Author + information
- Received February 24, 2017
- Revision received May 22, 2017
- Accepted May 25, 2017
- Published online July 17, 2017.
- J. Matthew Brennan, MD, MPHa,∗ (, )
- Laine Thomas, PhDa,b,
- David J. Cohen, MD, MScc,
- David Shahian, MDd,
- Alice Wang, MDb,
- Michael J. Mack, MDe,
- David R. Holmes, MDf,
- Fred H. Edwards, MDg,
- Naftali Z. Frankel, MSh,
- Suzanne J. Baron, MDc,
- John Carroll, MDi,
- Vinod Thourani, MDj,
- E. Murat Tuzcu, MDk,
- Suzanne V. Arnold, MDc,
- Roberta Cohnh,
- Todd Maserl,
- Brenda Schawel,
- Susan Strongl,
- Allen Stickfortl,
- Elizabeth Patrick-Lakel,
- Felicia L. Graham, MBAb,
- Dadi Dai, PhDb,
- Fan Li, MSa,
- Roland A. Matsouaka, PhDa,
- Sean O’Brien, PhDa,c,
- Fan Li, PhDa,
- Michael J. Pencina, PhDa,b and
- Eric D. Peterson, MD, MPHa,b
- aDuke University School of Medicine, Durham, North Carolina
- bDuke Clinical Research Institute, Durham, North Carolina
- cSaint Luke’s Mid America Heart Institute, Kansas City, Missouri
- dMassachusetts General Hospital, Harvard Medical School, Boston, Massachusetts
- eThe Heart Hospital Baylor Plano Research Center, Plano, Texas
- fMayo Clinic, Rochester, Minnesota
- gUniversity of Florida Health Science Center, Jacksonville, Florida
- hCaregiver Collaborator, Duke Clinical Research Institute, Durham, North Carolina
- iUniversity of Colorado Hospital, Aurora, Colorado
- jEmory University School of Medicine, Atlanta, Georgia
- kCleveland Clinic Abu Dhabi, Abu Dhabi, United Arab Emirates
- lPatient Collaborator, Duke Clinical Research Institute, Durham, North Carolina
- ↵∗Address for correspondence:
Dr. J. Matthew Brennan, Division of Cardiology, Duke University Medical Center, 2400 Pratt Street, Room 0311 Terrace Level, Durham, North Carolina 27705.
Background Randomized trials support the use of transcatheter aortic valve replacement (TAVR) for the treatment of aortic stenosis in high- and intermediate-risk patients, but the generalizability of those results in clinical practice has been challenged.
Objectives The aim of this study was to determine the safety and effectiveness of TAVR versus surgical aortic valve replacement (SAVR), particularly in intermediate- and high-risk patients, in a nationally representative real-world cohort.
Methods Using data from the Transcatheter Valve Therapy Registry and Society of Thoracic Surgeons National Database linked to Medicare administrative claims for follow-up, 9,464 propensity-matched intermediate- and high-risk (Society of Thoracic Surgeons Predicted Risk of Mortality score ≥3%) U.S. patients who underwent commercial TAVR or SAVR were examined. Death, stroke, and days alive and out of the hospital to 1 year were compared, as well as discharge home, with subgroup analyses by surgical risk, demographics, and comorbidities.
Results In a propensity-matched cohort (median age 82 years, 48% women, median Society of Thoracic Surgeons Predicted Risk of Mortality score 5.6%), TAVR and SAVR patients experienced no difference in 1-year rates of death (17.3% vs. 17.9%; hazard ratio: 0.93; 95% confidence interval [CI]: 0.83 to 1.04) and stroke (4.2% vs. 3.3%; hazard ratio: 1.18; 95% CI: 0.95 to 1.47), and no difference was observed in the proportion of days alive and out of the hospital to 1 year (rate ratio: 1.00; 95% CI: 0.98 to 1.02). However, TAVR patients were more likely to be discharged home after treatment (69.9% vs. 41.2%; odds ratio: 3.19; 95% CI: 2.84 to 3.58). Results were consistent across most subgroups, including among intermediate- and high-risk patients.
