were used to assess unadjusted and adjusted associations of DM with 1-year mortality post-TAVR, and to identify risk factors (predictors) for these diabetic patients.

Patients included a total of 47,643 treated with TAVR at 394 US hospitals. Of this, 29,794 (62.5%) patients had no DM and 17,849 (37.5%) had DM. Of these diabetic patients, 6,600 (37.0%) were insulin treated (IT DM); 11,249 (63%) were non-IT DM patients (with 8,031 patients receiving oral hypoglycemic therapy, 2,139 treated with diet, 47 receiving other noninsulin subcutaneous therapy, and 1,010 receiving no therapy).

Important patient demographics included, for the No-DM group (n=29,794), median age of 85 years and 50% male; and for the DM group (n=17,849), median age of 81 years and 55% male. DM patients were generally sicker, with a higher burden of comorbidities and predicted risk of mortality (by STS PROM score).

The investigators found that "30-day mortality was 5.0% in patients with DM (6.1% in IT DM and 4.4% in non-IT DM; P<0.001) versus 5.9% in patients without DM (P<0.001).  Overall, 1-year mortality was 21.8% in patients with DM (24.8% in IT DM and 20.1% in non-IT DM; P<0.001) versus 21.2% in patients without DM (P=0.274).  In a multivariable model, DM was associated with increased 1-year mortality (hazard ratio, 1.30; 95% CI, 1.13–1.49; P<0.001). Subgroup multivariable analysis showed stronger mortality association in IT diabetics (hazard ratio, 1.57; 95% CI, 1.28–1.91; P<0.001) than in non-IT diabetics (hazard ratio, 1.17; 95% confidence interval, 1.00–1.38; P=0.052)."

Numerous mortality risk factors for DM patients were identified. Thus, the investigators top three findings were: (1) DM was associated with increased 1-year mortality after adjustment for multiple baseline and procedural characteristics; (2) IT diabetics had increased 1-year mortality compared with non-IT diabetics; and (3) this increased mortality in IT diabetics drove the increased 1-year DM mortality overall compared to non-DM patients.

The investigators concluded that DM "does not confer a significant incremental short-term risk factor at TAVR, but does confer significant longer-term risk" specifically increased adjusted 1-year mortality compared to non-DM patients.

Alfredsson J, Stebbins A, Brennan JM, et al. Gait Speed Predicts 30-Day Mortality After Transcatheter Aortic Valve Replacement: Results from the Society of Thoracic Surgeons/American College of Cardiology Transcatheter Valve Therapy Registry. Circulation. 2016;133.

The aim of this study was to assess the impact of frailty (measured by 5-m gait speed) on mortality post-TAVR.

This study used data from the STS/ACC TVT registry for patients who underwent TAVR between November 2011 and June 2014, and were linked to Medicare claims (for 30-day outcomes).

The primary outcome was 30-day all-cause mortality. Secondary outcomes included in-hospital mortality, bleeding, acute kidney injury, and stroke.

The analysis method grouped patients into three cohorts based on gait speed: slowest walkers, >10 seconds for 5 m (<0.5 m/s); slow walkers, 6 to ≤10 seconds for 5 m (≥0.5–0.83 m/s); and normal walkers, ≤6 s for 5 m (≥0.83 m/s). The association between gait speed and 30-day mortality was analyzed as a continuous variable and across the 3 gait speed groups. Multivariate logistic regression models were used to assess the association of gait speed and 30-day mortality, adjusting for STS risk score, age, sex, access site, chronic lung disease, baseline renal impairment, and left ventricular ejection fraction < 30%.

Patients included a total of 8,039 from 256 hospitals, after multiple exclusions such as patients from low-volume hospitals, urgent/emergent TAVR procedures, procedures performed on patients <65 years of age ("because gait speed may not reflect pre-procedural frailty in these cases"), and missing data.

Important patient demographics included, for the overall study population (n=8,039), median age of 84 years and 49% male. Patients with the slowest gait speeds were more often female and had more heart failure, chronic lung disease, and comorbidity.

The investigators found that "gait speed independently predicts 30-day mortality after adjustment for STS-PROM score and key covariates, with every 0.2-m/s decrease in gait speed corresponding to an 11% increase in 30-day mortality" (adjusted OR 1.11, 95% CI 1.01–1.22). The slowest walkers had 35% higher 30-day mortality than normal walkers (adjusted OR 1.35, 95% CI 1.01–1.80), significantly longer hospital stays, and a lower probability of being discharged to home.

The investigators concluded that gait speed is independently associated with 30-day mortality after TAVR. This frailty measure should be "used for all patients referred for TAVR, given that it adds easy-to-obtain functional information beyond that reflected in established risk scores. Identification of frail patients with the slowest gait speeds facilitates pre-procedural evaluation and anticipation of a higher level of post- procedural care."

Arnold SV, Spertus JA, Vemulapalli S, et al. Association of Patient-Reported Health Status with Long-Term Mortality After Transcatheter Aortic Valve Replacement: Report From the STS/ACC TVT Registry. Circ Interv. 2015;8(12).

The aim of this study was to determine whether preprocedure health status is associated with survival after TAVR, using data from the TVT Registry to examine whether baseline KCCQ scores are associated with short- and long-term mortality after TAVR. Among 7769 patients from 286 sites in the TVT Registry, they examined the association between preprocedure (baseline) patient health status, as assessed by the KCCQ, and 1-year mortality after TAVR.  Median age for the analytic cohort was 84 years and 47% were men.  Median STS risk score was 7.0%.

For the overall results, compared with those with good health status before TAVR and after adjusting for a broad range of baseline covariates, patients with very poor health status had a 2-fold increased hazard of death over the first year after TAVR (adjusted HR 2.00, 95% CI 1.58–2.54), whereas those with poor and fair health status had intermediate outcomes (adjusted HR 1.54, 95% CI 1.22–1.95 and adjusted HR 1.20, 95% CI 0.94–1.55, respectively). The authors concluded that for all surgical risk categories in the TVT registry, "in a national, contemporary practice cohort, worse preprocedure patient health status, as assessed by the KCCQ, was associated with greater long-term mortality after TAVR."

Arnold SV, Spertus JA, Vemulapalli S, et al. Quality-of-Life Outcomes After Transcatheter Aortic Valve Replacement in an Unselected Population: A Report from the STS/ACC TVT Registry. JAMA Cardiol. 2017;2(4):409-416.

The aim of this study was to assess whether improvements in symptoms and quality of life demonstrated in TAVR clinical trials was generalizable to a broader, unselected population. This is important because "in the TVT registry, unselected patients treated with TAVR between 2011 and 2013 experienced a 1-year survival rate of 76%, similar to that observed in the pivotal trials. In this elderly population with extensive comorbidity and impaired health status, however, it is unlikely that prolonged survival alone (without improved health status) would be viewed as a desirable outcome."

This study used data from the STS/ACC TVT registry for patients who underwent TAVR between November 2011 and March 2016.

The primary outcome was health status assessed at 30 days and 1 year after TAVR with the Kansas City Cardiomyopathy Questionnaire-overall summary score (KCCQ-OS). The KCCQ-OS is a shortened, 12-item version of the KCCQ, a "patient-reported disease-specific health status survey developed to describe and monitor symptoms, functional status, and QOL in patients with heart failure." The KCCQ-OS combines 4 domains related to valvular heart disease: physical limitation, symptom frequency, quality of life, and social limitation (range 0–100 points, with higher scores indicating less symptom burden and better QOL).

"The KCCQ-OS was categorized as very poor (KCCQ-OS <25), poor (KCCQ-OS 25–49), fair (KCCQ-OS 50–74), and good QOL (KCCQ-OS ≥75). Changes in the KCCQ-OS of 5, 10, and 20 points correspond to small, moderate or large clinical improvements, respectively. In order to integrate QOL outcomes with survival, a favorable outcome at 1-year after TAVR was defined as survival with a reasonable QOL (KCCQ-OS score ≥60, roughly equivalent to New York Heart Association class I–II symptoms) without any meaningful worsening (decrease of ≥ 10 points in the KCCQ-OS score from baseline to 1 year)."

The analysis method evaluated changes in baseline KCCQ scores at 30-days and 1-year using paired t-tests. "Mean KCCQ-OS scores at 1-year were compared among key subgroups using ANCOVA. These comparisons were adjusted for baseline KCCQ-OS scores except for the analysis that was stratified by baseline KCCQ-OS scores. Rates of favorable outcome at 1-year were estimated for each subgroup and compared using chi-square tests. Factors associated with health status at 1 year after TAVR were determined using multivariable linear regression with generalized estimating equations to account for clustering of patients within sites."  Non-linearity and two-way interactions were further explored.

Patients included in the 30-day cohort totaled 31,636 from 406 hospitals who survived 30-days and completed the KCCQ at baseline and follow- up. The 1-year cohort included 7,014 patients from 169 hospitals who survived 1 year and completed the KCCQ at both baseline and follow-up (after excluding patients from 179 sites with <50% KCCQ completion rates).

Important patient demographics included, in the 30-day cohort (n=31,636), median age of 83 years 52% male, and 95% white; and in the 1-year cohort (n=7,014), median age of 84 years 51% male, and 95% white.

The study investigators found that "mean baseline KCCQ-OS was 42.3, indicating substantial health status impairment. Surviving patients had, on average, large improvements in health status at 30 days that persisted to 1 year, with a mean improvement in the KCCQ-OS of 27.6 points at 30 days and 31.9 points at 1 year." Predictors of worse health status at 1 year included worse baseline health status, older age, higher ejection fraction, and multiple other factors. "Overall, 62.3% of patients had a favorable outcome at 1 year (alive with reasonable quality of life [KCCQ-OS ≥60] and no significant decline [≥10 points] from baseline) with the lowest rates seen among patients with severe lung disease (51.4%), on dialysis (47.7%), or with very poor baseline health status (49.2%)."

The investigators concluded that patient health status improved substantially after TAVR on average, similar to improvements demonstrated in pivotal clinical trials. However, they noted that about "1 in 3 patients still had a poor outcome at 1 year after TAVR, half of which was due to death and half due to poor QOL," and thus they emphasized the need for continued improvements in patient selection, procedural technique, and post-procedural care.

Arsalan M, Szerlip M, Vemulapalli S, et al. Should Transcatheter Aortic Valve Replacement Be Performed in Nonagenarians? Insights From the STS/ACC TVT Registry. JACC. 2016;67(10).

The aim of this study was to compare the outcomes of nonagenarians (≥90 years of age) to younger patients (<90 years) undergoing TAVR. Whether TAVR benefits the very elderly is unknown as they are underrepresented in pivotal clinical trials. The concern is that they "might not survive the procedure as frequently, recover from the procedure as quickly, nor experience an improved functional outcome and quality of life."

This study used data from the STS/ACC TVT registry for patients who underwent TAVR between November 2011 and September 2014.

The analysis method compared in-hospital outcomes of patients ≥ 90 and < 90 years of age using the Pearson chi-square test for categorical variables and the Wilcoxon rank sum test for continuous variables. Cumulative incidences of death and nonfatal outcomes at 30 days and 1 year post TAVR were estimated for patients ≥ 90 and < 90 years of age. The 30-day observed to expected mortality ratios were calculated based on the baseline STS risk score. Unadjusted and adjusted effects of age on 30-day and 1-year mortality were assessed using Cox proportional hazards models. Multivariable models included covariates from a recent TVT model for in-hospital mortality (O’Brien 2015).

Primary outcomes were death, stroke, rehospitalization due to heart failure, aortic valve reintervention, MI and QOL at 30 days and 1 year. QOL was assessed with the 12-item Kansas City Cardiomyopathy Questionnaire (KCCQ-12, a shorter version of the full KCCQ).

Patients included in this study were all those who underwent TAVR between November 2011 and September 2014. There was a total of 24,025 patients, of which 3,773 (15.7%) were nonagenarians, from 329 participating hospitals.

Important patient demographics included median age of 92 years (nonagenarians) and 82 years (younger cohort). "Compared to patients under age 90, nonagenarians were more likely to be female and less likely to have high-risk features including prior non-aortic valve cardiac surgery procedure, diabetes, prior stroke, and prior myocardial infarction (MI), but, overall, had higher estimated surgical mortality (STS PROM scores, ≥90 versus <90: 9.2% versus 6.3%, P<0.001)."

The study investigators found that the 30-day and 1-year mortality was significantly higher among nonagenarians (≥90 versus < 90: 30-day, 8.8% versus 5.9%, P<0.001; 1-year, 24.8% versus 22.0%, P<0.001, absolute risk 2.8%, relative risk 12.7%). However, nonagenarians had a higher mean STS risk score (10.9% versus 8.1%, P<0.001) and therefore had similar ratios of observed to expected rates of 30-day death (≥90 versus < 90: 0.81, 95% CI 0.70–0.92 versus 0.72, 95% CI 0.67–0.78).  "There were no differences in the rates of stroke, aortic valve reintervention or myocardial infarction at 30-days or 1-year. Nonagenarians had lower (worse) median KCCQ-12 scores at 30-days; however, there was no significant difference at 1-year."

The investigators concluded that while nonagenarians have an increased, although clinically modest, risk for morbidity and mortality after TAVR simply based on their age, many experience prolonged survival and improved QOL.  Good patient selection is thus particularly important in the very elderly.

Baron SJ, Arnold SV, Herrmann HC, et al. Impact of Ejection Fraction and Aortic Valve Gradient on Outcomes of Transcatheter Aortic Valve Replacement. JACC. 2016;67(20):2349-2358.

The aim of this study was "to evaluate the impact of reduced left ventricular ejection fraction (LVEF) and low aortic valve gradient (AVG) on clinical outcomes after TAVR, and to determine whether the effect of AVG on outcomes is modified by LVEF." This is important because an association between these risks factors and long-term outcomes has been supported by prior studies, but the extent of their impact, and their possible interaction, is unclear.

This study used data from the STS/ACC TVT registry for patients who underwent TAVR between November 2011 and June 2014, and were linked to Medicare claims (for30-day and 1-year outcomes).

Patient outcomes included multiple in-hospital and 1-year outcomes, with no primary outcome specified. In-hospital outcomes included death, MI, stroke, new requirement for dialysis, and length of hospital stay. Clinical outcomes at 1 year included death, MI, stroke, and hospitalization for recurrent heart failure. Patient reported outcomes (for symptoms, functioning, and quality of life) used the KCCQ-OS (the short form) at baseline and 30-day follow-up.

The analysis method stratified the study cohort by LVEF and then AVG. "First, the cohort was divided into three groups according to LVEF using clinically relevant cut-points: Severe LV dysfunction (LVEF < 30%); Mild/Moderate LV Dysfunction (LVEF 30–50%); and preserved LV function (LVEF > 50%). Next, the cohort was divided into two groups according to mean AVG as assessed by pre-procedure echocardiography: Low AVG (mean AVG < 40mmHg); and High AVG (mean AVG ≥ 40 mmHg)." Unadjusted and adjusted analyses used the Kruskall-Wallis test, and Kaplan-Meier, Cox proportional hazards, generalized estimating equation, and Fine and Gray methods.

Patients included a total of 11,292 who underwent TAVR, after exclusions for patients with aborted procedures or missing data for LVEF or AVG.

Important patient demographics revealed significant differences in age, sex, and baseline clinical characteristics when categorized by LVEF and mean AVG (Tables 1a and 1b).

The study investigators found that "over the first year of follow-up after TAVR, patients with LV dysfunction and low AVG had higher rates of death and recurrent heart failure. After adjustment for other clinical factors, only low AVG was associated with higher mortality (HR 1.21, 95% CI 1.11–1.32, P < 0.001) and higher rates of heart failure (HR 1.52, 95% CI 1.36–1.69, P < 0.001), whereas the effect of LVEF was no longer significant.  There was no evidence of effect modification between AVG and LVEF with respect to either endpoint."  "At 30 day follow-up, health status as measured by the KCCQ-OS improved across all levels of LVEF; however, the absolute change in KCCQ-OS scores was greatest for patients with severe LV dysfunction and least for patients with normal LV function at baseline. In contrast, there were no significant differences between the low and high AVG group for any of the 30-day health status outcomes."

The investigators concluded that low AVG, but not LV dysfunction, was associated with higher rates of mortality and recurrent heart failure post-TAVR. While patients with severe aortic stenosis and low AVG (< 40 mmHg) may derive less long-term benefit from TAVR, the investigators believed that neither LV dysfunction nor low AVG alone should preclude consideration for TAVR in the absence of other indicators of poor prognosis.

Brennan JM, Holmes DR, Sherwood MW, et al. The association of transcatheter aortic valve replacement availability and hospital aortic valve replacement volume and mortality in the United States. Annals of Thoracic Surgery. 2014;98(6):2016-22.

The aim of this study was to assess whether the introduction of TAVR has affected hospitals’ SAVR and overall AVR case volumes and outcomes.  A central question the study sought to answer was: Would overuse of the TAVR procedure significantly reduce SAVR volumes?

This study used patient data from the STS adult cardiac surgery database (ACSD) and the TVT Registry to examine SAVR and TAVR procedures submitted between January 2008 and June 2013. Temporal trends in total case volume (SAVR plus TAVR), and observed and risk-adjusted in-hospital mortality rates were assessed among low-risk cases (STS risk score < 4%), intermediate-risk cases (4% to 8%), and high-risk cases (> 8%). A contemporary control was provided by non- TAVR centers.

The analysis method used the Wald test to compare observed to expected (O:E) ratios of risk-adjusted in-hospital mortality rates across three time intervals: (1) early premarket (quarter [Q]1 2008 to Q4 2009); (2) late premarket (Q1 2010 to Q3 2011); and (3) early postmarket (Q4 2011 to Q2 2013). Expected mortality was determined using STS risk scores. Trends in mortality rates were evaluated for the overall cohort and within subgroups determined by STS risk levels, date of procedure, and procedure type. Centers were further classified as "new" or "established," consistent with the 2012 Consensus statement and CMS NCD, based on whether they reported TAVR cases to the STS ACSD before or after FDA device approval in November 2011.

Patients included in this study were all those who underwent SAVR or TAVR from January 2008 to June 2013. A total of 215,767 SAVR and 11,436 TAVR procedures were evaluated from 801 sites (246 TAVR centers, 555 non-TAVR centers); 45 cases where then excluded as STS risk could not be calculated because of missing data.

Important patient demographics (sample size n=227,158) included mean age 73 ± 7 years and 63% men. A total of 149,307 AVR cases/patients (66%) were low clinical risk, 50,571 (22%) were intermediate risk, and 27,280 (12%) were high risk.

The study investigators found that "the total annual volume of AVR among 246 TAVR-performing hospitals increased from 19,578 to 33,004, with a 22% growth in SAVR volumes; non-TAVR hospital (n = 555) increases were more modest (16% growth). Expanded volumes at TAVR hospitals included increased SAVR use in low- and intermediate-risk cases, and TAVR use in high-risk cases. In parallel, in- hospital mortality for all AVR procedures at TAVR sites declined from 3.4% to 2.9% (O:E ratio 0.75 to 0.58, P < 0.001); the greatest declines were among intermediate- and high-risk SAVR patients. Owing to reduced SAVR mortality, TAVR centers experienced a significantly greater decline in O:E ratio for high-risk patient in-hospital mortality than non-TAVR centers (TAVR center O:E ratio, 0.81 to 0.61; non-TAVR center O:E ratio, 0.85 to 0.76; P < 0.001). After approval of TAVR for clinical use, a trend toward higher in-hospital mortality rates and O:E ratios for TAVR procedures was observed at new (but not at established) TAVR centers (O:E ratio, 0.41 to 0.67, P = 0.08)."

The investigators concluded that as TAVR has become available nationally, SAVR volumes have risen and overall AVR procedural mortality has declined, particularly among TAVR centers and high-risk SAVR patients. They stated that the increase in overall AVR volume could have been the result of increased: (a) diagnosis and referral of high-risk patients with symptomatic AV stenosis; (b) treatment of lower risk patients as result of the introduction of TAVR; or (c) inclusion of high-risk patients who may or may not derive therapeutic benefit.

Brennan JM, Thomas L, Cohen DJ, et al. Transcatheter Versus Surgical Aortic Valve Replacement. JACC. 2017;70(4).

The aim of this study was "to determine the safety and effectiveness of TAVR versus SAVR, particularly in intermediate- and high-risk patients, in a nationally representative real-world cohort." This would help assess the generalizability of RCTs that supported use of TAVR.

This study used data from the TVT registry for TAVR cases performed between January 2014 and September 2015, and from the STS National Database for SAVR cases performed between July 1, 2011 and December 31, 2013, that were linked to Medicare claims (for 1-year outcomes).

The analysis method used propensity scoring (defined as the probability calculated by logistic regression of receiving the treatment, here TAVR, given measured covariates derived by clinical input) in order to match TAVR and SAVR patient characteristics.  The goal was to allow outcomes of unselected, non-randomized TAVR and SAVR patients to be fairly compared using standard Cox proportional hazard models.

Primary outcomes included death, stroke, and days alive and out of hospital (DAOH) to 1 year, and discharge to home. Subgroup analyses was based on surgical risk, demographics, and comorbidities.

Patients included in the study were 17,910 TAVR and 22,618 SAVR patients who were available for propensity matching. Patients had severe, symptomatic aortic valve stenosis with intermediate or high surgical risk, underwent treatment with TAVR or SAVR in the U.S., and were considered eligible for either treatment. Key exclusions were: (1) patient characteristics that were thought to strongly favor one treatment or another; (2) patient who underwent subsequent aortic valve replacement during admission; and (3) hospitals submitting < 10 total SAVR or TAVR records during the study interval.

Important patient demographics after propensity matching included 4,732 SAVR and 4,732 TAVR patients, median age of 82 years, 48% women, and a median STS risk score of 5.6%.

The investigators found that "TAVR and SAVR patients experienced no difference in 1-year rates of death (17.3% vs. 17.9%; hazard ratio [HR] 0.93, 95% confidence interval [CI] 0.83–1.04) and stroke (4.2% vs. 3.3%; HR 1.18, CI 0.95–1.47), and no difference was observed in the proportion of DAOH to 1 year (rate ratio 1.00, CI 0.98–1.02). However, TAVR patients were more likely to be discharged home after treatment (69.9% vs. 41.2%; odds ratio 3.19, CI 2.84–3.58). Results were consistent across most subgroups, including among intermediate- and high-risk patients."

The investigators concluded that TAVR and SAVR, when performed in broad U.S clinical practice in a representative group of older, intermediate- and high-risk patients, resulted in similar rates of death, stroke, and DAOH to 1 year. "TAVR patients were more often discharged directly to home, reflecting a less demanding post-operative recovery. Results were consistent across most patient subgroups and across the spectrum of intermediate to high pre-operative surgical risk." This propensity matched cohort study thus supports that the positive findings of TAVR randomized trials are generalizable to wider clinical practice in real-world patients.

Carroll J, Vemulapalli S, Dai D et al. Procedural experience for transcatheter aortic valve replacement and relation to outcomes: The STS/ACC TVT registry. J Am Coll Cardiol. 2017.

The aim of this study was to assess the degree to which increasing experience during the introduction of the TAVR procedure, separated from other outcome determinants including patient and procedural characteristics, is associated with outcomes.

This study used data from TAVR patient cases submitted to the TVT Registry from November 2011 through November 2015. Continuing the training and practice patterns begun in early TAVR trials (such as PARTNER-I), there was general dissemination of knowledge throughout U.S. clinical practice through proctoring, with early operators (TAVR proceduralists) and institutions training subsequent operators and institutions. Additionally, device training and comprehensive support during procedures was provided by medical device companies.

Key study outcomes were designed to evaluate the association between TAVR procedural experience and important patient outcomes. Procedural experience was measured using cumulative hospital volume. "Site volume was chosen rather than operator volume given the combined multi-operator approach to TAVR performance involving both cardiology and cardiovascular surgery specialists." Patient outcomes were in-hospital risk-unadjusted and risk-adjusted outcomes: death, vascular complications, bleeding complications, and stroke, using standardized definitions including from the Valve Academic Research Consortium. In-hospital outcomes were assessed because of the large numbers and completeness of data available in the TVT registry. The study did not include 30-day and 1-year patient outcomes.

The analysis method used a case sequence approach rather than stratification of hospitals according to their cumulative case volume. This approach was adopted as many hospitals had small volumes, making a hospital-based analysis statistically challenging. Thus unadjusted and risk-adjusted outcomes were assessed as a function of an increasing number of procedures performed. The analysis used generalized linear and non-linear mixed-effect models, with a "three-level (patients, operators, and hospitals) hierarchical structure to account for inter-hospital variability, inter-operator variability nested within sites, intra-site clustering of TAVR volume, and patient case mix." This produced an average event rate for each consecutive case at an average site for a hypothetical "average" patient. The investigators sought 1) to describe both the learning curve associated with TAVR and any stable volume–outcome relationship; and 2) to include low-volume centers in the analysis without penalizing them for being within their learning curve.

Patients included in this study were all those who underwent TAVR and had data submitted to the TVT Registry from November 2011 through November 2015. There were a total of 42,988 procedures/patients from 395 sites and 1,915 individual operators.

Important patient demographics (sample size n=42,988) included mean age 83 years, 51% men, and 94% white. These were elderly patients with severe aortic stenosis and typically multiple comorbidities who were inoperable or high risk for SAVR.

The study investigators found that "increasing site volume was associated with lower in-hospital risk-adjusted outcomes, including mortality (p < 0.02), vascular complications (p < 0.003), and bleeding (p < 0.001) but was not associated with stroke (p = 0.14). From the first case to the 400th case in the volume–outcome model, risk-adjusted adverse outcomes declined, including mortality (3.57% to 2.15%), bleeding (9.56% to 5.08%), vascular complications (6.11% to 4.20%), and stroke (2.03% to 1.66%). Vascular and bleeding volume–outcome associations were nonlinear with a higher risk of adverse outcomes in the first 100 cases. An association of procedure volume with risk-adjusted outcomes was also seen in the subgroup having transfemoral access."

The investigators concluded that patient outcomes improved with increasing TAVR procedure experience. "After adjustment for patient factors, date of procedure, and specific procedural characteristics (including device iterations), an inverse association persisted between increasing case volume and lower in-hospital mortality, vascular complications, and bleeding." This association, whether deemed a prolonged learning curve or a manifestation of a volume–outcome relationship, suggested that concentrating experience in higher volume heart valve centers might be a means of improving outcomes." Further, "this association was most pronounced during the first 100 cases, indicating the effect of an early learning curve for TAVR."  Carroll and colleagues further state that "the number of sites needed in the United States to balance access and quality of TAVR outcomes cannot be definitively determined from the present study."

Chandrasekhar J, Dangas G, Yu J, et al. Sex-Based Differences in Outcomes with Transcatheter Aortic Valve Therapy. JACC. 2016;68 (25);2733-44.

The aim of this study was "to compare the in-hospital and 1-year outcomes in male and female subjects from the U.S. nationwide TAVR registry." This is important because subgroup analysis of randomized trials and small observational studies support an association of gender and outcomes after TAVR.

This study used data from the TVT Registry for patients who underwent TAVR between November 2011 and September 2014, and were linked to Medicare claims (for 1-year outcomes).

Patient outcomes included multiple in-hospital and 1-year outcomes, with no primary outcome specified. In-hospital outcomes included all- cause death, MI, stroke, major bleeding, and major vascular complications. One-year outcomes included time-to-event occurrence of death, MI, stroke, and clinically significant bleeding.

The analysis method compared outcomes between the two gender groups. In-hospital outcomes were assessed using logistic regression with generalized estimating equations to account for within-center clustering. One-year outcomes were assessed with the Kaplan-Meier method, or with cumulative incidence methods for nonfatal events. The Cox proportional hazards model was used for mortality; and Fine and Gray’s method for non-fatal events, with death as a competing risk. Covariates used to adjust outcomes in Cox models derived from previous TVT Registry studies.

Patients included a total of 23,652 who underwent TAVR in U.S. hospitals.  Of these, 11,808 (49.9%) were female and 11,844 (51.1%) male.

Important patient demographics included, for the female group (n=11,808), mean age of 82 years and 93% white; and for the male group (n=11,844), mean age of 82 years and 95% white. The female group "was older, with a lower prevalence of coronary artery disease, atrial fibrillation, and diabetes but a higher rate of porcelain aorta, lower glomerular filtration rate, and higher mean STS score (9.0% vs. 8.0%; all p < 0.001)."

The study investigators found that women were treated more often by using non-transfemoral access than men (45.0% vs. 34.0%). "Despite using smaller device sizes, women achieved valve cover index 8% more often than men (66% vs. 54%).  In-hospital vascular complications were higher in women (8.27% vs. 4.39%; adjusted hazard ratio [HR]: 1.70; 95% CI: 1.34 to 2.14; p<0.001) and a trend toward higher bleeding (8.01% vs 5.96%; adjusted HR: 1.19; 95% CI: 0.99 to 1.44; p = 0.06) was observed; however, 1-year mortality was lower (21.3% vs. 24.5%; adjusted HR: 0.73; 95% CI: 0.63 to 0.85; p < 0.001) in women than in men."

The investigators concluded that "female patients undergoing TAVR had a different risk profile compared with male patients. Notwithstanding a greater adjusted risk for in-hospital vascular complications, 1-year adjusted survival was superior in female patients."

Dodson JA, Williams MR, Cohen DJ, et al. Home Practice of Direct-Home Discharge and 30-Day Readmission After Transcatheter Aortic Valve Replacement in the Society of Thoracic Surgeons/American College of Cardiology Transcatheter Valve Therapy (STS/ACC TVT) Registry. JAHA. 2017;6(8).

The aim of this study was to evaluate the association of hospital readmissions and a hospital’s practice of discharging to home versus to skilled nursing facilities (SNFs) after TAVR. This is important because "nearly 17% of patients are readmitted within 30 days of discharge after TAVR," and where patients are discharged to may impact the likelihood of hospital readmission.

This study used data from the TVT Registry for patients who underwent TAVR between November 2011 and March 2015, and were linked to Medicare claims (for 30-day outcomes).

The primary outcome was hospital rates of 30-day readmission. A secondary outcome was hospital rates of 30-day mortality.

The analysis method was performed at the hospital level, with hospitals grouped into quartiles based on the frequency of direct-home discharge. Differences in patients, hospitals, treatment and regions were assessed using the Kruskal-Wallis test for continuous variables or chi-squared test for categorical variables. Kruskal-Wallis was used to evaluate the association between direct-home discharge and the outcomes of 30-day readmissions and 30-day mortality. Hierarchical logistic regression fixed-effects models were then used to assess the association of discharge location and 30-day readmission. Model covariates included "clinically plausible patient- or hospital-level characteristics" that might influence 30-day readmission.

Patients included a total of 18,568 who underwent TAVR at 329 hospitals, after multiple exclusions (for patients who died in hospital; were discharged to hospice, another acute care hospital, or a permanent nursing home; had missing data; or underwent TAVR at a hospital that performed < 5 TAVR cases).

Important patient demographics included, for the overall study, median age of 84 years, 51% male, and 95% white.

The investigators found that 69% of patients overall were discharged to home after TAVR. There were patient, treatment, hospital, and regional- level differences in characteristics among the hospital quartiles. "Hospitals in the highest quartile of direct home discharge (Q4) compared with hospitals in the lowest (Q1) were more likely to use femoral access (75.2% versus 60.1%, P<0.001), had fewer patients receiving transfusion (26.4% versus 40.9%, P<0.001), and were more likely to be located in the Southern United States (48.8% versus 18.3%, P<0.001). Median 30- day readmission rate was 17.9%. There was no significant difference in 30-day readmissions among quartiles (P=0.14), even after multivariable adjustment (odds ratio Q4 versus Q1=0.89, 95%CI 0.76-1.04; P=0.15). Factors most strongly associated with 30-day readmission were glomerular filtration rate, in-hospital stroke or transient ischemic attack, and non-femoral access."

The investigators concluded that there was no significant association between hospital practice of discharge to home versus SNF post-TAVR and 30-day hospital readmission. They stated that "further research is necessary to understand reasons for the significant regional variations in the practice of direct-home discharge as well as what proportion of readmissions are preventable."

Edwards FH, Cohen DJ, O’Brien SM, et al. Development and Validation of a Risk Prediction Model for In-Hospital Mortality After Transcatheter Aortic Valve Replacement. JAMA Cardiol. 2016;1(1):46-52.

The aim of this study was "to use a national population of patients undergoing TAVR to develop a statistical model that will predict in-hospital mortality after TAVR." Accurate risk prediction models exist for SAVR and should be developed for TAVR to allow comparison of risks for eligible patients choosing between these two procedures.

This study used data from the TVT Registry for patients who underwent TAVR between November 2011 and February 2014 for model development, and between March and October 2014 for model validation.

The analysis method used logistic regression to estimate the association between in-hospital mortality and baseline covariates.  An extensive list of possible covariates were initially selected by both expert opinion and statistical analysis. The final set of predictors were selected by stepwise pruning of this initial covariate list. Calibration involved plotting observed versus expected (O:E) mortality rates within prespecified subgroups and across quintiles of predicted risk among patients in the validation sample.

Patients included in the model development sample were 13,718 patients from 265 sites, of which 13,672 had sufficient data available. The final validation cohort included 6,868 patients from 314 sites.

Important patient demographics in the model development sample included 49% men and mean age 82 years; and in the validation sample, 52% men and mean age 82 years.

The study investigators found that the final model covariates – the best predictors of in-hospital mortality – (reported as ORs; 95%CIs) were: "age (1.13; 1.06-1.20), glomerular filtration rate per 5-U increments (0.93; 0.91-0.95), hemodialysis (3.25; 2.42-4.37), New York Heart

Association functional class IV (1.25; 1.03-1.52), severe chronic lung disease (1.67; 1.35-2.05), non-femoral access site (1.96; 1.65- 2.33), and procedural acuity categories 2 (1.57; 1.20-2.05), 3 (2.70; 2.05-3.55), and 4 (3.34; 1.59-7.02). Procedural acuity, also called "operative priority," is the patient’s "pre-procedure clinical state that determines the urgency of the procedure." Calibration analysis demonstrated no significant difference between the model (predicted vs observed) calibration line (−0.18 and 0.97 for intercept and slope, respectively) compared with the ideal calibration line."

The investigators concluded that this validated risk prediction model based on TVT Registry patient data "should be a valuable adjunct for patient counseling, local quality improvement, and national monitoring for appropriateness of selection of patients for TAVR."

Fadahunsi OA, Olowoyeye A, Ukaigwe A, et al. Incidence, Predictors, and Outcomes of Permanent Pacemaker Implantation Following Transcatheter Aortic Valve Replacement: Analysis from STS/ACC TVT Registry. JACC Cardiovasc Interv. 2016;9(21):2189-2199.

The aim of this study was "to evaluate the incidence, predictors, and clinical outcomes of permanent pacemaker (PPM) implantation following TAVR." This is important because "conduction abnormalities leading to PPM implantation are common complications following TAVR," and the predictors for and outcomes of PPM implantation are unclear.

This study used data from the TVT Registry for patients who underwent TAVR with a balloon-expandable Edwards SAPIEN valve (ESV) or self-expanding Medtronic CoreValve Revalving System (MCRS) between November 2011 and September 2014, and were linked to Medicare claims (for 30-day and 1-year outcomes).

Patient outcomes included multiple in-hospital, 30-day, and 1-year outcomes, with no primary outcome specified. Thirty-day and 1-year outcomes included mortality, heart failure admission, a composite of mortality or heart failure admission, and stroke or myocardial infarction.

The analysis method stratified the incidence of PPM implantation by valve type, access site, and procedural risk classification, and compared subgroups. Multivariate logistic regression was used to identify predictors of PPM implantation post-TAVR. Cumulative incidence methods were used to compare 30-day and 1-year outcomes between the PPM and no-PPM groups, treating death as a competing risk for non-fatal outcomes. "Unadjusted and adjusted associations of PPM implantation with 30-day and 1-year outcomes were assessed using Cox proportional hazards models for mortality and a composite of mortality or heart failure admission, and Fine and Gray’s proportional sub-distribution hazards models for nonfatal outcomes." Covariates derived from the TVT mortality risk model and baseline characteristics.

Patients included a total of 9,785 who underwent TAVR at 229 U.S. hospitals, and met study criteria. Multiple exclusions included prior implantation of a pacemaker or implantable cardioverter-defibrillator, intra-procedural pacemaker implantation, unsuccessful procedures, conversion to open procedures (the study did not take an "intention-to-treat" approach).

Important patient demographics included, for the PPM group (n=651), median age of 84 years and 52% male; and for the no- PPM group (n=9,134), median age of 84 years and 47% male. The PPM group was more likely to be male, and had a higher predicted risk of mortality (by STS risk score).

The study investigators found that PPM implantation was required within 30 days of TAVR in 6.7% of all patients (25.1% in those receiving self-expanding valves versus 4.3% for balloon-expanding valves). Predictors of PPM implantation were: age (per 5-year increment, OR 1.07, 95% CI 1.01 - 1.15), prior conduction defect (OR 1.93, 95% CI 1.63 to 2.29), and use of self-expanding valve (OR 7.56, 95% CI 5.98 to 9.56). PPM implantation was associated with increased 1-year mortality (24.1% versus 19.6%; HR 1.31, 95% CI 1.09 - 1.58), and a composite of mortality or heart failure admission at 1 year but not with heart failure admission alone.

