Author + information
- Received October 17, 2016
- Revision received November 20, 2016
- Accepted November 28, 2016
- Published online March 6, 2017.
- Frederick L. Grover, MDa,b,∗ (, )
- Sreekanth Vemulapalli, MDc,
- John D. Carroll, MDd,
- Fred H. Edwards, MDe,
- Michael J. Mack, MDf,
- Vinod H. Thourani, MDg,
- Ralph G. Brindis, MD, MPHh,
- David M. Shahian, MDi,
- Carlos E. Ruiz, MDj,
- Jeffrey P. Jacobs, MDk,l,
- George Hanzel, MDm,
- Joseph E. Bavaria, MDn,
- E. Murat Tuzcu, MDo,
- Eric D. Peterson, MD, MPHc,
- Susan Fitzgerald, RN, MSp,
- Matina Kourtis, MSq,
- Joan Michaels, RN, MSNp,
- Barbara Christensen, MSHA, RNp,
- William F. Seward, MAq,
- Kathleen Hewitt, MSN, RNp,
- David R. Holmes Jr., MDr,
- STS/ACC TVT Registry
- aDepartment of Surgery, School of Medicine, University of Colorado, Anschutz Medical Campus, Aurora, Colorado
- bDenver Department of Veterans Affairs Medical Center, Denver, Colorado
- cDuke University Medical Center, Durham, North Carolina
- dSchool of Medicine, University of Colorado, Anschutz Medical Campus, Aurora, Colorado
- eUniversity of Florida, Jacksonville, Florida
- fBaylor Scott and White Health, Plano, Texas
- gEmory University, Atlanta, Georgia
- hUniversity of California, San Francisco, California
- iMassachusetts General Hospital, Boston, Massachusetts
- jStructural and Congenital Heart Center, Hackensack University Medical Center and The Joseph M. Sanzari Children’s Hospital, Seton Hall Hackensack University–School of Medicine, Hackensack, New Jersey
- kDivision of Cardiac Surgery, Department of Surgery, Johns Hopkins University School of Medicine, Baltimore, Maryland
- lDivision of Cardiovascular Surgery, Department of Surgery, Johns Hopkins All Children’s Heart Institute, All Children’s Hospital and Florida Hospital for Children, St. Petersburg, Tampa, and Orlando, Florida
- mBeaumont Hospital, Royal Oak, Michigan
- nDivision of Cardiovascular Surgery, University of Pennsylvania, Philadelphia, Pennsylvania
- oCleveland Clinic, Abu Dhabi, United Arab Emirates
- pAmerican College of Cardiology, Washington, DC
- qThe Society of Thoracic Surgeons, Chicago, Illinois
- rMayo Clinic, Rochester, Minnesota
- ↵∗Address for correspondence:
Dr. Frederick L. Grover, University of Colorado School of Medicine, Anschutz Medical Campus, Department of Surgery, Academic Office 1, Room 6613, 12631 E. 17th Avenue, MSC310, Aurora, Colorado 80045.
Background The Society of Thoracic Surgeons (STS)/American College of Cardiology Transcatheter Valve Therapy (TVT) Registry captures all procedures with Food and Drug Administration–approved transcatheter valve devices performed in the United States, and is mandated as a condition of reimbursement by the Centers for Medicaid & Medicare Services.
Objectives This annual report focuses on patient characteristics, trends, and outcomes of transcatheter aortic and mitral valve catheter-based valve procedures in the United States.
Methods We reviewed data for all patients receiving commercially approved devices from 2012 through December 31, 2015, that are entered in the TVT Registry.
Results The 54,782 patients with transcatheter aortic valve replacement demonstrated decreases in expected risk of 30-day operative mortality (STS Predicted Risk of Mortality [PROM]) of 7% to 6% and transcatheter aortic valve replacement PROM (TVT PROM) of 4% to 3% (both p < 0.0001) from 2012 to 2015. Observed in-hospital mortality decreased from 5.7% to 2.9%, and 1-year mortality decreased from 25.8% to 21.6%. However, 30-day post-procedure pacemaker insertion increased from 8.8% in 2013 to 12.0% in 2015. The 2,556 patients who underwent transcatheter mitral leaflet clip in 2015 were similar to patients from 2013 to 2014, with hospital mortality of 2% and with mitral regurgitation reduced to grade ≤2 in 87% of patients (p < 0.0001). The 349 patients who underwent mitral valve-in-valve and mitral valve-in-ring procedures were high risk, with an STS PROM for mitral valve replacement of 11%. The observed hospital mortality was 7.2%, and 30-day post-procedure mortality was 8.5%.