Conclusions Among unselected intermediate- and high-risk patients, TAVR and SAVR resulted in similar rates of death, stroke, and DAOH to 1 year, but TAVR patients were more likely to be discharged home.
- safety and effectiveness
- surgical aortic valve replacement
- transcatheter aortic valve replacement
Aortic valve disease is the third most common cause of cardiovascular disease in the United States, affecting an estimated 2.5 million adults (5% of those affected are 65 years or older) (1,2). Severe untreated aortic valve stenosis substantially affects life expectancy and quality (3); however, patients with aortic valve disease are often older, with multiple comorbidities, making recovery from open surgical aortic valve replacement (SAVR) challenging (4). Over the past decade, transcatheter aortic valve replacement (TAVR) has emerged as a less invasive alternative to SAVR, thereby offering potential advantages for this older patient cohort (5). TAVR was approved by the U.S. Food and Drug Administration in 2011; since then, >80,000 commercial TAVR procedures have been performed in the United States in patients at intermediate, high, and prohibitive surgical risk (Matthew Brennan, February 4, 2017, personal communication).
To date, 3 high-quality randomized controlled trials have supported the use of TAVR in intermediate- and high-risk patients (6–8), but these clinical trials excluded important groups of patients with higher risk comorbidities and were conducted at a select group of high-volume valve centers. Consequently, whether these results are applicable to clinical practice has been questioned (9), and concerns regarding the safety and effectiveness of TAVR have been raised (10,11). These concerns are of increasing relevance because TAVR is applied to low- and intermediate-risk patients, in whom the risk of SAVR is less, and its long-term outcomes are well-documented (12).
To address these lingering questions, we used observational data from 2 large U.S. procedural registries to examine the real-world comparative effectiveness of TAVR versus SAVR in a nationally representative real-world cohort of older patients who may have been considered eligible for either TAVR or SAVR.
Study design and data sources
This was a multicenter, nonrandomized analysis of older patients with severe, symptomatic aortic valve stenosis at intermediate or high surgical risk who underwent treatment with TAVR or SAVR in the United States and may have been considered eligible for either treatment (on the basis of available data). Data for this analysis were drawn from 2 U.S. procedural registries: 1) SAVR data were drawn from the Society of Thoracic Surgeons (STS) National Database; and 2) TAVR data were drawn from the STS/American College of Cardiology TVT (Transcatheter Valve Therapy) Registry. The development and application of these registries have been described previously (13,14). More than 90% of cardiac surgery programs in the United States participate in the STS National Database, and participation in the TVT Registry is necessary for Medicare reimbursement. Notably, the involvement of a heart team is also necessary for Medicare reimbursement in the United States. For each registry, participants are required to submit 100% of their case records to the registry for quality assessment purposes. Missing data fields trigger critical warnings, and each registry has an independent data auditing program to ensure data accuracy. Records were linked to the Centers for Medicare and Medicaid Services fee-for-service administrative insurance claims files to create a longitudinal record including vital status and rehospitalization events, using validated techniques (15).
The most updated Medicare-linked files available were used from the TVT Registry and STS National Database. TVT Registry files are linked with Medicare claims by the Centers for Medicare and Medicaid Services twice each year, using updated files from the Centers for Medicare and Medicaid Services Chronic Conditions Data Warehouse. STS National Database files are linked with Medicare claims by the Duke Clinical Research Institute annually, using research-identifiable files from ResDAC (Minneapolis, Minnesota) (16). The availability of ResDAC files generally lags 12 to 18 months behind the date of service provision. Detailed clinical information and Medicare claims-based follow-up were available for 25,786 TAVR cases performed between January 1, 2014, and September 30, 2015, and 198,077 SAVR (or SAVR plus coronary artery bypass grafting) cases performed between July 1, 2011, and December 31, 2013.
