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
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
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).
A. 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:
- 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.
- 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:
- Cardiac surgeon and an interventional cardiologist experienced in the care and treatment of aortic stenosis who have:
- independently examined the patient face-to-face, evaluated the patient’s suitability for surgical aortic valve replacement (SAVR), TAVR or medical or palliative therapy;
- documented and made available to the other heart team members the rationale for their clinical judgment.
- Providers from other physician groups as well as advanced patient practitioners, nurses, research personnel and administrators.
- The heart team's interventional cardiologist(s) and cardiac surgeon(s) must jointly participate in the intra-operative technical aspects of TAVR.
- TAVR must be furnished in a hospital with the appropriate infrastructure that includes but is not limited to:
- On-site heart valve surgery and interventional cardiology programs,
- Post-procedure intensive care facility with personnel experienced in managing patients who have undergone open-heart valve procedures,
- 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:
- ≥ 50 open heart surgeries in the previous year prior to TAVR program initiation, and;
- ≥ 20 aortic valve related procedures in the 2 years prior to TAVR program initiation, and;
- ≥ 2 physicians with cardiac surgery privileges, and;
- ≥ 1 physician with interventional cardiology privileges, and;
- ≥ 300 percutaneous coronary interventions (PCIs) per year.
Qualifications to begin a TAVR program for heart teams without TAVR experience:
The heart team must include:
Cardiovascular surgeon with:
- ≥ 100 career open heart surgeries of which ≥ 25 are aortic valve related; and,
Interventional cardiologist with:
- Professional experience of ≥ 100 career structural heart disease procedures; or, ≥ 30 left-sided structural procedures per year; and,
- Device-specific training as required by the manufacturer.
Qualifications for hospital programs with TAVR experience:
The hospital program must maintain the following:
- ≥ 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; and,
- ≥ 2 physicians with cardiac surgery privileges; and,
- ≥ 1 physician with interventional cardiology privileges, and
- ≥300 percutaneous coronary interventions (PCIs) per year;and,
- 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:
- All-cause mortality;
- Transient Ischemic Attacks (TIAs);
- Major vascular events;
- Acute kidney injury;
- Repeat aortic valve procedures;
- New permanent pacemaker implantation;
- Quality of Life (QoL).
- 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.
- 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?
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:
- The heart team's interventional cardiologist(s) and cardiac surgeon(s) must jointly participate in the intra-operative technical aspects of TAVR.
- 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?
- The clinical study must adhere to the following standards of scientific integrity and relevance to the Medicare population:
- 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.
- The rationale for the study is well supported by available scientific and medical evidence.
- The study results are not anticipated to unjustifiably duplicate existing knowledge.
- 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.
- The study is sponsored by an organization or individual capable of completing it successfully.
- 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
- All aspects of the research study are conducted according to appropriate standards of scientific integrity.
- The study has a written protocol that clearly demonstrates adherence to the standards listed here as Medicare requirements.
- 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.
- 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).
- 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).
- 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.
- 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: firstname.lastname@example.org
Email subject line: "CED [NCD topic (i.e. TAVR)] [name of sponsor/primary investigator]"
See Appendix B for the NCD manual language.
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 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.
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).
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
| June 27, 2018
|| CMS posts a tracking sheet announcing the opening of the NCA. The initial 30-day public comment period begins.
| July 25, 2018
|| CMS convened a meeting of the Medicare Evidence Development & Coverage Advisory Committee (MEDCAC) regarding procedural volume requirements for hospitals and heart teams to begin and maintain TAVR programs.
| July 27, 2018
|| First public comment period ends.
| March 26, 2019
|| Proposed decision memorandum posted. 30-day public comment period begins.
| April 25, 2019
|| 30-day comment period ends. CMS receives 212 comments.
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.
||Indication Risk Stratum
(transfemoral access only)
|| Edwards SAPIEN
(transfemoral access only)
|| Edwards SAPIEN
||Alternate access labeling expansion
|| Medtronic CoreValve
|| High risk
|| Edwards SAPIEN XT
|| High risk and above
|| Medtronic CoreValve
|| High risk and above
||Edwards SAPIEN 3
||High risk and above
||Medtronic CoreValve Evolut R
||Native and valve-in-valve
|| High risk and above
|| Edwards SAPIEN XT
||High risk and above
|| Edwards SAPIEN XT
||Edwards SAPIEN 3
|| Intermediate risk
|| Medtronic CoreValve Evolut PRO
|| Native and valve-in-valve
||High risk and above
|| Edwards SAPIEN 3
|| High risk and above
|| Medtronic CoreValve, CoreValve Evolut R, and CoreValve PRO
|| Intermediate risk
|| Edwards Sapien 3 Ultra
|| Native and valve-in-valve
||Intermediate risk or above
||Boston Scientific LOTUS Edge Valve System
||High risk and above
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.
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.
B. Discussion of Evidence
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?
2. External Technology Assessments
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.
3. Internal Technology Assessment
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.
Randomized Controlled Trials, Meta-Analyses, Observational Studies
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
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
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."
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
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."
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 Apr 22;63(15):1522-8. doi: 10.1016/j.jacc.2014.01.036
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."
Studies on TAVR Case Volume and Mortality Outcome
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.
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."
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 Nov 1;3(11):1070-1078. doi: 10.1001/jamacardio.2018.3562.
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
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 4th 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."
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.
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."
Observational Studies Using the TVT Registry
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 Mort