Conclusions The TVT Registry is an innovative registry that that monitors quality, patient safety and trends for these rapidly evolving new technologies.
- procedural outcomes
- procedural risk
- transcatheter mitral clip
- transcatheter valves
- valve in valve
Following approval in November 2011 by the U.S. Food and Drug Administration (FDA) of the initial transcatheter aortic valve prosthesis, the transcatheter heart valve (Sapien, Edwards Lifesciences, Irvine, California), a National Coverage Decision (NCD) for transcatheter aortic valve replacement (TAVR) was developed by the U.S. Centers for Medicare & Medicaid Services (CMS). The CMS provides coverage for any patient receiving an FDA-approved device for a labelled indication. The NCD also mandated that the heart team and hospital must participate in a prospective nationally audited registry that consecutively enrolls these patients and that tracks certain specifications of the outcomes both for post-market surveillance of the device and also to determine the quality of outcomes at the individual centers (1).
As a result, The Society of Thoracic Surgeons (STS) and the American College of Cardiology (ACC) developed the Transcatheter Valve Therapy (TVT) Registry, working with the FDA, CMS, and industry manufacturers of these devices. The TVT Registry became operational on December 1, 2011, after being recommended by the FDA Cardiovascular Circulatory Device Panel in July 2011 (2). In January 2014, the self-expanding transcatheter valve (CoreValve, Medtronic, Minneapolis, Minnesota) was approved as the second TAVR valve in the United States. The first transcatheter mitral repair device was approved in the United States for degenerative mitral regurgitation (MitraClip device, Abbott Vascular, Menlo Park, California) in October 2013 (3). CMS also developed NCDs for the MitraClip device (4). With device iterations, newer generations of the valves (Sapien, SapienXT, Sapien3, and CoreValve) and CoreValve Evolut (Medtronic) have subsequently been added to the TVT Registry (Central Illustration). In part on the basis of data from the Registry, the FDA expanded labeling of TAVR devices to include alternate access for inoperable patients and valves in previously implanted surgical bioprostheses.
The TVT Registry issues annual progress reports (5–8) and the following report is an update of the procedures tracked in the TVT Registry since its inception in late 2011 through December 31, 2015.
The Registry currently has 3 modules—TAVR, transcatheter mitral leaflet clip (TMC), and transcatheter mitral valve-in-valve (TMViV)/transcatheter mitral valve-in-ring (TMViR). Patients who underwent a transcatheter aortic valve-in-valve (TAViV) implantation are part of the TAVR module. Each module captures demographic information, comorbidities, aortic and mitral specific baseline criteria, procedural details, outcomes such as mortality and morbidity (including in-hospital, 30 days, and 1 year), and quality of life. The Registry is HIPAA compliant, received an institutional review board waver of consent, and has a risk model for predicting in-hospital mortality that can be used for post-market surveillance and that enables each hospital to compare its results to national and peer group averages for quality improvement. Data are collected and stored by the ACC, which also provides participating hospital benchmark dashboards. Data are transmitted quarterly to the Duke Clinical Research Institute (DCRI), where an analytic copy is generated and kept from which statistical analyses for research and publications are performed. A Research and Publication Committee has been established to evaluate and approve research proposals and to monitor their progress. The data is independently audited for accuracy and electronically for data completeness. Certain endpoints are adjudicated by the DCRI, including stroke, transient ischemic attack, heart failure hospitalizations for TMC, and aortic and mitral reintervention. The TVT Registry Steering Committee overseeing the strategic direction of the Registry is composed of physicians appointed from the STS and ACC, FDA and CMS representatives, and staff members of STS and ACC. A Stakeholder Advisory Committee includes members from device manufacturers, patients, consumers, FDA and CMS, physicians, and health systems, and provides meaningful input regarding the needs and goals of each of the stakeholders.
TVT Registry hospital participants capture data that is used to direct in-hospital outcomes, and stroke and reoperation outcomes were adjudicated by a board-certified cardiologist at the DCRI using Valve Academic Research Consortium-2 (VARC-2) definitions. The adjudication process involved a review of specific queries and deidentified source records as needed.