Patients with characteristics that were thought to strongly favor 1 treatment or another were excluded (Online Figure 1). These characteristics included age <65 or >90 years, other major cardiac operations, history of endocarditis, emergency or salvage status, primary aortic insufficiency, hostile chest or porcelain aorta, moderate to severe mitral stenosis, and STS Predicted Risk of Mortality (PROM) score <3%. Subsequent aortic valve replacement procedures during the initial aortic valve replacement admission were excluded, and hospitals submitting <10 total SAVR or TAVR records during the study interval were also excluded. Following these exclusions, the population of interest included 17,910 TAVR and 22,618 SAVR patients who were available for propensity matching.
The Patient-Centered Outcomes Research Institute funded this study (grant CER-1306-04350), which the Institutional Review Board at the Duke University School of Medicine approved. The Duke Clinical Research Institute (Durham, North Carolina) was responsible for data management and statistical analysis, with oversight by a multidisciplinary research team that included patient and caregiver representatives, as well as statistical analysts and representatives from both the STS and American College of Cardiology.
Data definitions and outcomes
By design, data definitions are identical for most patient characteristics and outcomes across the STS National Database and the TVT Registry and are available for review in Online Table 1 and online (17,18). Outcomes from the index hospitalization were drawn from registry records. A list of potential outcomes available through Medicare claims were reviewed by a broad stakeholder panel that included patients, caregivers, clinicians, health science researchers, and statisticians. Primary outcomes of interest were chosen by consensus and included death, stroke, days alive and out of an acute care hospital facility (i.e., days alive and out of the hospital [DAOH]) to 1 year, and discharge home. Stroke and mortality were evaluated to 30 days and 1 year over a median follow-up period of 169.5 days for TAVR and 328 days for SAVR. Stroke was identified during the index procedural hospitalization using registry data. Following hospital discharge, stroke was identified using Medicare rehospitalization claims with a primary position International Classification of Diseases, Ninth Revision, Clinical Modification code of 434.x1, 436, 433.x1, 997.02, 437.1, 437.9, 430, 431, or 432.x.
An analytic sample was created using propensity score–based matching to correct for differences in characteristics of patients in the 2 registries. A propensity score, defined as the probability of receiving TAVR given measured covariates, was calculated using logistic regression. Detailed methods, including an extensive list of covariates identified by clinical input regarding factors thought to be related to both procedure selection and outcomes, and common to the 2 registries, are provided in Online Table 2. Overlap in the covariate distribution and propensity scores between study groups was assessed. Because patients at the tails (<5% and >95%) of the propensity distribution were thought to represent subjects with an overwhelming likelihood of treatment with 1 or the other of the 2 treatments, these patients were excluded (Online Figure 2). Propensity score matching was conducted in a 1:1 ratio, by greedy matching, using a caliper of 0.20 SDs in the linear predictor. The adequacy of the propensity model was confirmed by checking covariate balance before and after matching (Online Figure 3). Furthermore, to assess the potential for unmeasured confounding, the 2 treatments were compared using 2 falsification endpoints: lower extremity fracture and urinary tract infection. No statistically significant difference was observed for these outcomes to 1 year in the propensity-matched cohort (Online Figure 4).
Baseline characteristics of patients undergoing SAVR and TAVR were described and compared overall and within pre-specified subgroups on the basis of standardized differences (Online Figure 5). Cox proportional hazard models were used to compare outcomes of TAVR versus SAVR by hazard ratios and 95% confidence intervals (CIs). DAOH was modeled as count data using generalized estimating equations with a log link and a fixed offset (adjusting for differential follow-up time) to obtain rate ratios and 95% CIs. Models for treatment on outcomes were fit to the matched sample using a robust empirical variance to account for within-hospital clustering. Associations were estimated in pre-specified subgroups, along with 95% CIs and tests of interaction. Statistical significance was defined as p < 0.10, and significant values were evaluated for biological plausibility. All analyses were conducted using SAS version 9.4 (SAS Institute, Cary, North Carolina).
The propensity-matched cohort included 4,732 SAVR and 4,732 TAVR patients, with a median age of 82 years (interquartile range [IQR]: 77 to 85 years), 47.9% women, and a median STS PROM score of 5.6% (IQR: 4.2% to 8.2%). Baseline characteristics were well balanced across the 2 treatment groups (Table 1). Among TAVR patients, transfemoral access was used in 76%, and the valve prosthesis used was the CoreValve (Medtronic, Dublin, Ireland) in 33% and the Sapien (Edwards Lifesciences, Irvine, California) in 67%.