The investigators concluded that: "early PPM implantation is a common complication following TAVR" (6.7% overall, 25% for the self- expanding CoreValve), and "is associated with higher mortality and a composite of mortality or heart failure admission at 1 year."

Fanaroff AC, Manandhar P, Holmes DR, et al. Peripheral Artery Disease and Transcatheter Aortic Valve Replacement Outcomes. Circ Interv. 2017;10(10).

The aim of this study was to determine the prevalence of and outcomes associated with peripheral artery disease (PAD) in patients undergoing TAVR. This is important because: PAD is associated with increased cardiovascular mortality; PAD risk factors overlap with those for aortic stenosis; and significant PAD may contradict the preferred transfemoral TAVR approach.

This study used data from the TVT Registry for patients who underwent TAVR between November 2011 and September 2015, and were linked to Medicare claims (for 30-day and 1-year outcomes).

Study outcomes included multiple in-hospital and 1-year events with no primary outcome identified. In-hospital outcomes were all-cause mortality, stroke, MI, bleeding, and major vascular complications. One-year outcomes were all-cause mortality, all- cause readmission, stroke, and MI.

The analysis method separated patients that underwent TF access from those that underwent non-TF access (due to "marked differences in patient characteristics that resulted in the clinical decision to select one or the other access approach"). Cumulative incidence curves were constructed and unadjusted and adjusted analysis performed. Cox proportional hazards models were used to compare the risk- adjusted 1-year hazard ratio of all-cause death for patients with and without baseline PAD. Fine and Gray's proportional sub-distribution hazards model, which treats death as a competing risk, was used to compare the risk-adjusted 1-year hazards ratios of readmission, stroke, MI, and bleeding for patients with and without PAD.  Covariates for risk adjustment included all those in the TVT Registry in-hospital mortality risk score, along with other factors deemed clinically important by expert consensus.

Patients included a total of 27,440 treated with TAVR at 389 U.S. hospitals. Of these, 19,660 patients had TF access and 7,780 had non-TF access.

Important patient demographics revealed that in both access groups, the cohort with PAD was younger, more likely to be male, and generally sicker, including higher prevalence of coronary artery and cerebrovascular disease. (No overall study statistics were presented; rather these were broken down by access approach and further by presence or absence of PAD; Table 1.)

The study investigators found that "nearly 1 in 4 patients undergoing TAVR via TF access, and nearly half of patients undergoing TAVR via non-TF access, have PAD." At 1-year follow-up, "patients with PAD undergoing TAVR via TF access had a higher incidence of death (16.8 vs. 14.4%; adjusted HR 1.14, p = 0.01), readmission (45.5 vs, 42.1%; HR 1.11, p < 0.001), and bleeding (23.1 vs. 19.7%; HR 1.18, p < 0.001) compared with patients without PAD. Patients with PAD undergoing TAVR via non-TF access did not have significantly higher rates of 1-year mortality or readmission compared with patients without PAD."

The investigators concluded that "PAD is common among patients undergoing commercial TAVR via TF and non-TF access. Among patients undergoing TF TAVR, PAD is associated with a higher incidence of 1- year adverse outcomes compared with absence of PAD."

Grover FL, Vemulapalli S, Carroll JD, et al. 2016 Annual report of The Society of Thoracic Surgeons/American College of Cardiology Transcatheter Valve Therapy Registry. J Am Coll Cardiol. 2017 Mar 14;69(10):1215-1230.

The aim of this study was to focus on patient characteristics, trends, and outcomes of transcatheter aortic and mitral valve catheter-based valve procedures in the United States using the TVT Registry. Data for all patients receiving commercially approved devices from 2012 through December 31, 2015 was obtained from the TVT Registry.

For the overall results, the 54,782 patients with TAVR by the end of 2015 demonstrated decreases in expected risk of 30-day operative mortality STS PROM score of 7% to 6% and TAVR PROM score of 4% to 3% (both P < 0.0001) from 2012 to 2015. The median age decreased from 84 years in 2012 to 83 years in 2015 (P < 0.0001). The percent male increased from 52.6% in 2012 to 52.7% in 2015 (P < 0.0001) and the percent white decreased from 94.3% in 2012 to 94.1% in 2015 (P = 0.343.  Observed in-hospital mortality decreased from 5.7% in 2012 to 2.9% in 2015 (P < 0.0001), the 30-day mortality has decreased from 7.5% in 2012 to 4.6% in 2015 (P < 0.0001), and 1-year mortality decreased from 25.8% in 2012 to 21.6% in 2014 (P < 0.0001). Post-operative atrial fibrillation decreased over time from 6.9% in 2012 to 2013 to 3.7% in 2015, but conversely, 30-day new pacemaker insertion was 11.8% overall increasing from 8.8% in 2013 to 12% in 2015 (P < 0.0001).  Thirty-day aortic valve re-intervention changed from 0.4% in 2012 to 0.3% in 2015 (P = 0.627). The authors concluded that "the TVT Registry is an innovative registry that that monitors quality, patient safety and trends for these rapidly evolving new technologies."

Hansen, JW, Foy, A, Yadav, P, et al. Death and Dialysis After Transcatheter Aortic Valve Replacement. An Analysis of the STS/ACC TVT Registry. JACC Cardiovasc Interv. Sept. 2017.

The aim of this study was to determine the incidence of renal replacement therapy (RRT, which includes hemodialysis and peritoneal dialysis) after TAVR. This is important because "RRT may be an unacceptable outcome to some patients," yet the risk is unknown due to "wide discrepancy in the reported rate of RRT after TAVR (1.4% to 40%)." Studies to date have focused on acute kidney injury only, and have shown that "pre-procedural glomerular filtration rate (GFR) is independently predictive of AKI leading to death after TAVR."

This study used data from the TVT Registry for patients who underwent TAVR between November 2011 and September 2015, and were linked to Medicare claims (for 30-day and 1-year outcomes).

The primary outcomes were all-cause mortality, a new requirement of RRT, or a composite of both, at intervals of 30 days and 1 year after TAVR.

The analysis method grouped patient cohorts by chronic kidney disease (CKD) stage. CKD is the gradual loss of kidney function over time. CKD has 5 stages, defined by glomerular filtration rate (GFR): "CKD stage 1 is GFR >90 ml/min/m², stage 2 is GFR 60 to 89 ml/min/m², stage 3 is GFR of 30 to 59ml/min/m², stage 4 is GFR of 15 to 29 ml/min/m², stage 5 is GFR <15 ml/min/m². CKD is typically not clinically evident until GFR falls below 60 ml/min/m²." Patients with CKD stage 1 and 2 were combined and served as a control group. Primary outcomes were assessed as a function of pre-procedural GFR, both by CKD stage as well as on a continuous scale. Tests for non-linear relationships and Cox proportional hazards models were used.

Patients included a total of 44,778 treated with TAVR (CKD stages 1 and 2 combined, with 22,893 patients [51.13%]; stage 3 with 19,266 [43.03%]; stage 4 with 2,413 [5.39%]; and stage 5 with 206 [0.46%]).

Important patient demographics included, in the overall sample (n=44,778), mean age of 82 years, 51% male, and 95% white. Patients were generally sicker the greater the CKD stage.

The study investigators found that "in both unadjusted and adjusted analysis, pre-procedural GFR was associated with the outcomes of death and new RRT after TAVR" when GFR is<60 ml/min/m², and "increases significantly when GFR falls below 30 ml/min/m². Incremental increases in GFR of 5 ml/min/m² were statistically significant (unadjusted hazard ratio: 0.71; p<0.001) at 30 days, and continued to be significant at 1 year when pre-procedure GFR was<60ml/min/m². One in 3 CKD stage 4 patients will be dead within 1 year, with 14.6% (roughly 1 in 6) requiring dialysis.  In CKD stage 5, more than one-third of patients will require RRT within 30 days; nearly two-thirds will require RRT at 1 year."

The investigators concluded that pre-procedure GFR is associated with RRT and death following TAVR, and risk profiles should be used in "informed decision making" for patients with advanced CKD. "Recognition of this risk can focus future efforts on therapies to ultimately reduce the hazard of renal failure and need for post-procedure RRT."

Holmes DR, Brennan JM, Rumsfeld JS, et al. Clinical Outcomes at 1 Year Following Transcatheter Aortic Valve Replacement. JAMA. 2015;313 (10):1019-1028.

The aim of this study was to update a previous report of 30-day outcomes and present 1-year outcomes after TAVR in U.S. hospitals.

This study used data from the TVT Registry for patients who underwent TAVR between November 2011 and June 2013, and were linked to Medicare claims (for 30-day and 1-year outcomes).

Primary outcomes were death, stroke, and a composite of death or stroke at 30 days and 1 year, and days alive and out of the hospital (DAOH) at 1 year.  There were multiple secondary, including in-hospital, outcomes.

The analysis used Kaplan-Meier methods for unadjusted event rates up to 1 year. Cox proportional hazards models were used for analysis of adjusted mortality rates up to 1 year, and for subgroup analysis. Model covariates for adjustment of outcomes were selected from baseline patient characteristics to include comorbidities and risk factors. Cumulative incidence methods were used for non-fatal outcomes (e.g., stroke), treating death as a competing risk. Differences in outcomes across subgroups were measured using Fine and Gray’s method, with the same covariates used in the Cox mortality model.

Patients included a total of 12,182 treated with TAVR at 299 U.S. hospitals.

Important patient demographics included median age of 84 years, 48% male, and 95% white. The registry patients "were elderly and had multiple comorbidities, similar to prior TAVR studies." However, the median baseline STS risk score of 7.1% in this registry study was significantly lower than the 11.8% in the PARTNER A trial arm (high-risk but operable patients with STS PROM score > 10%) and 11.2% in the PARTNER B trial arm (inoperable patients determined by 2 surgeons), but was similar to the 7.3% median score in the CoreValve trial.

The study investigators found that 30-day mortality post-TAVR was 7.0% (95% CI, 6.5%-7.4%). "In the first year after TAVR, patients were alive and out of the hospital for a median of 353 days (IQR 312-359 days); 24.4% of survivors were re-hospitalized once and 12.5% were re- hospitalized twice." The 1-year all-cause mortality rate was 23.7% (95% CI, 22.8%-24.5%), the stroke rate was 4.1% (95% CI, 3.7%-4.5%), and the rate of the composite outcome of mortality and stroke was 26.0% (25.1%-26.8%). Predictors of 1-year mortality were: advanced age, male sex, end-stage renal disease requiring dialysis, severe chronic obstructive pulmonary disease, STS risk score greater than 15% (vs less than 8%), preoperative atrial fibrillation/flutter, and one procedural factor – non-transfemoral access ("which may be a surrogate for more advanced disease such as peripheral arterial disease or the inability to tolerate a more invasive procedure"). Women had a higher risk of stroke than men (HR 1.40, 95% CI, 1.15-1.71).

The investigators concluded that as 30-day mortality was only 7% and 1-year mortality 24%, the majority of mortality "does not represent periprocedural complications," highlighting the importance of "better prediction of the overall risks and benefits considering the existing comorbidities" of TAVR patients. Furthermore, "it may be possible to identify patients who may not benefit from this procedure and who should be counseled accordingly. For instance, in this study, small, very high-risk subsets of patients such as those aged 85 to 94 years, undergoing dialysis, and having an STS PROM score higher than 15% can be identified."

Holmes DR, Nishimura RA, Grover FL, et al. Annual Outcomes with Transcatheter Valve Therapy: From the STS/ACC TVT Registry. Annals of Thoracic Surgery. 2016;101(2):789-800.

The aim of this study was "to provide an overview on current U.S. TVT practice and trends. The emphasis is on demographics, in-hospital procedural characteristics, and outcomes of patients" undergoing TAVR. This study was a summary report, not a test of a specific hypothesis.

This study used data from the TVT Registry for patients who underwent TAVR between January 2012 and December 2014.

No primary outcome was specified. A wide range of demographics, procedural techniques (e.g., access site and valve type) and outcomes (e.g., in-hospital mortality, stroke, pacemaker implantations, TAVR conversions to surgery) were reported, with comparison between two time groups to describe trends.

In the analysis method baseline patient characteristics and in-hospital outcomes were summarized by percentages and compared across the subgroups using chi-square, Wilcoxon, or Kruskal-Wallis 2-sided tests.

Patients included a total of 26,414 treated with TAVR at 348 U.S. hospitals. Patients were divided into two groups, those with TAVR procedures 1) between January 1, 2012, and December 31, 2013; and 2) between January 1, 2014, and December 31, 2014.

Important patient demographics included, for the overall study population (n=26,414), median age of 84 (mean age of 82) years, 51% male, and 94% white.

The study investigators found that, comparing across the two time groups, "TAVR patients remain elderly," with multiple comorbidities "reflected by a high mean STS predicted risk of mortality (STS PROM) for surgical valve replacement (8.34%), were highly symptomatic (New York Heart Association functional class III/IV in 82.5%), frail (slow 5-m walk test in 81.6%), and had poor health status (median baseline Kansas City Cardiomyopathy Questionnaire score of 39.1). Procedure performance is changing, with an increased use of moderate sedation (from 1.6% to 5.1%) and increase in femoral access using percutaneous techniques (66.8% in 2014)." Unadjusted in-hospital mortality dropped from 5.3% to 4.4%.

Also of note, in the latest time period, 2014 (n=12,785), TAVR conversions to open heart surgery was 1.3%, and cardiopulmonary bypass was 2.9% (73% of which were emergent). Prior to discharge after TAVR, implantation of a new pacemaker or ICD was 11%; cardiac arrest was 4.3%; acute kidney injury, 2.2%, and a new requirement for dialysis, 1.7%.

The investigators concluded that "changes in baseline characteristics over the 2 timeframes were clinically minor, although statistically significant due to the size of the registry," and that mortality, myocardial infarction, kidney injury, and neurological complications remained low.

Hyman MC, Vemulapalli S, Szeto WY, et al. Conscious Sedation Versus General Anesthesia for Transcatheter Aortic Valve Replacement: Insights from the NDCR STS/ACC TVT Registry. Circulation. 2017;136(22).

The aim of this study was to compare patients undergoing TAVR with general anesthesia or conscious sedation for the primary outcome of in- hospital mortality. The investigators noted that "conscious sedation is used during TAVR with limited evidence as to the safety and efficacy."

This study used data from the STS/ACC TVT registry for patients who underwent elective, transfemoral TAVR between April 2014 and June 2015.

The primary outcome was in-hospital mortality. Secondary outcomes included 30-day mortality, in-hospital and 30-day death/stroke, procedural success, and intensive care unit and hospital length-of-stay.

The analysis used propensity-score methods, with "inverse probability of treatment weighting to control for potential differences in the type of patient selected for conscious sedation versus general anesthesia." The propensity analysis used 51 variables, including all variables in the validated in-hospital TVT Registry mortality risk model. Logistic regression models with generalized estimating equations and a random intercept were created "to assess the association between sedation type and the discrete end points of interest adjusted for all factors used in the propensity score analysis and incorporating the inverse probability of treatment weighted weights."

Patients included a total of 10,997 who underwent elective, transfemoral TAVR at 314 U.S. hospitals. Of these, 9,260 patients underwent general anesthesia and 1,737 underwent conscious sedation during TAVR.

Important patient demographics included, for the general anesthesia group (n=9,260) mean age of 82 years, 64% male, and 95% white; and for the conscious sedation group (n=1,737), 82 years, 64% male, and 92% white. Although there were some significant differences in patient characteristics between the two groups, the TVT Registry In-Hospital Mortality Risk score was not significantly different.

The study investigators found that compared with general anesthesia, conscious sedation was used in 1,737/10,997 (15.8%) of the TAVR cases with a significant trend of increasing usage over the time period studied (P for trend<0.001).  In unadjusted analyses, TAVR with conscious sedation was associated with lower in-hospital and 30-day mortality than TAVR with general anesthesia. This persisted in adjusted analysis for in-hospital mortality (1.5% versus 2.4%, P < 0.001) and 30-day mortality (2.3% versus 4.0%, P < 0.001). Conscious sedation was associated with improved secondary outcomes as well, including reductions in procedural inotrope requirement, and intensive care unit and hospital length of stay. Falsification end point analyses of vascular complications, bleeding, and new pacemaker/defibrillator implantation demonstrated no significant differences between groups after adjustment. However, conscious sedation was associated with lower procedural success (97.9% versus 98.6%, P < 0.001) after adjustment.

The investigators concluded that "TAVR with conscious sedation can be performed safely." They believe that conscious sedation may be associated with reduced rates of mortality, but that "comparative effectiveness analyses using observational data cannot definitively establish the superiority of one technique over the other."

Kochar A, Zhuokai L, Harrison JK, et al. Stroke and Cardiovascular Outcomes in Patients With Carotid Disease Undergoing Transcatheter Aortic Valve Replacement. Circ Cardiovasc Interv. 2018 Jun;11(6):e006322. doi: 10.1161/CIRCINTERVENTIONS.117.006322.

The aim of this study was to evaluate whether carotid artery disease (CD) is associated with an increased risk of stroke or mortality (30 day or 1 year) after TAVR.  Stroke is a complication of TAVR and is associated with reduced quality of life and higher mortality.  The investigators noted that "despite the development of embolic protection devices, there is a residual risk of stroke which may be related to other pathogeneses of stroke, such as carotid artery disease."

This study used data from the TVT Registry for patients who underwent TAVR from October 2013 to September 2015, and were linked to Medicare claims.

The primary outcome was the 1-year incidence of stroke. Secondary outcomes included 30-day incidence of stroke, and 30-day and 1-year incidence of all-cause mortality, a composite of stroke or all-cause mortality, MI, and bleeding.

The analysis method used TVT Registry criteria for determining CD status and severity: no CD (≤50 stenosis), moderate (50%–79%), severe (80%–99%), occlusive (100%). As CD severity was thus ordinal, the association between CD severity and patient characteristics was tested through the Spearman correlation coefficient for continuous variables and the Wilcoxon rank-sum test or Kruskal-Wallis test for categorical variables.

Cumulative incidence methods were used to assess 30-day and 1-year outcomes post-TAVR stratified by varying degrees of CD severity. For nonfatal outcomes, the cumulative incidence approach treated death as a competing risk. Adjustment for baseline covariates was performed with Cox proportional hazards models for mortality and composite outcomes. Fine and Gray proportional sub-distribution hazards models were used for stroke. The covariates were identified from the validated TVT risk prediction model for in-hospital mortality after TAVR.

Patients included in the study after linking with Medicare claims data (for outcomes) totaled 29,143 patients from 390 sites, of whom 6,410 patients (22%) had CD. When stratified by CD severity, 5001 (17.2%) were moderate CD, 940 (3.2%) were severe CD, and 469 (1.6%) were occlusive CD.

Important patient demographics included, for the No CD group (n=22,733), median age of 83 years, 51% male, 94% white; and for the CD group (n=6,410), median age of 82 years, 56% male, 96% white.  CD patients were generally sicker, with a higher burden of comorbidities and predicted risk of mortality (by STS risk score). The transfemoral approach (which is preferred) was used less frequently in patients with CD compared to patients without CD.

The investigators found that CD is common among TAVR patients, present in 1 of 5. There is "no association between the presence of CD and 30-day stroke (adjusted hazard ratio [HR], 1.16; 95% CI, 0.94–1.43) or mortality (adjusted HR, 1.10; 95% CI, 0.95–1.28). There was no association between CD and 1-year stroke (adjusted HR, 1.03; 95% CI, 0.86–1.24) or mortality (adjusted HR, 1.02; 95% CI, 0.93–1.12). There was no significant risk-adjusted association between severity of CD and 30-day or 1-year stroke or mortality."

The investigators concluded that CD was not associated with an increased risk of stroke or mortality (30-day or 1-year); thus post-TAVR stroke is likely due to mechanisms other than CD.

Kolte, D., Khera, S., Vemulapalli, S. Outcomes Following Urgent/Emergent Transcatheter Aortic Valve Replacement: Insights from the STS/ACC TVT Registry. JACC. March 2018.

The aim of this study was "to examine outcomes and identify independent predictors of mortality among patients undergoing urgent/emergent TAVR."

This study used data from the TVT Registry, linked to CMS claims, to identify patients who underwent urgent/emergent versus elective TAVR between November 2011 and June 2016. In the TVT registry, urgent status is defined as "procedure required during same hospitalization in order to minimize chance of further clinical deterioration;" emergent status is defined as "one in which there should be no delay in providing intervention."

The primary outcome was all-cause mortality (in-hospital, 30-day, and 1-year), derived from CMS claims. Secondary outcomes included numerous but important in-hospital outcomes defined using standardized consensus-derived criteria including those of the Valve Academic Research Consortium II.

The analysis method used the TVT Registry in-hospital mortality prediction model to calculate observed to expected (O:E) ratios for in-hospital mortality after urgent/emergent versus elective TAVR. Unadjusted 1-year mortality rates were compared using Kaplan-Meier methods. The 30- day and 1-year mortality rates after urgent/ emergent versus elective TAVR were compared using Cox proportional hazards models. Similarly, Cox proportional hazards models were used to identify independent predictors of 30-day and 1-year mortality in patients undergoing urgent/emergent TAVR. As typically done, a backward selection process on an extensive initial list of patient characteristics/covariates (i.e., pruning) was used to identify statistically significant predictors.

Important patient demographics included, in the urgent/emergent group, median age of 84 years, 52% male, 91% white; and in the elective group, median age of 84 years, 52% male, 93% white.  The urgent/emergent group were generally sicker, with a higher burden of comorbidities and predicted risk of mortality (by STS risk score), and a lower quality of life (by KCCQ score).

The study investigators found that "compared with elective TAVR, patients undergoing urgent/emergent TAVR had a higher rate of 30-day mortality (8.7% vs. 4.3%, adjusted hazard ratio (HR): 1.28, 95% CI: 1.10 to 1.48), and 1-year mortality (29.1% vs. 17.5%, adjusted HR: 1.20, 95% CI: 1.10 to 1.31). In patients undergoing urgent/emergent TAVR, non-femoral access and cardiopulmonary bypass were associated with increased risk, whereas use of balloon-expandable valve was associated with decreased risk of 30-day and 1-year mortality."

The investigators concluded that patients undergoing urgent/emergent TAVR had higher rates of mortality compared with elective TAVR, but tended to be sicker with worse prognosis at baseline. However, in-hospital mortality following urgent/emergent TAVR was significantly lower than that predicted by the TVT Registry model.  They thus believe that "urgent/emergent TAVR is feasible with acceptable outcomes and may be a reasonable option in a selected group of patients with severe aortic stenosis."

Mack MJ, Brennan JM, Brindis R, et al. Outcomes following transcatheter aortic valve replacement in the United States. JAMA. 2013 Nov 20;310(19):2069-77.

The aim of this study was to report the initial US commercial experience with TAVR using the TVT Registry. The results from all eligible US TAVR cases (n=7710) from 224 participating registry hospitals following the Edwards Sapien device commercialization (November 2011–May 2013) were analyzed.

For the overall results, the 7710 patients who underwent TAVR included 1559 (20%) cases that were inoperable and 6151 (80%) cases that were high-risk but operable. The median age was 84 years (interquartile range [IQR], 78-88 years); 3783 patients (49%) were women and the median STS risk score was 7%. The observed incidence of in-hospital mortality was 5.5% (95% CI 5.0%-6.1%). Other major complications included stroke (2.0%; 95% CI 1.7%-2.4%), dialysis-dependent renal failure (1.9%; 95% CI 1.6%-2.2%), and major vascular injury (6.4%; 95% CI 5.8%-6.9%). New-onset atrial fibrillation was observed in 6.0% (95% CI 5.5%-6.5%) and need for new pacemaker or implantable cardioverter-defibrillator in 6.6% (95% CI 6.1%-7.2%).  The incidence of 30-day death was 7.6% (95% CI 6.7%-8.6%) and aortic valve reintervention was 0.5% (95% CI 0.3%-0.8%). The authors concluded that "postapproval commercial introduction of this new technology with an early-generation device has yielded success rates and complication patterns that are similar to those documented in carefully performed randomized trials."

O’Brien SM, Cohen DJ, Rumsfeld JS, et al. Variation in Hospital Risk-Adjusted Mortality Rates Following Transcatheter Aortic Valve Replacement in the United States: A Report from the STS/ACC TVT Registry. Circ Cardiovasc Qual Outcomes. 2016;9:560-565.

The aim of this study was to develop a TAVR in-hospital mortality risk model and use it to evaluate variation in mortality rates across U.S. TAVR centers, consistent with the larger goal of assessing variation in TAVR procedural outcomes in broad community practice. This is important because while studies have documented 30-day and 1-year TAVR outcomes in the U.S., "the extent to which these outcomes vary across hospitals has not been reported."

This study used data from the TVT Registry for patients who underwent TAVR from November 2011 to October 2014.

The primary outcome was in-hospital mortality. This was chosen over 30-day mortality because "in-hospital mortality status was reported in 99.9% of records, whereas 30-day mortality status was missing in 20% at the time of model development initiation."

The analysis method used a Bayesian model to estimate hospital-specific risk-adjusted mortality rates.  This hierarchical logistic regression model adjusted for case mix by including 40 prespecified patient baseline factors (all those with a known or suspected association with mortality) and center-specific random intercepts. Notably, these factors did not include a frailty measure or the KCCQ as these data were deemed to be missing too frequently (60% and 61% respectively) at the time of model development. Calibration of the model was performed by "comparing observed versus expected mortality rates overall and within subgroups, based on deciles of predicted risk."

Patients included in the study totaled 22,248 who underwent TAVR from 318 U.S. hospitals, with a median of 58 cases per hospital.

Important patient demographics (n=22,248) included median age of 84 years, 50% male, and 91% white. "Patient characteristics varied substantially across hospitals. Even after excluding hospitals with <100 cases, there was more than a 5-fold difference across hospitals for the majority of baseline factors examined."

The investigators found that "a total of 1130 in-hospital deaths (5.1%) were observed. Reliability-adjusted risk-adjusted mortality rate estimates ranged from 3.4% to 7.7% with an interquartile range of 4.8% to 5.4%. A patient’s predicted odds of dying was 80% higher if treated by a hospital 1 standard deviation above the mean compared with a hospital 1 standard deviation below the mean (odds ratio =1.8; 95% credible interval, 1.4%–2.2%)."

The investigators concluded that there was "significant hospital-level variation in risk-adjusted TAVR mortality rates" among TAVR centers in the U.S.  The study supports that: 1) "institutional factors may play a role in affecting patient outcomes"; and 2) "given the low number of cases performed at many sites, measured mortality rates are expected to vary substantially by chance alone" – rendering it challenging to assess patient outcomes like mortality at low-volume centers.

Sorajja P, Kodali S, Reardon MJ, et al. Outcomes for the Commercial Use of Self-Expanding Prostheses in Transcatheter Aortic Valve Replacement: A Report From the STS/ACC TVT Registry. JACC Cardiovasc Interv. 2017;10(20):2090-2098.

The aim of this study was "to compare the outcomes of commercial TAVR with the repositionable Evolut R platform [Medtronic 2015] to those observed with the CoreValve device [Medtronic 2014]."

This study used data from the TVT Registry for patients with native aortic valve disease who underwent TAVR with a Medtronic self- expanding prosthesis (CoreValve or Evolut R, in the sizes of 23, 26, or 29 mm), between January 2014 and April 2016.

Study outcomes included a range of procedural, in-hospital, and 30-day outcomes (to include all-cause mortality, stroke, MI, device success, pacemaker dependence, degree of paravalvular regurgitation, hospital length-of-stay, quality-of-life, etc.), with no primary outcome specified.

The analysis method made unadjusted comparisons between the two groups (CoreValve and Evolut R) followed by adjusted comparisons using a Cox proportional hazards model for mortality. Covariates were derived from patient characteristics to include comorbidities and predicted risk of mortality (by STS risk score).

Patients included a total of 9,616 who underwent TAVR with a Medtronic self-expanding prosthesis in the U.S. Of these, 5,806 patients were treated with CoreValve and 3,810 were treated with Evolut R.

Important patient demographics included, for the CoreValve group, (n=5,806), mean age of 82 ± 8 years and 35% male; and for the Evolut R group (n=3,810), mean age of 81 ± 8 years and 38% male. There were some statistically significant differences observed between the two groups, including a lower predicted risk of mortality (by STS risk score) for the Evolut R group.

The study investigators found that "at 30 days, Evolut R patients had both lower mortality (3.7% vs. 5.3%; p < 0.001) and less need for a pacemaker (18.3% vs. 20.1%; p = 0.03)." The Evolut R TAVR group also had "greater device success (96.3% vs. 94.9%; p=0.001), and less need for a second prosthesis (2.2% vs. 4.5%; p< 0.001), less device migration (0.2%vs. 0.6%; p=0.01), a lower incidence of moderate/severe paravalvular regurgitation (post-procedure, 4.4% vs. 6.2%; p < 0.001), and shorter median hospital stay (4.0 vs. 5.0 days; p < 0.001)."  Quality-of- life was significantly improved for both the Evolut R and CoreValve groups.

The investigators concluded that "the Evolut R platform is associated with significant improvements in acute outcomes for patients undergoing TAVR" compared to the earlier CoreValve platform. They stated that "continual iteration of device technology and procedural techniques, therefore, remains essential for maximizing beneficial outcomes in patients with aortic stenosis who undergo TAVR."

Suri RM, Gulack BC, Brennan JM, et al. Outcomes of Patients with Severe Chronic Lung Disease Who Are Undergoing Transcatheter Aortic Valve Replacement. Annals of Thoracic Surgery. Available online 29 August 2015.

The aim of this study was "to determine the clinical outcomes after TAVR among patients with chronic lung disease (CLD) and to evaluate the safety of transaortic [Tao] versus transapical [TA] alternate access approaches in patients with varying severities of CLD." This is important because CLD is common in TAVR patients, and prior studies support that severe CLD "is associated with poor outcomes after cardiac surgical procedures including surgical aortic valve replacement."

This study used data from the STS/ACC TVT registry for patients with known CLD status who underwent TAVR between November 2011 and June 2014, and were linked to Medicare claims (for 1-year outcomes).

The primary outcome reported was 1-year mortality; there were multiple in-hospitaland 1-year outcomes assessed including stroke.

The analysis method grouped patients into three cohorts based on CLS status: mild, moderate, or severe (according to detailed STS database standard definitions). Clinical outcomes were evaluated across the CLD cohorts, and the risk-adjusted association between access route and post-TAVR mortality was determined among patients with severe CLD. Multivariable logistic regression was used to determine the adjusted association between severity of CLD and in-hospital outcomes, including mortality and stroke. Cox proportional hazards regression models were used to determine the adjusted association of the severity of CLD with 1-year mortality, and between alternative access site and 1-year mortality.  Covariates for adjustment in all models were baseline patient characteristics including comorbidities.

Patients included a total of 11,656 who underwent TAVR at 297 hospitals, after multiple exclusions such as unknown patient CLD status, and an access site other than TF, TA, or Tao.

Important patient demographics reported included, for the overall study population, median age of 84 years and 48% male. Moderate or severe CLD was present in 3,226 (27.7%) patients, including 1,662 (14.3%) with moderate CLD and 1,564 (13.4%) with severe CLD.

The investigators found that approximately 1 in 4 TAVR patients had moderate or severe CLD.  "Patients with moderate and severe CLD carried a higher predicted and observed risk of death up to 1-year, but their stroke risk was not increased. Compared with patients with no or mild CLD, patients with severe CLD had a higher rate of post-TAVR mortality to 1-year (32.3% versus 21.0%; adjusted hazard ratio [HR], 1.48; 95% confidence interval [CI], 1.31 to 1.66), as did those with moderate CLD (25.5%; adjusted HR, 1.16; 95% CI, 1.03 to 1.30). The adjusted rate of mortality was similar for transapical versus transaortic approaches to 1 year (adjusted HR, 1.17; 95% CI, 0.83 to 1.65)."

The investigators concluded that "moderate or severe CLD is associated with an increased risk of death to 1 year after TAVR in comparison with no or mild pulmonary disease; however, stroke risk does not appear to be elevated."  The risk of death for patients with severe CLD was similar for transapical and transaortic alternate-access approaches.

Thourani VH, Jensen HA, Babaliaros V, et al. Transapical and Transaortic Transcatheter Aortic Valve Replacement in the United States. Annals of Thoracic Surgery. 2015;100:1718–27.

The aim of this study was to assess in-hospital and 1-year outcomes of patients undergoing alternative access TAVR through the transapical (TA) or transaortic (TAo) approaches. This is important because the preferred transfemoral approach is often contradicted for patients otherwise eligible for TAVR.

This study used data from the TVT Registry for patients who underwent TAVR between from November 2011 and June 2014, and were linked to Medicare claims (for 30-day and 1-year outcomes).

The primary outcomes were 30-day and 1-year all-cause mortality. Secondary outcomes included in-hospital, 30-day, and 1-year stroke, and re- hospitalization for heart failure.

The analysis method grouped patients into 3 cohorts based on STS risk scores: less than 8%; 8% to 15%; greater than 15%. The risk-adjusted association between access route and mortality, stroke, and re- hospitalization for heart failure were assessed using Cox proportional hazards regression modeling. Covariates included in the Cox model for adjustment of outcomes were age, sex, renal failure, ejection fraction, prior aortic valve procedure, primary procedure indication, valve morphology, and atrial fibrillation/flutter.

Patients included a total of 4,953 who underwent TAVR with either TA or TAo access approaches. Of these, 4,085 patients underwent TA TAVR and 868 underwent TAo TAVR.

Important patient demographics included, for the overall study population (n=4,953), mean age of 83 ± 7 years and 41% male. Patients undergoing TAo TAVR were older, more likely to be female, and had higher STS PROM scores than patients undergoing TA TAVR.

The investigators found that "the median STS predicted risk of mortality was significantly higher among patients undergoing TAo (8.8 versus 7.4, p < 0.001). When compared with TA, TAo was associated with an increased risk of unadjusted 30-day mortality (10.3% versus 8.8%) and 1-year mortality (30.3% versus 25.6%, p=0.006). There were no significant differences between TAo and TA for in-hospital stroke rate (2.2%), major vascular complications (0.3%), and 1-year heart failure re-hospitalizations (15.7%)." Subgroup analysis of 1-year death, stroke, and heart failure rehospitalization supported the hypothesis that in high risk, inoperable patients, adjusted 1-year mortality was significantly higher for patients who underwent TAo (P=0.012); no other differences were seen between the two access groups.

The investigators concluded that "there were no risk-adjusted differences between TA and TAo access in mortality, stroke, or readmission rates as long as 1 year after TAVR."

Tuzcu EM, Kapadia SR, Vemulapalli S, et al. Transcatheter Aortic Valve Replacement of Failed Surgically Implanted Bioprostheses: The STS/ACC Registry J Am Coll Cardiol. 2018 Jul 24;72(4):370-382. doi: 10.1016/j.jacc.2018.04.074.

The aim of this study was "to evaluate the safety and effectiveness of valve-in-valve (ViV) TAVR for failed SAVR by comparing it with the benchmark of native valve (NV) TAVR."

This study used data from the TVT Registry for patients who underwent ViV- or NV-TAVR between November 2011 and June 2016.

In the analysis method, patients were "matched on sex, inoperable/extreme risk designation, hostile chest [factors that prohibit redo thoracic surgery] or porcelain aorta, 5-m-walk time, and STS Predicted Risk of Mortality (PROM) for reoperation in a 1:2 fashion to patients undergoing NV-TAVR." Primary outcomes were unadjusted and adjusted measures including in-hospital, 30-day and 1-year mortality, stroke, hospitalization for heart failure, and aortic valve reintervention.

Time-to-event (Kaplan-Meier) methods were used to evaluate unadjusted mortality rates to 1 year. Cumulative incidence methods were used for analysis of non-fatal events (stroke, heart failure, and aortic valve reintervention) to account for the probability that death could preclude an event from occurring (unlike standard time-to-event methods which assume a death-free environment). Unadjusted and adjusted hazard ratios comparing mortality risks across subgroups were evaluated with the Wald test.

Patients included in the study after linking with 1-year Medicare claims data (for outcomes) and patient matching between the groups, totaled 3,409 (n = 1,150, ViV-TAVR and n = 2,259, NV-TAVR).

Important patient demographics included, in the ViV-TAVR group, median age 79 years, 61% male, 95% white, STS score 6.9%; and in the NV-TAVR group, median age 84 years, 61% male, 96% white, STS score 6.8%. There were significant difference between the matched groups. The ViV-TAVR group more frequently had New York Heart Association functional class III or IV symptoms, mitral or tricuspid regurgitation, permanent pacemaker, lower left ventricular ejection fraction, and previous multiple cardiac surgeries. Patients in the NV-TAVR group were older, had higher rates of diabetes, coronary artery disease, prior percutaneous coronary intervention, and peripheral vascular disease, and more frequently required a non-transfemoral approach.