For 1-year outcomes, Medicare in-hospital administrative claims files were used for detection of rehospitalization events during the study interval using the following International Classification of Diseases-Ninth Revision-Clinical Modification codes: for stroke, 433.x1, 434.x1, 997.02, 436, 437.1, 437.9, 430, 431, and 432.x; for heart failure, 398.x, 402.x1, 404.x1, 404.x3, and 428.x; and for aortic valve reintervention, 35.11, 35.21, 35.22, 35.01, 35.05, 35.06, and 35.09. For rehospitalization events, follow-up was censored at the time of death, end of fee-for-service coverage, loss of Part A or B coverage, or end of the follow-up period, whichever occurred first. Following hospital discharge, death was identified using the Medicare Denominator file. Patient follow-up was considered censored at the end of the most recent follow-up period (September 2015).
Baseline patient characteristics and in-hospital outcomes were summarized by percentages and compared across subgroups using chi-square, Wilcoxon, or Kruskal-Wallis 2-sided tests, as appropriate. For all analyses, p values <0.05 were considered statistically significant, and all analyses were performed at the DCRI using SAS software, version 9.3 (SAS Institute, Cary, North Carolina).
As of December 31, 2015, there were 418 sites that perform TAVR (Figure 1), 176 TMC centers (Figure 2), 98 TMViV and TMViR sites (Figure 3), and 292 centers that perform TAViV. Figure 4 demonstrates the geographic dispersion of TAVR centers.
By the end of 2015, there were 54,782 TAVR procedures entered into the Registry, which increased annually from 4,627 in 2012 to 24,808 in 2015 (p < 0.0001), and 1,898 patients who underwent TAViV. From 2013 to 2015, 3,745 TMC procedures were performed. In addition, from 2013 to 2015, 349 patients had a TMViV or TiViR procedure.
TAVR baseline characteristics
Baseline characteristics of the 54,782 TAVR patients are shown in Table 1. The median age of 83 years was remarkably consistent over all 4 years, with some increase in patients ≤80 years of age (Figure 5). There was a relatively even distribution of male and female patients (52% vs. 48%, respectively). Overall, the population was predominantly white (94%) and was 3% to 4% of Black/African American and 3% to 4% Hispanic. The median Society of Thoracic Surgeons Predicted Risk of Mortality (STS PROM) was 7% for the 4-year period, although in 2015 it decreased to 6% (p < 0.0001). The TVT Registry overall predicted in-hospital mortality was 4%, and decreased to 3% (p < 0.0001) in 2015. The 5-m walk test was positive for frailty in 52.8% of the 79% of patients who were tested in 2014 and 57.5% of the 85% who were tested in 2015. These values were relatively consistent over the 4 years, except for slight decreases in 2015.
TAVR operative characteristics
TAVR operative characteristics are shown in Table 1. General anesthesia was used in 97.6% of patients in 2012 and decreased to 82.6% in 2015. Conversely, moderate sedation increased from 2.2% to 16.6% over the same period. Overall, 74.9% of patients received a balloon-expandable valve, with a high of 96.9% in 2013; this decreased to 66.4% in 2015 after the commercial approval of a self-expanding valve, which accounted for 32.6% of TAVRs in 2015. The femoral artery access site for valve insertion progressively increased from 75.9% in 2012 to 86.6% in 2015. Transapical access has decreased from 14.5% in 2012 to 6.1% in 2015. Other access sites were only used in 6.8% patients in 2015.
TAVR outcomes are listed in Table 2. TAVR in-hospital deaths have progressively decreased from a high of 5.7% in 2012 to 2.9% in 2015 (p < 0.0001), and the 30-day mortality has decreased from 7.5% in 2012 to 4.6% in 2015 (p < 0.0001) for an overall 4-year average of 5.7% (Figure 6). The 1-year mortality rate was 22.6% overall and decreased from 25.8% in 2012 to 21.6% in 2014 (p < 0.0001) (Figure 7). Overall in-hospital stroke rate was 2.1%, and slightly decreased over the 4-year period from 2.2% to 2.0% (p < 0.05). The 30-day stroke was 2.1% overall decreasing from 2.3% to 1.9% in 2015 (p = 0.03). The 1-year stroke rate obtained by linkage with CMS administrative data was 3.8% overall (Figure 7).