On average, TAVR patients spent 31 h (IQR: 24 to 57 h) in the intensive care unit and a total of 4 days (IQR: 3 to 6 days) in the hospital during the index admission, whereas SAVR patients spent an average of 68 h (IQR: 37 to 119 h) in the intensive care unit and 8 days (IQR: 6 to 11 days) in the hospital during the index admission. In-hospital mortality was lower among TAVR patients than SAVR patients (3.0% vs. 5.0%; p < 0.001), while the incidence of stroke was no different (2.5% vs. 2.7%; p = 0.40). Compared with SAVR patients, TAVR patients experienced a higher incidence of new pacemaker or implantable cardioverter-defibrillator placement (12.8% vs. 6.3%; p < 0.001) and major vascular complications (4.2% vs. 0.4%; p < 0.001) but a lower incidence of blood transfusions (packed red blood cells: TAVR, 0 U [IQR: 0 to 0 U]; SAVR, 2 U [IQR: 0 to 4 U]; p < 0.001) and new requirement for hemodialysis (1.7% vs. 3.2%; p < 0.001) during the initial hospitalization.
Discharge home was more common among TAVR patients than SAVR patients (69.9% vs. 41.2%), overall and within each subgroup that was studied (Online Figure 6). Discharge to an extended care facility, transitional care unit, or rehabilitation unit was more common among SAVR patients (41.2% vs. 20.5%; p < 0.01).
Death and stroke
No difference in death at 1 year was observed with TAVR versus SAVR (17.3% vs. 17.9%; p = 0.40), although a lower early risk for mortality was observed with TAVR (Central Illustration). A similar 1-year risk for death was observed across most subgroups of interest (Figure 1A); however, those with prior cardiac surgery experienced a lower 1-year risk for mortality when treated with TAVR versus SAVR (p for interaction = 0.09).
The risk for stroke was highest in the first 30 days following treatment and was identical between TAVR and SAVR (2.8% vs. 2.8%; p = 0.13) patients. An increase in the incidence of stroke was observed among TAVR (vs. SAVR) patients between 30 days and 1 year, with a progressive divergence of the stroke event curves. Nevertheless, the overall risk for stroke remained low during this interval (0.5% vs. 1.4%) (Figure 2), and the overall difference in risk for stroke was not significant to 1 year (TAVR vs. SAVR hazard ratio: 1.18; 95% CI: 0.95 to 1.47). Patients with home oxygen experienced a lower risk for stroke to 1 month with TAVR versus SAVR (p for interaction = 0.06) (Figure 1B), but by 1 year, neither treatment was favored in these patients (Online Figure 8).
Days alive and out of hospital
In the first year following hospital discharge, ≥80% of patients were alive and out of an acute care hospital for at least 11 of 12 months (Figure 3). The proportion of DAOH was similar between TAVR and SAVR patients (rate ratio: 1.0; 95% CI: 0.98 to 1.02), a result that was consistent across all subgroups (Figure 1C).
Influence of pre-operative surgical risk
After verifying covariate balance across 3 risk levels of STS PROM (3% to 5%, 5% to 8%, and ≥8%), a stratified analysis was performed. Increasing pre-operative surgical risk was associated with a lower likelihood of discharge home, fewer DAOH, a higher risk for stroke, and a higher risk for death to 1 year; however, the relative treatment effect (TAVR vs. SAVR) was consistent for each outcome of interest across the spectrum of intermediate to high baseline surgical risk (STS PROM) (Table 2).
In a broad cohort of older U.S. patients with severe aortic stenosis who were eligible for treatment with either TAVR or SAVR, no significant difference was observed in death, stroke, or DAOH to 1 year. TAVR patients were more often discharged directly home, reflecting a less demanding post-operative recovery. Results were consistent across most patient subgroups and across the spectrum of intermediate to high pre-operative surgical risk. These findings are largely consistent with those observed in pivotal randomized clinical trials and support the safety and effectiveness of TAVR in real-world intermediate- and high-risk patients.