The study investigators found that "unadjusted analysis revealed lower 30-day mortality (2.9% vs. 4.8%; p < 0.001), stroke (1.7% vs. 3.0%; p = 0.003), and heart failure hospitalizations (2.4% vs. 4.6%; p < 0.001) in the ViV-TAVR compared with NV-TAVR group. Adjusted analysis revealed lower 30-day mortality (HR: 0.503; 95% CI: 0.302 to 0.839; p = 0.008), lower 1-year mortality (HR: 0.653; 95% CI: 0.505 to 0.844; p = 0.001), and [fewer] hospitalizations for heart failure (HR: 0.685; 95% CI: 0.500 to 0.939; p = 0.019) in the ViV-TAVR group." "Adjusted analysis of aortic valve reintervention showed no difference between ViV-TAVR and NV-TAVR at 30 days or 1 year."

The investigators concluded that rates of 1-year all-cause mortality, stroke, and hospitalization for heart failure (but not of aortic valve reintervention) were significantly lower for ViV-TAVR patients compared with the matched NV-TAVR patients. Thus, ViV-TAVR "is a safe and effective procedure in patients with failed SAVR who are at high risk for repeat surgery."

4. Medicare Evidence Development & Coverage Advisory Committee (MEDCAC)

On July 25, 2018, CMS convened a meeting of the Medicare Evidence Development & Coverage Advisory Committee (MEDCAC) to discuss the body of evidence, hear presentations, consider public comments, and make recommendations to CMS regarding the appraisal of the state of currently available evidence for procedural volume requirements for SAVR, TAVR, PCI and other relevant structural heart disease procedures as they relate to TAVR programs. The MEDCAC panel also discussed whether volume requirements create unintended barriers to accessing TAVR. The meeting began with a CMS presentation describing the meeting focus, the history and the evolution of TAVR, including the current NCD and FDA approvals. The panel then heard presentations from five invited guests, 10 scheduled speakers and eight members of the public.

The panel, including nine voting members and two additional panelists, voted on nine questions (listed below), the last of which included three parts. The panel voted using a scale of one to five, with one representing a low confidence vote and five representing a high confidence vote and an average voting score of 2.5 representing intermediate confidence. CMS recorded the scores of the voting panel members and calculated the average score for each question. The panel also considered one discussion question.

Hospital Requirements to Begin TAVR Programs

1. How confident are you that there is sufficient evidence that a certain threshold of SAVR procedural volumes must be required for hospitals without previous TAVR experience to begin TAVR programs?
   Average score: 3.78


2. How confident are you that there is sufficient evidence that a certain threshold of PCI procedural volumes must be required for hospitals without previous TAVR experience to begin TAVR programs?
   Average score: 3.44


3. How confident are you that the benefits of meeting procedural (i.e., SAVR, PCI) volume requirements to begin a TAVR program outweigh the harms of limiting access to TAVR to only hospitals that meet volume requirements?
   Average score: 3.11

Hospital Requirements to Maintain TAVR Programs

4. How confident are you that there is sufficient evidence that a certain threshold of SAVR procedural volumes must be required for hospitals with TAVR experience to maintain TAVR programs?
   Average score: 3.56


5. How confident are you that there is sufficient evidence that a certain threshold of PCI procedural volumes must be required for hospitals with TAVR experience to maintain TAVR programs?
   Average score: 3.33


6. How confident are you that the benefits of meeting procedural (i.e., SAVR, TAVR, PCI) volume requirements to maintain a TAVR program outweigh the harms of limiting access to TAVR to only hospitals that meet volume requirements?
   Average score: 3.67

Operator Requirements to Begin TAVR Programs

7. To begin performing TAVR, how confident are you that there is sufficient evidence that a certain threshold of SAVR and TAVR procedural volumes must be required for the principle cardiovascular surgeon on a TAVR heart team?
   Average score: 4.33


8. To begin performing TAVR, how confident are you that there is sufficient evidence that a certain threshold of structural heart disease procedural volumes must be required for the principle interventional cardiologist on a TAVR heart team?
   Average score: 4.22

Heart Team Requirements to Maintain TAVR Programs

9. To maintain proficiency, how confident are you that there is sufficient evidence that a certain threshold of TAVR procedural volumes must be required for:
   1. The principle cardiovascular surgeon on a TAVR heart team?
      Average score: 3.33
   2. The principle interventional cardiologist on a TAVR heart team?
      Average score: 4.11
   3. The combined experience of the principle cardiovascular surgeon and interventional cardiologist on a TAVR heart team?
      Average score: 3.78

Information about the meeting, including the agenda, presentations from speakers, transcripts, and results of the voting questions are available on the CMS Website at: https://www.cms.gov/medicare-coverage-database/details/medcac-meeting-details.aspx?MEDCACId=75.

Nishimura RA, Otto CM, Bonow RO, et al. 2017 AHA/ACC focused update of the 2014 AHA/ACC guideline for the management of patients with valvular heart disease: a report of the American College of Cardiology/American Heart Association Task Force on Clinical Practice Guidelines. Circulation. 2017;135:e1159–e1195.

This is a focused update to the "2014 AHA/ACC Guideline for the Management of Patients With Valvular Heart Disease" (2014 VHD guideline) to incorporate findings from several randomized controlled trials (RCTs) that were published since its release. Clinical trials presented at annual professional society scientific meetings were reviewed in addition to peer-reviewed published literature from October 2013 through November 2016.

The guidelines document was approved for publication by the governing bodies of the ACC and the AHA and was endorsed by the Association for Thoracic Surgery (AATS), American Society of Echocardiography (ASE), Society for Cardiovascular Angiography and Interventions (SCAI), Society of Cardiovascular Anesthesiologists (SCA), and Society of Thoracic Surgeons (STS).

The Class of Recommendation (COR) indicates the strength of the recommendation and estimates the magnitude of benefit versus risk.

• Class 1 (Strong): Is recommended. Should be performed/administered.
• Class IIa (Moderate): Is reasonable. Can be useful/effective/beneficial.
• Class IIb (Weak): May/might be reasonable. Usefulness/effectiveness is unknown/unclear/uncertain or not well established.
• Class III: No Benefit (Moderate): Is not recommended. Is not indicated/useful/effective/beneficial.
• Class III: Harm (Strong): Potentially harmful/Causes harm. Should not be performed/administered/other.

The Level of Evidence (LOE) rates the quality of the evidence based on the type, quantity, and consistency of the data from clinical trials and other sources.

• Level A
  • High-quality evidence from more than 1 RCT
  • Meta-analyses of high quality RCTs
  • One or more RCTs corroborated by high-quality registry studies
• Level B-Randomized (R)
  • Moderate-quality evidence from 1 or more RCTs
  • Meta-analyses of moderate-quality RCTs
• Level B-nonrandomized (NR)
  • Moderate-quality evidence from 1 or more well-designed, well-executed nonrandomized studies, observational studies, or registry studies
  • Meta-analysis of such studies
• Level C-Limited Data (LD)
  • Randomized or nonrandomized observational or registry studies with limitations of design or execution
  • Meta-analyses of such studies
  • Physiological or mechanistic studies in human subjects
• Level C-Expert Opinion (EO)
  • Consensus of expert opinion based on clinical experience  The following Class 1 Level A and B recommendations were put forward: CLASS 1
• Symptomatic severe AS and prohibitive surgical risk: TAVR is recommended for symptomatic patients with severe AS (Stage D) and a prohibitive risk for surgical AVR who have a predicted post-TAVR survival greater than 12 months. (LOE: A)
• Symptomatic severe AS and high surgical risk: Surgical AVR or TAVR is recommended for symptomatic patients with severe AS (Stage D) and high risk for surgical AVR, depending on patient-specific procedural risks, values, and preferences. (LOE: A)
• Symptomatic severe AS and intermediate or low surgical risk: Surgical AR is recommended for symptomatic patients with severe AS (Stage D) and asymptomatic patients with severe AS (Stage C) who meet an indication for AVR when surgical risk is low or intermediate. (LOE: B-NR)

Otto CM, Kumbhani DJ, Alexander KP et al. 2017 ACC expert consensus decision pathway for transcatheter aortic valve replacement in the management of adults with aortic stenosis: a report of the American College of Cardiology Task Force on Clinical Expert Consensus Documents. J Am Coll Cardiol. 2017; 69:1313-46.

The clinical expert consensus pathway provides additional details and practical guidance about TAVR for native valve severe aortic stenosis with point-of-care checklists and algorithms. It is separated into four sections: preprocedure evaluation of the patient being considered for TAVR; imaging modalities and measurements; key issues in performing the TAVR procedure; recommendations for patient follow-up after TAVR. The pathway starts when a patient is being considered for TAVR based on the indication for AVR and choice of valve type. The primary objective of the document is to provide a framework for the several steps involved in managing patients undergoing TAVR. It outlines key steps in patient selection and evaluation, imaging modalities and measurements, issues in performing the TAVR procedure, and provides recommendations for post-TAVR management.

The decision pathways document highlights the importance of: shared decision making including the heart team, referring physician, patients and their family; risk category assessment (STS risk estimate, frailty, major organ system dysfunction, and procedure-specific impediments); integrated benefit-risk of TAVR and shared decision making.  The consensus reports that TAVR is best achieved by a multidisciplinary, collaborative Heart Valve Team, where cardiologists with expertise in valvular heart disease, structural interventional cardiologists, imaging specialists, cardiovascular surgeons, cardiovascular anesthesiologists, and cardiovascular nursing professionals are included in the team.

Bavaria JE, Tommaso CL, Brindis RG, Carroll JD, et al. 2018 AATS/ACC/SCAI/STS Expert consensus systems of care document: Operator and institutional recommendations and requirements for transcatheter aortic valve replacement: A joint report of the American Association for Thoracic Surgery, the American College of Cardiology, the Society for Cardiovascular Angiography and Interventions, and the Society of Thoracic Surgeons.  J Am Coll Cardiol. 2018 Jul 18. pii: S0735-1097(18)35377-4.

This multisocietal Expert Consensus Systems of Care document was commissioned by the American Association for Thoracic Surgery (AATS), the American College of Cardiology (ACC), the Society for Cardiovascular Angiography and Interventions (SCAI), and the Society of Thoracic Surgeons (STS). Expert Consensus Systems of Care documents are intended to summarize the position of these partnering organizations on the availability, delivery, organization, and quality of cardiovascular care. The AATS, ACC, SCAI, and STS have joined together to provide recommendations for institutions and individuals to assess their potential for instituting and/or maintaining a high-quality TAVR program. The first multisocietal document on institutional and operator requirements for performing TAVR was published in 2012 and is now updated to reflect the current evolution in practice and quality benchmarks.

The writing group has included a multimodal approach to quality measurement that allows the requirements to evolve in anticipation of newer treatment modalities; expansion to younger and lower-risk populations; and emerging evidence regarding patient outcomes, cost, cost- effectiveness, and durability. Since publication of the original document in 2012, the consensus document states that TAVR indications have been extended into groups of patients who are eligible for SAVR at intermediate to high risk; TAVR has also become an alternative to reoperation for those with severe bioprosthetic aortic valve degeneration.

Continuing education requirements:
It is expected that the MDT will participate in appropriate CME.

*Vascular arterial interventions include TEVAR/EVAR, carotid stenting, renal artery stenting, iliac and femoral artery stenting, coarctation stenting, and acute limb ischemia related interventions.
† For the purposes of this hospital volume requirement SAVR is defined to include all aortic valve replacement (mechanical, bioprosthesis, homograft, autograft [Ross], composite valve graft or root replacement) or aortic valve repair procedures, including concomitant valve resuspension for acute aortic dissection and valve-sparing aortic root replacement. Simple adjuvant aortic valve procedures, e.g., suturing closed regurgitant aortic valves in an LVAD patient, excising a papillary fibroelastoma or thrombus, etc., are not counted.
ACC indicates American College of Cardiology; AUC, appropriate use criteria; FDA, Food and Drug Administration; NCD, National Coverage Decision; NCDR, National Cardiovascular Data Registry; NQF, National Quality Forum; EVAR, endovascular aneurysm repair (or endovascular aortic repair); PCI, percutaneous coronary intervention; SAVR, surgical aortic valve replacement; STS, Society of Thoracic Surgeons; TAVR, transcatheter aortic valve replacement; TEVAR, thoracic endovascular aortic/aneurysm repair; TVT, Transcatheter Valve Therapies

Aortic Stenosis Writing Group, Bonow RO, Brown AS, Gillam LD, et al. ACC/AATS/AHA/ASE/EACTS/HVS/SCA/SCAI/SCCT/SCMR/STS 2017 Appropriate Use Criteria for the Treatment of Patients With Severe Aortic Stenosis: A Report of the American College of Cardiology Appropriate Use Criteria Task Force, American Association for Thoracic Surgery, American Heart Association, American Society of Echocardiography, European Association for Cardio-Thoracic Surgery, Heart Valve Society, Society of Cardiovascular Anesthesiologists, Society for Cardiovascular Angiography and Interventions, Society of Cardiovascular Computed Tomography, Society for Cardiovascular Magnetic Resonance, and Society of Thoracic Surgeons.   J Am Soc Echocardiogr. 2018 Feb;31(2):117-147.

The American College of Cardiology collaborated with the American Association for Thoracic Surgery, American Heart Association, American Society of Echocardiography, European Association for Cardio-Thoracic Surgery, Heart Valve Society, Society of Cardiovascular Anesthesiologists, Society for Cardiovascular Angiography and Interventions, Society of Cardiovascular Computed Tomography, Society for Cardiovascular Magnetic Resonance, and Society of Thoracic Surgeons to develop and evaluate Appropriate Use Criteria (AUC) for the treatment of patients with severe aortic stenosis. The purpose of the Appropriate Use Criteria was to address the topic of AS and its treatment options, including SAVR and TAVR to determine the appropriate use of treatment options in selected patient scenarios.

A number of common patient scenarios experienced in daily practice were developed by experts in the field representing multiple subspecialty societies along with assumptions and definitions for those scenarios, which were all created using guidelines, clinical trial data, and expert opinion in the field of aortic stenosis. The writing group identified 95 clinical scenarios based on patient symptoms and clinical presentation, and up to 6 potential treatment options for those patients. A separate, independent rating expert panel was asked to score each clinical scenario as "Rarely Appropriate," "May Be Appropriate," or "Appropriate."  The authors report that "after considering factors such as symptom status, left ventricular (LV) function, surgical risk, and the presence of concomitant coronary or other valve disease, the rating panel determined that either SAVR or TAVR was appropriate in most patients with symptomatic AS at intermediate or high surgical risk; however, situations commonly arose in clinical practice in which the indications for SAVR or TAVR were less clear, including situations in which one form of valve replacement would appear reasonable when the other was less so, as do other circumstances in which neither intervention was the suitable treatment option." As an example of a clinical scenario, TAVR rather than SAVR was considered an appropriate intervention in symptomatic patients with frailty, since these factors could pose increased surgical risk that would not captured in STS-PROM risk scoring (porcelain aorta or hostile chest), and/or significant comorbidities, including lung or liver disease, malignancy, and dementia.

Walters DL, Webster M, Pasupati S, et al. Position statement for the operator and institutional requirements for a transcatheter aortic valve implantation (TAVI) program.  Heart Lung Circ. 2015 Mar;24(3):219-23.

The Cardiac Society of Australia and New Zealand (CSANZ) and the Australia and New Zealand Society of Cardiac and Thoracic Surgeons (ANZSCTS) joined together to provide recommendations for institutions and individual operators to assess their ability to initiate and maintain a transcatheter valve program. The authors acknowledged that multi-society consensus statements have been produced in the US and Europe and that these statements were reviewed as part of the development of the Australian and New Zealand guidelines. The position paper endorsed the important role of a multi-disciplinary "Heart Team" in selecting patients for TAVR as fundamental to the establishment of a successful program. The core members of a Heart Team are an interventional cardiologist and a cardiac surgeon supported by a TAVR nurse case manager / coordinator.

For TAVR interventional cardiologist who have never performed TAVR, the following pre-requisites were suggested: 100 structural procedures lifetime or 20 left sided structural per year of which at least 10 should be balloon aortic valvuloplasty. The interventional cardiologist should have been trained and proctored on the devices being used. For an operator who has never implanted a transcatheter valve, a minimum of 10 proctored cases, in which the primary and secondary operators are working as a team, is recommended. Additional cases may be required depending on the assessment of the proctor.

For the cardiac surgeon, the following experience and training is recommended as follows:

• 100 surgical AVR career, at least 10 of which are "high-risk" with an STS score > 6, or
• 25 surgical AVR per year or
• 50 surgical AVR in two years and
• at least 20 AVR in last year prior to TAVR initiation

For a cardiac surgeon who has never implanted a transcatheter valve, a minimum of 10 proctored cases, in which the primary and secondary operators are working as a team, is recommended. Additional cases may be required depending on the assessment of the proctor.

For institutional requirements, the following activity levels for institutions undertaking TAVR programs were suggested:

• Institutional interventional program
  • 1000 catheter studies/400 PCI per year
• Institutional surgical program
  • 50 Total AVR per year of which at least 10 aortic valve replacement (AVR) should be high-risk (STS score > 6)
  • Minimum of two institutionally-based cardiac surgeons in program

The following minimum volume and outcomes requirements were recommended for approved TAVR programs:

• Program volume of 20 TAVR per year or 40 per two years
• 30-day all-cause mortality < 10%
• 30-day all-cause neurologic events including transient ischemic attack (TIAs) < 10%
• Major vascular complication rate < 10%
• 80% one-year survival rate for patients after the program has been running for two years (two-year average)
• All cases should be submitted to a prospective national database registry.

Table 2. Key TAVR Trials

Table 3: Adverse Events in Key TAVR Trials and Registry Studies

ACM denotes all-cause mortality, C control (historic), CI confidence interval (classical statistics) or credible interval (Bayesian statistics) [CI presented when reported], DAOH days alive and out of hospital, FDA Food and Drug Administration, HR hazard ratio, ITT intention-to-treat analysis, AT as-treated analysis, KCCQ Kansas City cardiomyopathy questionnaire, Non-I non-inferiority trial design, NR not reported, OMT (or O) optimal medical therapy, P p-value, RCT randomized controlled trial, Rehosp rehospitalization, SAVR (or S) surgical aortic-valve replacement, STS Society of Thoracic Surgeons, Super superiority trial design, TAVR (or T) transcatheter aortic-valve replacement, TVT transcatheter valve therapy.

Public comments sometimes cite the published clinical evidence and give CMS useful information.  Public comments that give information on unpublished evidence such as the results of individual practitioners or patients are less rigorous and therefore less useful for making a coverage determination.

CMS uses the initial public comments to inform its proposed decision.  CMS responds in detail to the public comments on a proposed decision when issuing the final decision memorandum.  All comments that were submitted without personal health information may be viewed in their entirety by using the following link https://www.cms.gov/medicare-coverage-database/details/nca-view-public-comments.aspx?NCAId=293.

Initial Comment Period: 6/27/2018 - 7/27/2018

During the initial 30-day public comment period, we received 98 comments. Of these 98 comments, three were omitted from publication on the CMS website due to excessive personal health information content, and two commenters posted twice. Commenters offered a variety of suggestions for modification to the existing policy, many of which focused on removing or reducing the current procedural volume requirements in favor of a policy that emphasizes TAVR program quality over procedural volume quantity. Some commenters supported maintaining volume requirements and some recommended CMS modify the policy based on the 2018 AATS/ACC/SCAI/STS Expert Consensus document (Bavaria et al, 2018). Many commenters also stressed the importance of ensuring patients, particularly in minority populations, with aortic stenosis are aware of, understand and have access to TAVR and all other treatment options for aortic stenosis.

The majority of comments were provided by physicians, other healthcare professionals and hospitals/health systems. Two comments were submitted by professional societies. One by the Association of Black Cardiologists (ABC) and the other on behalf of the American Association for Thoracic Surgery (AATS), American College of Cardiology (ACC), Society for Cardiovascular Angiography and Interventions (SCAI), and Society of Thoracic Surgeons (STS). Additional professional groups that offered comments were AdvaMed and the Heart Valve Disease Policy Task Force. Four comments were received from TAVR device manufacturers including Abbott, Boston Scientific Corporation, Edwards and Medtronic.

Second Comment Period: 3/26/2019 – 4/25/2019

During the 30-day comment period following the release of the proposed NCD and decision memorandum, CMS received 212 comments.  Of these 212, 35 were omitted from publication on the CMS website due to personal health information content.  The comments were generally supportive of CMS’ proposed NCD but a number of commenters recommended changes to the patient evaluation criteria and volume requirements.  Overall, commenters showed support for transitioning from volume requirements to outcome metrics.  Some support transitioning now while others agree that the measures are not yet developed or available.  Many comments supported the heart team and facility infrastructure requirements.  A few commenters suggested removing ‘symptomatic’ from the patient indications.  One commenter suggested that an NCD is no longer needed for TAVR.  While most commenters supported continued data collection through CED, one commenter, representing 14 organizations, supported the discontinuance of CED.  Detailed summaries with CMS responses are included below.

The majority of comments were provided by cardiologists and interventional cardiologists and other healthcare professionals.  There were four comments from professional associations, including America’s Health Insurance Plans (AHIP), the Federation of American Hospitals (FAH), AdvaMed, and professional societies, including the Society of Thoracic Surgeons (STS), the American College of Cardiology (ACC), the American Association for Thoracic Surgery (AATS), and the Society for Cardiovascular Angiography & Interventions (SCAI).  Four comments were from medical technology manufacturers, including Abbott, Boston Scientific, Edwards Lifesciences and Medtronic.  Four comments were from patient advocacy groups, including the Alliance for Aging Research, and many comments were from patients or caregivers sharing their positive personal experiences.  Twelve comments were submitted without providing an affiliation.

Seven articles were submitted with public comments that were not available or considered for the proposed decision.  These articles have been reviewed and included in the analysis section.

Access

Comment

Many commenters expressed support for general coverage of and increasing access to TAVR.  These commenters cited various barriers to access including geographical location (long travel time to a hospital that meets the NCD requirements) and provider preference. Other commenters stressed that TAVR offers a valuable treatment option for patients with aortic stenosis. 

Response

CMS agrees that ensuring patients have appropriate access to TAVR is important.  Having access also includes ensuring that patients have good quality care.  This decision is a balance of broad access but also ensuring Medicare patients have good quality care. 

Comment

We received many comments regarding volume requirements.  The comments ranged from advocating removing volume requirements because it limits access to the TAVR procedure to replacing volume requirements with outcomes and quality measures.  One commenter asserted that hospitals not be restricted from providing TAVR and another recommended eliminating all volume and other requirements so all patients can access TAVR. Two commenters called for CMS to make TAVR accessible to all patients regardless of other factors.

Response

As noted above, CMS believes it is important for beneficiaries to have access to TAVR.  We believe, based on the evidence reviewed, that volume and other requirements set forth in the NCD are important to ensure Medicare beneficiaries have access to TAVR in programs that deliver high quality care and ensure the best health outcomes for Medicare beneficiaries.

Comment

Some commenters asserted that annual procedural volume requirements place access burdens on minority and rural beneficiary populations and are barriers to smaller hospitals that serve them.  Three commenters supported ensuring access to TAVR in community programs with established or active surgical and interventional cardiology programs, two of whom noted the importance of access in public hospitals that provide charity care to many patients.

Response

We understand from stakeholders that the existing volume requirements may limit some hospitals from establishing a TAVR program, particularly those in rural areas and those serving large minority populations.  Because we believe that it is important for beneficiaries to have access to quality TAVR programs we have updated the volume requirements from the 2012 levels.  These new volume requirements are consistent with evidence that supports the need for procedural volume thresholds, while allowing for flexibility in meeting them.

Comment

One commenter requested access to TAVR valves for use in both the aortic valve and mitral valve.

Response 

We appreciate the comment.  Mitral valve is outside the scope of this NCD.

Comment

One commenter noted that barriers to accessing TAVR for prospective patients result in excess morbidity and mortality which, when due to undertreatment, likely exceed any difference in TAVR outcomes between high and low volume centers.  Another commenter noted that as the FDA expands approved indications for TAVR, more patients will be eligible for TAVR causing more patients to have to wait longer for treatment if volume requirements must be met.

Response

We understand the concerns commenters presented around the need for access and the anticipated limitations as FDA approved indications expand and more patients become eligible for TAVR and we believe it is important for beneficiaries to have access to TAVR.  We also believe, based on the evidence reviewed, that the volume and other requirements set forth in the NCD are important to ensure Medicare beneficiaries have access to TAVR in programs that deliver high quality care and ensure the best health outcomes for Medicare beneficiaries.

Comment

Several commenters noted that the proposed procedural volume requirements would increase the number of TAVR programs and access to TAVR.  One commenter asserted that there is no evidence of an access problem and that the proposals would result in new centers opening in already well serviced or saturated areas.  Another commenter recommended that access should be to high quality programs.  One commenter suggested tracking patient access or loss of access to inform a future reconsideration of the policy.

Response

As noted above, CMS agrees that ensuring patients have access to quality TAVR programs is important. The evidence submitted supported that there are many factors, including volume criteria, that predict good patient outcomes.  While volume may not be the sole measure to predict good patient outcomes, there is evidence to demonstrate that it is a good predictor of outcomes (Carroll 2017).  We added a CED question to ascertain whether we can lower procedural volume requirements when an outcome measure is fully developed. 

The evidence reviewed, including new evidence submitted, supports that volume requirements are part of the equation of predicting good quality.  Specifically the number of PCI, considered by commenters to be one of the more restrictive criteria, was reduced consistent with the multisociety consensus statement.  Further, the MEDCAC had greater than intermediate confidence that there is sufficient evidence to support procedural volumes and procedural volume requirements are consistent with the AATS/ACC/SCAI/STS updated professional society consensus statements.  We note that the AATS/ACC/SCAI/STS professional society consensus statements are based on the current state of evidence from trials, and studies based on the TVT registry, considered along with current practice patterns which was a major factor in this decision.  Therefore, based on the totality of the evidence, including all of the public comments and articles sent in response to the proposed NCD, we are maintaining the procedural volume requirements we proposed. While the procedural volume requirements continued to be required, there is greater flexibility in meeting these requirements.  We believe the hospital and heart team operator volume requirements reflect our intent to strike a balance between ensuring hospitals have the experience and capabilities to handle complex structural heart disease cases with an evolving evidence base and reducing the burden of unnecessary requirements on hospitals and patients which will allow broader access to quality TAVR programs. 

Indications

Comment

Many commenters supported maintaining coverage for FDA-approved indications. 

Response

We agree with commenters that coverage of TAVR for FDA-approved indications should continue.  

Comment

Several commenters requested or noted that "symptomatic" may need to be removed from the description of covered indications because FDA-approval for this expanded indication may be likely in the future.

Response

The decision to pursue aortic valve replacement should reflect an appropriate balance of risks and benefits, and mortality increases substantially after symptoms develop from severe aortic stenosis (Kitai 2011).  The current AHA/ACC guideline recommends surgical AVR in patients with severe aortic stenosis who report symptoms (Nishimura 2014). 

Comment

Commenters recommended covering low risk patients. 

Response

Since we are continuing coverage of TAVR for FDA-approved indications, low risk patients with symptomatic aortic stenosis will be eligible for coverage if the FDA approves this indication. 

Comment

One commenter encouraged CMS to keep coverage decisions abreast of current clinical trial data and FDA approvals.

Response

As noted above, under this NCD, we are continuing coverage for patients with symptomatic aortic stenosis for FDA-approved indications.  The NCD would not need to be reconsidered in order to cover patients with symptomatic aortic stenosis if the FDA approves new indications.  NCDs are established following an extensive evidence review which includes all publicly available evidence on current clinical trials.

Surgeon Examination

Comment:

Numerous commenters expressed support for the proposal to require one cardiac surgeon.  There were many reasons cited supporting a single cardiac surgeon examination.  Some commenters believed that the second surgeon requirement was unnecessary and eliminating this burden would be more efficient and save time for patients.  Another commenter supported requiring one cardiac surgeon to independently evaluate and sign off and a second cardiac surgeon to review the case and document rationale for the heart team. 

Response  

In the final NCD, CMS requires a cardiac surgeon and an interventional cardiologist to evaluate the patient face to face to determine suitability for TAVR, SAVR or any other therapy that is in the best interest of the patient.  We modified the proposed NCD from only a cardiac surgeon to include an interventional cardiologist because, as the public comments and evidence demonstrates, TAVR is a technically complex procedure with an evolving evidence base, and patients benefit from the multidisciplinary review, rather than a single physician.  We agreed with public commenters that to ensure the best patient health outcomes an interventional cardiologist as well as a cardiac surgeon should evaluate the patient suitability for TAVR.  This approach is also supported by the Consensus Statement.  It is important to have these evaluations documented and submitted to the heart team.  We believe this approach integrates multiple perspectives into a balanced, patient-centered care plan and encourages evidence-based medical care.

Comment

Some commenters suggested that CMS require an interventional cardiologist or cardiologist to also evaluate patients for suitability, stressing the importance of the decision being shared and unbiased. The same commenters also encouraged CMS to require that the surgeon and interventional cardiologist or cardiologist be part of the heart team.  One commenter asserted that this joint evaluation be performed in a multidisciplinary valve clinic.

Response 

As stated in the above response, we modified the proposed NCD from a single cardiac surgeon to also include an interventional cardiologist because as the public comments and evidence demonstrate, TAVR is a technically complex procedure with an evolving evidence base, and patients benefit from the multidisciplinary review.   We agreed with public commenters that to ensure the best patient health outcomes an interventional cardiologist as well as a cardiac surgeon must be part of the heart team and should evaluate the patient suitability for a TAVR.  We believe this approach integrates multiple perspectives into a balanced, patient-centered care plan and encourages evidence-based medical care.

Comment

Several commenters expressed support for allowing one TAVR operator to perform the evaluation, either a cardiac surgeon or an interventional cardiologist, and not limiting this requirement to a cardiac surgeon.  One commenter noted that a consultation by a cardiac surgeon is not needed for transfemoral TAVR procedures.

Response

We agree with the public comment that an interventional cardiologist should evaluate the patient.  However, we revised the final NCD to require both an interventional cardiologist and cardiac surgeon to evaluate the patient, document their evaluation, and present their rationale to the heart team.  We believe this final decision streamlines the evaluation of patients with aortic stenosis, places appropriate emphasis on multidisciplinary decision making, and is responsive to strong public support for the heart team concept.

In making this revision, we reviewed the multispecialty Consensus Statement that recommends a cardiac surgeon and another physician evaluate the patient as well as public comment that an interventional cardiologist should be involved in the evaluation.  We note that while an evaluation by an interventional cardiologist from the heart team is a new requirement in the final NCD, an interventional cardiologist in practice already performs an independent evaluation prior to participation in the procedure and should not be an additional burden.  TAVR is a complex intervention with an evolving evidence base.  We believe that both an interventional cardiologist and a cardiac surgeon bring distinctive perspectives to the heart team, and the combination fosters a balanced consideration of treatment options that is especially valuable at the extremes of surgical risk.  The multi-disciplinary heart team is a patient-centered concept that supports evidence-based medical decision making and promotes the best possible outcome. 

While not a requirement of this final NCD, we encourage continuation of the current practice wherein heart team interventional cardiologists and cardiac surgeons consult other specialists (e.g., cardiac imaging, cardiac anesthesia, and palliative care) as needed to refine treatment recommendations that are tailored to the circumstances of each patient. While not a requirement of this final NCD, we note that patients and referring physicians may continue to seek a second, independent opinion if desired. Finally, we strongly encourage heart teams to engage referring physicians in medical decision-making because referring providers are often able to offer a broader perspective that is grounded in a longitudinal patient relationship. 

Comment

Several commenters called for equipoise between TAVR and SAVR specific to this examination for suitability.  One commenter asserted that the NCD should encourage the comprehensive, collaborative evaluation and treatment of all patients with aortic stenosis at centers that provide both treatments, void of intrinsic bias.  Three commenters requested that a provider with TAVR expertise should be required to consult on SAVR cases.  Two commenters noted that all aortic stenosis patients should be evaluated by a heart team with a cardiac surgeon and interventional cardiologist or by physicians certified to perform TAVR to determine treatment.

Response 

We appreciate the recommendations specific to all aortic stenosis patients and SAVR.  We believe the heart team is important when considering all treatment options. However, the subject of this NCD is TAVR and we are unable to establish requirements specific to all aortic stenosis patients or SAVR with this decision.  Coverage of SAVR is at the discretion of the Medicare Administrative Contractors (MACs). 

Comment:

Several commenters expressed support for the revised language requiring suitability for SAVR, TAVR, medical or palliative therapy instead of only SAVR.  One commenter requested that CMS remove the requirement for the examination to be face-to-face so as to allow for coverage to continue as telemedicine advances.

Response

We appreciate the supportive comments.  We appreciate the comment to remove the face-to-face requirement for this examination based on continued advancements in telemedicine. However, face-to-face encounters contribute important context that informs patient assessment for this important medical decision.  We believe this visit must continue to take place in person.  To reduce patient burden, many programs have established a TAVR clinic that co-locates cardiac surgeons and interventional cardiologists and coordinates schedules between them. 

Heart Team

Comment

Eight commenters supported continued inclusion of the heart team in the proposed NCD with one commenter specifically requesting that the heart team should be used to determine which patients would benefit from SAVR as well, not just TAVR. One commenter noted that "advanced patient practitioners" is not a recognized term and should be changed to "advanced practice providers."

Response

We appreciate comments supporting the heart team and believe it continues to be an important element of this NCD.  We believe this decision further highlights the critical role of the heart team.  We remind readers that this analysis is focused on TAVR and we are unable to establish requirements specific to SAVR with this decision.  Coverage determinations for SAVR under § 1862(a)(1)(A) will be made by the MACs. Information on how to request an NCD is available at https://www.cms.gov/Medicare/Coverage/DeterminationProcess/howtorequestanNCD.html.

With regards to the term "advanced patient practitioners," we note that this term is used in the multispecialty society consensus statement.  As such, we have not revised it to "advanced practice providers." 

Shared Decision Making

Comment

Numerous commenters expressed support for the concept of shared decision making (SDM).  Several commenters encouraged CMS to work with stakeholders to ensure validated SDM tools are developed and implemented.  Other commenters clearly noted their support for requiring SDM, but recognized that TAVR decision aids are not available yet.  Two commenters asserted that there are currently publicly available SDM aids.  Several commenters requested that SDM use be mandated.  Two commenters recommended that the NCD clearly state that patient preferences must be considered.  One commenter requested that SDM be required only if it applies to all aortic valve disease patients and that instead of requiring a specific tool, encourage tools to comply with standards for high quality.  One commenter asserted that medical decisions, based on best available research, be made by the patient and their doctor, not bureaucrats.

Response

We recognize the importance of shared decision making in many clinical scenarios and have required shared decision making in other NCDs (for example, implantable cardiac defibrillators: https://www.cms.gov/medicare-coverage-database/details/ncd-details.aspx?NCDId=110&ncdver=4&NCAId=39&bc=ACAAAAAAQAAA&).  As part of SDM in current NCDs, we prefer to require the use of an evidence-based decision aid or tool.  We support patient shared decision making in AVR but there is not a fully developed tool available at this time. We note there are tools in development. For example, the Patient-Centered Outcomes Research Institute (PCORI) funded research conducted by Brennan and colleagues (CER-1306-04350/ NCT02266251), created and assessed a personalized decision assistance tool designed to evaluate important health outcomes comparing SAVR to TAVR replacement for operable patients with aortic valve disease considering aortic valve replacement, and to develop and assess a personalized risk assessment tool designed to evaluate expected health outcomes with TAVR for inoperable patients considering aortic valve replacement.

We strongly encourage standardized decision aids or tools [the National Quality Forum (NQF) has published standards for decision aids (www.qualityforum.org/Projects/c-d/Decision_Aids/Final_Report.aspx)] to facilitate the decision making process between a patient and physician and will be monitoring this space closely.

Joint Participation of Interventional Cardiologists and Cardiac Surgeons
Comment

Ten commenters supported our proposals to maintain the requirement that an interventional cardiologist and cardiac surgeon participate jointly in the intraoperative aspects of TAVR.  One specifically noted the importance of this requirement given the reduction in volume requirements to begin TAVR programs. 

Response

We appreciate the comments. 

Comment

Ten commenters challenged requiring operators to jointly participate and asserted that only one operator is needed.  These commenters offered various suggestions including only requiring two operators when needed (for example, non-transfemoral access sites), allowing an interventional cardiologist to perform TAVR without a cardiac surgeon in the room and allowing either a cardiothoracic surgeon or interventional cardiologist to perform TAVR alone. Another commenter noted that a cardiac surgeon is not required to perform TAVR so one should not be required to jointly participate in the procedure and instead only be required to be on the premises and rapidly available. Three commenters expressed support for two operators, but recommended that the combination of specifically one interventional cardiologist and one cardiac surgeon not be required.  Instead these commenters recommended allowing any combination of two operators (two interventional cardiologists, two cardiac surgeons or one of each) and also recommended that surgical support be available during TAVR procedures. 