Post-operative atrial fibrillation decreased over time from 6.9% in 2012 to 2013 to 3.7% in 2015, but conversely, 30-day new pacemaker insertion was 11.8% overall increasing from 8.8% in 2013 to 12% in 2015 (p < 0.0001) (Table 2). Acute kidney injury requiring dialysis occurred in 1.4% of patients overall and decreased from 1.7% in 2012 to 0.9% in 2015 (p < 0.0001).
There was no major bleeding in 91.6% of patients overall, improving from 87.1% in 2012 to 93.1% in 2015 (p < 0.001). Packed red blood cells or whole blood transfusions were required in 30.1% of patients overall and decreased from 44.7% in 2012 to 22.0% in 2015 (p < 0.001), representing a 50% reduction in use of blood products. Major vascular access site complications from 2013 to 2015 occurred in only 1% to 2% of patients, and remained relatively constant (Figure 8).
Overall, among the 90% of patients with echocardiograms, aortic regurgitation at discharge or 30 days was none/trace in 64.1% of patients, mild in 29.0%, and moderate/severe in 6.9%; this improved over the 4-year period, with moderate/severe aortic regurgitation decreasing from 10.8% in 2012 to 6.2% in 2015 (p < 0.0001) (Figure 9). The post-procedure aortic valve gradient at discharge (documented in 87% of patients) was <10 mm Hg in 46.5% of patients in 2012 and improved to 57.3% in 2015. The post-procedure aortic valve gradient was between 10 and 20 mm Hg in 45.8% of patients in 2012, decreasing to 36.4% in 2015; it was >20 mm Hg in 7.7% of patients in 2012, decreasing to 6.3% in 2015 (Figure 10). Aortic valve reintervention at 30 days was 0.3%, with very little variation. The 30-day valve-related readmission, derived from the subset of patients with CMS linkage, was 0.9% and was relatively stable over the 4-year time period. In-hospital length of stay, from admission to discharge or death, was 5 days, decreasing from 6 days in 2012 to 4 days in 2015 (p < 0.0001).
Mitral leaflet clip procedures—baseline data
TMC procedures were first entered into the TVT Registry in November 2013 after FDA approval for commercial use for patients with prohibitive risk for open mitral valve repair, and baseline data for these procedures is shown in Table 3. The overall median age was 81 years, decreasing from 82 years in 2014 to 81 years in 2015 (p < 0.01). A majority of patients were male (54.5%), and this was also fairly consistent throughout the 3-year period. The STS PROM for mitral valve repair was 6.1% from 2014 to 2015. A hostile chest was present in 7.6% of patients, and home oxygen use was 14.5% overall for 2014 to 2015. A porcelain aorta was present in 2.1% of patients over the time period of 2014 to 2015 and decreased to 1.6% in 2015.
Mitral leaflet clip procedure outcomes: 2014 to 2015
Mitral leaflet clip outcomes are summarized in Table 4. In-hospital deaths occurred in an average of 2.3% of patients overall, with a slight reduction from 2.9% in 2014 to 2.1% in 2015 (p = nonsignificant [NS]). Thirty-day mortality occurred in 5% of patients overall, with a 4.7% rate in 2015 (p = NS). In-hospital stroke was present in 0.5% of patients overall and decreased to 0.3% in 2015 (p < 0.05). The 30-day stroke rate was 0.7% overall, decreasing to 0.6% in 2015 (p = NS). Overall morbidity was low, with new atrial fibrillation in 1.1% and acute renal injury requiring dialysis in 0.7%. The major vascular access site complication rate was 0.2% overall and remained consistent.
The most important periprocedural morbidity was single leaflet device attachment, which was identified in 1.4% of patients and was consistent over time, as was the 30-day single leaflet device attachment (1.5%). The need for closure of the atrial septostomy following mitral clip delivery was 1.4%. In-hospital mitral valve reintervention was 0.6%, and 30-day mitral valve reintervention was 0.8%, both of which were consistent over the study period. Readmission within 30 days due to heart failure occurred in 2.4% of patients overall, and was consistent over time. Mitral regurgitation at discharge or 30 days was none, trace, mild, or moderate regurgitation (grade ≤2) and occurred in 85.2% in 2014 and 86.6% in 2015. A mitral valve gradient of <5 mm Hg occurred in 73.8% of patients overall, with no significant difference over time.