In 3 previous randomized clinical trials among patients with severe aortic stenosis and intermediate or high surgical risk, TAVR has demonstrated similar (or superior) outcomes to 1 year when compared with SAVR. In high-risk patients, Cohort A of the randomized PARTNER (Placement of Aortic Transcatheter Valves) trial (n = 699, 25 U.S. centers) demonstrated similar rates of death to 1 year with a first-generation balloon-expandable TAVR prosthesis versus SAVR (24.2% vs. 26.8; p = 0.44) but with an increased risk for stroke or transient ischemic attack (8.3% vs. 4.3%; p = 0.04) (6). In a lower risk cohort, the U.S. CoreValve trial (n = 795, 45 U.S. centers) demonstrated lower rates of death to 1 year with a self-expanding TAVR prosthesis (14.2% vs. 19.1%; p = 0.04 for superiority), without an increased risk for stroke (8.8% vs. 12.6%; p = 0.10) (8). These results were consistent across most subgroups of patients. Among intermediate-risk patients in the PARTNER 2A trial (n = 2,032, 57 centers), patients randomized to a balloon-expandable second-generation TAVR prosthesis (vs. SAVR) experienced similar rates of death (12.3% vs. 12.9%; p = 0.69) or stroke (8.0% vs. 8.1%; p = 0.88) at 1 year (7). Again, no significant subgroup interactions were observed.
Despite favorable results in carefully controlled randomized trials, the generalizability of trial results to real-world patients has been questioned by some, because of systematic exclusion from clinical trials of patients with certain high-risk comorbidities (e.g., hemodialysis, recent stroke [<6 months prior], very low left ventricular ejection fraction [<20%]) (9). Responding to these concerns, nonrandomized evaluations of TAVR versus SAVR have been performed (19,20) with mixed results, particularly among intermediate-risk patients. A propensity-adjusted comparison of intermediate-risk TAVR patients from the SAPIEN 3 registry cohort (n = 963) versus SAVR patients from the PARTNER 2A trial (n = 747) demonstrated lower 1-year risks for mortality (7.4% vs. 13.0%) and stroke (4.6% vs. 8.2%) with TAVR using the third-generation Sapien 3 balloon-expandable prosthesis versus SAVR (21). By contrast, in a propensity-matched analysis of 5,997 intermediate-risk patients undergoing TAVR versus SAVR as part of GARY (German Aortic Valve Registry), patients treated with TAVR had a substantially higher 1-year risk for mortality versus SAVR (15.5% vs. 10.9%; p = 0.002) (11).
The results of our analyses are largely consistent with those of the pivotal randomized clinical trials. In a broad cohort of both intermediate- and high-risk older patients, the 1-year incidences of mortality and stroke are similar to those previously published for both intermediate- and high-risk patients. Consistent with the PARTNER and PARTNER 2A trial results, we observed a similar comparative risk for mortality for TAVR and SAVR among both intermediate- and high-risk patients. In contrast to results from GARY, we did not observe an increased risk for mortality among intermediate-risk patients. In our study, we used detailed phenotypic information from both the STS National Database and the TVT Registry to both exclude patients who would not have been considered for both procedures and closely match the remaining eligible patients; many of these variables were not available in other observational datasets. The availability of these additional data elements may account for differences between our study outcomes and both the GARY and SAPIEN 3 results, allowing a more accurate approximation of the existing randomized trial results.