Response

We appreciate the comments as well as recommendations.  The NCD continues the requirement from the 2012 NCD for both an interventional cardiologist and cardiac surgeon to jointly participate in the intraoperative technical aspects of TAVR. During our evidence review there was not any data demonstrating equivalent or improved outcomes with a single operator or alternate operator combination.  Further this requirement is supported by the Societies public comment. 

Hospital Infrastructure

Comment

Several commenters expressed their support of the hospital infrastructure requirements, specifically the requirement that TAVR programs have onsite heart valve surgery and interventional cardiology programs.

Response

We agree and appreciate the commenters’ support for our decision.

Procedural Volume Requirements

General Comments
Comment

Numerous commenters supported the proposed changes to the procedural volume requirements.  Two commenters expressed support and noted the need for requiring minimum volume requirements. 

Response

Thank you for your comment.

Comment

One commenter disagreed with the proposals noting that the requirements are too loose.

Response

We believe our decision strikes a balance between ensuring hospitals have the experience and capabilities to handle complex structural heart disease cases while limiting the burden and barriers unnecessary requirements may have on both hospitals and patients. 

Comment

Some commenters requested changes to the volume requirements so lower volume hospitals could perform TAVR procedures. 

Response

As described in the analysis section, the evidence reviewed, including new evidence submitted, supports that volume requirements are part of the equation of predicting good quality.  Specifically the number of PCI, considered by commenters to be one of the more restrictive criteria, was reduced consistent with the multisociety consensus statement.  Further, the MEDCAC had greater than intermediate confidence that there is sufficient evidence to support procedural volumes and procedural volume requirements are consistent with the AATS/ACC/SCAI/STS updated professional society consensus statements.  We note that the AATS/ACC/SCAI/STS professional society consensus statements are based on the current state of evidence from trials, and studies based on the TVT registry, considered along with current practice patterns which was a major factor in this decision.  Therefore, based on the totality of the evidence, including all of the public comments and articles sent in response to the proposed NCD, we are maintaining the procedural volume requirements we proposed. While the procedural volume requirements continued to be required, there is greater flexibility in meeting these requirements.  We believe the hospital and heart team operator volume requirements reflect our intent to strike a balance between ensuring hospitals have the experience and capabilities to handle complex structural heart disease cases with an evolving evidence base and reducing the burden of unnecessary requirements on hospitals and patients which will allow broader access to quality TAVR programs. 

Comment

One commenter requested that CMS allow numbers from different facilities with the same operators to be combined to meet volume requirements.

Response

We believe procedural volume requirements are important to ensure individual hospitals have the experience and capabilities to handle complex structural heart disease cases.  As such, multiple facilities may not combine numbers to meet the hospital volume requirements set forth in this NCD.

Comment

One commenter noted that the proposed changes could result in significantly poorer outcomes for patients with critical aortic stenosis and another commenter asserted that the proposals would increase the number of low volume TAVR programs at increased risk for suboptimal outcomes. 

Response

As stated earlier, the evidence submitted supported that there are many factors, including volume criteria that predict good patient outcomes.  We believe the procedural volume requirements are generally consistent with the AATS/ACC/SCAI/STS updated professional society consensus statement.  While the procedural volume requirements continue to be required, there is greater flexibility in meeting these requirements.  We believe the hospital and heart team operator volume requirements reflect our intent to strike a balance between ensuring hospitals have the experience and capabilities to handle complex structural heart disease cases with an evolving evidence base and reducing the burden of unnecessary requirements on hospitals and patients which will allow broader access to quality TAVR programs. 

Comment

One commenter expressed concern that volume requirements may cause hospitals to encourage patients to undergo certain procedures in order to meet thresholds instead of providing the best treatment option. 

Response

While we understand this commenter’s concern, we also note that physicians are obligated to ensure patients undergo the best treatment plan for each individual patient’s situation and should assist them in obtaining this treatment. We believe the new flexibility to meet global AVR volume requirements without specific SAVR versus TAVR numbers (but with a minimal volume required for TAVR) will mitigate incentives for providers to favor one procedure over the other. Furthermore, a reduction in the volume requirement would likely increase access but at the expense of patient safety.

Comment

One commenter noted that the proposed volume requirements are inconsistent with 1) the expert consensus document; 2) recently published findings on inverse association of mortality with hospital and operator TAVR volume, 3) new data on inverse mortality with hospital SAVR volume, 4) major limitations in assessing TAVR and SAVR performance and outcomes when volumes are low.  This commenter requested that the volume requirements be increased as supported by Vemulapalli et al. (2019) who describe a clear relationship between hospital and operator procedural volume and 30-day risk adjusted mortality as well as Salemi et al. (2019) who demonstrated an inverse relationship between operator volume and in-hospital outcomes.  Another commenter stated that Vemulapalli et al. (2019) use statistical arguments that lack medical meaning as performing statistical comparisons on huge data sets is known to show statistical significance for minimal absolute differences in measured variables.  This commenter noted that the article by Vemulapalli et al. (2019) appears to include mortality and 30-day outcome data that is not relevant or statistically credible due to the absence of data on a large number of patients.  This commenter contended that the study does not demonstrate a volume/outcome relationship.

Response

As stated earlier in this document and in the analysis section, the evidence submitted supported that there are many factors, including volume criteria, that predict good patient outcomes.  While volume may not be the sole measure to predict good patient outcomes, there is evidence to demonstrate that it is a good predictor of outcomes (Carroll 2017).  We added a CED question to ascertain whether we can lower procedural volume requirements when an outcome measure is fully developed. 

The evidence reviewed, including new evidence submitted, supports that volume requirements are part of the equation of predicting good quality.  Specifically the number of PCI, considered by commenters to be one of the more restrictive criteria, was reduced consistent with the multisociety consensus statement.  Further, the MEDCAC had greater than intermediate confidence that there is sufficient evidence to support procedural volumes and procedural volume requirements are consistent with the AATS/ACC/SCAI/STS updated professional society consensus statement.  We note that the AATS/ACC/SCAI/STS professional society consensus statement are based on the current state of evidence from trials, and studies based on the TVT registry, considered along with current practice patterns which was a major factor in this decision.  Therefore, based on the totality of the evidence, including all of the public comments and articles sent in response to the proposed NCD, we are maintaining the procedural volume requirements we proposed. While the procedural volume requirements continue to be required, there is greater flexibility in meeting these requirements.  We believe the hospital and heart team operator volume requirements reflect our intent to strike a balance between ensuring hospitals have the experience and capabilities to handle complex structural heart disease cases with an evolving evidence base and reducing the burden of unnecessary requirements on hospitals and patients which will allow broader access to quality TAVR programs. 

Qualifications to Begin for Programs Without TAVR Experience

Comment

Several commenters expressed support for the proposed requirements to begin a TAVR program.

Response

Thank you for your comments.

Comment

One commenter requested that CMS reduce any preexisting requirements to facilitate access to TAVR.  Another commenter disagreed with reduced volume requirements to begin a TAVR program. 

Response

We believe our decision strikes a balance between ensuring hospitals have the experience and capabilities to handle complex structural heart disease cases while limiting the burden and barriers unnecessary requirements may have on both hospitals and patients. 

Comment

We had several comments that made recommendations on our volume requirements.  These comments ranged from specific number of open heart surgeries, SAVRs, and aortic valve procedures.  Further, the commenters requested that we broadly define cardiac surgical aortic valve.

Response

We appreciate the thoughtful comments.  As stated previously, the evidence reviewed, including new evidence submitted, supports that volume requirements are part of the equation of predicting good quality.  The MEDCAC had greater than intermediate confidence that there is sufficient evidence to support procedural volumes and procedural volume requirements are consistent with the AATS/ACC/SCAI/STS updated professional society consensus statements.  We note that the AATS/ACC/SCAI/STS professional society consensus statement is based on the current state of evidence from trials, and studies based on the TVT registry, considered along with current practice patterns which was a major factor in this decision.  Therefore, based on the totality of the evidence, including all of the public comments and articles sent in response to the proposed NCD, we are maintaining the procedural volume requirements we proposed. While the procedural volume requirements continued to be required, there is greater flexibility in meeting these requirements. 

Comment

Two commenters asserted that requiring 10 aortic valve procedures a year equates to having no substantial cardiac surgical program or meaningful valvular heart disease program to start a TAVR program and reflects inadequate expertise and infrastructure.

Response

We appreciate the comment.  We believe the hospital volume requirements to begin a TAVR program reflect our intent to strike a balance between ensuring hospitals have the experience and capabilities to handle complex structural heart disease cases with an evolving evidence base and reducing the burden of unnecessary requirements on hospitals and patients which will allow broader access to quality TAVR programs. 

Comment

A few commenters recommended specific volume requirements for surgeons and interventional cardiologists.  The numbers were different depending on the commenter.  For example, one commenter recommended that a surgeon should be required to perform 75-100 open heart cases in the previous 12 months with at least 25 being aortic cases before beginning a TAVR program.  This commenter also recommended that an interventional cardiologist perform at least 50 TAVRs instead of 50 structural heart disease cases before beginning a TAVR program. 

Response

We appreciate the recommended modifications to the proposed volume requirements for heart team physicians in new programs.  As noted in the above response, based on the totality of the evidence, including all of the public comments and articles sent in response to the proposed NCD, we are maintaining the procedural volume requirements we proposed. While the procedural volume requirements continued to be required, there is greater flexibility in meeting these requirements.  We believe the volume requirements to begin a TAVR program reflect our intent to strike a balance between ensuring operators have the experience and capabilities to handle complex structural heart disease cases with an evolving evidence base and reducing the burden of unnecessary requirements on hospitals and patients which will allow broader access to quality TAVR programs. 

Requirements to Maintain for Programs with TAVR Experience

Comment

Two commenters expressed support for the proposed volume requirements for hospitals with TAVR experience.  One commenter expressed support for programs maintaining a combined volume of TAVR and SAVR procedures. 

Response

We appreciate the comments supporting this provision of the NCD.

Comment

One commenter requested that CMS require "aortic valve related procedures" as used in the proposed requirements to begin a TAVR program instead of "AVRs (TAVR or SAVR)" to describe requirements for hospitals to maintain TAVR programs.

Response

We believe it is important for hospitals with TAVR experience to specifically perform AVRs, not just aortic valve related procedures. The flexibility in this language is intended to allow facilities without TAVR experience to begin TAVR programs based on their experience with aortic valve procedures.  Programs already performing AVRs must continue to do so as we believe AVRs, with a minimum number of TAVR procedures, is important to ensure hospitals have the experience and capabilities to handle complex structural heart disease cases.

Comment

We had many commenters make a variety of recommendations regarding the volume procedure requirements.  These commenters ranged from requiring 30 TAVR procedures to 100 TAVR procedures annually.  There were also a number of comments that made suggestions on the types of procedures to meet the volume requirements.  

Response

As stated in the analysis, the evidence reviewed, including new evidence submitted, supports that volume requirements are part of the equation of predicting good quality.  Further, the MEDCAC had greater than intermediate confidence that there is sufficient evidence to support procedural volumes and procedural volume requirements are consistent with the AATS/ACC/SCAI/STS updated professional society consensus statements.  We note that the AATS/ACC/SCAI/STS professional society consensus statement is based on the current state of evidence from trials, and studies based on the TVT registry, considered along with current practice patterns which was a major factor in this decision.  Therefore, based on the totality of the evidence, including all of the public comments and articles sent in response to the proposed NCD, we are maintaining the procedural volume requirements we proposed. While the procedural volume requirements continue to be required, there is greater flexibility in meeting these requirements.  We believe the hospital volume requirements reflect our intent to strike a balance between ensuring hospitals have the experience and capabilities to handle complex structural heart disease cases with an evolving evidence base and reducing the burden of unnecessary requirements on hospitals and patients which will allow broader access to quality TAVR programs. 

Comment

A commenter noted that the proposals could allow only TAVRs to be performed in a hospital without any SAVRs which could undermine the role of the cardiac surgeon and surgical expertise available in the hospital. 

Response

The NCD requires TAVR to be furnished in a hospital with appropriate infrastructure that includes an on-site heart valve surgery program. As such we expect hospitals will be performing and maintaining proficiency in SAVR as well as TAVR.

Comment

One commenter asserted that increased volume requirements are needed to ensure a large enough number of TAVR procedures to serve as a denominator to properly measure quality of clinical outcomes.  Another commenter noted that 20 TAVRs per year with a three year rolling period would allow for appropriate statistical power to measure composite outcomes rigorously which exceeds the number of cases the STS requires to receive a star rating for SAVR.

Response

As stated earlier in this document and in the analysis section, the evidence submitted supported that there are many factors, including volume criteria, that predict good patient outcomes.  While volume may not be the sole measure to predict good patient outcomes, there is evidence to demonstrate that it is a good predictor of outcomes (Carroll 2017).  We added a CED question to ascertain whether we can lower procedural volume requirements when an outcome measure is fully developed. 

Percutaneous Coronary Intervention (PCI) Volume Requirements

Comment

Many commenters strongly supported the reduction of PCIs from 400 to 300 per year. 

Response

We appreciate the supportive comment.

Comment

Many commenters requested revisions to the PCI requirement.  These comments stated that the number was too high, would limit the number of TAVR programs, and there was no link between the number of PCIs and defining a quality TAVR program.  One commenter recommended basing the PCI requirement on the patient area served and supported fewer than 300 per year.  Some commenters recommended 200 PCIS annually.   Further, some commenters asserted that a PCI volume requirement is inappropriate altogether.  Another commenter suggested incorporation of cardiac catheterization lab procedural volumes at 500 per year to include PCIs and structural procedures requiring a skill set more similar to TAVR. 

Response

Based on the evidence, we believe including a PCI volume requirement is important to ensure that hospital programs have sufficient experience with TAVR access issues and procedural complications which may require vascular arterial interventions. Further, the evidence reviewed, including new evidence submitted, supports that volume requirements are part of the equation of predicting good quality.  Specifically the number of PCI, considered by commenters to be one of the more restrictive criteria, was reduced consistent with the multisociety consensus statement.  Further, the MEDCAC had greater than intermediate confidence that there is sufficient evidence to support procedural volumes and procedural volume requirements are consistent with the AATS/ACC/SCAI/STS updated professional society consensus statement.  We note that the AATS/ACC/SCAI/STS professional society consensus statement is based on the current state of evidence from trials, and studies based on the TVT registry, considered along with current practice patterns which was a major factor in this decision.  Therefore, based on the totality of the evidence, including all of the public comments and articles sent in response to the proposed NCD, we are maintaining the procedural volume requirements we proposed.  We also note that all individual hospitals must meet the requirements set forth in this final decision and exceptions or waivers for those that do not are not allowable under this policy.

Individual Operator Requirements

Comment

One commenter agreed that the proposed operator training requirements are adequate. 

Response

Thank you for the comment.

Comment

A commenter requested that the volume requirements for individual operators be reduced.  Several commenters asserted that volume requirements should be specific to each operator instead of team based to ensure inexperienced operators do not perform TAVR.  Some of these commenters recommended that primary operators meet requirements of 100 career structural heart disease procedures and 30 left sided procedures to begin and 20 TAVRs per year or 40 over two years to maintain.  One commenter recommended including individual operator volume requirements of 40-50 TAVRs per year.  Citing Vemulapalli et al. (2019), another commenter recommended that TAVR operators maintain a volume of 40 TAVRs per year.

One commenter noted a need for formal operator training requirements to ensure sufficient understanding of imaging modalities and root/peripheral anatomy as well as a need for expertise in large bore access and managing related complications.  One commenter also called for dedicated training requirements for operators or for CMS to at least define "structural heart procedure."  Another commenter recommended reviewing literature on the proficiency of TAVR implanters based on volumes, noting that empiric quality and safety data should be used to establish the most appropriate volume requirements for animplanter.  One commenter asserted that it is not CMS’ role to define procedural volume requirements for board certified practitioners.  Instead criteria should be defined by certifying boards and the judgment of individual board-certified operators.

Response

The operator requirements are based on stakeholder input, published evidence and the society consensus statement.  Since the issuance of the 2012 NCD, there has been no new evidence assessing individual interventional cardiologist experience volume requirement effects on TAVR patient outcomes. As noted in the above responses, based on the totality of the evidence, including all of the public comments and articles sent in response to the proposed NCD, we are maintaining the procedural volume requirements we proposed. While the procedural volume requirements continue to be required, there is greater flexibility in meeting these requirements.  We believe the volume requirements to begin a TAVR program reflect our intent to strike a balance between ensuring operators have the experience and capabilities to handle complex structural heart disease cases with an evolving evidence base and reducing the burden of unnecessary requirements on hospitals and patients which will allow broader access to quality TAVR programs.

Volume Requirement Misalignment

Comment

Some commenters expressed concerns with the increase in procedural volume requirements for hospitals to begin TAVR programs and then to maintain their TAVR programs.  One commenter recommended that the number of procedures to begin should either increase or the number to maintain should decrease.  Another commenter called for better alignment of initial and maintenance aortic valve related procedure volume requirements. Two commenters noted that by relaxing the requirements to begin a TAVR program and then expanding the requirements to maintain, programs will not be able to maintain necessary volume requirements.  One commenter noted a five fold increase in volume requirements for aortic valve related procedures between requirements to begin a TAVR program and requirements to maintain a TAVR program. The same commenter made several recommendations on specific volume numbers CMS should adopt in the NCD. 

Response

We appreciate these comments pointing out the potential for new TAVR programs to become unsustainable due to the increase in procedural volume requirements between beginning a TAVR program and maintaining a TAVR program.  As described in the analysis section, the evidence reviewed, including new evidence submitted, supports that volume requirements are part of the equation of predicting good quality.  Specifically the number of PCI, considered by commenters to be one of the more restrictive criteria, was reduced consistent with the multisociety consensus statement.  Further, the MEDCAC had greater than intermediate confidence that there is sufficient evidence to support procedural volumes and procedural volume requirements are consistent with the AATS/ACC/SCAI/STS updated professional society consensus statement.  We note that the AATS/ACC/SCAI/STS professional society consensus statement is based on the current state of evidence from trials, and studies based on the TVT registry, considered along with current practice patterns which was a major factor in this decision.  Therefore, based on the totality of the evidence, including all of the public comments and articles sent in response to the proposed NCD, we are maintaining the procedural volume requirements we proposed. While the procedural volume requirements continued to be required, there is greater flexibility in meeting these requirements.  We believe the hospital and heart team operator volume requirements reflect our intent to strike a balance between ensuring hospitals have the experience and capabilities to handle complex structural heart disease cases with an evolving evidence base and reducing the burden of unnecessary requirements on hospitals and patients which will allow broader access to quality TAVR programs. 

CED

Comment

Commenters expressed support for continued coverage of FDA-approved clinical trials, which are covered under CED, with one commenter requesting that CMS permit updates in design and platforms that may not reflect commercial requirements when performed as part of an FDA-approved trial.  Another commenter expressed support for the proposed CED questions and requested clarification around what "composite metric" CMS is referring to in CED question iv, noting that it appears to be the STS/ACC 30-day mortality/morbidity metric that is in development. 

Response

We appreciate the supportive comments for CED.  With regards to proposed CED question iv, the question is not specific to a particular metric, just that the question must be addressed through a composite metric. 

Comment

One commenter, representing 14 organizations, asserted that continuing CED is an unnecessary regulatory burden and cost to providers and contrary to CMS’ focus on "patients over paperwork."  This commenter requested that CMS end CED because available evidence is sufficient to conclude that TAVR improves health outcomes for beneficiaries with symptomatic severe aortic stenosis for FDA-approved indications.  This commenter further challenged CMS’ explanation that CED should continue in order to generate more evidence specific to minority populations, stating that continuing CED will instead continue to limit access to TAVR for minority populations.

Response

We disagree that CED should end at this time.  Specifically, we note that while evidence is supportive of TAVR for specific patient populations, the indications for TAVR are changing, and there is ongoing need to track the incidence of paraventricular regurgitation and the need for permanent pacemaker implantation.  We concluded from our analysis that gaps in the current evidence base lead to uncertainty about the overall impact of TAVR on beneficiary outcomes when furnished outside of the setting of evidence development or clinical trial protocols, and data should continue to be collected through a minimum of one year post TAVR.

We also noted in our analysis that while recent trials in low surgical risk patients are encouraging, the long-term durability of the latest-generation TAVR valves in these patients remains unknown (while SAVR valves have been demonstrated to last 10-20 years depending on valve type).  Additionally, we believe that continued data collection and research that includes registry reporting for all risk groups is necessary for the underlying analysis that will support development and validation of emerging, risk-adjusted outcome measures which we (along with all stakeholders) hope will replace volume requirements for evaluation of TAVR proficiency of operators, heart teams and hospitals. 

Registry Participation

Comment

Many commenters supported continued registry participation as proposed.  One commenter supported adding a publicly reportable composite metric in the registry that explores procedure-related actors and patient health outcomes. 

Response

We appreciate the supportive comments.

Comment

One commenter requested that CMS ensure the registry includes a way to track regulatory compliance and ensure patient safety.  Specifically this commenter requested the registry be sensitive to new observations and subject to change by regulatory or scientific findings; monitoring be used to ensure evidence for new subpopulations is of equal or greater quality and allow for analyses of these subpopulations; and the process should be flexible and capture additional monitoring about provider reported complications.

Response

We agree that the registry plays an important role in regulatory compliance and that patient safety is paramount.  Under the 2012 NCD and with this final determination, CMS approves registries that meet the requirements set forth in the NCD. We have not added requirements around regulatory compliance and patient safety because the registry's purpose for CMS is to generate evidence on the evidence gaps cited in the NCD. 

Comment

Two commenters requested that CMS require completeness and accuracy of data so as to support quality measurement and answer CED questions.  Another commenter recommended that CMS consider how to effectively streamline data collection for the registry while ensuring that CED questions can be answered.  A different commenter recommended that CMS adopt principles to improve the value and governance of approved registries noting a need to limit the size and scope of registries so they are less burdensome and more transparent and useful.  Another commenter suggested using existing health information for registry/CED requirements to minimize paperwork requirements that could limit access by deterring providers from offering TAVR.  One commenter stated that the STS/ACC TVT Registry is committed to improving by decreasing the burden of data collection while improving TAVR standards and outcomes. 

Response

We agree that the registry plays an important role in regulatory compliance and that patient safety is paramount.  We also agree with commenters that completeness and accuracy are important to the integrity of the registry.  We recognize the costs associated with implementation and management of registries and aim to minimize burden related to registry participation.  Under the 2012 NCD and with this final determination, CMS approves registries that meet the requirements set forth in the NCD.  This NCD does not establish further restrictions or requirements on the registries.  We have not added requirements on completeness and accuracy of data in the registry or limitations to the size and scope of the registry.  We believe these issues are best addressed by TAVR thought leaders and stakeholders.  We encourage approved registries to engage in continuous quality improvement efforts to improve data completeness and accuracy while streamlining collection so as to minimize burden for participants.  We are pleased to hear that the STS/ACC TVT Registry is committed to decreasing the burden of data collection while improving TAVR standards and outcomes.  When available, a composite outcome measure will contribute valuable information that informs improvements in patient outcomes.

Comment

One commenter supported an all women registry due to the different outcomes seen in women versus men as well as inclusion of women in clinical trials and the ability to separately analyze data by gender.  Another commenter requested that the 5 minute walk test and KCCQ be removed as they are unnecessary.

Response

CMS supports the generation of information specific to all minority populations and encourages registries to collect and analyze data to explore outcomes specific to these populations.  CMS reminds readers that the NCD requires registries to track quality of life outcomes and registries determine the tools that best meet this requirement.  At the time of writing, the STS Registry used TAVR Data Collection Form v2.1 which included fields for the KCCQ-12 pre-procedure and at follow up intervals.  The form also includes fields for the 5 meter walk test at the same intervals (https://www.ncdr.com/WebNCDR/docs/default-source/tvt-public-page-documents/2-1_tvt_tavrdcf.pdf?sfvrsn=2).

Outcomes/Quality Measures

Comment

Many commenters expressed support for the use of outcomes or quality measures specific to TAVR to provide an assessment of hospital performance.  Many commenters supported discontinuation of the procedural volume requirements for Medicare coverage purposes once measures are developed.  Some commenters specifically requested that hospitals be required to report on their TAVR outcomes to determine, ensure and maintain quality.  One commenter requested that public reporting of hospital based health outcomes data be used instead of annual procedural volume requirements.  One commenter requested clarification around the timing for replacing procedural volume requirements with outcome measures and suggested it be accomplished during a future reconsideration of the NCD to remove CED requirements. 

One commenter recommended that hospitals meet and maintain robust quality metrics including 30-day risk-adjusted mortality and complications and one year outcomes including survival and patient reported outcomes.  This commenter also noted it is ideal to use a 3 star public reporting outcome metric harmonized with outcomes reporting for SAVR. 

Response

CMS agrees that validated outcome measures may be an appropriate alternative to procedural volume requirements when establishing quality standards for TAVR programs.  We will continue to follow the STS/ACC TVT Registry’s progress in developing TAVR specific outcome measures.  As we state in this document, CMS may reconsider this NCD to review replacing procedural volume criteria with an outcome metric, however it is too premature to predict any timing around a future reconsideration.  The CMS Innovation Center is similarly exploring application of a composite outcome measure to the Bundled Payments for Care Improvement Advanced TAVR Clinical Episode, which could also be captured through the STS/ACC TVT Registry.

Comment

One commenter requested that CMS do more to assure optimal patient care and continuous quality improvement given the expected increase in new sites with lower procedural volume requirements.  This commenter recommended that hospitals be required to use benchmark quality reports from the TVT registry and implement quality improvement actions if low performance for consecutive quarters is documented. 

Response

As noted above, validated outcome measures may be an appropriate alternative to procedural volume requirements in determining TAVR program proficiency.  Any other quality based modifications to the NCD requirements could be considered should a future NCD reconsideration address the establishment of TAVR outcome measures.

Comment

Two commenters suggested CMS require an external certification process by an approved credentialing organization (like the Joint Commission) to ensure TAVR programs meet requirements and quality standards.

Response

Thank you for the comment.  At this time, we are not requiring credentialing. 

Public Reporting

Comment

Some commenters expressed support for public reporting of TAVR data.  Four commenters specifically requested that CMS grant the public access to data on TAVR, two commenters requested that CMS require public reporting of outcome measures and two other commenters noted the importance of open reporting of outcomes as essential to high quality and successful TAVR programs. 

One commenter stated that the NCD must specify mechanisms for patients and families to have confidence that a hospital meets reasonable performance metrics.  One commenter requested that, should the registry requirement continue, CMS require the registry to make in-hospital and 30-day adjusted mortality data publicly available at the hospital level.  This commenter noted that the New York Department of Health reports hospital outcomes for TAVR so the registry can and should as well, despite previous indications by the registry that a hospital’s data can be made public only with the permission of the hospital.  One commenter requested that Hospital Compare include hospital specific outcomes data on in-hospital and 30-day risk adjusted mortality for SAVR and TAVR.  Another commenter requested CMS require outcome studies for each specific medical facility and require that information be provided to patients upon request.  One commenter suggested requiring a flexible public reporting requirement and noted that the registry will start public reporting in 2020, similar to 3-star system used for SAVR, with commercial transfemoral TAVR volume, in-hospital and 30-day risk adjusted mortality.

Response

CMS supports making valid and reliable outcome data publicly available because data transparency facilitates patient engagement and promotes patient centered care.  Our decision requires that the results of questions i. through iv. in A.7 be reported publicly as described in CED criterion k in section B of the NCD.  We note that criterion k includes a requirement that study results, including all prespecified outcomes, be made public within 12 months of a study’s primary completion date. 

CMS will not add further reporting requirements through this final decision memorandum.  CMS also does not add reporting requirements to the Hospital Compare program through the NCD process. However, we note that approved registries and TAVR programs have the ability to publicly release facility specific TAVR data independent of CMS payment rules.  CMS encourages the TVT registry to refine valid and reliable outcome measures, which may be used for multiple purposes: (1) to inform patient treatment choices; (2) to meet TAVR program quality requirements in the future; and (3) to apply to CMS payment models like the BPCI Advanced payment model.

Aortic Stenosis Treatment Options

Comment

Numerous commenters requested equipoise between all aortic stenosis treatment options.  Several commenters asserted that the NCD should focus on aortic stenosis and address all aortic valve replacement therapies instead of focusing on TAVR only. One commenter noted the need to address barriers to treatment for aortic stenosis.  Some commenters, focusing on patient evaluation requirements, stated that an interventional cardiologist should consult on SAVR patients and all aortic valve patients.  Several commenters noted the importance of the heart team for all patients considering aortic valve replacement.  One commenter noted that patients should have equal access to both TAVR and SAVR and another called for equitable access to all treatment options, quality and transparency.  Another commenter requested that SAVR be held to the same standards and outcomes as TAVR.  One commenter requested that CMS require hospitals to participate in a national registry that provides regular site performance and national benchmarking reports for both TAVR and SAVR.  One commenter questioned whether isolated SAVR hospitals should receive Medicare reimbursement.

Response

We appreciate the recommendations specific to all aortic stenosis patients and SAVR, however, the subject of this NCD is TAVR and we are unable to establish requirements specific to all aortic stenosis patients or SAVR with this decision.  Coverage of SAVR is at the discretion of the MACs. 

Additional Comments

Comment

One commenter asserted that the decision should consider that TAVR is less invasive, less insulting to the brain, safer and likely less expensive.  Another commenter stated that reimbursement rules must consider physiology differences in women.  One commenter noted that CMS should give substantial weight to the studies based on information from the STS/ACC TVT Registry that report favorable patient outcomes of the TAVR procedure.

Response

CMS performs an extensive literature review to consider all available and relevant data when developing NCDs which may include the types of data noted by commenters.

Comment

One commenter questioned whether bills for Medicare Advantage plan participants undergoing TAVR under this NCD should be sent to the beneficiary’s Medicare Advantage plan or traditional Medicare.

Response

Because TAVR is covered under an NCD, all TAVR claims for Medicare Advantage plan participants should be submitted to the applicable Medicare Advantage plan. Further information on this is available in the Medicare Claims Processing Manual (Pub. 100-04), chapter 32, section 290.4.

Comment

One commenter requested that CMS pay for permanent pacemaker procedures after TAVR, if required, because it is a complication that is not under the TAVR operators’ control and can occur with or without anticipation.

Response

This analysis and final NCD are limited to conditions of coverage for TAVR.  Coverage of cardiac pacemakers is addressed in NCD 20.8.

Comment

One commenter asserted that CMS must be more diligent in monitoring programs that are performing TAVR as the commenter is aware of programs that do not meet the NCD requirements.

Response

All requirements set forth in this NCD must be met to be eligible for Medicare coverage of TAVR.  Hospitals are still responsible for ensuring that they meet the NCD requirements. 

VIII. CMS Analysis

National coverage determinations are determinations by the Secretary with respect to whether or not a particular item or service is covered nationally by Medicare (§1869(f)(1)(B) of the Act). In order to be covered by Medicare, an item or service must fall within one or more benefit categories contained within Part A or Part B, and must not be otherwise excluded from coverage. Moreover, with limited exceptions, the expenses incurred for items or services must be reasonable and necessary for the diagnosis or treatment of illness or injury or to improve the functioning of a malformed body member (§1862(a)(1)(A) of the Act).

In addition to §1862(a)(1)(A) of the Act, a second statutory provision may permit Medicare payment for items and services in some circumstances. That statute, section 1862(a)(1)(E) of the Act, provides, in pertinent part, that:

Section 1142 of the Act describes the authority of the AHRQ to conduct and support research on outcomes, effectiveness, and appropriateness of services and procedures to identify the most effective and appropriate means to prevent, diagnose, treat, and manage diseases, disorders, and other health conditions. That section includes a requirement that the Secretary assure that AHRQ research priorities under Section 1142 appropriately reflect the needs and priorities of the Medicare program.

CED is a paradigm whereby Medicare covers items and services on the condition that they are furnished in the context of approved clinical studies or with the collection of additional clinical data. In making coverage decisions involving CED, CMS decides after a formal review of the medical literature to cover an item or service only in the context of an approved clinical study or when additional clinical data are collected to assess the appropriateness of an item or service for use with a particular beneficiary.

The 2014 CED Guidance Document is available at https://www.cms.gov/medicare-coverage-database/details/medicare-coverage-document- details.aspx?MCDId=27.

When making national coverage determinations, we evaluate the evidence related to our analytic questions based on the quality, strength and totality of evidence presented in the reviewed literature. As part of this evaluation, it is important to consider whether the evidence is relevant to the Medicare beneficiary population. In determining the generalizability of the results of the body of evidence to the Medicare population, we consider, at minimum, the age, race and gender of the study participants.

Evidence Review Summary:

For this reconsideration, CMS focused on the following questions:

• Is the evidence sufficient to conclude that TAVR improves health outcomes for Medicare beneficiaries with cardiac symptoms and severe aortic stenosis who are not candidates for SAVR?
• Is the evidence sufficient to conclude that TAVR improves health outcomes for Medicare beneficiaries with cardiac symptoms and severe aortic stenosis who are candidates for SAVR, and are at either high or intermediate surgical risk?

Is the evidence sufficient to conclude that TAVR improves health outcomes for Medicare beneficiaries with cardiac symptoms and severe aortic stenosis who are not candidates for SAVR?

There are two major randomized trials (PARTNER 1B and PARTNER 2B) and one major non-randomized study (CoreValve B) that evaluated TAVR in patients with cardiac symptoms and severe aortic stenosis who are not candidates for surgical aortic valve replacement (are inoperable). The PARTNER 2B and CoreValve B trials are new since our 2012 NCD. While the PARTNER 1 trials were reviewed in depth in our 2012 NCD, we summarize those trials and our analysis of key health outcomes, assessing both benefits and harms, as this is the basis for our discussion of all subsequent TAVR trials which are the focus of this NCD reconsideration.

PARTNER consisted of two parallel, prospective, multi-center, randomized, active-treatment-controlled trials (PARTNER 1A and 1B), both of which used first-generation, balloon-expandable transcatheter valves. "Severe aortic stenosis" was determined by a standard echocardiographic definition; and in 93% of all patients cardiac symptoms were severe (NYHA class III or IV PARTNER 1B, which was completed and published first, was a superiority trial demonstrating that in patients who were not surgical candidates, transfemoral TAVR substantially reduced (by 20 percentage points) 1-year death from any cause, the primary outcome in the intention-to-treat analysis (Leon 2010).

For secondary outcomes, compared to the medical treatment group, the TAVR group had significantly reduced cardiac symptoms (by NYHA classification), improved function (by the 6-min walk test), and fewer hospital readmissions (Leon 2010). Other quality of life measures were not reported. TAVR patients however had significantly greater rates of major vascular complications (1 of 6 patients in absolute terms), major bleeding (1 of 5 patients), and of strokes, including major strokes (approximately double the rate for TAVR versus medically treated patients).

A second study, the Medtronic CoreValve® U.S. Pivotal Trial (Popma 2014; herein referred to as the "CoreValve B"), evaluating TAVR in inoperable patients with severe aortic stenosis and cardiac symptoms was the non-randomized companion study to the CoreValve A randomized trial (discussed below). CoreValve B compared inoperable patients who underwent TAVR with a lower-profile, self-expanding valve to a historic control. The CoreValve B investigators added major stroke to the PARTNER primary outcome of death from any cause, creating a composite 1-year primary outcome, due to widespread concern about the elevated stroke rate associated with TAVR seen in PARTNER.

The non-randomized CoreValve B study demonstrated a substantial reduction in this primary outcome compared to the historical control. Secondary outcomes allowed comparison to inoperable PARTNER B and TVT registry patients. Comparisons to PARTNER 1B however should be made with caution as PARTNER 1B patients may have been somewhat sicker, with an overall mean STS PROM risk score of 11.6% compared to 10.3% in the CoreValve B study; moreover, PARTNER 1B used the more rigorous intention-to-treat, while CoreValve B used as- treated, statistical analysis.

With these caveats in mind, this study’s 1-year rate of death from any cause was 24%, compared to 31% in PARTNER 1B. The need for permanent pacemaker placement at 1 year was 26% however compared to 5% in PARTNER 1B, and it is unclear whether this could have impacted differences in mortality. The 30-day mortality rate in turn was 8%, compared 7% in inoperable TVT registry patients who were treated contemporaneously with transfemoral TAVR using the SAPIEN valve (Mack 2013).

The rate of major stroke in CoreValve B TAVR patients was 2% and 4% at 30 days and 1 year, respectively, compared to the 5% and 8% rates  in PARTNER 1B. The rate of major vascular complications was just above 8% at 30 days and 1 year, compared to 16% and 17% rates in PARTNER 1B; however, major or life-threatening bleeding was very high at 43%, compared to 22% in PARTNER 1B. The rate of paravalvular aortic regurgitation was just above 4% at 1 year, compared to a nearly 11% rate in PARTNER 1B.