TMViV and TMViR procedures—baseline characteristics
Baseline characteristics for TMViV and TMViR procedures are shown in Table 5. From August 2013 through 2015, a total of 349 patients underwent a TMViV (insertion into a degenerated mitral valve prosthesis) or TMViR (insertion in patients with severe mitral regurgitation following repairs using annular rings) procedure, with relative incidences of 76.1% and 23.9% of patients, respectively. The median patient age was 76 years, and patients were predominantly female (61%). The median STS PROM for this group was 11% overall and was consistent over the years. A hostile chest was present in 13.4% of patients overall with no significant trends, and home oxygen use was 14.8% with porcelain aorta present in 2.0% of patients.
TMViV and TMViR procedures—outcomes
TMViV and TMViR in-hospital outcomes are shown in Table 6. Overall, 70.1% of patients had transapical access and 24.4% had transseptal access. There was a trend toward more transseptal access in 2015 (28.2%). In-hospital mortality was 7.2%, and mortality at 30 days was 8.5%, comparing favorably with the STS PROM of 11.1%. Post-procedure morbidity was low in these complex patients, with post-operative atrial fibrillation occurring in 4.0%, cardiac arrest in 3.4%, stroke in 1.1%, and acute kidney injury requiring dialysis in 2.9% of patients. VARC-2 major bleeding occurred in 4.9%, and the rate of life-threatening bleeding was 4.6%. Blood transfusions were administered to 40.4% of patients, and only 0.6% of patients had a major vascular access site complication. Closure of the atrial septostomy was required in 5.7% of patients. In the 84% who had echocardiograms, the post-procedure incidence of mitral insufficiency was none in 48.1%, trace/trivial/mild in 33%, and moderate/moderate-severe/severe in only 2.6%. The post-procedure median mitral valve gradient was 6 mm Hg. In-hospital left ventricular outflow obstruction developed in only 1.4%.
The FDA issued an announcement titled “Strengthening Our National System for Medical Device Post-market Surveillance” in April 2013 (9), revealing several significant steps taken to shorten the approval process for medical devices on the basis of strong post-market surveillance. They noted that the input and active participation from many other stakeholders is necessary to strengthen post-market surveillance and that a national system cannot be implemented and achieved by the FDA alone, and they advocated for device-specific registries in selective product areas supplemented by additional data sources, such as adverse event reports and administrative and claims data. These would enhance post-market surveillance, thereby enabling the FDA to be more efficient in the approval process for life-saving devices. The use of registries for FDA surveillance ultimately will benefit our patients, the industry manufacturers, those of us that care for patients, and our hospitals, by decreasing the time from development to commercial approval (10).
In this paper, we have reviewed the overall experience and trends for patients undergoing TAVR, the most recent data on TMC, and trends including recent baseline data and outcomes for the TMViV and TMViR procedures.
Transcatheter aortic valve replacement
As noted, the number of TAVR sites enrolled in TVT Registry has continued to increase. Trends have revealed a significant decrease in the patients’ disease severity, likely reflecting the change in procedure approval criteria (from patients who were inoperable for surgical aortic valve replacement to high risk) during the time reviewed. However, most patients undergoing TAVR remain elderly and have a major burden of comorbid conditions.
Importantly, actual TAVR in-hospital, 30-day, and 1-year mortality significantly decreased over time. The 1-year mortality continues to be high, and further investigation into the predictors of which patients are unlikely to benefit from the procedure, both in terms of survival and quality of life, needs to occur. Stroke, however, remained relatively constant throughout the time period of TAVR. The amount of patients requiring an intraprocedure or post-procedure pacemaker increased beginning in 2014, likely resulting from commercial approval of self-expanding TAVR devices. There have been modest but significant decreases in acute kidney injury and major bleeding over time, the latter likely due to improved technology such as the introduction of the newer valve models having smaller sheath sizes (minimum of 14-F). This enabled a greater percentage of patients to receive femoral/iliac cannulation versus transapical, and also resulted in less iliofemoral artery injury and better patient selection and operative techniques. Moderate/severe aortic regurgitation has decreased over time, with acceptable gradients at discharge. The 30-day aortic valve reintervention rate was very low. The improved results are likely related to greater experience and a lower-risk patient population, as well as improving technology and anesthetic techniques.