Notably, the rates of stroke reported to 1 year in this cohort are roughly 50% lower than those reported to 1 year in each of the reported clinical trials, including the intermediate-risk PARTNER 2A clinical trial. The reduced strokes rates observed here are consistent with those reported by others, including the nonrandomized SAPIEN 3 intermediate-risk analysis (21); the reason for this finding is unclear. This observation may represent an underascertainment or underreporting of stroke events, because dedicated post-operative neurology evaluations that were available in pivotal trials were likely to reveal a higher incidence of both clinically significant and insignificant strokes. Similar to results from the PARTNER trial, we observed a nonsignificant but progressive increase in the 1-year risk for stroke among TAVR patients in our cohort. The cause (and clinical importance) of this observation is unknown. No such increase in stroke risk was observed in either the U.S. CoreValve High Risk trial or the PARTNER 2A trial. However, this finding warrants further investigation. To evaluate alternative strategies to address the excess risk for stroke following TAVR, both the ATLANTIS (Anti-Thrombotic Strategy After Trans-Aortic Valve Implantation for Aortic Stenosis) and the GALILEO (Global Study Comparing a Rivaroxaban-Based Antithrombotic Strategy to an Antiplatelet-Based Strategy after Transcatheter Aortic Valve Replacement to Optimize Clinical Outcomes) trials are randomizing patients post-TAVR to various post-TAVR anticoagulation strategies (22). Finally, we observed a significantly lower risk for stroke at 1 month among patients with home oxygen. We hypothesize that this finding is related to an increase in underlying aortic calcification (from tobacco exposure) among patients with home oxygen use. In these patients, avoiding direct manipulation of the ascending aorta with TAVR (vs. SAVR) may lead to a lower stroke incidence.
Importantly, we did not see significant differences in treatment effects across most patient subgroups, including within the intermediate- and high-risk strata. These results are generally consistent with available randomized data (23); however, our analysis does suggest that patients with significant lung disease and prior cardiac surgery may derive additional benefit from TAVR (vs. SAVR) for selected outcomes.
First, this was not a randomized treatment comparison, and bias (particularly through imbalances in patient frailty) may have influenced our results.
Second, we found that nearly one-half of the patients in the United States had a very high likelihood of receiving treatment with either SAVR (31.5%) or TAVR (14.8%), making it unlikely that the alternative treatment was considered a reasonable option in nearly one-half of patients. Consequently, the results reported here are intended to evaluate treatment effects among those generally considered eligible for either procedure, excluding patients with extremely high or low propensities for TAVR.
Third, although results of our subgroup analyses have demonstrated general parity of treatment effects across patient subgroups, it is likely that certain comorbidity combinations may favor 1 treatment over another. The importance of developing decision assistance tools to help optimize individualized patient care cannot be overstated.
Fourth, because of differences in the mechanisms of Medicare linkage from the STS National Database and TVT Registry, there was an offset in the interval of inclusion for SAVR (July 1, 2011, to December 31, 2013) and TAVR (January 1, 2014, to September 30, 2015). Consequently, the results reported here may underestimate the safety and effectiveness of SAVR if surgical outcomes significantly improved during the offset interval.
Fifth, the outcomes presented here were selected from a list of available outcomes by a broad stakeholder panel that included patients and caregivers; however, there was general agreement that quality of life is an important metric for consideration when choosing between these 2 procedures. Quality-of-life data, physical functioning, and New York Heart Association functional class at follow-up were not available for SAVR patients and therefore could not be presented in our study. Likewise, several important surrogate outcomes were not available, such as degree of aortic valve insufficiency and left ventricular remodeling. Also, expectations regarding long-term valve durability are key to treatment decisions, especially among younger patients and those at lower pre-operative surgical risk. An evaluation of the need for valve reintervention will be important as this cohort matures over time.
Sixth, cause of death was thought to be an important consideration that could not be addressed in our study because of a lack of necessary data.
Finally, it is important to recognize that the treatment of aortic valve disease is a rapidly developing field, with frequent modifications in device technology for both minimally invasive SAVR and TAVR. The data reported here are the most contemporary available in the United States and reflect outcomes of patients treated following the interval of early adoption of TAVR technology in the United States; however, recent TAVR device modifications have lowered device delivery profiles, improved prosthesis-annular apposition, and improved device repositioning capabilities. These modifications have lowered the incidence of procedural complications, including periprocedural stroke, acute vascular complications, device malposition, and perivalvular aortic insufficiency. As TAVR and SAVR devices continue to evolve, the relative risks and benefits of these 2 procedures may change. Diligent monitoring of outcomes will continue to help direct future device innovation in this field.