The apparent marked reduction in most adverse events in this study compared to the early PARTNER 1B trial, in similar (if not equal) populations, may be the result of a combination of factors including a lower-profile device and delivery system, better valve fitting, vascular CT evaluation for patient selection and pre-procedural planning, and greater operator and heart-team experience.

PARTNER 2B also evaluated TAVR in inoperable patients (Webb 2015). As with PARTNER 1, PARTNER 2 consisted of two parallel, prospective, multicenter, randomized, active-treatment controlled trials in patients with symptomatic, severe aortic stenosis, who were at high surgical risk (trial A), or inoperable (trial B). Similar to CoreValve, the PARTNER 2 trials evaluated a new TAVR device (SAPIEN XT) that had a lower profile than that used in the original PARTNER 1 trials (SAPIEN).

PARTNER 2B demonstrated that TAVR with SAPIEN XT was noninferior to TAVR with SAPIEN for the composite of death from any cause, major stroke, or rehospitalization at 1 year (the primary outcome), and for all subcomponents of this composite, and had significantly fewer vascular complications (Webb 2015). There was no difference between the two groups in other secondary outcomes, including patient symptoms and function, and paravalvular regurgitation.

The absolute percentages for outcomes of the PARTNER 2B SAPIEN XT group were more similar to those of the TAVR patients in the contemporaneous CoreValve B, than in the earlier PARTNER 1B trial, except for permanent pacemaker placement (8% for 2B, 26% for CoreValve B, 5% for 1B) and paravalvular regurgitation (20% for 2B, 4% for CoreValve B, 11% for 1B); this may reflect overall improvement in second-generation devices although again one must be cautious about comparisons across these trials. While pacemaker placement has not been associated with increased mortality or stroke at 5-year follow up, moderate to severe, and even mild, paravalvular regurgitation has been (Mack 2015, Kapadia 2015, Kodali 2014, Athappan 2013).

The totality of evidence from these trials and studies – the original PARTNER 1B randomized trial (Leon 2010); secondary analyses of its data, and long-term outcome follow-up studies of 1B patients (Kapadia 2015); the subsequent CoreValve B study (Popma 2014) and PARTNER 2B trial (Webb 2015) – support benefits of TAVR for highly selected patients with severe aortic stenosis and cardiac symptoms who are not candidates for surgical aortic valve replacement.

We note findings in incidence of stroke (Kapadia et al., 2014), paraventricular regurgitation or permanent pacemaker placement for these patients since our prior decision. Based on considerations of benefits and harms, we find the recent evidence supportive of this promising technology. These considerations support the need for continuing evidence development and data collection. We note that the evidence is insufficient for minority populations. We also await reports on longer-term outcomes for benefits and harms, including quality of life, for our beneficiaries. We continue to believe that the current coverage under CED offers the appropriate balance of quality and access, while simultaneously stimulating innovation of devices, procedural techniques, and indications for use (for subpopulations and patients with various comorbidities), and so we are continuing coverage with evidence development.

Is the evidence sufficient to conclude that TAVR improves health outcomes for Medicare beneficiaries with cardiac symptoms and severe aortic stenosis who are candidates for SAVR, and are at either high or intermediate surgical risk?

Symptomatic high surgical risk patients

There is one major and one smaller trial evaluating TAVR in patients at high surgical risk. The larger one, PARTNER 1A, was a noninferiority trial demonstrating that in patients who were high risk but still considered to be surgical candidates, TAVR resulted in a rate of death from any cause at 1 year (the primary outcome) similar to that of open-heart surgical aortic valve replacement (Smith 2011).

For secondary outcomes, the TAVR group achieved improvements in cardiac symptoms (by NYHA classification) and function (by the 6-minute walk test) at 1 year that were similar to the surgical control group. As in the PARTNER 1B trial, the TAVR group in PARTNER 1A more frequently had major strokes and major vascular complications than the control group at 1 year (although the difference in major stroke was not statistically significant). The surgical control group in PARTNER 1A had significantly higher rates of major bleeding and new-onset atrial fibrillation.

The PARTNER 1A trial was reviewed in depth in our 2012 NCD. Since then, multiple secondary analyses of PARTNER 1A trial data and follow-up patient outcome studies have contributed to the evidence base on the use of TAVR in patients at high surgical risk. Follow up studies demonstrated the noninferiority of TAVR compared to surgery for the primary outcome of death from any cause at 2 years (Kodali 2012) and 5 years (Mack 2015).

As for valve durability, we agree with comments by Mack, "although 5 years is still too short a time to expect differences in durability between transcatheter and surgical prostheses, especially in this very elderly patient population, it is reassuring that structural valve deterioration in the TAVR groups have not been reported" (Mack 2015); and by Reardon, "the lack of structural valve deterioration at 5 years [for both PARTNER 1A and 1B patients] is highly encouraging but is not adequate time to judge the durability of biological valves. Long-term follow-up remains imperative to establish the long-term durability of this technology in patients where survival beyond 5 years is expected" (Reardon 2015).

Stroke risk was also encouraging at 5 years. Mack reports that "although periprocedural stroke or transient ischemic attack were more common with TAVR than SAVR at 30 days (5.5% vs 2.4%; p=0.4 [intention-to-treat population]), by 5 years this difference had dissipated" (Mack 2015).

Paravalvular regurgitation remained a problem for TAVR. While the findings of non-inferiority of TAVR for the primary outcome of death from any cause remained stable, even mild aortic regurgitation post-procedure was associated with increased mortality at both 2-year (Kodali 2012) and 5-year follow-up (72% for moderate-to-severe aortic regurgitation and 57% for mild aortic regurgitation; Mack 2015).

Quality of life assessments beyond symptoms and function were not reported in the initial PARTNER trial publications, but were embedded in the trial design and reported separately. Based on the KCCQ summary score, for PARTNER 1A (high-risk) patients, there was no difference in quality of life between TAVR and SAVR at 1 year (Reynolds 2012). Quality of life assessments were not reported at 5 years (Mack 2015).

The smaller CHOICE trial randomized patients whom the investigators deemed to be "high-risk" (considering multiple patient factors; the STS PROM score was 6% overall) to transfemoral TAVR with the SAPIEN XT balloon-expandable valve or transfemoral TAVR with the self- expanding CoreValve (Abdel-Wahab 2014). The balloon-expandable group had a higher rate of device success (proper placement and immediate functioning of the device) and a lower rate of paravalvular regurgitation, which remained stable at 1-year follow up (Abdel-Wahab 2015). The authors reported no difference in 1-year clinical outcomes of death from any cause, any stroke, or hospital readmissions – although the study was not powered to detect statistical differences in clinical outcomes. The findings of no difference between these devices was supported by observational multicenter and registry studies (Abdel-Wahab 2014 and 2015, Van Belle 2014).

The evidence from the PARTNER 1A trial, longer-term outcome follow-up studies of 1A patients, the smaller CHOICE trial, and observational studies, support benefits of TAVR (with either a balloon-expandable or self-expanding device) or patients with severe aortic stenosis and cardiac symptoms who are at high surgical risk. We note findings in incidence of stroke, paraventricular regurgitation and permanent pacemaker placement for these patients since our prior decision. Based on considerations of benefits and harms, we find the recent evidence supportive of this promising technology. These considerations indicate the need for continuing evidence development and data collection. We note that the evidence is insufficient for minority populations.  We also await reports on longer-term outcomes for benefits and harms, including quality of life, for our beneficiaries. We continue to believe that the current coverage under CED offers the appropriate balance of quality and access, while simultaneously stimulating innovation of devices, procedural techniques, and indications for use (for subpopulations and patients with various comorbidities), and so we are continuing CED. We believe this is consistent with the real world study findings, for example, Arnold et al. (2017) reported that 1 in 3 patients had a poor outcome at 1 year.

Symptomatic intermediate surgical risk patients

Early research in higher risk patients was followed by three major multicenter, randomized controlled trials focused on intermediate-risk patients: CoreValve A, PARTNER 2A, and SURTAVI.

The first to be published, CoreValve A (Adams 2014), again evaluated a new, lower-profile, self-expanding bioprosthetic valve. Although some have categorized CoreValve A randomized trial patients as "high risk" (Gargiulo 2016), CoreValve investigators represented these patients as being at "increased surgical risk," as determined by clinical judgment. CoreValve A investigators explicitly did not use the PARTNER 1A threshold of an STS PROM score of 10% or higher as a guideline for study inclusion; rather, the STS risk score was considered but no range prespecified. The final study population had an overall STS PROM score of 7%, compared to 12% for PARTNER 1A "high risk" patients (see Table 2 for STS PROM scores of key trials). Hence we place CoreValve in this "intermediate risk" category.  However, we agree that this "intermediate risk" category generally does not represent "the ‘middle ground’ between low- and high-risk patients but rather a subset of higher-risk patients" (Eagle 2016).

CoreValve A demonstrated that TAVR was noninferior to surgical aortic valve replacement for the primary outcome of death from any cause at 1 year. CoreValve A found that TAVR was also noninferior to surgery in important secondary outcomes such as cardiac symptoms (using the NYHA classification) and quality of life (using the KCCQ and the Medical Outcomes Study 12-item Short Form General Health Survey, or SF-12). Exploratory subgroup analysis suggested a reduction in the rate of major adverse cardiovascular and cerebrovascular events (MACCE); this included no increase in the risk of major stroke (an improvement over PARTNER 1A, in which the rate of major stroke was approximately double for the TAVR group). The relative contributions to an improvement in stroke from factors such as a lower-profile device, less sick patients, and greater operator experience in CoreValve compared to PARTNER 1A are unclear. Similarly, the TAVR risk of major vascular complications was lower in CoreValve A than PARTNER 1A (comparing raw data), but remained substantially higher than CoreValve A surgical controls. There was a significantly higher rate of TAVR moderate or severe paravalvular regurgitation compared to controls. CoreValve A TAVR patients also had a greater need for new permanent pacemaker implantation, for reasons that are unclear. The surgical controls had increased risk for major bleeding (inherent to open-heart surgery), acute kidney injury (possibly related to major bleeding), and new-onset or worsening of atrial fibrillation. Subsequent studies demonstrated that primary and secondary outcomes for CoreValve A trial patients, and device durability, remained stable at 2-year (Reardon 2015) and 3-year (Deeb 2016) follow-up.

PARTNER 2A demonstrated that in patients at intermediate surgical risk, TAVR was noninferior to surgical controls for a composite of death from any cause or disabling stroke at 2 years (the primary outcome), and for all subcomponents of this composite. The TAVR group had fewer strokes compared to the surgical group as well (again possibly due to better devices, techniques and operator experience since the earlier PARTNER trial), but this was not statistically significant. For secondary outcomes, the PARTNER 2A TAVR and surgical groups had similar improvements in cardiac symptoms at 2 years (by NYHA classification) and function (by the 5-meter walk test). The TAVR group had decreased life-threatening or disabling bleeding, acute kidney injury, new atrial fibrillation, and hospital length of stay, but increased major vascular complications – a pattern among these trials that again appears to reflect inherent differences in vascular and surgical interventions. There was no difference between TAVR and surgical groups in permanent pacemaker implantation (unlike in CoreValve, where TAVR was substantially higher). The 2A TAVR rate of paravalvular regurgitation however was high at 26% (similar to 2B).

SURTAVI demonstrated that in patients at intermediate surgical risk (but lower risk by STS score than previous intermediate risk trials, see Table 2), TAVR with a self-expanding prosthetic valve was noninferior to surgery for the composite of death from any cause or major stroke at 2 years (the primary outcome). For secondary outcomes, consistent with the pattern seen in prior trials, surgery had higher rates of acute kidney injury, atrial fibrillation, and need for blood transfusion, while TAVR had higher rates of permanent pacemaker placement and paravalvular regurgitation.

The evidence from these three major multicenter, randomized controlled trials – CoreValve A, PARTNER 2A, and SURTAVI – secondary and propensity score analyses, and (albeit limited) outcome follow-up studies, combine to support the benefits of TAVR for patients with severe aortic stenosis and cardiac symptoms who are at intermediate surgical risk. We note findings in incidence of paraventricular regurgitation and permanent pacemaker placement for these patients since our prior decision. Based on considerations of benefits and harms, we find the recent evidence supportive of this promising technology. These considerations indicate the need for continuing evidence development and data collection. We note that the evidence is insufficient for minority populations. We also await reports on longer-term outcomes for benefits and harms, including quality of life, for our beneficiaries. We continue to believe that the current coverage under CED offers the appropriate balance of quality and access, while simultaneously stimulating innovation of devices, procedural techniques, and indications for use (for subpopulations and patients with various comorbidities), and so we are continuing coverage with evidence development. We believe this is consistent with the real world study findings, for example, Arnold et al. (2017) reported that 1 in 3 patients had a poor outcome at 1 year.

Symptomatic low surgical risk patients

Earlier TAVR research in higher and intermediate risk patients was recently followed by two multicenter, randomized controlled trials on low-risk patients (average STS score: 1.9%): PARTNER 3 (Mack 2019) and Evolut (Popma 2019).  As with the earlier trials, patients in these low surgical risk trials had symptomatic, severe aortic stenosis.

PARTNER 3 demonstrated that TAVR with a latest-generation, balloon-expandable valve was non-inferior to SAVR, and in a second prespecified analysis, superior to SAVR, for the primary composite outcome of death from any cause, stroke, or rehospitalization at 1 year (Mack 2019).  Evolut in turn demonstrated that TAVR with a latest-generation self-expanding valve was non-inferior to SAVR for the primary composite outcome of death from any cause or disabling stroke at 2 years (Popma 2019).

We agree, as do the PARTNER 3 authors, with the Evolut authors’ opinion that "Definitive conclusions regarding the advantages and disadvantages of TAVR as compared with surgery await long-term clinical and echocardiographic follow-up" which for Evolut is planned to continue through 10 years for all patients.  In this regard, a major limitation of both low-risk trials is that TAVR valve durability remains unknown past 1 and 2 years respectively.  We noted that surgical valves have been documented to last 10-20 years depending on valve type (with latest-generation surgical valves lasting longer).  This consideration is of particular importance to low-risk patients who are likely to live longer compared to higher-risk patients.   

These two low-risk trials were published in March 2019.  The evidence base is less established and assessments continue.  CED is appropriate for this low-risk population for the same reasons as discussed for higher-risk populations, but also due to the added importance of determining long-term TAVR valve durability for these low-risk, longer-living patients.  

Since the finalization of the 2012 TAVR NCD, TAVR programs have been established in over 500 hospitals across the country. Many of the requirements in the 2012 NCD were included to ensure that hospitals and providers with limited or no TAVR experience followed distinct and specific requirements based on requirements that had been incorporated in the promising pivotal clinical studies completed at the time. As TAVR has continued to grow and evolve, both in terms of utilization and FDA approved indications, hospitals and providers have grown more comfortable with the procedure as well as the appropriate parameters for patient evaluation and selection and hospital infrastructure and capabilities. As such we proposed to modify many of the requirements set forth in the 2012 TAVR NCD specific to section A, treatment of symptomatic aortic valve stenosis when furnished according to an FDA approved indication, to be more flexible and less burdensome for hospitals, clinicians, and patients given the experience gained since 2012.

Patient Population:  The decision to pursue aortic valve replacement should reflect an appropriate balance of benefits and harms, and mortality increases substantially after symptoms develop from severe aortic stenosis (Kitai 2011).  The current AHA/ACC guideline recommends surgical AVR in patients with severe AS who report symptoms Nishimura 2014).  The 2017 ACC expert consensus decision pathway (Otto 2017) stresses the importance of an accurate patient diagnosis and staging of aortic stenosis, where all patients being considered for TAVR should have severe symptomatic aortic stenosis (Stage D). Furthermore, all FDA approved indications for TAVR require patients to have symptomatic aortic stenosis.  Many public comments addressed the question of whether TAVR is appropriate for asymptomatic patients.  Our determination on the clinical utility of TAVR (i.e., whether the intervention has a statistically significant and clinically meaningful improvement on patient health outcomes) for asymptomatic patients with severe aortic stenosis awaits all of the below:

(1) completion and publication of well-designed and powered trials in asymptomatic patients with severe aortic stenosis (we are aware of at least one that is ongoing);

(2) based on the trial evidence above, a change in the FDA-approved label as an approved indication; and,

(3) similar to patients in higher surgical risk groups, long term data on health outcomes based on the ongoing data collection in the TVT registry.  

Therefore, we will maintain the 2012 requirement where TAVR is covered for the treatment of symptomatic aortic valve stenosis when furnished according to a Food and Drug Administration (FDA)-approved indication when the procedure is furnished with a complete aortic valve and implantation system that has received FDA premarket approval (PMA) for that system's FDA-approved indication.

Patient Evaluation: In the 2012 NCD we required two cardiac surgeons to independently examine patients face-to-face and evaluate their suitability for open AVR surgery. As noted above, the 2018 AATS/ACC/SCAI/STS Consensus Statement (Bavaria 2018) has been updated to require one cardiac surgeon only citing the "established role of TAVR with published AUC (Bonow 2017) and greater experience in the assessment of risk for SAVR".  We proposed that one cardiac surgeon must independently examine the patient face-to-face, and evaluate the patient’s suitability for SAVR, TAVR or medical or palliative therapy.  We believed this modification was appropriate given the advancements and progress made since 2012 as TAVR has become more widely performed.  

However, in response to numerous, well-considered public comments on this topic, we are further modifying our requirement taking into account: (1) the accumulated experience of the TAVR surgeons and interventionalists, (2) the wide acceptance of the heart team approach, and (3) concern for improving access while maintaining quality of care. 

As such, we have revised our decision to establish a heart team based approach to the evaluation of a patient’s suitability for SAVR, TAVR or medical or palliative therapy. These revisions require both an interventional cardiologist and cardiac surgeon to evaluate the patient, document their evaluation, and present their rationale to the heart team.  We believe this final decision streamlines the evaluation of patients with aortic stenosis, places appropriate emphasis on multidisciplinary decision making, and is responsive to strong public support for the heart team concept.

In making this revision, we reviewed the multispecialty Consensus Statement that recommends a cardiac surgeon and another physician evaluate the patient as well as public comment that an interventional cardiologist should be involved in the evaluation.  We note that while an evaluation by an interventional cardiologist from the heart team is a new requirement in the final NCD, an interventional cardiologist in practice already performs an independent evaluation prior to participation in the procedure and should not be an additional burden.  TAVR is a complex intervention with an evolving evidence base.  We believe that both an interventional cardiologist and a cardiac surgeon bring distinctive perspectives to the heart team, and the combination fosters a balanced consideration of treatment options that is especially valuable at the extremes of surgical risk.  The multi-disciplinary heart team is a patient-centered concept that supports evidence-based medical decision making and promotes the best possible outcomes. 

While not a requirement of this final NCD, we encourage continuation of the current practice wherein heart team interventional cardiologists and cardiac surgeons consult other specialists (e.g., cardiac imaging, cardiac anesthesia, and palliative care) as needed to refine treatment recommendations that are tailored to the circumstances of each patient. Additionally and while not a requirement of this final NCD, we note that patients and referring physicians may continue to seek a second, independent opinion if desired. Finally, we strongly encourage heart teams to engage referring physicians in medical decision-making because referring providers are often able to offer a broader perspective that is grounded in a longitudinal patient relationship. 

TAVR Heart Team Composition: The multidisciplinary heart team is a critical element in the success of all TAVR programs. The premarket pivotal studies, 2017 ACC expert consensus decision pathway (Otto 2017), and the 2018 AATS/ACC/SCAI/STS Consensus Statement (Bavaria 2018), discussed in detail above, include specific parameters around the composition of the heart team. We proposed to maintain the heart team concept specified in the 2012 decision. Consistent with the above noted evidence, we proposed that Medicare beneficiaries are more likely to experience the best achievable outcomes when TAVR is furnished while the patient (preoperatively and postoperatively) is under the care of a heart team. Similar to the aforementioned societal documents and pivotal studies, we proposed that the heart team is to be comprised of a cohesive, multidisciplinary team of medical professionals which includes a cardiac surgeon and an interventional cardiologist experienced in the care and treatment of aortic stenosis and includes providers from other physician specialties as well as advanced patient practitioners, nurses, research personnel and administrators.  Public comments were generally supportive of our proposals, which we are now finalizing.

Joint Participation of Heart Team Operators: We proposed to make no changes to the joint participation of heart team operators and to continue to require that the interventional cardiologist(s) and cardiac surgeon(s) jointly participate in the intra-operative technical aspects of TAVR.  We are finalizing this proposal.

Facility Infrastructure Requirements: We proposed to remove several hospital infrastructure requirements that we believe are now unnecessarily prescriptive given the experience hospitals and heart teams have with TAVR and existing site initiation requirements from the medical device companies.  Specifically, we proposed to remove specifications around the cardiac catheterization lab or hybrid operating room/catheterization lab, non-invasive imaging technology and sufficient space to accommodate equipment. We proposed to maintain the existing requirements for hospitals to have an on-site heart valve surgery program, post-procedure intensive care facility with personnel experienced in managing patients who have undergone open-heart valve procedures and appropriate procedural volumes as further specified below. The hospital program volume requirements, detailed below, proposed specific PCI volume requirements which require an active interventional cardiology program. Therefore, we proposed that TAVR should be performed in a hospital with appropriate infrastructure which includes an on-site interventional cardiology program.  Public comments were generally supportive, and we are finalizing these proposals.

Hospital Volume Requirements and Heart Team Volume Requirements: Stakeholders have varying opinions regarding procedural volume requirements for hospitals and heart team operators to both begin and maintain TAVR programs. The 2012 NCD includes procedural volume requirements reflective of the 2012 SCAI/AATS/ACCF/STS multisociety expert consensus statement on TAVR operator and institutional requirements (Tommasso 2012). The updated consensus statement (Bavaria 2018) revisits and revises the procedural volume requirements for both hospitals and operators. These updates are based in part on consensus with the support of data collected and analyzed by authoring organizations, AATS/ACC/SCAI/STS.  This data was used is several peer reviewed publications, many summarized in the evidence section.  We note that some stakeholders are challenging the consensus statements basis for volume requirements. 

Stakeholders have cited concerns around access to TAVR, including in the recent round of public comments on the proposed NCD and summarized in the public comment section.  Concerns highlight both geographic barriers when prospective patients must travel long distances to participating hospitals, as well as socioeconomic and patient preference barriers when prospective patients are unable to use their trusted provider and hospital of choice because the hospital cannot meet the volume requirements. We recognize access and travel apply to many procedures.  We have taken all of that into consideration but the solution to access cannot compromise quality.  To assist CMS in determining whether to retain volume requirements, we held a MEDCAC.  The MEDCAC had greater than intermediate confidence that there is sufficient evidence supporting procedural volumes.

After reviewing all of the evidence, we continue to believe volume requirements are needed to ensure good patient outcomes.  This finding holds even after a full evaluation of papers not considered in the proposed NCD, but cited in public comments.  Specifically on procedural volume-outcome relationship, Vemulapalli 2019, Russo 2019, Mao 2018, and Salemi 2019.  We added summaries of these papers to our evidence section.  We note that the Russo 2019 study supports our goal of transitioning from volume to outcome measures.  The authors state that their findings "support that good outcomes are not merely a function of quantity, but are influenced by a constellation of factors, including technological advancements, best practices, collaborative knowledge programs, and organizational culture."  This study raised many interesting questions, but could not instruct changing the specific volume we proposed in our decision or resolve disagreements with other contemporary studies.

Salemi et al (2019) opined, with respect to a requirement that heart teams/centers must complete "at least 20 such procedures annually or 40 procedures biannually," that while "these numbers appear to be generally consistent with our study results in general terms, perhaps these guidelines are more appropriately directed toward specific operators."  They believed this highlights "the importance of proctoring and partnerships with experienced TAVR practitioners during the early adoption period."

Russo et al (2019) found a "learning curve" termination at 200 cases when analyzing all generations of balloon-expandable valves, and opined that their study’s overall findings "support that good outcomes are not merely a function of quantity, but are influenced by a constellation of factors, including technological advancements, best practices, collaborative knowledge programs, and organizational culture."

Finally, Vemulapalli et al (2019) found a statistically significant difference between the lowest and highest hospital volume quartiles in terms of 30-day risk-adjusted mortality. Because of few deaths in either group (i.e., a small event rate), the relative difference in mortality was high but the absolute difference was still small (as pointed out by a public commenter).  This study supports the hypothesis that centers with higher procedural volume have slightly better outcomes in terms of patient survival.

However, this slight improvement in survival could be offset by an unintended consequence of raising procedural volume requirements.  Other analyses (discussed in the analysis and public comment section of this NCD) indicate that raising volume requirements would result in substantially fewer TAVR programs in both rural and urban areas. Having fewer TAVR programs would likely lead to further delays in treatment for certain patients for whom AVR is indicated and who prefer TAVR.  We know from other research that delays in AVR treatment entails its own mortality (e.g., Malaisrie 2014, also mentioned in public comments).

It is unclear whether, if we were to raise procedural volume requirements, the decrease in mortality (hypothesized by the Vemulapalli 2019 study) would be offset by the increased mortality resulting from delayed treatment due to having substantially fewer TAVR centers. Thus, the Vemulapalli 2019 study does not convincingly inform raising procedural volume requirements in our NCD.

The evidence submitted supported that there are many factors, including volume criteria that predict good patient outcomes.  While volume may not be the sole measure to predict good patient outcomes, there is evidence to demonstrate that it is a good predictor of outcomes (Carroll 2017).  We added a CED question to ascertain whether we can lower volume procedural requirements when an outcome measure is fully developed.  We discuss outcome measures more below. 

The evidence reviewed, including new evidence submitted, supports that volume requirements are part of the equation of predicting good quality.  Further, the MEDCAC had greater than intermediate confidence that there is sufficient evidence to support procedural volumes and volume procedural requirements are consistent with the AATS/ACC/SCAI/STS updated professional society consensus statements.  We note that the AATS/ACC/SCAI/STS professional society consensus statements are based on the current state of evidence from trials, and studies based on the TVT registry, considered along with current practice patterns which was a major factor in this decision.  Therefore, based on the totality of the evidence, including all of the public comments and articles sent in response to the proposed NCD, we are maintaining the procedural volume requirements we proposed. While the procedural volume requirements continued to be required, there is greater flexibility in meeting these requirements.  We believe the hospital and heart team operator volume requirements reflect our intent to strike a balance between ensuring hospitals have the experience and capabilities to handle complex structural heart disease cases with an evolving evidence base and reducing the burden of unnecessary requirements on hospitals and patients. 

We are finalizing the requirements described below for hospitals and heart teams without TAVR experience to begin a TAVR program followed by requirements for hospitals with TAVR experience to maintain a TAVR program:

Hospital Requirements to Begin TAVR Programs
As stated in the proposed decision, the MEDCAC had greater than intermediate confidence that there is sufficient evidence to support hospital procedural volumes for SAVR and PCI in hospitals without TAVR experience to maintain TAVR programs, noting that the requirements to begin a TAVR program outweigh the harms of limiting access to TAVR to only hospitals that meet volume requirements.

Institutional SAVR Program. The 2012 requirements, based upon the AATS/ACC/SCAI/STS Expert Consensus document at that time, specified ≥ 50 total AVRs in the previous year prior to TAVR, including ≥ 10 high-risk patients and ≥ 2 physicians with cardiac surgery privileges. The institutional SAVR program volume requirements are designed to ensure that hospitals without an established TAVR program have experience with complications which may arise during TAVR including conversion to SAVR. When reassessing this requirement, we endeavored to strike a balance between ensuring hospitals have the experience and capabilities to handle complex structural heart disease cases, and limiting the burden of unnecessary requirements on hospitals and patients.

We are finalizing the annual volume of cases (≥ 50) in the previous year prior to establishing a TAVR program but have provided flexibility on how that is met.  We are requiring that those cases be open heart surgeries, instead of AVRs only as required in the 2012 NCD, with ≥ 20 aortic valve related procedures in the 2 years prior to TAVR program initiation.  Consistent with the updated AATS/ACC/SCAI/STS Expert Consensus document, we maintain the ≥ 2 physicians with cardiac surgery privileges requirement, ensuring that hospital programs will have sufficient experience and capabilities on hand to deal with complications as well as complex structural heart disease cases. 

Institutional PCI Experience. Consistent with the 2018 AATS/ACC/SCAI/STS Expert consensus document, we are requiring hospital programs to perform ≥ 300 PCIs per year. The hospital program volume requirement, detailed above, noted specific PCI volume requirements which require an active interventional cardiology program.  We note there was significant public comment on the number of PCI procedures required under this NCD.  However, as noted above based on the totality of the evidence, including public comment, we believe this is the right volume requirement to ensure the best patient outcomes. 

Accordingly, we add a requirement that TAVR program for heart teams without TAVR experience should include at least one physician with interventional cardiology privileges to ensure that hospital programs have sufficient experience on hand to deal with TAVR access issues and procedural complications which may require arterial vascular interventions.

Heart Team Requirements to Begin TAVR Programs
The MEDCAC had relatively high confidence that there is sufficient evidence supporting a threshold of SAVR and PCI procedural volumes for the principle cardiovascular surgeon on a TAVR heart team and a structural heart disease procedural volume for the principle interventional cardiologist on a TAVR heart team.

Cardiovascular Surgeon Experience. We maintain the career experience of ≥ 100 cases but provide flexibility on how this is met.  We are requiring that those cases may be open heart surgeries, instead of AVRs only as required in the 2012 NCD, of which ≥ 25 are aortic valve related.

In addition to the career procedure requirements, the 2012 NCD specified that the cardiovascular surgeon had to perform.≥ 50 AVRs in 2 years and ≥ 20 AVRs in the last year prior to TAVR initiation.  During the MEDCAC, Dr. Martin Leon’s presentation for Advanced Medical Technology Association (AdvaMed) discussed in detail some of the requirements to be considered a TAVR center.  Elements Dr. Leon noted specific to the cardiovascular surgeon include the demonstration of procedural proficiency in cardiac surgery (specifically aortic valve disease management and therapy) and centers must have a functional Heart Valve Team with multidisciplinary expertise and a designated heart valve clinic for case screening. Given this existing industry standard for initiating a TAVR program, we remove requirements for AVR procedures needing to be conducted in the two years prior to TAVR program.

Interventional Cardiologist Experience. The 2012 NCD included a requirement that the interventional cardiologist on the TAVR heart team have professional experience with 100 structural heart disease procedures lifetime or 30 left-sided structural procedures per year of which 60% should be BAV, where atrial septal defect and patent foramen ovale closure were not considered left-sided procedures. The 2018 AATS/ACC/SCAI/STS expert consensus document specifies participation in at least 100 transfemoral TAVR cases with at least 50 cases as primary operator. However, since the issuance of the 2012 NCD, there has been no new evidence assessing the impact of individual interventional cardiologist volume requirements on TAVR patient outcomes. Therefore, this final NCD does not change the 2012 requirements but provides greater flexibility in how the left-sided structural procedures experience is met through removing the stipulation that "60% of any left-sided procedure should be BAV and the exclusion of atrial septal defect and patent foramen ovale closure." Finally, the 2012 decision also included a requirement for device-specific training as required by the manufacturer.  We maintain this requirement. 

Hospital Requirements to Maintain TAVR Programs
The MEDCAC had greater than intermediate confidence that there is sufficient evidence to support hospital procedural volumes for SAVR and PCI in hospitals with TAVR experience to maintain TAVR programs, noting that the requirements to begin a TAVR program outweigh the harms of limiting access to TAVR to only hospitals that meet volume requirements.

Institutional AVR Program. This 2012 NCD requirements, based upon the AATS/ACC/SCAI/STS expert consensus document at that time, specified ≥ 20 AVRs per year or ≥ 40 AVRs every 2 years and ≥ 2 physicians with cardiac surgery privileges. The 2012 NCD and expert consensus document did not contain specific hospital TAVR volume requirements for maintaining TAVR programs.

The updated consensus document includes a requirement for ≥ 50 TAVRs per year and ≥ 30 SAVR. As stated previously in this decision memo, the institutional SAVR program volume requirements are designed to ensure that hospitals maintain experience sufficient to address complications which may arise during TAVR including conversion to SAVR. Preliminary analyses from the STS-ACC TVT Registry data that was presented at the MEDCAC provided some evidence in support of a clinically meaningful association of higher mortality and other major complications with site annual volume below a threshold of 50 procedures/year (Bavaria and Tommaso Presentation, slide 14-17, MEDCAC, 2018, https://www.cms.gov/medicare-coverage-database/details/medcac-meeting-details.aspx?MEDCACId=75). During the MEDCAC, the observed expansion of TAVR in favor of SAVR and a predicted continued expansion were noted by several presenters. Volume requirements including both TAVR and SAVR should consider the shifting preference of approaches to AVR.  We believe an AVR volume requirement with a minimum number of TAVR procedures provides a balance of ensuring hospitals have maintained proficiency in TAVR and the experience and capabilities to handle complex structural heart disease cases while limiting the burden and barriers unnecessary requirements may have on hospitals and patients alike. Therefore we finalized that hospitals maintain ≥ 50 AVRs (TAVR or SAVR) per year including ≥ 20 TAVR procedures in the prior year; or, ≥ 100 AVRs (TAVR or SAVR) every 2 years, including ≥ 40 TAVR procedures in the prior 2 years. Lastly, consistent with the updated AATS/ACC/SCAI/STS expert consensus document, we maintain the ≥ 2 physicians with cardiac surgery privileges requirement.

Institutional PCI Experience. For the reasons stated above, under the "Requirements to Begin a TAVR Program" section, we are maintaining what we proposed and requiring these hospital programs to perform ≥ 300 PCIs per year and add a requirement that heart teams in TAVR programs with TAVR experience should include at least one physician with interventional cardiology privileges.  We note that while we are maintaining our proposed NCD ≥ 300 PCIs per year requirement, this is lower than what was in the 2012 NCD. 

Heart Team Requirements to Maintain TAVR Programs
The MEDCAC had greater than intermediate confidence that there is sufficient evidence supporting a threshold of TAVR procedural volumes for the combined experience of the cardiovascular surgeon and the interventional cardiologist on a TAVR heart team to maintain TAVR Programs. Based on stakeholder input, published evidence and the society consensus statement, we proposed to remove procedural volume requirements for the heart team.

Outcome Measures: There has been wide stakeholder support for a paradigm shift in which TAVR proficiency would be determined by outcome measures rather than procedural volume.  A shift from volume to outcomes was supported by multiple presentations throughout the MEDCAC and in public comments.  We continue to believe that outcome measures may be an appropriate alternative, and acknowledge that volume requirements were always intended as the best available proxy until validated, mature, and agreed-upon outcome measures emerged.  However, we also note that even an outcome measure based approach to quality assurance may be limited at low volume centers for statistical reasons, and compliance with the other programmatic requirements may be similarly impractical at low volumes.

The STS continues to develop and maintain hospital performance measures including aortic valve replacement surgery outcomes (https://www.sts.org/quality-safety/performance-measures). The following two STS measures have been endorsed by the National Quality Forum (NQF). The risk-adjusted operative mortality for aortic valve replacement measure was NQF endorsed (#0120) on December 6, 2011. It captures the percent of patients undergoing AVR who die, including both all deaths occurring during the hospitalization in which the procedure was performed, even if after 30 days, and those deaths occurring after discharge from the hospital, but within 30 days of the procedure. Additionally, the STS has an AVR composite score which was NQF endorsed (#2561) on November 7, 2014. It is comprised of absence of operative mortality and major morbidity. Mortality is risk-adjusted and defined as death during the same hospitalization as surgery or after discharge but within 30 days of the procedure. Major morbidity is risk-adjusted and defined as having reoperations for any cardiac reason, renal failure, deep sternal wound infection, prolonged ventilation/intubation, or cerebrovascular accident/permanent stroke.

Separately, the TVT Registry collection of peri-procedural and 30-day data allows for benchmarking of certain outcome measures.  Additionally, the STS/ACC TVT Registry, in partnership with statisticians and investigators at Duke Clinical Research Institute, developed a risk model to report a 30-day composite score as a TAVR Quality Metric on which they sought public comment ending on August 29, 2018.  The 30-day composite consists of ordered categories based on the best possible outcome (e.g. alive and free of major complications) to the worst possible outcome (30-day death) during hospitalization and the 30-day follow-up period.  This composite includes 30-day death, 30-day stroke, 30-day life-threatening/major bleed, acute kidney injury (stage III), or 30-day moderate-to-severe paravalvular leak.

We recognize that the STS/ACC TVT Registry has not made this measure final. However, we understand that efforts continue to further develop and refine such outcome measures and encourage continued progress toward the establishment of widely-supported TAVR outcome measures. The CMS Innovation Center is also exploring application of a composite outcome measure to the Bundled Payments for Care Improvement Advanced TAVR Clinical Episode, which could also be captured through the STS/ACC TVT Registry.  While we are not finalizing any outcome measure as a replacement for volume requirements, we are adding a new CED question that explores the relationship between procedure-related factors and patient health outcomes.  We are monitoring this area closely and may reconsider this NCD in the future to consider replacing or modifying procedural volume criteria in conjunction with outcome measures when appropriate measures emerge.