The 2015 report from the GARY (German Aortic Valve Registry) (11) on 15,964 TAVR procedures performed from 2011 to 2013 revealed similar outcomes to TAVR patients in the TVT Registry. The average age of both populations was low 80s, and the German STS PROM was 5%, whereas TVT was 7%. In-hospital mortality for GARY was 5.2% as opposed to 3.9% in the TVT Registry. Cerebrovascular accident occurred in 1.5% of GARY TAVR patients compared with 2.1% of patients in the TVT Registry.
At this point little is known about the ultimate durability of the transcatheter valves. There have been no definite negative signs detected by the TVT Registry to date; however, the longest follow-up is only 3 years. The TVT Registry is currently conducting a research study to explore leaflet mobility and evidence of thrombosis. Makkar et al. (12) found evidence of reduced leaflet motion using 4-dimensional computed tomography in 40% of patients in a clinical trial and noted a reduction in this phenomenon with warfarin. However, there appeared to be very little if any clinical correlation with this finding.
The FRANCE-2 (FRench Aortic National CoreValve and Edwards-2) Registry (13) reported 3-year follow-up of 4,201 TAVR patients enrolled in 2010 to 2011 and found a 3-year all-cause mortality of 42% and a cardiovascular mortality of only 17.5%, with severity of aortic valve gradients and regurgitation up to 3 years indicating adequate valve performance.
Aortic valve in valve procedures
Overall, 2,893 TAViV procedures have been entered into the database from 2012 to 2015. A detailed paper on the TVT analysis of this procedure will be forthcoming, and therefore, the results are not reported here.
Sorajja et al. (14) reported the TVT Registry experience with patients undergoing TMC procedures from November 2013 through August 2014. For the purposes of this report, we reviewed 3,745 patients’ data with a special focus on 2015 to see if there were any trends since their publication that might reflect on the outcomes after initial experience with the procedure. These were only minor differences, such as a slight decrease in age. The STS PROM for mitral valve repair remained constant, and in-hospital and 30-day observed mortality moderately decreased. Correction of mitral regurgitation by this technique was limited to the group of patients who were classified to be at prohibitive risk for an open procedure. Mitral regurgitation was significantly reduced in 87% of patients to grade ≤2.
TMViV and TMViR procedures
There were 349 TMViV and TMViR (76.1% and 23.9%, respectively) procedures performed with transcatheter aortic valves from 2013 through 2015. The patients were predominantly at high risk. In-hospital and 30-day mortality were considerably lower than the STS PROM. Very few patients experienced post-operative left ventricular outflow tract obstruction, in-hospital stroke, or required dialysis. No patients required an in-hospital mitral valve reintervention; among 83.7% of patients who had available echocardiograms, moderate to severe mitral regurgitation was unusual and gradients were low. The early results in this rather high-risk group of patients, who were able to avoid an open reoperation, are encouraging, and this procedure appears to be an attractive treatment option for this group of patients. Of interest is the slight increase in the past year of the transseptal approach from 14.6% to 28.2%. These data compare very favorably to those reported by Eleid et al. (15), who reviewed 33 patients with TMViV and 9 with TMViR procedures, with an STS PROM of 13.2% versus 11% for TVT with similar post-operative echocardiographic results.
Data completeness and accuracy have been an issue, as it frequently is in newly developed registries. A data quality program exists promoting data completeness and accuracy by including electronic data checks; proactive training and orientation by TVT staff with participating hospitals and their data managers; and data completeness assessment and reporting to sites, giving each site feedback to facilitate data cleaning. An annual NCDR Data Managers Meeting is held to network and discuss strategies for accurate and complete data. The TVT Steering Committee monitors data quality and, most recently, completed the first pilot audit. A separate task force is conducting an in-depth analysis of data quality, completeness, and accuracy of the TVT data. Several data quality improvement opportunities have been identified, such as improving 1-year follow-up ascertainment, and these are being addressed. Data collection can be burdensome for data collectors, and work is underway to simplify the number of data elements collected, focusing particularly on those that are critically important to adequately measure quality of care, enable risk adjustment of outcomes, and provide post-market surveillance of the new devices. Data completeness has slowly improved over the 4 years of data collection; in-hospital data completeness is relatively good, but the 30-day and 1-year data collection needs improvement.