We used propensity score methods to compare 1-year outcomes of TAVR versus SAVR in a large, real-world cohort of older U.S. patients with aortic valve stenosis who were at intermediate or high surgical risk. Importantly, our results confirm and extend the observations of existing randomized studies in this field. Compared with SAVR, TAVR patients experienced a lower incidence of in-hospital mortality and were more often discharged directly to home. At 1-year follow-up, death, stroke, and DAOH were similar for the 2 treatments in the overall cohort and across most patient subgroups, including those within the spectrum of intermediate to high surgical risk.
COMPETENCY IN MEDICAL KNOWLEDGE: Among unselected patients at intermediate or high surgical risk, TAVR and SAVR were associated with similar rates of stroke, DAOH, and death at 1 year, but those undergoing TAVR were more often discharged home, reflecting easier post-operative recovery.
TRANSLATIONAL OUTLOOK: Continued evolution of TAVR and SAVR technologies may change the relative risks and benefits of these procedures and influence clinical decision making.
The authors thank Erin Campbell, MS, for her editorial contributions to this report. Ms. Campbell did not receive compensation for her assistance, apart from her employment at the institution where this study was conducted.
For a supplemental Methods section as well as supplemental figures and tables, please see the online version of this article.
The research reported in this article was funded through a Patient-Centered Outcomes Research Institute award (CER-1306-04350). The statements presented in this article are solely the responsibility of the authors and do not necessarily represent the views of the Patient-Centered Outcomes Research Institute, its Board of Governors, or Methodology Committee. Dr. Cohen has received research grant support from Edwards Lifesciences, Medtronic, Boston Scientific, and Abbott Vascular; and has received consulting income from Edwards Lifesciences and Medtronic. Dr. Carroll is a site investigator in PARTNER 2 (Placement of Aortic Transcatheter Valves) and the Medtronic low-risk trial. Dr. Tuzcu is a member of the executive committee of the PARTNER trial. All other authors have reported that they have no relationships relevant to the contents of this paper to disclose.
- Abbreviations and Acronyms
- confidence interval
- days alive and out of the hospital
- interquartile range
- Predicted Risk of Mortality
- surgical aortic valve replacement
- Society of Thoracic Surgeons
- transcatheter aortic valve replacement
- Received February 24, 2017.
- Revision received May 22, 2017.
- Accepted May 25, 2017.
- 2017 American College of Cardiology Foundation
- Goldbarg S.H.,
- Elmariah S.,
- Miller M.A.,
- Fuster V.
- Bonow R.O.,
- Mann D.L.,
- Zipes D.P.,
- Libby P.
- Nishimura R.A.,
- Otto C.M.,
- Bonow R.O.,
- et al.
- Holmes D.R. Jr..,
- Mack M.J.,
- Kaul S.,
- et al.
- Jilaihawi H.,
- Chakravarty T.,
- Weiss R.E.,
- Fontana G.P.,
- Forrester J.,
- Makkar R.R.
- Summers M.R.,
- Cremer P.C.,
- Jaber W.A.
- ↵Werner N, Zahn R, Beckmann A, et al. Patients at intermediate surgical risk undergoing isolated interventionally or surgical aortic valve replacement for severe symptomatic aortic valve stenosis. One year results from the German Aortic Valve Registry (GARY). Presented at: American Heart Association Scientific Sessions; November 14, 2016; New Orleans, LA.
- Brennan J.M.,
- Edwards F.H.,
- Zhao Y.,
- O’Brien S.M.,
- Douglas P.S.,
- Peterson E.D.
- Shahian D.M.,
- Jacobs J.P.,
- Edwards F.H.,
- et al.
- Carroll J.D.,
- Edwards F.H.,
- Marinac-Dabic D.,
- et al.
- Society of Thoracic Surgeons
- Society of Thoracic Surgeons and American College of Cardiology Transcatheter Valve Therapy Registry
- ↵Anti-Thrombotic Strategy After Trans-Aortic Valve Implantation for Aortic Stenosis (ATLANTIS). Available at: https://clinicaltrials.gov/ct2/show/NCT02664649. Accessed February 20, 2017.
- Siontis G.C.,
- Praz F.,
- Pilgrim T.,
- et al.