Research Questions
In 2012, CMS posed questions regarding the evidence for TAVR for the treatment of symptomatic aortic valve stenosis which CED registries and studies were to address (see Appendix C for the current 20.32 NCD). Based on our concerns at the time, we required additional data to be collected via clinical study participation (see https://www.cms.gov/medicare-coverage-database/details/ncd-details.aspx?NCDId=355).

We assessed the extent to which the published literature, including completed CED studies, addressed the following questions. (Each approved study had to address one or more aspects of one or more of the CED questions below.) The following outcomes were to be critically evaluated through a minimum of one year:

• stroke;
• all-cause mortality;
• transient ischemic attacks (TIAs);
• major vascular events;
• acute kidney injury;
• repeat aortic valve procedures;
• change in quality of life pre- and post-TAVR.

Since the 2012 NCD, there have been 26 approved clinical studies of TAVR and one national registry under CED. Seven of the 26 studies have reached completion. After reviewing the totality of the new evidence, we assessed how it addresses each of the outcomes.

All-Cause Mortality, Stroke, and Transient Ischemic Attacks (TIAs). We consider stroke, TIA, and all-cause mortality together as these have appropriately emerged as a composite primary outcome of choice for trials and observational studies alike. Initial trials (e.g., PARTNER 1) focused on all-cause mortality alone. However, more recent trials (e.g., SURTAVI) have included major or disabling stroke in a composite with all-cause mortality at 2 years, out of recognition that patients may view disabling stroke as a worse outcome than death (2012 NCD).  We are also aware of ongoing or planned trials that include a composite of any stroke (but not TIA) and all-cause mortality, with the goal of increasing event rates and thus statistical precision given expectations of decreased event rates generally as the TAVR procedure becomes safer with newer generation devices and techniques, and better patient selection. While cardiovascular mortality is an interesting secondary outcome, all-cause mortality remains better as part of a composite primary outcome is it accounts for competing causes of death, which is especially important for our fragile, older and/or disabled Medicare population.

The results of key randomized controlled trials for patients with severe, symptomatic aortic stenosis and various levels of surgical risk (see Table 2), observational studies using TVT registry data (e.g., Grover 2017, Brennan 2017), and well-designed and well-executed meta-analyses (Arora, 2016; Carnero-Alcázar, 2017; Lazkani, 2018; Singh, 2018; Villablanca, 2016; Siontis 2019), combine to demonstrate the following:

First, when comparing TAVR to SAVR, the risk for stroke all-cause mortality, stroke and transient ischemic attack was similar between TAVR and SAVR patients at 1 year to 2 years and beyond of follow up after the procedure.

Second, using STS/TVT registry data, the 1-year rate of TAVR stroke remained relatively constant, while the 1-year rate of TAVR mortality had a statistically significant decrease, throughout the time period of TAVR between 2012 and 2015 across all risk groups.

We believe all-cause mortality, stroke, and TIA remain important outcomes for trials and observational studies alike and thus these data should remain an integral part of CMS reporting requirements under CED. Therefore, we will continue the data collection requirement for all-cause mortality, stroke, and TIA from the 2012 NCD.

Major vascular events. A pattern has emerged across the key TAVR trials for various patient surgical risk groups (see Table 2) demonstrating that major vascular complications tend to be more common in TAVR than in SAVR.  These trial findings have remained stable in outcome follow-up studies up to 5 years (e.g., Mack 2015; Kapadia 2015; Gleason 2018). Similarly, several meta-analyses (Carnero-Alcázar, 2017; Lazkani, 2018; Villablanca, 2016) showed that vascular access complications of unspecified length of follow-up were generally more frequent in the TAVR group compared to SAVR (although in the latest trials in low-risk patients, there was no statistical difference between groups).  As presented in more detail in our discussion of TAVR patient risk categories, TAVR devices and delivery systems have improved over the past several years. However, vascular complications remain an important consideration for many patients and physicians weighing procedural options, and so we retain this category as part of CMS reporting requirements under CED.

Acute kidney injury. Acute kidney injury (AKI) is a common adverse event after both TAVR and SAVR, and predicts worse outcomes, including death and hospital readmission (Hansen 2017; Brown 2016). A pattern has emerged across key TAVR trials, including recent trials in low-risk patients (see Table 2) demonstrating that AKI is less common in TAVR than in SAVR. This trend is supported in meta-analyses (Carnero-Alcázar 2017; Siontis 2019).  Another meta-analysis (Villablanca 2016) however found no difference in AKI of unspecified follow-up between TAVR and SAVR patients.

Our 2012 CED requirement to report AKI data both reflected and stimulated research focused on AKI. The end point of kidney disease is end- stage renal disease (ESRD) requiring renal replacement therapy (RRT). Like major or disabling stroke, RRT may be viewed by many of our beneficiaries as an unacceptable outcome, to some perhaps equivalent to or worse than death (Hansen 2017; 2012 NCD). We note that this view may evolve as RRT itself becomes less burdensome and produces better patient outcomes, and that RRT is currently the focus of multiple CMS Innovation models. The glomerular infiltration rate (GFR) is the most common method of assessing kidney disease. Studies have demonstrated that pre-procedural GFR predicts AKI leading to RRT or death after TAVR (Meneguz-Moreno 2017; Hansen 2017). Given the data on AKI that have been accumulated in trials and the TVT registry, we are maintaining reporting of AKI; however, we proposed to add pre-procedural GFR and post-procedural new RRT as data requirements as well.

Repeat aortic valve procedures. Paravalvular regurgitation (PVR) is a complication of AVR, both TAVR and SAVR. PVR can cause symptoms such as shortness of breath, chest pain, and fatigue, has been associated with increased mortality (Mack 2015), and can lead to the need for a repeat aortic valve procedure. Surgically implanted prosthetic valves last 10-20 years with bioprosthetic valves typically lasting 12-15 years; mechanical valves can last longer but require long-term anticoagulation. This compares to the 12-year life expectancy of an average 60 year-old after AVR (van Geldorp 2009; Applegate 2017). While the durability of first-generation TAVR valves has been established at 5-year follow up, the long-term durability remains unknown.

The totality of evidence from the key trials (see table 2), meta-analyses and TVT registry studies support that PVR is a more frequent complication of TAVR than of SAVR. Follow-up studies of trial patients report stability of PVR findings at 5 years (Mack 2015, Kapadia 2015, Gleason 2018). Other studies support that PVR assessed at 1-year (Lazkani, 2018), > 2-years (Lazkani, 2018), or of unspecified duration (Carnero- Alcázar, 2017; Singh, 2018) of follow-up was more frequent in patients treated with TAVR.  The exception to this pattern is the recent PARTNER 3 trial using a third-generation balloon-expandable valve in low-risk patients, which found no statistical difference in PVR between TAVR and SAVR groups.  Based on TVT registry data, which may better reflect results achievable in wide community practice, moderate-to-severe aortic regurgitation of unspecified follow-up has decreased over time from 2012 to 2015 among all risk groups (Grover, 2017).  Moderate-to-severe aortic regurgitation was more frequent in the TAVR group compared to SAVR, but the overall trend has been decreasing.

As for the risk of repeat aortic valve intervention, and focusing on the more recent studies using newer-generation TAVR devices , the PARTNER 2 trial found no significant difference between TAVR and SAVR at 1-year (Leon, 2016) and 2-years (Leon, 2016) of follow up after the procedure; however, the SURTAVI trial found that reintervention occurred more frequently in the TAVR group at 1-year (Reardon, 2017) and 2-years (Reardon, 2017) of follow up after the procedure.  The difference between the results of the SURTAVI and the PARTNER 2 trials could be due to the small numbers involved in the analysis of the aortic valve reintervention data. Using TVT registry data on patients undergoing TAVR for all risk groups, 30-day aortic valve reintervention rate was very low and demonstrated little variation up to 2015, but the data was still sparse and the numbers remained small (Grover, 2017).  The recent Evolut trial in low-risk patients (Popma 2019) reported no significant difference between TAVR and SAVR groups.  For all patient risk groups, the evidence for repeat aortic valve reintervention is sparse and prone to fluctuation due to the small numbers of aortic valve reintervention procedures and the evolving device technologies and device delivery approaches. Further long-term data is needed to clarify the question of durability of the TAVR device as measured by the frequency of repeat aortic valve reintervention across all patient risk groups.  Thus we continue data collection on the question of repeat aortic valve interventions from the 2012 NCD.

Quality of Life (QoL). Quality of life assessments beyond symptoms and function were not reported in the initial PARTNER trial publications; however, a separate study found no difference in quality of life at 1 year between TAVR and SAVR PARTNER 1A (high-risk) patients (Reynolds 2012) based on the KCCQ summary score. Quality of life was not reported at 5-year follow up (Mack 2015). A substudy of PARTNER 2A (intermediate-risk) patients found "no significant differences between TAVR and SAVR in any health status measure [including the KCCQ] at 1 or 2 years" (Baron 2017b). The SURTAVI trial also showed no difference between TAVR and SAVR groups in the KCCQ summary score at 12 months of follow up (Reardon 2017).  The recent low-risk trials reported conflicting results on quality of life between the TAVR and SAVR groups (Mack 2019, Popma 2019).

Several TVT registry investigators have focused on the importance of quality of life in addition to survival (e.g., Holmes 2015; Arnold 2015, 2017, 2018; Lazar 2010). They point out that patients who live longer as a result of TAVR but who experience no change or a decrease in quality of life may not benefit from the procedure, while TAVR patients who do not live longer but who experience substantially improved quality of life may benefit.

The totality of evidence suggests that quality of life appears to improve after TAVR compared to its baseline, but it is unclear if the improvements are sustained in the long term compared to SAVR for patients across different surgical risk categories or if the changes are related to evolving device technology. Data from registries and clinical trials are needed to further assess long term quality of life for TAVR patients across all risk categories, especially given the trend of increasing use of TAVR in lower-risk patients who are likely to live longer. Long-term quality of life analyses should include comparison of TAVR to baseline measurements and to SAVR, and should use validated and reliable disease-specific and general health status measurement tools, including the KCCQ, which is currently included in the TVT registry, and the NIH Patient-Reported Outcomes Measurement Information System® (PROMIS®). We continue the quality of life data collection requirements from the 2012 NCD.

New Permanent Pacemaker Placement. The frequency of the need for new PPM placement after TAVR compared to SAVR appears to depend on the type of TAVR device used. The procedures themselves (SAVR or TAVR) can injure the cardiac conduction system in heart walls, creating electrical conduction abnormalities.  A permanent pacemaker is then required to maintain normal heart rhythm.

Assessing the newer-generation devices used in recently published trials, TAVR using a balloon-expandable prosthetic valve demonstrated no difference in the rate of new PPM implantation compared to SAVR at 30 days, 1 year and 2 years (Leon 2016).  However, TAVR using a self-expanding valve demonstrated a substantially greater rate of new PPM implantation compared to SAVR at 30 days (the sponsor did not publish 1-year or 2-year results for new PPM placement).  The pattern of PPM implantation being more of a problem for self-expanding than for balloon-expanding valves continued in the recent low-risk trials (Mack 2019, Popma 2019).  Based on TVT registry data, the frequency of new PPM implantation after TAVR increased across all risk groups beginning in 2014, likely resulting from commercial approval of self-expanding TAVR devices (Grover, 2017).

Unlike paravalvular regurgitation, which is associated with increased mortality after TAVR, the long-term relationship of PPM implantation and mortality after TAVR is unknown. Even if there were no long-term mortality risk from PPM implantation itself, the burden of a pacemaker may depend on a patient’s overall life expectancy after TAVR. Patients with shorter horizons may be more concerned about survival and overall quality of life in years gained after the procedure than about living with a pacemaker. Conversely, for patients with lower mortality risk and longer horizons, permanently living with a pacemaker may weigh heavier in their decision and lead to selection of an alternative device or procedure. More data from registries and trials alike is thus needed on rates of PPM implantation for competing procedures and devices, including for new ones under development. We thus proposed to add a new requirement to report PPM implantation rates up to at least 1 year after TAVR to the TVT registry.

Conclusion. After reviewing the published literature, we believe that gaps in the current evidence base lead to uncertainty about the overall impact of TAVR on beneficiary outcomes when furnished outside of the setting of evidence development or clinical trial protocols.  Therefore we require that data continue to be collected through a minimum of one year post TAVR regarding for: (A) TAVR for the treatment of symptomatic aortic valve stenosis when furnished according to an FDA approved indication via a National Registry and (B) TAVR for uses that are not expressly listed as an FDA approved indication should continue to address the questions below via clinical studies.

Based on the evidence review, we require that the following outcomes be critically evaluated through a minimum of 1 year:

• stroke;
• all-cause mortality;
• transient ischemic attacks (TIAs);
• major vascular events;
• acute kidney injury;
• repeat aortic valve procedures;
• new permanent pacemaker implantation;
• change in quality of life pre- and post-TAVR.

Given the generally supportive evidence from new studies published since our proposed decision memorandum, and absence of any new, definitive contradictory evidence, we are finalizing the above requirements.

Registry Requirement:

In addition to the one-year outcomes detailed above, CMS had additional questions regarding TAVR for the treatment of symptomatic aortic valve stenosis when furnished according to an FDA approved indication (see Appendix C for the current 20.32 NCD). Based on our concerns at the time, we required additional data to be collected via a registry under the CED paradigm (see https://www.cms.gov/medicare-coverage-database/details/nca-details.aspx?NCAId=257&NCDId=355).

The CMS-approved registry needed to collect all data necessary and have a written executable analysis plan in place to address the following questions (to appropriately address some questions, Medicare claims or other outside data may be necessary):

• When performed outside a controlled clinical study, how do outcomes and adverse events compare to the pivotal clinical studies?
• How do outcomes and adverse events in subpopulations compare to patients in the pivotal clinical studies?
• What is the long term ( ≥ 5 year) durability of the device?
• What are the long term ( ≥ 5 year) outcomes and adverse events?
• How do the demographics of registry patients compare to the pivotal studies?

Since the 2012 NCD, there has been one CMS-approved national registry, the TVT Registry. The TVT Registry was established by the ACC and STS in coordination with industry stakeholders, CMS, and FDA as a platform for CMS coverage with evidence development and FDA post-market approval studies, and to provide clinical performance feedback to registry participants for quality improvement.

The registry collects and standardizes information including patient demographics, comorbidities, functional status including neurocognitive function, quality of life, procedural details (e.g., device and access used), and discharge locations, as well as in-hospital outcomes such as mortality and stroke. Longer-term (30-day and 1-year or longer) patient outcomes are determined through clinical follow up or via linkage with CMS claims data, and reflect the range of primary and secondary outcomes seen in TAVR clinical trials. The standardization of data includes harmonization with internationally-used criteria from the Value Academic Research Consortium where possible.

We assessed the extent to which the TVT Registry published literature has addressed the following five questions for the registry data collection requirement in the 2012 NCD. Based on our analysis, the peer-reviewed publications were directly related to these questions. In addition, we have summarized the publications and have assessed to what extent they have answered these questions.

We believe there is encouraging support for the benefits of TAVR achieved in clinical trials are generalizable to broader community practice, for TAVR sites and heart teams meeting registry criteria.  Registry patients appear to be benefitting from TAVR overall, with a risk-benefit profile not unlike those seen in clinical trials. Moreover, patient outcomes for the U.S. registry appear similar to those in German and French TAVR registries (Walther 2015, Gilard 2016), although we have not analyzed those other registries in any depth.

The latest published summary of demographic and clinical characteristics and annual outcomes and trends of U.S. TVT registry patients (from 2012 through 2015) presents a picture similar to trial patients (Grover 2017). As discussed in more detail in the Evidence section of this decision, the 54,782 registry patients included in this study were elderly (mean age of 83 years), 52% male, 94% white, with multiple comorbidities, a TVT risk score (based on TAVR inpatient mortality) of 4% overall that significantly decreased to 3% in 2015, and an STS mortality risk score (the probability of death in 30 days post procedure) of 7% overall which significantly decreased to 6% in 2015 (see Table 2 for STS scores of key trials). The preferred transfemoral artery access was used in 74% overall, significantly increasing to 87% in 2015.

Registry patients have had similar or better outcomes compared to their trial counterparts. Although no direct comparison to trial patients was made, Grover and colleagues reported that for registry patients, unadjusted TAVR in-hospital, 30-day, and 1-year mortality significantly decreased from 2012 to 2015 (with in-hospital mortality, likely related to the procedure, dropping from 6% to 3%). They state that "1-year mortality continues to be high [22%], and further investigation into the predictors of which patients are unlikely to benefit from the procedure, both in terms of survival and quality of life, needs to occur." Stroke has remained relatively stable (in-hospital and 30-day at 2.1% and 1-year at less than 4%). Paravalvular regurgitation, major bleeding, vascular complications, acute kidney injury, and new atrial fibrillation have all decreased over time. A similar picture of relative outcomes was seen in the previous TVT Registry annual outcomes report (Holmes 2016).

As for device durability, Grover and colleagues conclude that, "At this point, little is known about the ultimate durability of the transcatheter valves.  There have been no definite negative signs detected by the TVT Registry to date; however, the longest follow-up is only 3 years" (Grover 2017). This is consistent with analyses of 5-year follow-up of PARTNER 1 trial patients (Mack 2015, Kapadia 2015). We agree with Reardon’s conclusion that "the lack of structural valve deterioration at 5 years [for both PARTNER 1A and 1B patients] is highly encouraging but is not adequate time to judge the durability of biological valves. Long-term follow-up remains imperative to establish the long- term durability of this technology in patients where survival beyond 5 years is expected" (Reardon 2015).

Brennan and colleagues in turn directly compared TAVR and SAVR in similar (propensity-score matched) patients at high and intermediate surgical risk to test whether the non-inferiority of TAVR determined in trials is seen in real-world clinical practice as well (Brennan 2017). Using data from the TVT Registry (for the TAVR cohort) and the STS National Database (for the SAVR cohort), the investigators found that in unselected (non-trial) intermediate- and high-risk patients who may have been considered eligible for either treatment, "TAVR and SAVR resulted in similar rates of death, stroke, and DAOH [days alive and out of hospital] to 1 year, but TAVR patients were more likely to be discharged to home" (Brennan 2017).

Brennan’s comparison of TVT Registry to STS National Database patients, while not the same thing as a direct comparison between similar TAVR trial and TVT registry patients, does offer an apples-to-apples comparison of TAVR versus SAVR in real-world patients, and supports the hypothesis that trial results are generalizable.

Despite the encouraging evidence that TAVR performed in non-trial settings benefits Medicare patients, we acknowledge that there is no one study or collection of studies that lay out the comparison of pre-procedural patient characteristics and short-term, and longer-term (5-year), patient outcomes and device durability between trial and registry patients across the spectrum of patient risk categories. We believe that participation in the TVT registry, in parallel with ongoing and planned clinical trials, remains a good combination to achieve the appropriate access while stimulating innovation of devices and procedural techniques, and evolution of indications for use in various.

At the same time, we encourage efforts to render the TVT registry less burdensome, more efficient, and more useful to participating practitioners, while allowing for continual improvement in TAVR standards and quality, hence patient outcomes. We thus support the efforts of the societies that run the registry to continually improve it.

In sum, we believe that evidence gaps remain which result in uncertainty about the overall impact of TAVR on beneficiary outcomes when furnished outside of the setting of evidence development or clinical trial protocols.

Despite great progress, important gaps remain in the evidence base. Key questions that if answered could fill these gaps include:

• What are the outcomes (e.g., survival, quality of life, complications, device durability, ancillary needs such as for pacemakers, etc.) for ongoing trials TAVR pivotal studies? What are the long term (5-year) survival and device durability outcomes for each surgical risk group?  Are the outcomes of TVT Registry patients similar to those observed in pivotal trials?
• What is the echocardiographic, CT and/or MR assessment of transcatheter aortic valvular performance, deterioration and durability as compared to surgical AVR?
• Within patient populations (defined by risk level) for which TAVR has demonstrated a benefit, what are the pre-procedural patient characteristics (including comorbidities), and procedure-related factors, that predict outcomes? Can standardized, patient- and family-friendly, evidence-based risk assessment tools improve patient-physician shared decision making? What subgroups of patients within a given population may benefit substantially more or less from the procedure?
• How can complications associated with various TAVR devices and delivery systems, such as paravalvular regurgitation, need for permanent pacemaker implantation, and vascular events, be further reduced in severity and frequency?

We are thus finalizing our conclusion that CED supports appropriate, investigator-led research to answer the above questions. We anticipate the combined input of research on TAVR that includes: randomized, controlled trials; follow-up studies on trial patient outcomes; secondary analyses and meta- analyses; and studies of non-trial, registry patients to include comparison of outcomes to their trial counterparts. Therefore, we are maintaining the CMS-approved registry data collection requirement for addressing the following questions (to appropriately address some questions, Medicare claims or other outside data may be necessary):

• When performed outside a controlled clinical study, how do outcomes and adverse events compare to the pivotal clinical studies?
• What is the long term durability of the device?
• What are the long term outcomes and adverse events?
• What morbidity and procedure-related factors contribute to TAVR patients outcomes?

Considerations for Further Research:
Endorsed in the 2018 AATS/ACC/SCAI/STS Expert Consensus guidelines (Bavaria 2018), current professional guidelines call for clinicians to use shared decision making (SDM) when comparable, but distinctly different, treatment options exist for valvular heart disease. The 2018 Expert Consensus criteria (Bavaria 2018) for new or existing TAVR programs include the "[u]se of an SDM process incorporating patient preference" (Bavaria 2018).

We recognize the importance of shared decision making in many clinical scenarios and have required shared decision making in other NCDs (for example, implantable cardiac defibrillators: https://www.cms.gov/medicare-coverage-database/details/ncd-details.aspx?NCDId=110).  As part of SDM, we have required the use of an evidence-based decision aid or tool.  We support patient shared decision making in AVR but there is not a fully developed tool available at this time. We note there are tools in development. For example, the Patient-Centered Outcomes Research Institute (PCORI) funded research conducted by Brennan and colleagues (CER-1306-04350/ NCT02266251), created and assessed a personalized decision assistance tool designed to evaluate important health outcomes comparing SAVR to TAVR replacement for operable patients with aortic valve disease considering aortic valve replacement, and to develop and assess a personalized risk assessment tool designed to evaluate expected health outcomes with transcatheter aortic valve replacement for inoperable patients considering aortic valve replacement.

We strongly encourage standardized decision aids or tools [the National Quality Forum (NQF) has published standards for decision aids (www.qualityforum.org/Projects/c-d/Decision_Aids/Final_Report.aspx)] to facilitate the decision making process between a patient and physician and will be monitoring this space closely.

Additional research and evidence are needed on patients 90 years of age and older. Arsalan et al. (2016) reported that 15.7% of patients in the registry were ≥90 years and had significantly higher surgical risk, 30 day and 1 year mortality.

Gender
There is conflicting evidence regarding gender disparities. Compared to men, women had improved survival after TAVR (Williams 2014; Conrotto 2015; O’Connor 2015), but a statistically significant higher risk of stroke (Williams 2014; O’Connor 2015), major bleeding (Conrotto 2015; O’Connor 2015), and major vascular complications (Williams 2014; Conrotto 2015; O’Connor 2015). Chieffo and colleagues (2018) reported that intermediate to high-risk women enrolled in the WIN-TAVI (Women’s INternational Transcatheter Aortic Valve Implantation) registry, the first ever all-women contemporary TAVR registry, experienced a low incidence of 1-year mortality and stroke. Yet, men are more frequently treated with TAVR than women (Leon Presentation, slide 17, MEDCAC, 2018, https://www.cms.gov/medicare-coverage-database/details/medcac-meeting-details.aspx?MEDCACId=75).  In light of the disparate health outcomes for women and the differential growth in TAVR, CMS recognizes that gender disparities in TAVR administration persists. We believe that the innovative all-women WIN-TAVI registry should begin to address gender disparities.

Race
The US population consists of 15% African Americans and 17% Hispanic, and only 4% of TAVRs are performed in African Americans and 4.3% in Hispanics (Tommaso presentation, slide 45, MEDCAC, 2018, https://www.cms.gov/medicare-coverage-database/details/medcac-meeting-details.aspx?MEDCACId=75). Using STS/TVT registry data, the 3.8% of Black/African-Americans undergoing TAVR has not changed from 2012 to 2015 (Grover 2016). Caucasians are more likely to receive TAVR than African Americans (Leon presentation, slide 17, MEDCAC, 2018, https://www.cms.gov/medicare-coverage-database/details/medcac-meeting-details.aspx?MEDCACId=75; Sleder 2017),  The TAVR MEDCAC panel acknowledged that there is an under diagnosis and treatment of aortic stenosis in the African-American population regardless of the aortic valve therapy approach (e.g., SAVR, TAVR, etc.).  With respect to treatment outcomes, compared to the Caucasian population, African Americans had similar rates of post-TAVR mortality (McNeely 2018; Alqahtani 2018; Minha 2015), stroke, permanent pacemaker implantation, vascular complications, and acute kidney injury (Alqahtani 2018). We are finalizing changes introducing greater flexibility to the heart team and hospital volume requirements and expiration of hospital volume requirements for established TAVR programs. These changes aim to provide appropriate patient access while ensuring hospitals and heart teams have the experience and capabilities to handle complex structural heart disease cases, with an evolving evidence base and reducing unintended barriers to TAVR.  Additional future steps for the stakeholder community, as discussed during the MEDCAC, include the development of a greater understanding of patient barriers to TAVR use and utilizing that information to inform awareness campaigns directed toward patients and physicians. The lack of evidence in minority populations is a challenge to developing appropriate decision aids or tools for patient engagement and decision making as well.

IX. Conclusion/Final Decision

The Centers for Medicare & Medicaid Services (CMS) will cover Transcatheter Aortic Valve Replacement (TAVR) for the treatment of symptomatic aortic valve stenosis through Coverage with Evidence Development (CED).

1. The procedure is furnished with a complete aortic valve and implantation system that has received FDA premarket approval (PMA) for that system's FDA approved indication.
2. The patient (preoperatively and postoperatively) is under the care of a heart team: a cohesive, multi-disciplinary, team of medical professionals. The heart team concept embodies collaboration and dedication across medical specialties to offer optimal patient-centered care. The heart team includes the following:
   1. Cardiac surgeon and an interventional cardiologist experienced in the care and treatment of aortic stenosis who have:
      1. independently examined the patient face-to-face, evaluated the patient’s suitability for surgical aortic valve replacement (SAVR), TAVR or medical or palliative therapy;
      2. documented and made available to the other heart team members the rationale for their clinical judgment.
   2. Providers from other physician groups as well as advanced patient practitioners, nurses, research personnel and administrators.
3. The heart team's interventional cardiologist(s) and cardiac surgeon(s) must jointly participate in the intra-operative technical aspects of TAVR.
4. TAVR must be furnished in a hospital with the appropriate infrastructure that includes but is not limited to:
   1. On-site heart valve surgery and interventional cardiology programs,
   2. Post-procedure intensive care facility with personnel experienced in managing patients who have undergone open-heart valve procedures,
   3. Appropriate volume requirements per the applicable qualifications below:

   There are two sets of qualifications; the first set outlined below is for hospital programs and heart teams without previous TAVR experience and the second set is for those with TAVR experience.

   Qualifications to begin a TAVR program for hospitals without TAVR experience:

   The hospital program must have the following:

   1. ≥ 50 open heart surgeries in the previous year prior to TAVR program initiation, and;
   2. ≥ 20 aortic valve related procedures in the 2 years prior to TAVR program initiation, and;
   3. ≥ 2 physicians with cardiac surgery privileges, and;
   4. ≥ 1 physician with interventional cardiology privileges, and;
   5. ≥ 300 percutaneous coronary interventions (PCIs) per year.

   Qualifications to begin a TAVR program for heart teams without TAVR experience:

   The heart team must include:

   1. Cardiovascular surgeon with:
      1. ≥ 100 career open heart surgeries of which ≥ 25 are aortic valve related; and,
   2. Interventional cardiologist with:
      1. Professional experience of ≥ 100 career structural heart disease procedures; or, ≥ 30 left-sided structural procedures per year; and,
      2. Device-specific training as required by the manufacturer.

   Qualifications for hospital programs with TAVR experience:

   The hospital program must maintain the following:

   1. ≥ 50 AVRs (TAVR or SAVR) per year including ≥ 20 TAVR procedures in the prior year ; or,
   2. ≥ 100 AVRs (TAVR or SAVR) every 2 years, including ≥ 40 TAVR procedures in the prior 2 years; and,
   3. ≥ 2 physicians with cardiac surgery privileges; and,
   4. ≥ 1 physician with interventional cardiology privileges, and
   5. ≥300 percutaneous coronary interventions (PCIs) per year;and,


5. The heart team and hospital are participating in a prospective, national, audited registry that: 1) consecutively enrolls TAVR patients; 2) accepts all manufactured devices; 3) follows the patient for at least one year; and, 4) complies with relevant regulations relating to protecting human research subjects, including 45 CFR Part 46 and 21 CFR Parts 50 & 56. 

   The following outcomes must be tracked by the registry; and the registry must be designed to permit identification and analysis of patient, practitioner and facility level variables that predict each of these outcomes:

   1. Stroke;
   2. All-cause mortality;
   3. Transient Ischemic Attacks (TIAs);
   4. Major vascular events;
   5. Acute kidney injury;
   6. Repeat aortic valve procedures;
   7. New permanent pacemaker implantation;
   8. Quality of Life (QoL).


6. The registry shall collect all data necessary and have a written executable analysis plan in place to address the following questions (to appropriately address some questions, Medicare claims or other outside data may be necessary). Specifically, for the CED question iv, this must be addressed through a composite metric. For the below CED questions (i-iv), the results must be reported publicly as described in CED criterion k.
   1. When performed outside a controlled clinical study, how do outcomes and adverse events compare to the pivotal clinical studies?
   2. What is the long term durability of the device?
   3. What are the long term outcomes and adverse events?
   4. What morbidity and procedure-related factors contribute to TAVR patients outcomes?

   Consistent with section 1142 of the Act, the Agency for Healthcare Research and Quality (AHRQ) supports clinical research studies that CMS determines meet the above-listed standards and address the above-listed research questions.

B. TAVR is covered for uses that are not expressly listed as an FDA-approved indication when performed within a clinical study that fulfills all of the following:

1. The heart team's interventional cardiologist(s) and cardiac surgeon(s) must jointly participate in the intra-operative technical aspects of TAVR.
2. As a fully-described, written part of its protocol, the clinical research study must critically evaluate not only each patient's quality of life pre- and post-TAVR (minimum of 1 year), but must also address at least one of the following questions:
   • What is the incidence of stroke?
   • What is the rate of all-cause mortality?
   • What is the incidence of new permanent pacemaker implantation?
   • What is the incidence of transient ischemic attacks (TIAs)?
   • What is the incidence of major vascular events?
   • What is the incidence of acute kidney injury?
   • What is the incidence of repeat aortic valve procedures?


3. The clinical study must adhere to the following standards of scientific integrity and relevance to the Medicare population:
   1. The principal purpose of the study is to test whether the item or service meaningfully improves health outcomes of affected beneficiaries who are represented by the enrolled subjects.
   2. The rationale for the study is well supported by available scientific and medical evidence.
   3. The study results are not anticipated to unjustifiably duplicate existing knowledge.
   4. The study design is methodologically appropriate and the anticipated number of enrolled subjects is sufficient to answer the research question(s) being asked in the National Coverage Determination.
   5. The study is sponsored by an organization or individual capable of completing it successfully.
   6. The research study is in compliance with all applicable Federal regulations concerning the protection of human subjects found in the Code of Federal Regulations (CFR) at 45 CFR Part 46. If a study is regulated by the Food and Drug Administration (FDA), it is also in compliance with 21 CFR Parts 50 and 56. In addition, to further enhance the protection of human subjects in studies conducted under CED, the study must provide and obtain meaningful informed consent from patients regarding the risks associated with the study items and /or services, and the use and eventual disposition of the collected data
   7. All aspects of the research study are conducted according to appropriate standards of scientific integrity.
   8. The study has a written protocol that clearly demonstrates adherence to the standards listed here as Medicare requirements.
   9. The study is not designed to exclusively test toxicity or disease pathophysiology in healthy individuals. Such studies may meet this requirement only if the disease or condition being studied is life threatening as defined in 21 CFR §312.81(a) and the patient has no other viable treatment options.
   10. The clinical research studies and registries are registered on the www.ClinicalTrials.gov website by the principal sponsor/investigator prior to the enrollment of the first study subject. Registries are also registered in the Agency for Healthcare Quality (AHRQ) Registry of Patient Registries (RoPR).
   11. The research study protocol specifies the method and timing of public release of all prespecified outcomes to be measured including release of outcomes if outcomes are negative or study is terminated early. The results must be made public within 12 months of the study’s primary completion date, which is the date the final subject had final data collection for the primary endpoint, even if the trial does not achieve its primary aim. The results must include number started/completed, summary results for primary and secondary outcome measures, statistical analyses, and adverse events. Final results must be reported in a publicly accessibly manner; either in a peer-reviewed scientific journal (in print or on-line), in an on-line publicly accessible registry dedicated to the dissemination of clinical trial information such as ClinicalTrials.gov, or in journals willing to publish in abbreviated format (e.g., for studies with negative or incomplete results).
   12. The study protocol must explicitly discuss beneficiary subpopulations affected by the item or service under investigation, particularly traditionally underrepresented groups in clinical studies, how the inclusion and exclusion criteria affect enrollment of these populations, and a plan for the retention and reporting of said populations on the trial. If the inclusion and exclusion criteria are expected to have a negative effect on the recruitment or retention of underrepresented populations, the protocol must discuss why these criteria are necessary.
   13. The study protocol explicitly discusses how the results are or are not expected to be generalizable to affected beneficiary subpopulations. Separate discussions in the protocol may be necessary for populations eligible for Medicare due to age, disability or Medicaid eligibility.

Consistent with section 1142 of the Act, the Agency for Healthcare Research and Quality (AHRQ) supports clinical research studies that meet the above-listed standards and address the above-listed research questions.

The principal investigator must submit the complete study protocol, identify the relevant CMS research question(s) that will be addressed, and cite the location of the detailed analysis plan for those questions in the protocol, plus provide a statement addressing how the study satisfies each of the standards of scientific integrity (a. through m. listed above), as well as the investigator's contact information, to the address below. The information will be reviewed, and approved studies will be identified on the CMS Website.

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

Email address for protocol submissions: clinicalstudynotification@cms.hhs.gov
Email subject line: "CED [NCD topic (i.e. TAVR)] [name of sponsor/primary investigator]"

See Appendix B for the NCD manual language.

When making national coverage determinations, CMS evaluates relevant clinical evidence to determine whether or not the evidence is of sufficient quality to support a finding that an item or service is reasonable and necessary.  The overall objective for the critical appraisal of the evidence is to determine to what degree we are confident that: 1) the specific assessment questions can be answered conclusively; and 2) the intervention will improve health outcomes for patients.

We divide the assessment of clinical evidence into three stages: 1) the quality of the individual studies; 2) the generalizability of findings from individual studies to the Medicare population; and 3) overarching conclusions that can be drawn from the body of the evidence on the direction and magnitude of the intervention’s potential risks and benefits.

The methodological principles described below represent a broad discussion of the issues we consider when reviewing clinical evidence.  However, it should be noted that each coverage determination has its unique methodological aspects.

Assessing Individual Studies

Methodologists have developed criteria to determine weaknesses and strengths of clinical research.  Strength of evidence generally refers to: 1) the scientific validity underlying study findings regarding causal relationships between health care interventions and health outcomes; and 2) the reduction of bias.  In general, some of the methodological attributes associated with stronger evidence include those listed below:

• Use of randomization (allocation of patients to either intervention or control group) in order to minimize bias.
• Use of contemporaneous control groups (rather than historical controls) in order to ensure comparability between the intervention and control groups.
• Prospective (rather than retrospective) studies to ensure a more thorough and systematical assessment of factors related to outcomes.
• Larger sample sizes in studies to demonstrate both statistically significant as well as clinically significant outcomes that can be extrapolated to the Medicare population.  Sample size should be large enough to make chance an unlikely explanation for what was found.
• Masking (blinding) to ensure patients and investigators do not know to that group patients were assigned (intervention or control).  This is important especially in subjective outcomes, such as pain or quality of life, where enthusiasm and psychological factors may lead to an improved perceived outcome by either the patient or assessor.

Regardless of whether the design of a study is a randomized controlled trial, a non-randomized controlled trial, a cohort study or a case-control study, the primary criterion for methodological strength or quality is to the extent that differences between intervention and control groups can be attributed to the intervention studied.  This is known as internal validity.  Various types of bias can undermine internal validity.  These include:

• Different characteristics between patients participating and those theoretically eligible for study but not participating (selection bias).
• Co-interventions or provision of care apart from the intervention under evaluation (performance bias).
• Differential assessment of outcome (detection bias).
• Occurrence and reporting of patients who do not complete the study (attrition bias).