The first pilot audit report indicates an overall agreement rate of 83.4% for audited procedures that occurred between October 1, 2013, and June 30, 2014. The aggregate agreement rate for follow-up data was not as high as baseline and warrants improvement. Linkage with the CMS administrative database enhances the ability of the TVT Registry to access long-term outcomes. An audit of 2015 data is currently being conducted.
Additional new foci for TVT registry
In anticipation of newer developing technology and what is expected of the TVT Registry, data elements are being developed to capture transcatheter tricuspid replacement and for metrics to report appropriate use criteria for TAVR procedures.
Thus far, 20 papers have been published (listed in the Online Appendix), 20 are currently in the process of being written, and 20 new proposals are being evaluated by the Research and Publications Committee. We are increasing the analytic capacity for the research projects and are working on linking the STS and TVT databases for relevant research studies.
The TVT Registry Steering Committee is committed to the development of a series of risk models that will be derived entirely from TVT Registry data. Using this rich source of “real-world” clinical information from virtually all commercial TAVR cases in the United States, several models are now in various stages of development.
In-hospital mortality was chosen as the outcome for the initial model because it is a clearly defined outcome with clinical importance and a high degree of completeness. A model based on over 13,000 TAVR patients was developed for use at the patient level (16). Discrimination and calibration indexes are more favorable than previously reported models used for TAVR. This model is available as a mobile application and is a valuable clinical tool for patient counseling.
Another model to evaluate in-hospital mortality has been developed for risk-adjusted comparisons at the hospital level (17). This model is used by the TVT Registry to provide objective information for hospitals to compare their local results to risk-adjusted national benchmarks.
Subsequent models to predict 30-day and long-term mortality are planned. A model to predict the probability of stroke is in the final stages of development, and nonfatal outcome models will be added in the near future, including a composite model. Existing models will be updated with additional data at regular intervals. One of the most promising areas of statistical modeling deals with predicting benefits as well as risks, including quality-of-life measures. This type of risk/benefit analysis is particularly important in the TAVR population and will therefore be an area of major emphasis in the future.
The rapid launch and progress of the TVT Registry has yielded important trends for patient outcomes and clinical care of TAVR patients. The TVT Registry continues to mature, capturing real-world patient characteristics, appropriateness, and outcomes for TAVR, TAViV, TMC, and TMViV; it is now preparing for transcatheter tricuspid valve technology. An in-hospital mortality risk model has been developed, and other models are in development that will help individual hospital programs assess their performance compared with national peer groups, thereby serving as a tool for quality improvement.
Trends over time have been captured for current FDA-approved devices and have shown overall progressive improvement in outcomes, and in TAVR, a decrease in predicted risk corresponding to changes in the CMS-NCD. The TMViV and TMViR procedures appear promising for high-risk patients with degenerated surgical mitral bioprostheses or recurrent mitral insufficiency following annular ring repair, at least on the basis of initial short-term outcomes. In patients with severe mitral regurgitation of their native mitral valve, who are at prohibitive risk for surgical mitral valve repair, TMC procedures were shown to reduce severe mitral regurgitation to grade 2 or less at a relatively low risk.
There is a continued focus on using a multidimensional approach to improve data completeness and accuracy through education, electronic data checks, and most recently, external audits. Long-term follow-up is augmented by linking TVT to CMS administrative data. The TVT Registry has served as a focus to bring providers, the CMS, the FDA, industry, and the public together, enhancing communication and cooperation. The primary goals are to improve patient care and outcomes, decrease the overall time for post-approval studies, facilitate continued evidence development, and provide enhanced device surveillance.
COMPETENCY IN SYSTEMS-BASED PRACTICE: A registry of 54,782 TAVR procedures performed from 2012 through 2015 found decreases in early and 1-year mortality but increased post-procedure pacemaker insertion. A much smaller number of patients underwent catheter-based mitral valve interventions.
TRANSLATIONAL OUTLOOK: As technology and outcomes improve, progressively lower-risk patients are likely to become candidates for these catheter-based valve interventions.
For a list of other TVT Registry publications, please see the online version of this article.
Dr. Grover is the recipient of a Thoracic Surgery Foundation grant for medical mission work in Nepal, funded by Edwards. Dr. Vemulapalli has received research grants from Abbott Vascular, the American College of Cardiology, and The Society of Thoracic Surgeons; and has served as a consultant for Novella and Premiere Research. Dr. Mack has served as the co-principal investigator for the Partner 3 trial and COAPT trial. Dr. Thourani has served as a consultant for Edwards. Dr. Ruiz has received institutional educational grants from Medtronic; and institutional grants from Philips. Dr. Hanzel has served as an unpaid proctor for the Watchman device from Boston Scientific. Dr. Bavaria is a primary investigator for Edwards Lifesciences and Medtronic; and a sub primary investigator for St. Jude Medical and Boston Scientific. Dr. Peterson has served as the co-principal investigator of the Data Coordinating Center for The Society of Thoracic Surgeons-American College of Cardiology Transcatheter Aortic Valve Replacement. All other authors have reported that they have no relationships relevant to the contents of this paper to disclose.
This article is copublished in the Journal of the American College of Cardiology and The Annals of Thoracic Surgery.
- Abbreviations and Acronyms
- American College of Cardiology
- Duke Clinical Research Institute
- National Coverage Decision
- Society of Thoracic Surgeons
- STS PROM
- Society of Thoracic Surgeons Predicted Risk of Mortality
- transcatheter aortic valve-in-valve
- transcatheter aortic valve replacement
- transcatheter mitral leaflet clip
- transcatheter mitral valve-in-ring
- transcatheter mitral valve-in-valve
- transcatheter valve therapy
- Valve Academic Research Consortium
- Received October 17, 2016.
- Revision received November 20, 2016.
- Accepted November 28, 2016.
- American College of Cardiology Foundation and The Society of Thoracic Surgeons
- ↵Centers for Medicare & Medicaid Services. National coverage determination (NCD) for transcatheter aortic valve replacement (TAVR) (20.32). 2013. Available at: https://www.cms.gov/medicare-coverage-database/details/ncd-details.aspx?NCDId=355&ncdver=1&NCAId=257&ver=4&NcaName=Transcatheter+Aortic+Valve+Replacement+(TAVR)&bc=ACAAAAAACAAAAA%3D%3D&. Accessed August 21, 2016.
- Rumsfeld J.S.,
- Holmes D.R.,
- Stough Gattis W.,
- et al.
- ↵Centers for Medicare & Medicaid Services. Proposed decision memo for transcatheter mitral valve repair (TMVR) (CAG-00438N). Available at: https://www.cms.gov/medicare-coverage-database/details/nca-proposed-decision-memo.aspx?NCAId=273&NcaName=Transcatheter+Mitral+Valve+(TMV)+Procedures&bc=ACAAAAAAAgAAAA%3D%3D&. Accessed August 21, 2016.
- ↵U.S. Food and Drug Administration. FDA expands approved use of Sapien artificial heart valve. October 19, 2012. Available at: http://www.fda.gov/NewsEvents/Newsroom/PressAnnouncements/ucm323478.htm. Accessed August 21, 2016.
- Carroll J.D.,
- Edwards F.H.,
- Marinac-Dabic D.,
- et al.
- Holmes D.R.,
- Nishimura R.A.,
- Grover F.L.,
- et al.
- ↵Center for Devices and Radiological Health. Strengthening our national system for medical device postmark surveillance update and next steps. 2013. Available at: http://www.fda.gov/downloads/MedicalDevices/Safety/CDRHPostmarketSurveillance/UCM348845.pdf. Accessed July 16, 2016.
- Carroll J.D.,
- Shuren J.,
- Jensen T.S.,
- et al.
- Walther T.,
- Hamm C.W.,
- Schuler G.,
- et al.
- Gilard M.,
- Eltchaninoff H.,
- Donzeau-Gouge P.,
- et al.
- Sorajja P.,
- Mack M.,
- Vemulapalli S.,
- et al.
- Eleid M.F.,
- Cabalka A.K.,
- Williams M.R.,
- et al.
- Edwards F.H.,
- Cohen D.J.,
- O’Brien S.M.,
- et al.
- O’Brien S.M.,
- Cohen D.J.,
- Rumsfeld J.S.,
- et al.