In principle, rankings of research design have been based on the ability of each study design category to minimize these biases.  A randomized controlled trial minimizes systematic bias (in theory) by selecting a sample of participants from a particular population and allocating them randomly to the intervention and control groups.  Thus, in general, randomized controlled studies have been typically assigned the greatest strength, followed by non-randomized clinical trials and controlled observational studies.  The design, conduct and analysis of trials are important factors as well.  For example, a well-designed and conducted observational study with a large sample size may provide stronger evidence than a poorly designed and conducted randomized controlled trial with a small sample size.  The following is a representative list of study designs (some of that have alternative names) ranked from most to least methodologically rigorous in their potential ability to minimize systematic bias:

Randomized controlled trials
Non-randomized controlled trials
Prospective cohort studies
Retrospective case control studies
Cross-sectional studies
Surveillance studies (e.g., using registries or surveys)
Consecutive case series
Single case reports

When there are merely associations but not causal relationships between a study’s variables and outcomes, it is important not to draw causal inferences.  Confounding refers to independent variables that systematically vary with the causal variable.  This distorts measurement of the outcome of interest because its effect size is mixed with the effects of other extraneous factors.  For observational, and in some cases randomized controlled trials, the method in that confounding factors are handled (either through stratification or appropriate statistical modeling) are of particular concern.  For example, in order to interpret and generalize conclusions to our population of Medicare patients, it may be necessary for studies to match or stratify their intervention and control groups by patient age or co-morbidities.

Methodological strength is, therefore, a multidimensional concept that relates to the design, implementation and analysis of a clinical study.  In addition, thorough documentation of the conduct of the research, particularly study selection criteria, rate of attrition and process for data collection, is essential for CMS to adequately assess and consider the evidence.

Generalizability of Clinical Evidence to the Medicare Population

The applicability of the results of a study to other populations, settings, treatment regimens and outcomes assessed is known as external validity.  Even well-designed and well-conducted trials may not supply the evidence needed if the results of a study are not applicable to the Medicare population.  Evidence that provides accurate information about a population or setting not well represented in the Medicare program would be considered but would suffer from limited generalizability.

The extent to that the results of a trial are applicable to other circumstances is often a matter of judgment that depends on specific study characteristics, primarily the patient population studied (age, sex, severity of disease and presence of co-morbidities) and the care setting (primary to tertiary level of care, as well as the experience and specialization of the care provider).  Additional relevant variables are treatment regimens (dosage, timing and route of administration), co-interventions or concomitant therapies, and type of outcome and length of follow-up.

The level of care and the experience of the providers in the study are other crucial elements in assessing a study’s external validity.  Trial participants in an academic medical center may receive more or different attention than is typically available in non-tertiary settings.  For example, an investigator’s lengthy and detailed explanations of the potential benefits of the intervention and/or the use of new equipment provided to the academic center by the study sponsor may raise doubts about the applicability of study findings to community practice.

Given the evidence available in the research literature, some degree of generalization about an intervention’s potential benefits and harms is invariably required in making coverage determinations for the Medicare population.  Conditions that assist us in making reasonable generalizations are biologic plausibility, similarities between the populations studied and Medicare patients (age, sex, ethnicity and clinical presentation) and similarities of the intervention studied to those that would be routinely available in community practice.

A study’s selected outcomes are an important consideration in generalizing available clinical evidence to Medicare coverage determinations.  One of the goals of our determination process is to assess health outcomes.  These outcomes include resultant risks and benefits such as increased or decreased morbidity and mortality.  In order to make this determination, it is often necessary to evaluate whether the strength of the evidence is adequate to draw conclusions about the direction and magnitude of each individual outcome relevant to the intervention under study.  In addition, it is important that an intervention’s benefits are clinically significant and durable, rather than marginal or short-lived.  Generally, an intervention is not reasonable and necessary if its risks outweigh its benefits.

If key health outcomes have not been studied or the direction of clinical effect is inconclusive, we may also evaluate the strength and adequacy of indirect evidence linking intermediate or surrogate outcomes to our outcomes of interest.

Assessing the Relative Magnitude of Risks and Benefits

Generally, an intervention is not reasonable and necessary if its risks outweigh its benefits.  Health outcomes are one of several considerations in determining whether an item or service is reasonable and necessary.  CMS places greater emphasis on health outcomes actually experienced by patients, such as quality of life, functional status, duration of disability, morbidity and mortality, and less emphasis on outcomes that patients do not directly experience, such as intermediate outcomes, surrogate outcomes, and laboratory or radiographic responses.  The direction, magnitude, and consistency of the risks and benefits across studies are also important considerations.  Based on the analysis of the strength of the evidence, CMS assesses the relative magnitude of an intervention or technology’s benefits and risk of harm to Medicare beneficiaries.

[XXX.X]

A. General

Transcatheter aortic valve replacement (TAVR - also known as TAVI or transcatheter aortic valve implantation) is used in the treatment of aortic stenosis. A bioprosthetic valve is inserted percutaneously using a catheter and implanted in the orifice of the aortic valve.

B. Nationally Covered Indications

The Centers for Medicare & Medicaid Services (CMS) covers transcatheter aortic valve replacement (TAVR) under Coverage with Evidence Development (CED) with the following conditions:

1. The procedure is furnished with a complete aortic valve and implantation system that has received FDA premarket approval (PMA) for that system's FDA approved indication.
2. The patient (preoperatively and postoperatively) is under the care of a heart team: a cohesive, multi-disciplinary, team of medical professionals. The heart team concept embodies collaboration and dedication across medical specialties to offer optimal patient-centered care. The heart team includes the following:
   1. Cardiac surgeon and an interventional cardiologist experienced in the care and treatment of aortic stenosis who have:
      1. independently examined the patient face-to-face, evaluated the patient’s suitability for surgical aortic valve replacement (SAVR), TAVR or medical or palliative therapy;
      2. documented and made available to the other heart team members the rationale for their clinical judgment.
   2. Providers from other physician groups as well as advanced patient practitioners, nurses, research personnel and administrators.

3. The heart team's interventional cardiologist(s) and cardiac surgeon(s) must jointly participate in the intra-operative technical aspects of TAVR.


4. TAVR must be furnished in a hospital with the appropriate infrastructure that includes but is not limited to:
   1. On-site heart valve surgery and interventional cardiology programs,
   2. Post-procedure intensive care facility with personnel experienced in managing patients who have undergone open-heart valve procedures,
   3. Appropriate volume requirements per the applicable qualifications below:

There are two sets of qualifications; the first set outlined below is for hospital programs and heart teams without previous TAVR experience and the second set is for those with TAVR experience.

5. The heart team and hospital are participating in a prospective, national, audited registry that:

   1) consecutively enrolls TAVR patients; 2) accepts all manufactured devices; 3) follows the patient for at least one year; and, 4) complies with relevant regulations relating to protecting human research subjects, including 45 CFR Part 46 and 21 CFR Parts 50 & 56. 

   The following outcomes must be tracked by the registry; and the registry must be designed to permit identification and analysis of patient, practitioner and facility level variables that predict each of these outcomes:

   1. Stroke;
   2. All-cause mortality;
   3. Transient Ischemic Attacks (TIAs);
   4. Major vascular events;
   5. Acute kidney injury;
   6. Repeat aortic valve procedures;
   7. New permanent pacemaker implantation;
   8. Quality of Life (QoL).


6. The registry shall collect all data necessary and have a written executable analysis plan in place to address the following questions (to appropriately address some questions, Medicare claims or other outside data may be necessary). Specifically, for the CED question iv, this must be addressed through a composite metric. For the below CED questions (i-iv), the results must be reported publicly as described in CED criterion k.
   1. When performed outside a controlled clinical study, how do outcomes and adverse events compare to the pivotal clinical studies?
   2. What is the long term durability of the device?
   3. What are the long term outcomes and adverse events?
   4. What morbidity and procedure-related factors contribute to TAVR patients outcomes?

   Consistent with section 1142 of the Act, the Agency for Healthcare Research and Quality (AHRQ) supports clinical research studies that CMS determines meet the above-listed standards and address the above-listed research questions.

B. TAVR is covered for uses that are not expressly listed as an FDA-approved indication when performed within a clinical study that fulfills all of the following:

1. The heart team's interventional cardiologist(s) and cardiac surgeon(s) must jointly participate in the intra-operative technical aspects of TAVR.
2. As a fully-described, written part of its protocol, the clinical research study must critically evaluate not only each patient's quality of life pre- and post-TAVR (minimum of 1 year), but must also address at least one of the following questions:
   • What is the incidence of stroke?
   • What is the rate of all-cause mortality?
   • What is the incidence of new permanent pacemaker implantation?
   • What is the incidence of transient ischemic attacks (TIAs)?
   • What is the incidence of major vascular events?
   • What is the incidence of acute kidney injury?
   • What is the incidence of repeat aortic valve procedures?


3. The clinical study must adhere to the following standards of scientific integrity and relevance to the Medicare population:
   1. The principal purpose of the study is to test whether the item or service meaningfully improves health outcomes of affected beneficiaries who are represented by the enrolled subjects.
   2. The rationale for the study is well supported by available scientific and medical evidence.
   3. The study results are not anticipated to unjustifiably duplicate existing knowledge.
   4. The study design is methodologically appropriate and the anticipated number of enrolled subjects is sufficient to answer the research question(s) being asked in the National Coverage Determination.
   5. The study is sponsored by an organization or individual capable of completing it successfully.
   6. The research study is in compliance with all applicable Federal regulations concerning the protection of human subjects found in the Code of Federal Regulations (CFR) at 45 CFR Part 46. If a study is regulated by the Food and Drug Administration (FDA), it is also in compliance with 21 CFR Parts 50 and 56. In addition, to further enhance the protection of human subjects in studies conducted under CED, the study must provide and obtain meaningful informed consent from patients regarding the risks associated with the study items and /or services, and the use and eventual disposition of the collected data.
   7. All aspects of the research study are conducted according to appropriate standards of scientific integrity.
   8. The study has a written protocol that clearly demonstrates adherence to the standards listed here as Medicare requirements.
   9. The study is not designed to exclusively test toxicity or disease pathophysiology in healthy individuals. Such studies may meet this requirement only if the disease or condition being studied is life threatening as defined in 21 CFR §312.81(a) and the patient has no other viable treatment options.
   10. The clinical research studies and registries are registered on the www.ClinicalTrials.gov website by the principal sponsor/investigator prior to the enrollment of the first study subject. Registries are also registered in the Agency for Healthcare Quality (AHRQ) Registry of Patient Registries (RoPR).
   11. The research study protocol specifies the method and timing of public release of all prespecified outcomes to be measured including release of outcomes if outcomes are negative or study is terminated early. The results must be made public within 12 months of the study’s primary completion date, which is the date the final subject had final data collection for the primary endpoint, even if the trial does not achieve its primary aim. The results must include number started/completed, summary results for primary and secondary outcome measures, statistical analyses, and adverse events. Final results must be reported in a publicly accessibly manner; either in a peer-reviewed scientific journal (in print or on-line), in an on-line publicly accessible registry dedicated to the dissemination of clinical trial information such as ClinicalTrials.gov, or in journals willing to publish in abbreviated format (e.g., for studies with negative or incomplete results).
   12. The study protocol must explicitly discuss beneficiary subpopulations affected by the item or service under investigation, particularly traditionally underrepresented groups in clinical studies, how the inclusion and exclusion criteria effect enrollment of these populations, and a plan for the retention and reporting of said populations on the trial. If the inclusion and exclusion criteria are expected to have a negative effect on the recruitment or retention of underrepresented populations, the protocol must discuss why these criteria are necessary.
   13. The study protocol explicitly discusses how the results are or are not expected to be generalizable to affected beneficiary subpopulations. Separate discussions in the protocol may be necessary for populations eligible for Medicare due to age, disability or Medicaid eligibility.

Consistent with section 1142 of the Act, the Agency for Healthcare Research and Quality (AHRQ) supports clinical research studies that meet the above-listed standards and address the above-listed research questions.

The principal investigator must submit the complete study protocol, identify the relevant CMS research question(s) that will be addressed, and cite the location of the detailed analysis plan for those questions in the protocol, plus provide a statement addressing how the study satisfies each of the standards of scientific integrity (a. through m. listed above), as well as the investigator's contact information, to the address below. The information will be reviewed, and approved studies will be identified on the CMS Website.

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

Email address for protocol submissions: clinicalstudynotification@cms.hhs.gov
Email subject line: “CED [NCD topic (i.e. TAVR)] [name of sponsor/primary investigator]”

C. Nationally Non-Covered Indications

TAVR is not covered for patients in whom existing co-morbidities would preclude the expected benefit from correction of the aortic stenosis.

D. Other

NA

(This NCD last reviewed June 2019.)

Item/Service Description 

Transcatheter aortic valve replacement (TAVR - also known as TAVI or transcatheter aortic valve implantation) is used in the treatment of aortic stenosis. A bioprosthetic valve is inserted percutaneously using a catheter and implanted in the orifice of the aortic valve.

Indications and Limitations of Coverage 

B.   Nationally Covered Indications

The Centers for Medicare & Medicaid Services (CMS) covers transcatheter aortic valve replacement (TAVR) under Coverage with Evidence Development (CED) with the following conditions:

1. TAVR is covered for the treatment of symptomatic aortic valve stenosis when furnished according to a Food and Drug Administration (FDA)-approved indication and when all of the following conditions are met 
   
   
   1. The procedure is furnished with a complete aortic valve and implantation system that has received FDA premarket approval (PMA) for that system's FDA approved indication.
   
   
   
   2. Two cardiac surgeons have independently examined the patient face-to-face and evaluated the patient's suitability for open aortic valve replacement (AVR) surgery; and both surgeons have documented the rationale for their clinical judgment and the rationale is available to the heart team.
   
   
   
   3. The patient (preoperatively and postoperatively) is under the care of a heart team: a cohesive, multi-disciplinary, team of medical professionals. The heart team concept embodies collaboration and dedication across medical specialties to offer optimal patient-centered care.

TAVR must be furnished in a hospital with the appropriate infrastructure that includes but is not limited to:

1. On-site heart valve surgery program,
2. Cardiac catheterization lab or hybrid operating room/catheterization lab equipped with a fixed radiographic imaging system with flat-panel fluoroscopy, offering quality imaging,
3. Non-invasive imaging such as echocardiography, vascular ultrasound, computed tomography (CT) and magnetic resonance (MR),
4. Sufficient space, in a sterile environment, to accommodate necessary equipment for cases with and without complications,
5. Post-procedure intensive care facility with personnel experienced in managing patients who have undergone open-heart valve procedures,
6. Appropriate volume requirements per the applicable qualifications below.

There are two sets of qualifications; the first set outlined below is for hospital programs and heart teams without previous TAVR experience and the second set is for those with TAVR experience.

Qualifications to begin a TAVR program for hospitals without TAVR experience:

The hospital program must have the following:

Qualifications to begin a TAVR program for heart teams without TAVR experience:

The heart team must include:

1. Cardiovascular surgeon with:
   1. ≥ 100 career AVRs including 10 high-risk patients; or,
   2. ≥ 25 AVRs in one year; or,
   3. ≥ 50 AVRs in 2 years; and which include at least 20 AVRs in the last year prior to TAVR initiation; and,
2. Interventional cardiologist with:
   1. Professional experience with 100 structural heart disease procedures lifetime; or,
   2. 30 left-sided structural procedures per year of which 60% should be balloon aortic valvuloplasty (BAV). Atrial septal defect and patent foramen ovale closure are not considered left-sided procedures; and,
3. Additional members of the heart team such as echocardiographers, imaging specialists, heart failure specialists, cardiac anesthesiologists, intensivists, nurses, and social workers; and,
4. Device-specific training as required by the manufacturer.

Qualifications for hospital programs with TAVR experience:

The hospital program must maintain the following:

1. ≥ 20 AVRs per year or ≥ 40 AVRs every 2 years; and,
2. ≥ 2 physicians with cardiac surgery privileges; and,
3. ≥ 1000 catheterizations per year, including ≥ 400 percutaneous coronary interventions (PCIs) per year.

Qualifications for heart teams with TAVR experience:

The heart team must include:

1. cardiovascular surgeon and an interventional cardiologist whose combined experience maintains the following:
1. ≥ 20 TAVR procedures in the prior year, or,
2. ≥ 40 TAVR procedures in the prior 2 years; and,
2. Additional members of the heart team such as echocardiographers, imaging specialists, heart failure specialists, cardiac anesthesiologists, intensivists, nurses, and social workers.

4.   The heart team's interventional cardiologist(s) and cardiac surgeon(s) must jointly participate in the intra-operative technical aspects of TAVR.

5.   The heart team and hospital are participating in a prospective, national, audited registry that: 1) consecutively enrolls TAVR patients; 2) accepts all manufactured devices; 3) follows the patient for at least one year; and, 4) complies with relevant regulations relating to protecting human research subjects, including 45 CFR Part 46 and 21 CFR Parts 50 & 56. The following outcomes must be tracked by the registry; and the registry must be designed to permit identification and analysis of patient, practitioner and facility level variables that predict each of these outcomes:

The registry should collect all data necessary and have a written executable analysis plan in place to address the following questions (to appropriately address some questions, Medicare claims or other outside data may be necessary):

Consistent with section 1142 of the Act, the Agency for Healthcare Research and Quality (AHRQ) supports clinical research studies that CMS determines meet the above-listed standards and address the above-listed research questions.

2. TAVR is covered for uses that are not expressly listed as an FDA-approved indication when performed within a clinical study that fulfills all of the following.
   1. The heart team's interventional cardiologist(s) and cardiac surgeon(s) must jointly participate in the intra-operative technical aspects of TAVR.
   
   
   
   2. As a fully-described, written part of its protocol, the clinical research study must critically evaluate not only each patient's quality of life pre- and post-TAVR (minimum of 1 year), but must also address at least one of the following questions:
      • What is the incidence of stroke?
      • What is the rate of all-cause mortality?
      • What is the incidence of transient ischemic attacks (TIAs)?
      • What is the incidence of major vascular events?
      • What is the incidence of acute kidney injury?
      • What is the incidence ofrepeat aortic valve procedures?
   3. The clinical study must adhere to the following standards of scientific integrity and relevance to the Medicare population:
      1. The principal purpose of the research study is to test whether a particular intervention potentially improves the participants' health outcomes.
      2. The research study is well supported by available scientific and medical information or it is intended to clarify or establish the health outcomes of interventions already in common clinical use.
      3. The research study does not unjustifiably duplicate existing knowledge.
      4. The research study design is appropriate to answer the research question being asked in the study.
      5. The research study is sponsored by an organization or individual capable of executing the proposed study successfully.
      6. The research study is in compliance with all applicable Federal regulations concerning the protection of human subjects found in the Code of Federal Regulations (CFR) at 45 CFR Part 46. If a study is regulated by the Food and Drug Administration (FDA), it is also in compliance with 21 CFR Parts 50 and 56. In particular, the informed consent includes a straightforward explanation of the reported increased risks of stroke and vascular complications that have been published for TAVR.
      7. All aspects of the research study are conducted according to appropriate standards of scientific integrity (see http://www.icmje.org).
      8. The research study has a written protocol that clearly addresses, or incorporates by reference, the standards listed here as Medicare coverage requirements.
      9. The clinical research study is not designed to exclusively test toxicity or disease pathophysiology in healthy individuals. Trials of all medical technologies measuring therapeutic outcomes as one of the objectives meet this standard only if the disease or condition being studied is life threatening as defined in 21 CFR §312.81(a) and the patient has no other viable treatment options.
      10. The clinical research study is registered on the www.ClinicalTrials.gov website by the principal sponsor/investigator prior to the enrollment of the first study subject.
      11. The research study protocol specifies the method and timing of public release of all pre-specified outcomes to be measured including release of outcomes if outcomes are negative or study is terminated early. The results must be made public within 24 months of the end of data collection. If a report is planned to be published in a peer reviewed journal, then that initial release may be an abstract that meets the requirements of the International Committee of Medical Journal Editors (http://www.icmje.org). However a full report of the outcomes must be made public no later than three (3) years after the end of data collection.
      12. The research study protocol must explicitly discuss subpopulations affected by the treatment under investigation, particularly traditionally underrepresented groups in clinical studies, how the inclusion and exclusion criteria affect enrollment of these populations, and a plan for the retention and reporting of said populations on the trial. If the inclusion and exclusion criteria are expected to have a negative effect on the recruitment or retention of underrepresented populations, the protocol must discuss why these criteria are necessary.
      13. The research study protocol explicitly discusses how the results are or are not expected to be generalizable to the Medicare population to infer whether Medicare patients may benefit from the intervention. Separate discussions in the protocol may be necessary for populations eligible for Medicare due to age, disability or Medicaid eligibility.

Consistent with section 1142 of the Act, AHRQ supports clinical research studies that CMS determines meet the above-listed standards and address the above-listed research questions.

4.    The principal investigator must submit the complete study protocol, identify the relevant CMS research question(s) that will be addressed, and cite the location of the detailed analysis plan for those questions in the protocol, plus provide a statement addressing how the study satisfies each of the standards of scientific integrity (a. through m. listed above), as well as the investigator's contact information, to the address below. The information will be reviewed, and approved studies will be identified on the CMS Website.

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

C.    Nationally Non-Covered Indications

TAVR is not covered for patients in whom existing co-morbidities would preclude the expected benefit from correction of the aortic stenosis.

D.    Other

NA

(This NCD last reviewed May 2012.)

Abdel-Wahab M, Mehilli J, Frerker C, et al. Comparison of balloon-expandable vs self-expandable valves in patients undergoing transcatheter aortic valve replacement: the CHOICE randomized clinical trial. JAMA. 2014;311(15):1503-14. doi: 10.1001/jama.2014.3316.

Abdel-Wahab M, Neumann FJ, Mehilli J, et al. 1-Year Outcomes After Transcatheter Aortic Valve Replacement With Balloon-Expandable Versus Self-Expandable Valves: Results From the CHOICE Randomized Clinical Trial. J Am Coll Cardiol. 2015;66(7):791-800. doi: 10.1016/j.jacc.2015.06.026.

Abramowitz Y, Vemulapalli S, Chakravarty T, et al. Clinical Impact of Diabetes Mellitus on Outcomes After Transcatheter Aortic Valve Replacement: Insights from the STS/ACC TVT Registry. Circ Interv. 2017;10(11).

Ad N, Holmes SD, Shuman DJ, et al. The effect of initiation of a transcatheter aortic valve replacement program in the treatment of severe aortic stenosis. Seminars in Thoracic and Cardiovascular Surgery. 2016;28(2):353-360. PMID: 28043443.

Adams DH, Popma JJ, Reardon MJ, et al. Transcatheter aortic-valve replacement with a self- expanding prosthesis. N Engl J Med. 2014;370(19):1790-8. doi: 10.1056/NEJMoa1400590.

Alfredsson J, Stebbins A, Brennan JM, et al. Gait Speed Predicts 30-Day Mortality After Transcatheter Aortic Valve Replacement: Results from the Society of Thoracic Surgeons/American College of Cardiology Transcatheter Valve Therapy Registry. Circulation. 2016;133.

Alli O, Booker JD, Lennon RJ, et al. Transcatheter aortic valve implantation: assessing the learning curve. JACC Cardiovascular Interventions. 2012 Jan;5(1):72-9. PMID: 22197410.

Alli O, Rihal CS, Suri RM, et al. Learning curves for transfemoral transcatheter aortic valve replacement in the PARTNER-I trial: Technical performance. Catheter Cardiovasc Interv. 2016;87 (1):154-62. doi: 10.1002/ccd.26120. PMID: 26256280.

Alqahtani F, Aljohani S, Almustafa A, et al. Comparative outcomes of transcatheter aortic valve replacement in African American and Caucasian patients with severe aortic stenosis. Catheterization Cardiovascular Interventions. 2018;91(5):932-937. PMID: 28941139.

Ando T, Akintoye E, Telila T, et al. Hospital outcomes of transcatheter versus surgical aortic valve replacement in female in the United States. Catheter Cardiovasc Interv. 2018;91(4):813-819. doi: 10.1002/ccd.27362. PMID: 28990736.

Ando T, Takagi H; Representing ALICE (All-Literature Investigation of Cardiovascular Evidence) Group. Comparison of late mortality after transcatheter aortic valve implantation versus surgical aortic valve replacement: Insights from a meta-analysis. European Journal of Internal Medicine. 2017 May;40:43-49. doi: 10.1016/j.ejim.2017.01.023. PMID: 28162851.

Aortic Stenosis Writing Group, Bonow RO, Brown AS, Gillam LD, et al. ACC/AATS/AHA/ASE/EACTS/HVS/SCA/SCAI/SCCT/SCMR/STS 2017 Appropriate Use Criteria for the treatment of patients with severe aortic stenosis: A report of the American College of Cardiology Appropriate Use Criteria Task Force, American Association for Thoracic Surgery, American Heart Association, American Society of Echocardiography, European Association for Cardio-Thoracic Surgery, Heart Valve Society, Society of Cardiovascular Anesthesiologists, Society for Cardiovascular Angiography and Interventions, Society of Cardiovascular Computed Tomography, Society for Cardiovascular Magnetic Resonance, and Society of Thoracic Surgeons. J Am Soc Echocardiogr. 2018;31(2):117-147. doi: 10.1016/j.echo.2017.10.020.

Applegate PM, Boyd WD, Applegate II RL, et al. Is it the time to reconsider the choice of valves for cardiac surgery: mechanical or bioprosthetic? J Biomed Res. 2017; 31(5): 373-376. doi: 10.7555/JBR.31.20170027. PMID: 28958994.

Arnold SV, Reynolds MR, Wang K, et al. Health status after transcatheter or surgical aortic valve replacement in patients with severe aortic stenosis at increased surgical risk: results from the CoreValve US Pivotal Trial. JACC Cardiovasc Interv. 2015;8(9):1207-1217. doi: 10.1016/j.jcin.2015.04.018.

Arnold SV, Spertus JA, Vemulapalli S, et al. Association of patient-reported health status with long- term mortality after transcatheter aortic valve replacement: Report from the STS/ACC TVT registry. Circ Cardiovasc Interv. 2015 Dec;8(12):e002875. doi: 10.1161/CIRCINTERVENTIONS.115.002875. PMID: 26643740.

Arnold SV, Spertus JA, Vemulapalli S, et al. Quality-of-life outcomes after transcatheter aortic valve replacement in an unselected population: A report from the STS/ACC Transcatheter Valve Therapy Registry. JAMA Cardiol. 2017 Apr 1;2(4):409-416. doi: 10.1001/jamacardio.2016.5302. PMID: 28146260.

Aronow WS, Ahn C, Kronzon I, et al. Association of coronary risk factors and use of statins with progression of mild valvular aortic stenosis in older persons. Am JCardiol. 2001 Sep 15;88(6):693-5. PMID: 11564402.

Arora S, Misenheimer JA, Jones W, et al. Transcatheter versus surgical aortic valve replacement in intermediate risk patients: a meta-analysis. Cardiovascular Diagnosis and Therapy. 2016 Jun;6 (3):241-9. doi: 10.21037/cdt.2016.03.04.

Arora S, Misenheimer JA, Ramaraj R. Transcatheter aortic valve replacement: Comprehensive review and present status. Tex Heart Inst J. 2017 Feb 1;44(1):29-38. doi: 10.14503/THIJ-16-5852.

Arora S, Strassle PD, Ramm CJ, et al. Transcatheter versus surgical aortic valve replacement in patients with lower surgical risk scores: A systematic review and meta-analysis of early outcomes. Heart Lung and Circulation. 2017 Aug;26(8):840-845. doi: 10.1016/j.hlc.2016.12.003. PMID: 28169084.

Arora S, Vaidya SR, Strassle PD, et al. Meta-analysis of transfemoral TAVR versus surgical aortic valve replacement. Catheterizations and Cardiovascular Interventions: official journal of the Society of Cardiac Angiography and Interventions. 2018 Mar 1;91(4):806-812. doi: 10.1002/ccd.27357.PMID: 29068166.

Arsalan M, Szerlip M, Vemulapalli S, et al. Should Transcatheter Aortic Valve Replacement Be Performed in Nonagenarians? Insights From the STS/ACC TVT Registry. JACC. 2016;67(10).

Asch FM, Vannan MA, Singh S, et al. Hemodynamic and echocardiographic comparison of the lotus and corevalve transcatheter aortic valves in patients with high and extreme surgical risk: An analysis from the REPRISE III randomized controlled trial. Circulation. 2018 Jun 12;137(24):2557-2567. doi: 10.1161/CIRCULATIONAHA.118.034129.

Athappan G, Patvardhan E, Tuzcu EM, et al. Incidence, predictors, and outcomes of aortic regurgitation after transcatheter aortic valve replacement: meta-analysis and systematic review of literature. J Am Coll Cardiol. 2013;61(15):1585-95. doi: 10.1016/j.jacc.2013.01.047.

Attias D, Maillet JM, Copie X, et al. Prevalence, clinical characteristics and outcomes of high-risk patients treated for severe aortic stenosis prior to and after transcatheter aortic valve implantation availability. European Journal of Cardiothoracic Surgery. 2015;47(5):e206-12. PMID: 25687453.

Badheka AO, Patel NJ, Panaich SS, et al. Effect of hospital volume on outcomes of transcatheter aortic valve implantation. American Journal of Cardiology. 2015;116(4):587-94. PMID: 26092276.

Bakaeen FG, Chu D, Ratcliffe M, et al. Severe aortic stenosis in a veteran population: treatment considerations and survival. Ann Thorac Surg. 2010 Feb;89(2):453-8. doi: 10.1016/j.athoracsur.2009.10.033.

Baron SJ, Arnold SV, Herrmann HC, et al. Impact of Ejection Fraction and Aortic Valve Gradient on Outcomes of Transcatheter Aortic Valve Replacement. JACC. 2016;67(20):2349-2358.

Baron SJ, Arnold SV, Reynolds MR, et al. Durability of quality of life benefits of transcatheter aortic valve replacement: Long-term results from the CoreValve US extreme risk trial. Am Heart J. 2017 Dec;194:39-48. doi: 10.1016/j.ahj.2017.08.006. PMID: 29223434.

Baron SJ, Arnold SV, Wang K, et al. Health status benefits of transcatheter vs surgical aortic valve replacement in patients with severe aortic stenosis at intermediate surgical risk: results from the PARTNER 2 randomized clinical trial. JAMA Cardiol. 2017;2(8):837-845. doi: 10.1001/jamacardio.2017.2039.

Bavaria JE, Tommaso CL, Brindis RG, et al. 2018 AATS/ACC/SCAI/STS expert consensus systems of care document: operator and institutional recommendations and requirements for transcatheter aortic valve replacement: A joint report of the American Association for Thoracic Surgery, the American College of Cardiology, the Society for Cardiovascular Angiography and Interventions, and the Society of Thoracic Surgeons. Journal of the American College of Cardiology. 2018. pii: S0735- 1097(18)35377-4. doi: 10.1016/j.jacc.2018.07.002.

Berry C, Lloyd SM, Wang Y, Macdonald A, Ford I. The changing course of aortic valve disease in Scotland: temporal trends in hospitalizations and mortality and prognostic importance of aortic stenosis. Eur Heart J. 2013;34:1538-47.

Bob-Manuel T, Sharma A, Nanda A, et al. A review of racial disparities in transcatheter aortic valve replacement (TAVR): accessibility, referrals and implantation. Annals of Translational Medicine.2018;6(1):10. PMID: 29404356.

Bonow RO, Carabello BA, Chatterjee K, et al. ACC/AHA 2006 guidelines for the management of patients with valvular heart disease: a report of the American College of Cardiology/American Heart Association Task Force on Practice Guidelines (Writing Committee to Revise the 1998 Guidelines for the Management of Patients with Valvular Heart Disease) developed in collaboration with the Society of Cardiovascular Anesthesiologists endorsed by the Society for Cardiovascular Angiography and Interventions and the Society of Thoracic Surgeons. J Am Coll Cardiol. 2006;48(3):e1-e148.

Brennan JM, Edwards FH, Zhao Y, et al. Long-term survival after aortic valve replacement among high-risk elderly patients in the United States: insights from the Society of Thoracic Surgeons Adult Cardiac Surgery Database, 1991 to 2007. Circulation 2012 Sep 25;126(13):1621-9. PMID:22907936.

Brennan JM, Holmes DR, Sherwood MW, et al. The association of transcatheter aortic valve replacement availability and hospital aortic valve replacement volume and mortality in the United States. Annals of Thoracic Surgery. 2014;98(6):2016-22; discussion 2022. PMID: 25443009.

Brennan JM, Thomas L, Cohen DJ, et al. Transcatheter versus surgical aortic valve replacement: propensity-matched comparison. J Am Coll Cardiol. 2017;70(4):439-450. doi:10.1016/j.jacc.2017.05.060.

Brown JR, Hisey WM, Marshall EJ, et al. Acute kidney injury severity and long-term readmission and mortality after cardiac surgery. Ann Thorac Surg. 2016;102:8.

Carabello BA, Paulus WJ. Aortic stenosis. Lancet. 2009;373:956–66.

Carnero-Alcázar M, Maroto LC, Cobiella-Carnicer J, et al. Transcatheter versus surgical aortic valve replacement in moderate and high-risk patients: a meta-analysis. European Journal of Cardiothoracic Surgery. 2017 Apr 1;51(4):644-652. doi: 10.1093/ejcts/ezw388. PMID: 28007879.

Carroll JD, Vemulapalli S, Dai D, et al. Procedural experience for transcatheter aortic valve replacement and relation to outcomes: The STS/ACC TVT Registry. Journal of the American College of Cardiology. 2017;70(1):29-41. PMID: 28662805.

Chakos A, Wilson-Smith A, Arora S, et al. Long term outcomes of transcatheter aortic valve implantation (TAVI): a systematic review of 5-year survival and beyond. Annals of Cardiothoracic Surgery. 2017 Sep;6(5):432-443. doi: 10.21037/acs.2017.09.10. PMID: 29062738.

Chandrasekhar J, Dangas G, Yu J, et al. Sex-based differences in outcomes with transcatheter aortic valve therapy: TVT registry From 2011 to 2014. J Am Coll Cardiol. 2016 Dec 27;68(25):2733-2744. doi: 10.1016/j.jacc.2016.10.041. PMID: 28007135.

Chieffo A, Petronio AS, Mehilli J, et al. 1-Year clinical outcomes in women after transcatheter aortic valve replacement: Results from the first WIN-TAVI registry. JACC Cardiovasc Interv. 2018;11 (1):1-12. doi: 10.1016/j.jcin.2017.09.034.

Clarke S, Wilson ML, Terhaar M. Using clinical decision support and dashboard technology to improve Heart Team efficiency and accuracy in a transcatheter aortic valve implantation (TAVI) program. Stud Health Technol Inform. 2016;225:98-102. PMID: 27332170.

Cheitlin MD, Armstrong WF, Aurigemma GP, et al. ACC/AHA/ASE 2003 guideline update for the clinical application of echocardiography: summary article: a report of the American College of Cardiology/American Heart Association Task Force on Practice Guidelines (ACC/AHA/ASE Committee to Update the 1997 Guidelines for the Clinical Application of Echocardiography). J Am Coll Cardiol 2003;42:954 –70. PMID: 12957449.

Coeytaux RR, Williams JW Jr, Gray RN, Wang A. Percutaneous heart valve replacement for aortic stenosis: state of the evidence. Ann Intern Med. 2010 Sep 7;153(5):314-24. doi: 10.7326/0003-4819- 153-5-201009070-00267.

Conrotto F, D'Ascenzo F, Presbitero P, et al. Effect of gender after transcatheter aortic valve implantation: a meta-analysis. Ann Thorac Surg. 2015;99(3):809-16. doi: 10.1016/j.athoracsur.2014.09.089.

Coylewright M, Palmer R, O'Neill ES, et al. Patient-defined goals for the treatment of severe aortic stenosis: a qualitative analysis. Health Expect. 2016 Oct;19(5):1036-43. doi: 10.1111/hex.12393.

Crestanello JA, Popma JJ, Adams DH, et a. Long-term health benefit of transcatheter aortic valve replacement in patients with chronic lung disease. JACC Cardiovasc Interv. 2017;10(22):2283-2293. doi: 10.1016/j.jcin.2017.07.025.

Cribier A, Eltchaninoff H, Bash A, et al. Percutaneous transcatheter implantation of an aortic valve prosthesis for calcific aortic stenosis: first human case description. Circulation. 2002 Dec 10;106 (24):3006-8.

Cruz Rodriguez B, Acharya P, Salazar-Fields C, et al. Comparison of frequency of referral to cardiothoracic surgery for aortic valve disease in Blacks, Hispanics, and Whites. American Journal of Cardiology. 2017;120(3):450-455. PMID: 28583680.

Dauerman HL, Reardon MJ, Popma JJ, et al. Early recovery of left ventricular systolic function after CoreValve transcatheter aortic valve replacement. Circ Cardiovasc Interv. 2016;9(6). pii: e003425. doi: 10.1161/CIRCINTERVENTIONS.115.003425.

de Biasi AR, Paul S, Nasar A, et al. National analysis of short-term outcomes and volume-outcome relationships for transcatheter aortic valve replacement in the era of commercialization. Cardiology. 2016;133(1):58-68. PMID: 26489084.

Deeb GM, Reardon MJ, Chetcuti S, et al. 3-Year outcomes in high-risk patients who underwent surgical or transcatheter aortic valve replacement. J Am Coll Cardiol. 2016;67(22):2565-74. doi: 10.1016/j.jacc.2016.03.506.

Dodson JA, Williams MR, Cohen DJ, et al. Home Practice of Direct-Home Discharge and 30-Day Readmission After Transcatheter Aortic Valve Replacement in the Society of Thoracic Surgeons/American College of Cardiology Transcatheter Valve Therapy (STS/ACC TVT) Registry. JAHA. 2017;6(8).

D'Onofrio A, Salizzoni S, Agrifoglio M, et al. Medium term outcomes of transapical aortic valve implantation: results from the Italian Registry of Trans-Apical Aortic Valve Implantation. Annals of Thoracic Surgery. 2013;96(3):830-5; discussion 836. PMID: 23870695.

Eagle KA, Weinberg RL. In severe aortic stenosis, TAVR was noninferior to conventional surgery for death or disabling stroke at 2 years. Ann Intern Med. 2016;165(4):JC21. doi: 10.7326/ACPJC-2016- 165-4-021.

Edwards FH, Cohen DJ, O'Brien SM, et al. Development and Validation of a Risk Prediction Model for In-Hospital Mortality After Transcatheter Aortic Valve Replacement. JAMA Cardiol. 2016;1 (1):46-52.

Elmaraezy A, Ismail A, Abushouk AI, et al. Efficacy and safety of transcatheter aortic valve replacement in aortic stenosis patients at low to moderate surgical risk: a comprehensive meta-analysis. BMC Cardiovascular Disorders 2017 Aug 24;17(1):234. doi: 10.1186/s12872-017-0668-1. PMID: 28836953.

Fadahunsi OA, Olowoyeye A, Ukaigwe A, et al. Incidence, Predictors, and Outcomes of Permanent Pacemaker Implantation Following Transcatheter Aortic Valve Replacement: Analysis from STS/ACC TVT Registry. JACC Cardiovasc Interv. 2016;9(21):2189-2199.

Fanaroff AC, Manandhar P, Holmes DR, et al. Peripheral Artery Disease and Transcatheter Aortic Valve Replacement Outcomes. Circ Interv. 2017;10(10).

Feldman TE, Reardon MJ, Rajagopal V, et al. Effect of mechanically expanded vs self-expanding transcatheter aortic valve replacement on mortality and major adverse clinical events in high-risk patients with aortic stenosis: The REPRISE III randomized clinical trial. JAMA. 2018;319(1):27-37. doi: 10.1001/jama.2017.19132.

Forrest JK, Mangi AA, Popma JJ, et al. early outcomes with the Evolut PRO repositionable self- expanding transcatheter aortic valve with pericardial wrap. JACC Cardiovasc Interv. 2018;11(2):160- 168. doi: 10.1016/j.jcin.2017.10.014.

Gargiulo G, Sannino A, Capodanno D, et al. Transcatheter Aortic Valve Implantation Versus Surgical Aortic Valve Replacement: A Systematic Review and Meta-analysis. Ann Intern Med. 2016;165 (5):334-44. doi: 10.7326/M16-0060. Epub 2016 Jun 7.

Gleason TG, Reardon MJ, Popma JJ, et al. Self-expanding transcatheter aortic valve replacement or surgical valve replacement in high-risk patients: 5-year outcomes. J Am Coll Cardiol. 2018. pii: S0735-1097(18)38290-1. doi: 10.1016/j.jacc.2018.08.2146.

Grover FL, Vemulapalli S, Carroll JD, et al. 2016 Annual report of the Society of Thoracic Surgeons/American College of Cardiology Transcatheter Valve Therapy Registry. J Am Coll Cardiol. 2017 Mar 14;69(10):1215-1230. doi: 10.1016/j.jacc.2016.11.033. PMID: 27956264.

Hahn RT, Pibarot P, Stewart WJ, et al. Comparison of transcatheter and surgical aortic valve replacement in severe aortic stenosis: a longitudinal study of echocardiography parameters in cohort A of the PARTNER trial (placement of aortic transcatheter valves). J Am Coll Cardiol. 2013; 61: 2514±2521. https://doi.org/10.1016/j.jacc.2013.02.087 PMID: 23623915.

Hansen, JW, Foy, A, Yadav, P, et al. Death and Dialysis After Transcatheter Aortic Valve Replacement. An Analysis of the STS/ACC TVT Registry. JACC Cardiovasc Interv. Sept. 2017.

Health Quality Ontario. Transcatheter Aortic Valve Implantation for treatment of aortic valve stenosis: A health technology assessment. Ont Health Technol Assess Ser. 2016 Nov 1;16(19):1-94.

Henn MC, Percival T, Zajarias A, et al. Learning alternative access approaches for transcatheter aortic valve replacement: Implications for new transcatheter aortic valve replacement centers. The Annal of Thoracic Surgery. 2017;103(5):1399-1405. PMID: 27765175.

Hermiller JB Jr, Yakubov SJ, Reardon MJ, et al. Predicting Early and late mortality after transcatheter aortic valve replacement. J Am Coll Cardiol. 2016;68(4):343-52. doi: 10.1016/j.jacc.2016.04.057.

Herrmann HC, Thourani VH, Kodali SK, et al. One-year clinical outcomes with SAPIEN 3 transcatheter aortic valve replacement in high-risk and inoperable patients with severe aortic stenosis. Circulation. 2016;134(2):130-40. doi: 10.1161/CIRCULATIONAHA.116.022797.

Hollenberg SM. Valvular heart disease in adults: Etiologies, classification, and diagnosis. FP Essent. 2017 Jun;457:11-16. PMID: 28671804.

Holmes DR Jr, Brennan JM, Rumsfeld JS, et al. Clinical outcomes at 1 year following transcatheter aortic valve replacement. JAMA. 2015 Mar 10;313(10):1019-28. doi: 10.1001/jama.2015.1474. PMID: 25756438.

Holmes DR Jr, Nishimura RA, Grover FL, et al. Annual outcomes with Transcatheter Valve Therapy: From the STS/ACC TVT registry. Ann Thorac Surg. 2016 Feb;101(2):789-800. PMID: 27175453.

Hyman MC, Vemulapalli S, Szeto WY, et al. Conscious Sedation Versus General Anesthesia for Transcatheter Aortic Valve Replacement: Insights from the NDCR STS/ACC TVT Registry. Circulation. 2017;136(22).

Iung B, Vahanian A. Epidemiology of acquired valvular heart disease. Can J Cardiol. 2014 Sep;30 (9):962-70. doi: 10.1016/j.cjca.2014.03.022. PMID: 24986049.

Jensen HA, Condado JF, Devireddy C, et al. Minimalist transcatheter aortic valve replacement: The new standard for surgeons and cardiologists using transfemoral access? The Journal of Thoracic and Cardiovascular Surgery. 2015;150(4):833-9. PMID: 26318351.

Kapadia SR, Leon MB, Makkar RR, et al. 5-year outcomes of transcatheter aortic valve replacement compared with standard treatment for patients with inoperable aortic stenosis (PARTNER 1): a randomised controlled trial. Lancet. 2015 Jun 20;385(9986):2485-91. doi: 10.1016/S0140-6736(15) 60290-2. PMID: 25788231.

Kapadia SR, Tuzcu EM, Makkar RR, et al. Long-term outcomes of inoperable patients with aortic stenosis randomly assigned to transcatheter aortic valve replacement or standard therapy. Circulation. 2014 Oct 21;130(17):1483-92. doi: 10.1161/CIRCULATIONAHA.114.009834. PMID: 25205802.

Kappetein AP, Head SJ, Généreux P, et al. Updated standardized endpoint definitions for transcatheter aortic valve implantation: the Valve Academic Research Consortium-2 consensus document. Journal of Thoracic and Cardiovascular Surgery. 2013;145(1):6-23.

Khan SU, Lone AN, Saleem MA, et al. Transcatheter vs surgical aortic-valve replacement in low- to intermediate-surgical-risk candidates: A meta-analysis and systematic review. Clinical Cardiology. 2017 Nov;40(11):974-981. doi: 10.1002/clc.22807. PMID: 29168984.

Kim LK, Minutello RM, Feldman DN, et al. Association between transcatheter aortic valve implantation volume and outcomes in the United States. The American Journal of Cardiology. 2015;116(12):1910-5. PMID: 26508710.

Kitai T, Honda S, Okada Y, et al. Clinical outcomes in non-surgically managed patients with very severe versus severe aortic stenosis. Heart. 2011;97(24):2029-2032. DOI: 10.1136/heartjnl-2011-300137.

Kleiman NS, Maini BJ, Reardon MJ, et al. Neurological events following transcatheter aortic valve replacement and their predictors: A report from the CoreValve Trials. Circ Cardiovasc Interv. 2016;9 (9). pii: e003551. doi: 10.1161/CIRCINTERVENTIONS.115.003551.

Kochar A, Zhuokai L, Harrison JK, et al. Stroke and Cardiovascular Outcomes in Patients With Carotid Disease Undergoing Transcatheter Aortic Valve Replacement. Circ Cardiovasc Interv. 2018 Jun;11(6):e006322. doi: 10.1161/CIRCINTERVENTIONS.117.006322.

Kodali SK, Pibarot P, Douglas PS, et al. Paravalvular regurgitation after transcatheter aortic valve replacement with the Edwards sapien valve in the PARTNER trial: characterizing patients and impact on outcomes. Eur Heart J. 2015;36(7):449-56. doi: 10.1093/eurheartj/ehu384. Epub 2014 Oct 1.

Kodali S, Thourani VH, White J, et al. Early clinical and echocardiographic outcomes after SAPIEN 3 transcatheter aortic valve replacement in inoperable, high-risk and intermediate-risk patients with aortic stenosis. Eur Heart J. 2016 Jul 21;37(28):2252-62. doi: 10.1093/eurheartj/ehw112.

Kodali SK, Williams MR, Smith CR, et al. Two-year outcomes after transcatheter or surgical aortic- valve replacement. N Engl J Med. 2012;366(18):1686-95. doi: 10.1056/NEJMoa1200384. Epub 2012 Mar 26.

Kolte, D., Khera, S., Vemulapalli, S. Outcomes Following Urgent/Emergent Transcatheter Aortic Valve Replacement: Insights from the STS/ACC TVT Registry. JACC. March 2018.

Kularatna S, Byrnes J, Mervin MC, Scuffham PA. Health technology assessments reporting cost- effectiveness of transcatheter aortic valve implantation. Int J Technol Assess Health Care. 2016 Jan;32(3):89-96. doi: 10.1017/S0266462316000180.

Landes U, Barsheshet A, Finkelstein A, et al. Temporal trends in transcatheter aortic valve implantation, 2008-2014: patient characteristics, procedural issues, and clinical outcome. Clinical Cardiology. 2017;40(2):82-88. PMID: 27783882.

Lazkani M, Singh N, Howe C, et al. An updated meta-analysis of TAVR in patients at intermediate risk for SAVR. Cardiovascular Revascularization Medicine: including molecular interventions. 2018 Apr 20. pii: S1553-8389(18)30129-5. doi: 10.1016/j.carrev.2018.04.001. PMID: 29748086.

Leon MB, Piazza N, Nikolsky E, et al. Standardized endpoint definitions for Transcatheter Aortic Valve Implantation clinical trials: a consensus report from the Valve Academic Research Consortium. J Am Coll Cardiol. 2011;57:253–269.

Leon MB, Smith CR, Mack M, et al. Transcatheter aortic-valve implantation for aortic stenosis in patients who cannot undergo surgery. N Engl J Med. 2010;363:1597-607.

Leon MB, Smith CR, Mack MJ, et al. Transcatheter or surgical aortic-valve replacement in intermediate-risk patients. N Engl J Med. 2016;374(17):1609-20. doi: 10.1056/NEJMoa1514616.

Lindman BR, Maniar HS, Jaber WA, et al. Effect of tricuspid regurgitation and the right heart on survival after transcatheter aortic valve replacement: insights from the Placement of Aortic Transcatheter Valves II inoperable cohort. Circ Cardiovasc Interv. 2015;8(4). pii: e002073. doi: 10.1161/CIRCINTERVENTIONS.114.002073.

Liu Z, Kidney E, Bem D, et al. Transcatheter aortic valve implantation for aortic stenosis in high surgical risk patients: A systematic review and meta-analysis. PLoS One. 2018 May 10;13 (5):e0196877.

Lytvyn L, Guyatt GH, Manja V, et al. Patient values and preferences on transcatheter or surgical aortic valve replacement therapy for aortic stenosis: a systematic review. BMJ Open. 2016 Sep 29;6 (9):e014327. doi: 10.1136/bmjopen-2016-014327. PMID: 27687903.

Mack MJ, Brennan JM, Brindis R, et al. Outcomes following transcatheter aortic valve replacement in the United States. JAMA. 2013 Nov 20;310(19):2069-77. doi: 10.1001/jama.2013.282043. PMID: 24240934.

Mack MJ, Leon MB, Smith CR, et al. 5-year outcomes of transcatheter aortic valve replacement or surgical aortic valve replacement for high surgical risk patients with aortic stenosis (PARTNER 1): a randomised controlled trial. Lancet. 2015 Jun 20;385(9986):2477-84. doi: 10.1016/S0140-6736(15) 60308-7. PMID: 25788234.

Mack M, Leon M, Thourani V, et al. Transcatheter aortic-valve replacement with a balloon- expandable valve in low-risk patients. N Engl J Med. 2019 March 16; DOI: 10.1056/NEJMoa1814052.

Maganti K, Rigolin VH, Sarano ME, et al. Valvular heart disease: diagnosis and management. Mayo Clin Proc. 2010 May;85(5):483-500. doi: 10.4065/mcp.2009.0706. PMID: 20435842.

Mao J, Redberg RF, Carroll JS, et al. Association between hospital surgical aortic valve replacement volume and transcatheter aortic valve replacement outcomes. JAMA Cardiol. 2018;3(11): 1070-1078. Doi: 10.1001/jamacardio.2018.3562.

McNeely C, Zajarias A, Fohtung R, et al. Racial comparisons of the outcomes of transcatheter and surgical aortic valve implantation using the medicare database. Am J Cardiol. 2018;122(3):440-445. doi: 10.1016/j.amjcard.2018.04.019.

Minha S, Barbash IM, Magalhaes MA, et al. Outcome comparison of African-American and Caucasian patients with severe aortic stenosis subjected to transcatheter aortic valve replacement: a single-center experience. Catheterization and Cardiovascular Interventions. 2015;85(4):640-7. PMID: 24782407.

Minha S, Waksman R, Satler LP, et al. Learning curves for transfemoral transcatheter aortic valve replacement in the PARTNER-I trial: Success and safety. Catheter Cardiovasc Interv. 2016;87 (1):165-75. doi: 10.1002/ccd.26121. PMID: 26425793.

Mrsic Z, Hopkins SP, Antevil JL, Mullenix PS. Valvular heart disease. Prim Care. 2018 Mar;45 (1):81-94. doi: 10.1016/j.pop.2017.10.002. PMID: 29406946.

Murphy ES, Lawson RM, Starr A, Rahimtoola SH. Severe aortic stenosis in patients 60 years of age or older: left ventricular function and 10-year survival after valve replacement. Circulation. 1981 Aug;64(2 Pt 2):II184-8. PMID: 7249320.

Mylotte D, Osnabrugge RLJ, Windecker S, et al. Transcatheter aortic valve replacement in Europe: adoption trends and factors influencing device utilization. J Am Coll Cardiol. 2013 Jul 16;62(3):210- 219. doi: 10.1016/j.jacc.2013.03.074.

Neyt M, Van Brabandt H, Devriese S, et al. A cost-utility analysis of transcatheter aortic valve implantation in Belgium: focusing on a well-defined and identifiable population. BMJ Open. 2012 May 4;2(3). pii: e001032. doi: 10.1136/bmjopen-2012-001032.

Nielsen HH, Klaaborg KE, Nissen H, et al. A prospective, randomised trial of transapical transcatheter aortic valve implantation vs. surgical aortic valve replacement in operable elderly patients with aortic stenosis: the STACCATO trial. EuroIntervention. 2012 Jul 20;8(3):383-9. doi: 10.4244/EIJV8I3A58. PMID: 22581299.

Nishimura RA, Otto CM, Bonow RO, et al. 2017 AHA/ACC focused update of the 2014 AHA/ACC guideline for the management of patients with valvular heart disease: a report of the American College of Cardiology/American Heart Association Task Force on Clinical Practice Guidelines. Circulation. 2017;135:e1159–e1195.

Nishimura RA, Otto CM, Bonow RO, et al. 2014 AHA/ACC guideline for the management of patients with valvular heart disease: a report of the American College of Cardiology/American Heart Association Task Force on Practice Guidelines. Circulation. 2014 Jun 10;129(23):2440-92.

Nkomo VT, Gardin JM, Skelton TN, et al. Burden of valvular heart diseases: a population-based study. Lancet. 2006;368:1005-11.

O'Brien SM, Cohen DJ, Rumsfeld JS, et al. Variation in Hospital Risk-Adjusted Mortality Rates Following Transcatheter Aortic Valve Replacement in the United States: A Report from the STS/ACC TVT Registry. Circ Cardiovasc Qual Outcomes. 2016;9:560-565.

O'Connor SA, Morice MC, Gilard M, et al. Revisiting sex equality with transcatheter aortic valve replacement outcomes: A collaborative, patient-level meta-analysis of 11,310 patients. J Am Coll Cardiol. 2015;66(3):221-228. doi: 10.1016/j.jacc.2015.05.024.

Oh JK, Little SH, Abdelmoneim SS, et al. Regression of paravalvular aortic regurgitation and remodeling of self-expanding transcatheter aortic valve: An observation from the CoreValve U.S. Pivotal Trial. JACC Cardiovasc Imaging. 2015;8(12):1364-1375. doi: 10.1016/j.jcmg.2015.07.012.

Onorati F, D'Errigo P, Barbanti M, et al. Different impact of sex on baseline characteristics and major periprocedural outcomes of transcatheter and surgical aortic valve interventions: Results of the multicenter Italian OBSERVANT Registry. Journal of Thoracic and Cardiovascular Surgery. 2014 May;147(5):1529-39. PMID: 23856202.

Osnabrugge RL, Arnold SV, Reynolds MR, et al. Health status after transcatheter aortic valve replacement in patients at extreme surgical risk: results from the CoreValve U.S. trial. JACC Cardiovasc Interv. 2015;8(2):315-323. doi: 10.1016/j.jcin.2014.08.016.

Osnabrugge RL, Mylotte D, Head SJ, et al. Aortic stenosis in the elderly: disease prevalence and number of candidates for transcatheter aortic valve replacement: a meta-analysis and modeling study. J Am Coll Cardiol. 2013 Sep 10;62(11):1002-12. doi: 10.1016/j.jacc.2013.05.015.

Otto CM, Kumbhani DJ, Alexander KP et al. 2017 ACC expert consensus decision pathway for transcatheter aortic valve replacement in the management of adults with aortic stenosis: a report of the American College of Cardiology Task Force on Clinical Expert Consensus Documents. J Am Coll Cardiol. 2017; 69:1313-46.

Otto CM, Kuusisto J, Reichenbach DD, et al. Characterization of the early lesion of 'degenerative' valvular aortic stenosis. Histological and immunohistochemical studies. Circulation. 1994 Aug;90 (2):844-53. PMID: 7519131.

Otto CM, Mickel MC, Kennedy JW, et al. Three-year outcome after balloon aortic valvuloplasty. Insights into prognosis of valvular aortic stenosis. Circulation. 1994 Feb;89(2):642-50. PMID: 8313553.

Panaich SS, Patel NJ, Patel N, et al. Hospital volume and in-hospital outcomes after transcatheter aortic valve implantation. Am J Cardiol. 2016;117(6):1028-9. doi: 10.1016/j.amjcard.2015.12.020. PMID: 26796196.

Panoulas VF, Francis DP, Ruparelia N, et al. Female-specific survival advantage from transcatheter aortic valve implantation over surgical aortic valve replacement: Meta-analysis of the gender subgroups of randomised controlled trials including 3758 patients. Int J Cardiol. 2018;250:66-72. doi: 10.1016/j.ijcard.2017.05.047.

Patel HJ, Herbert MA, Paone G, et al. The midterm impact of transcatheter aortic valve replacement on surgical aortic valve replacement in Michigan. The Annals of Thoracic Surgery. 2016;102(3):728- 734. PMID: 27154157.

Pellikka PA, Sarano ME, Nishimura RA, et al. Outcome of 622 adults with asymptomatic, hemodynamically significant aortic stenosis during prolonged follow-up. Circulation. 2005;111 (24):3290-3295.

Piazza N, Kalesan B, van Mieghem N, et al. A 3-center comparison of 1-year mortality outcomes between transcatheter aortic valve implantation and surgical aortic valve replacement on the basis of propensity score matching among intermediate-risk surgical patients. JACC Cardiovasc Interv. 2013 May;6(5):443-51. doi: 10.1016/j.jcin.2013.01.136.

Pibarot P, Hahn RT, Weissman NJ, et al. Association of paravalvular regurgitation with 1-year outcomes after transcatheter aortic valve replacement with the SAPIEN 3 valve. JAMA Cardiol. 2017;2(11):1208-1216. doi: 10.1001/jamacardio.2017.3425.

Popma JJ, Adams DH, Reardon MJ, et al. Transcatheter aortic valve replacement using a self- expanding bioprosthesis in patients with severe aortic stenosis at extreme risk for surgery. J Am Coll Cardiol. 2014;63(19):1972-81. doi: 10.1016/j.jacc.2014.02.556. PMID: 24657695.

Popma JJ, Deeb G, Yakubov S, et al. Transcatheter aortic-valve replacement with a self-expanding valve in low-risk patients. N Engl J Med. 2019 March 16; DOI: 10.1056/NEJMoa1816885.

Popma JJ, Reardon MJ, Khabbaz K, et al. Early clinical outcomes after transcatheter aortic valve replacement using a novel self-expanding bioprosthesis in patients with severe aortic stenosis who are suboptimal for surgery: Results of the Evolut R U.S. Study. JACC Cardiovasc Interv. 2017;10 (3):268-275. doi: 10.1016/j.jcin.2016.08.050.

Rahimtoola SH. Catheter balloon valvuloplasty for severe calcific aortic stenosis: a limited role. J Am Coll Cardiol. 1994 Apr;23(5):1076-8. PMID: 8144771.

Rajamannan NM, Evans FJ, Aikawa E, et al. Calcific aortic valve disease: not simply a degenerative process: a review and agenda for research from the National Heart and Lung and Blood Institute Aortic Stenosis Working Group. Executive summary: calcific aortic valve disease-2011 update. Circulation. 2011;124:1783-91.

Ray S. Changing epidemiology and natural history of valvular heart disease. Clin Med (Lond). 2010 Apr;10(2):168-71. PMID: 20437994.

Reardon MJ, Adams DH, Coselli JS, et al. Self-expanding transcatheter aortic valve replacement using alternative access sites in symptomatic patients with severe aortic stenosis deemed extreme risk of surgery. J Thorac Cardiovasc Surg. 2014;148(6):2869-76.e1-7. doi: 10.1016/j.jtcvs.2014.07.020.

Reardon MJ, Adams DH, Kleiman NS, et al. 2-Year outcomes in patients undergoing surgical or self- expanding transcatheter aortic valve replacement. J Am Coll Cardiol. 2015;66(2):113-21. doi: 10.1016/j.jacc.2015.05.017.

Reardon MJ, Kleiman NS, Adams DH, et al. Outcomes in the randomized CoreValve US Pivotal High Risk Trial in patients with a Society of Thoracic Surgeons Risk Score of 7% or Less. JAMA Cardiol. 2016;1(8):945-949. doi: 10.1001/jamacardio.2016.2257.

Reardon MJ, Van Mieghem NM, Popma JJ, et al. Surgical or transcatheter aortic-valve replacement in intermediate-risk patients. N Engl J Med. 2017;376(14):1321-1331. doi: 10.1056/NEJMoa1700456.

Reynolds MR, Magnuson EA, Lei Y, et al. Health-related quality of life after transcatheter aortic valve replacement in inoperable patients with severe aortic stenosis. Circulation. 2011; 124: 1964±1972. https://doi.org/10.1161/CIRCULATIONAHA.111.040022 PMID: 21969017.

Reynolds MR, Magnuson EA, Wang K, et al. Health-related quality of life after transcatheter or surgical aortic valve replacement in high-risk patients with severe aortic stenosis: results from the PARTNER (Placement of AoRTic TraNscathetER Valve) Trial (Cohort A). J Am Coll Cardiol. 2012; 60: 548±558. https://doi.org/10.1016/j.jacc.2012.03.075 PMID: 22818074.

Ross J Jr, Braunwald E. Aortic stenosis. Circulation. 1968 Jul;38(1 Suppl):61-7. PMID: 4894151.

Russo MJ, McCabe JM, Thourani VH, et al. Case volume and outcomes after TAVR with balloon-expanded prostheses. Insights from TVT Registry. J Am Coll Cardiol. 2019;73:427-440. doi: 10.1016/j.jacc.2018.11.031.

Salemi A, Sedrakyan A, Mao J, et al. Individual operator experience and outcomes in transcatheter aortic valve replacement. J Am Coll Cardiol Intv. 2019;12:90-7. doi: 10.1016/j.jcin.2018.10.030.

Schwarz F, Baumann P, Manthey J, et al. The effect of aortic valve replacement on survival. Circulation. 1982 Nov;66(5):1105-10. PMID: 7127696.

Siemieniuk RA, Agoritsas T, Manja V, et al. Transcatheter versus surgical aortic valve replacement in patients with severe aortic stenosis at low and intermediate risk: systematic review and meta- analysis. BMJ (clinical research ed.). 2016 Sep 28;354:i5130. doi: 10.1136/bmj.i5130. PMID: 27683246.

Singh K, Carson K, Rashid MK, et al. Transcatheter aortic valve implantation in intermediate surgical risk patients with severe aortic stenosis: A systematic review and meta-analysis. Heart Lung and Circulation. 2018 Feb;27(2):227-234. doi: 10.1016/j.hlc.2017.02.032. PMID: 28473216.

Siontis GC, Overtchouk P, Cahill TJ, et al. Transcatheter aortic valve implantation vs. surgical aortic valve replacement for treatment of symptomatic severe aortic stenosis: an updated meta-analysis. European Heart Journal. 2019;0:1–11. doi.org/10.1093/eurheartj/ehz275.

Siordia JA Jr, Loera JM, Scanlon M, et al. Three-year survival comparison between transcatheter and surgical aortic valve replacement for intermediate- and low-risk patients. Innovations (Philadelphia, Pa.). 2018 May/Jun;13(3):153-162. doi: 10.1097/IMI.0000000000000507. PMID:29912140.

Sleder A, Tackett S, Cerasale M, et al. Socioeconomic and racial disparities: a Case-control study of patients receiving transcatheter aortic valve replacement for severe aortic stenosis. Journal of Racial and Ethnic Health Disparities. 2017;4(6):1189-1194. PMID: 28039604.

Smith CR, Leon MB, Mack MJ, et al. Transcatheter versus surgical aortic-valve replacement in high- risk patients. N Engl J Med. 2011;364:2187–98.

Søndergaard L, Steinbrüchel DA, Ihlemann N, et al. Two-year outcomes in patients with severe aortic valve stenosis randomized to transcatheter versus surgical aortic valve replacement: The All- Comers Nordic aortic valve intervention randomized clinical trial. Circ Cardiovasc Interv. 2016 Jun;9 (6). pii: e003665. doi: 10.1161/CIRCINTERVENTIONS.115.003665. PMID: 27296202.

Sorajja P, Kodali S, Reardon MJ, et al. Outcomes for the commercial use of self-expanding prostheses in transcatheter aortic valve replacement: A report from the STS/ACC TVT registry. JACC Cardiovasc Interv. 2017 Oct 23;10(20):2090-2098. doi: 10.1016/j.jcin.2017.07.027. PMID: 29050627.

Suri RM, Gulack BC, Brennan JM, et al. Outcomes of Patients with Severe Chronic Lung Disease Who Are Undergoing Transcatheter Aortic Valve Replacement. Annals of Thoracic Surgery. Available online 29 August 2015.

Suri RM, Minha S, Alli O, et al. Learning curves for transapical transcatheter aortic valve replacement in the PARTNER-I trial: Technical performance, success, and safety. J Thorac Cardiovasc Surg. 2016;152(3):773-780.e14. doi: 10.1016/j.jtcvs.2016.04.028. PMID: 27215927.

Takagi H, Mitta S, Ando T, et al. Long-term survival after transcatheter versus surgical aortic valve replacement for aortic stenosis: A meta-analysis of observational comparative studies with a propensity-score analysis. Catheterizations and Cardiovascular Interventions: official journal of the Society of Cardiac Angiography and Interventions. Catheter Cardiovasc Interv. 2018;00:1–12. https://doi.org/10.1002/ccd. 27521 doi: 10.1002/ccd.27521. PMID: 29388316.

Tam DY, Vo TX, Wijeysundera HC, et al. Transcatheter vs surgical aortic valve replacement for aortic stenosis in low-intermediate risk patients: A meta-analysis. The Canadian Journal of Cardiology. 2017 Sep;33(9):1171-1179. doi: 10.1016/j.cjca.2017.06.005. PMID: 28843328.

Tice JA, Sellke FW, Schaff HV. Transcatheter aortic valve replacement in patients with severe aortic stenosis who are at high risk for surgical complications: summary assessment of the California Technology Assessment Forum. J Thorac Cardiovasc Surg. 2014 Aug;148(2):482-91.e6. doi: 10.1016/j.jtcvs.2013.09.056. PMID: 24252939.

Thyregod HGH, Møller CH, Søndergaard L, et al. Transcatheter versus optimal medical treatment and surgical aortic valve replacement for aortic valve stenosis. 2015. Cochrane Library site accessed October 22, 2018: https://www.cochrane.org/CD010488/VASC_transcatheter-versus-optimal- medical-treatment-and-surgical-aortic-valve-replacement-aortic-valve.

Thourani VH, Jensen HA, Babaliaros V, et al. Transapical and Transaortic Transcatheter Aortic Valve Replacement in the United States. Annals of Thoracic Surgery. 2015;100:1718-27.

Tommaso CL, Bolman RM III, Feldman T, et al. Multisociety (AATS, ACCF, SCAI, and STS) Expert Consensus Statement: Operator and Institutional Requirements for Transcatheter Valve Repair and Replacement, Part 1: Transcatheter Aortic Valve Replacement. J Am Coll Cardiol. 2012;59(22). doi: 10.1016/j.jacc.2012.02.016.

Turina J, Hess O, Sepulcri F, et al. Spontaneous course of aortic valve disease. Eur Heart J. 1987 May;8(5):471-83. PMID: 3609042.

Tuzcu EM, Kapadia SR, Vemulapalli S, et al. Transcatheter Aortic Valve Replacement of Failed Surgically Implanted Bioprostheses: The STS/ACC Registry J Am Coll Cardiol. 2018 Jul 24;72 (4):370-382. doi: 10.1016/j.jacc.2018.04.074.

United States Census Bureau 2011. Available at: http://www.census.gov/popest/data/national/asrh/2011/index.html. Accessed November 16, 2012.

Van Belle E, Juthier F, Susen S, et al. Postprocedural aortic regurgitation in balloon-expandable and self-expandable transcatheter aortic valve replacement procedures: analysis of predictors and impact on long-term mortality: insights from the FRANCE2 Registry. Circulation. 2014;129(13):1415-27. doi: 10.1161/CIRCULATIONAHA.113.002677. Epub 2014 Feb 24.

Van Brabandt H, Neyt M, Hulstaert F. Transcatheter aortic valve implantation (TAVI): risky and costly. BMJ. 2012 Jul 31;345:e4710. doi: 10.1136/bmj.e4710.

Van Geldorp MW, WR EJ, Kappetein AP, et al. Patient outcome after aortic valve replacement with a mechanical or biological prosthesis: weighing lifetime anticoagulant-related event risk against reoperation risk. J Thorac Cardiovasc Surg. 2009;137(4):881-6, 886e1-5. doi: 10.1016/j.jtcvs.2008.09.028. Epub 2009 Feb 25.

Vaquerizo B, Bleiziffer S, Wottke M, et al. Impact of transcatheter aortic valve implantation on surgical aortic valve. International Journal of Cardiology. 2017;243:145-149. PMID: 28536003.

Varadarajan P, Kapoor N, Bansal RC, et al. Clinical profile and natural history of 453 nonsurgically managed patients with severe aortic stenosis. Ann Thorac Surg. 2006 Dec;82(6):2111-5. PMID: 17126120.

Vemulapalli S, Carroll JD, Mack MJ, et al. Procedural volume and outcomes for transcatheter aortic-valve replacement. N Engl J Med. 2019; doi: 10.1056/NEJMsa1901109.

Verma DR, Pershad Y, Pershad A, et al. Impact of institutional volume and experience with CT interpretation on sizing of transcatheter aortic valves: A multicenter retrospective study. Cardiovascular Revascularization Medicine. 2016;17(8):566-570. PMID: 27641485.

Vilela AT, Grande A, Palma JH, et al. Transcatheter valve implantation versus aortic valve replacement for aortic stenosis in high-risk patients. 2015. Cochrane Library Site accessed October 22, 2018. https://www.cochrane.org/CD010304/VASC_transcatheter-valve-implantation-versus- aortic-valve-replacement-aortic-stenosis-high-risk-patients.

Villablanca PA, Mathew V, Thourani VH, et al. A meta-analysis and meta-regression of long-term outcomes of transcatheter versus surgical aortic valve replacement for severe aortic stenosis. International Journal of Cardiology. 2016 Dec 15;225:234-243. doi: 10.1016/j.ijcard.2016.10.003. PMID: 27732927.

Walters DL, Webster M, Pasupati S, et al. Position statement for the operator and institutional requirements for a transcatheter aortic valve implantation (TAVI) program. Heart Lung Circ. 2015;24(3):219-23. doi: 10.1016/j.hlc.2014.09.009. PMID: 25488705.

Wang A, Harrison JK, Bashore TM. Balloon aortic valvuloplasty. Prog Cardiovasc Dis. 1997 Jul- Aug;40(1):27-36.

Wang Y, Zhou Y, Zhang L, et al. Midterm outcome of transcatheter versus surgical aortic valve replacement in low to intermediate risk patients: A meta-analysis of randomized controlled trials. Journal of Cardiology. 2018 Jun;71(6):534-539. doi: 10.1016/j.jjcc.2017.10.020. PMID: 29655505.

Wassef AWA, Alnasser S, Rodes-Cabau J, et al. Institutional experience and outcomes of transcatheter aortic valve replacement: Results from an international multicentre registry. International Journal of Cardiology. 2017 Oct 15;245:222-227. PMID: 28760395.

Wassef AWA, Rodes-Cabau J, Liu Y, et al. The learning curve and annual procedure volume standards for optimum outcomes of transcatheter aortic valve replacement: Findings from an international registry. JACC Cardiovasc Interv. 2018;11(17):1669-1679. doi: 10.1016/j.jcin.2018.06.044. PMID: 30190058.

Webb JG, Doshi D, Mack MJ, et al. Randomized evaluation of the SAPIEN XT transcatheter heart valve system in patients with aortic stenosis who are not candidates for surgery. JACC Cardiovasc Interv. 2015;8(14):1797-806. doi: 10.1016/j.jcin.2015.08.017.

Williams M, Kodali SK, Hahn RT, et al. Sex-related differences in outcomes after transcatheter or surgical aortic valve replacement in patients with severe aortic stenosis: Insights from the PARTNER Trial (Placement of Aortic Transcatheter Valve). J Am Coll Cardiol. 2014;63(15):1522-8. doi: 10.1016/j.jacc.2014.01.036.

Williams JW, Coeytaux R, Wang A, et al. Percutaneous Heart Valve Replacement. Technical Brief No. 2. (Prepared by Duke Evidence-based Practice Center under Contract No. 290-02-0025.) Rockville, MD: Agency for Healthcare Research and Quality. August 2010. Available at: www.effectivehealthcare.ahrq.gov/reports/final.cfm.

Witberg G, Lador A, Yahav D, et al. Transcatheter versus surgical aortic valve replacement in patients at low surgical risk: A meta-analysis of randomized trials and propensity score matched observational studies. Catheterizations and Cardiovascular Interventions: official journal of the Society of Cardiac Angiography and Interventions. 2018 Feb 1;00:1–9. PMID: 29388308.

Yakubov SJ, Adams DH, Watson DR, et al. 2-yearoutcomes after iliofemoral self-expanding transcatheter aortic valve replacement in patients with severe aortic stenosis deemed extreme risk for surgery. J Am Coll Cardiol. 2015;66(12):1327-34. doi: 10.1016/j.jacc.2015.07.042.