Author + information
- Received November 14, 2018
- Revision received February 12, 2019
- Accepted March 5, 2019
- Published online June 3, 2019.
- Molly Szerlip, MDa,∗∗ (, )
- Alan Zajarias, MDb,∗,
- Sreekanth Vemalapalli, MDc,
- Matthew Brennan, MDc,
- Dadi Dai, PhDc,
- Hersh Maniar, MDb,
- Brian R. Lindman, MDd,
- Ralph Brindis, MDe,
- John D. Carroll, MDf,
- Mohanad Hamandi, MDa,
- Fred H. Edwards, MDg,
- Fred Grover, MDf,
- Sean O’Brien, PhDc,
- Eric Peterson, MDc,
- John S. Rumsfeld, MD, PhDf,
- Dave Shahian, MDh,
- E. Murat Tuzcu, MDi,
- David Holmes, MDj,
- Vinod H. Thourani, MDk and
- Michael Mack, MDa
- aBaylor Scott and White Health, Plano, Texas
- bWashington University School of Medicine, St. Louis, Missouri
- cDuke University School of Medicine, Durham, North Carolina
- dVanderbilt University School of Medicine, Nashville, Tennessee
- eUniversity of California School of Medicine, San Francisco, California
- fUniversity of Colorado School of Medicine, Denver, Colorado
- gUniversity of Florida School of Medicine, Jacksonville, Florida
- hHarvard Medical School, Boston, Massachusetts
- iCleveland Clinic Foundation, Cleveland, Ohio
- jMayo Clinic, Rochester, Minnesota
- kMedStar Heart and Vascular Institute and Georgetown University School of Medicine, Washington, DC
- ↵∗Address for correspondence:
Dr. Molly Szerlip, 4716 Alliance Boulevard, Pavilion 2, Suite 340, Plano, Texas 75093.
Background In patients with end-stage renal disease (ESRD), surgical aortic valve replacement is associated with higher early and late mortality, and adverse outcomes compared with patients without renal disease. Transcatheter aortic valve replacement (TAVR) offers another alternative, but there are limited reported outcomes.
Objectives The purpose of this study was to determine the outcomes of TAVR in patients with ESRD.
Methods Among the first 72,631 patients with severe aortic stenosis (AS) treated with TAVR enrolled in the Society of Thoracic Surgeons (STS)/American College of Cardiology (ACC) TVT (Transcatheter Valve Therapies) registry, 3,053 (4.2%) patients had ESRD and were compared with patients who were not on dialysis for demographics, risk factors, and outcomes.
Results Compared with the nondialysis patients, ESRD patients were younger (76 years vs. 83 years; p < 0.01) and had higher rates of comorbidities leading to a higher STS predicted risk of mortality (median 13.5% vs. 6.2%; p < 0.01). ESRD patients had a higher in-hospital mortality (5.1% vs. 3.4%; p < 0.01), although the observed to expected ratio was lower (0.32 vs. 0.44; p < 0.01). ESRD patients also had a similar rate of major vascular complications (4.5% vs. 4.6%; p = 0.86), but a higher rate of major bleeding (1.4% vs. 1.0%; p = 0.03). The 1-year mortality was significantly higher in dialysis patients (36.8% vs. 18.7%; p < 0.01).
Conclusions Patients undergoing TAVR with ESRD are at higher risk and had higher in-hospital mortality and bleeding, but similar vascular complications, when compared with those who are not dialysis dependent. The 1-year survival raises concerns regarding diminished benefit in this population. TAVR should be used judiciously after full discussion of the risk-benefit ratio in patients on dialysis.
Aortic stenosis (AS) is the most common valvular lesion encountered in patients with end-stage renal disease (ESRD) (1). The coexistence of cardiovascular disease and other comorbidities in patients with ESRD elevates the risk of 30-day mortality after surgical aortic valve replacement (SAVR) to as high as 17.3% (2,3). In addition, the 1-year mortality in elderly ESRD patients after isolated SAVR for symptomatic AS has been documented between 34% and 53%, raising concerns about the utility of this treatment option (4). With such high mortality, alternative treatment options that may offer improved outcome, including a survival advantage, should be examined.
Transcatheter aortic valve replacement (TAVR) provides symptomatic improvement and a survival advantage for patients with severe symptomatic AS who cannot undergo SAVR or are at high risk for perioperative complications and mortality (5–8). Up to 50% of patients undergoing TAVR have pre-existing renal impairment, and worsening renal function has been associated with increasing mortality (9). Due to elevated procedural and 30-day mortality with SAVR, patients with ESRD are a population who may potentially benefit from TAVR, although the outcome of this procedure is unclear as ESRD patients were excluded from enrollment in pivotal clinical trials. There are a few series reporting acceptable results, but these series are very small and most larger studies show worse outcomes in TAVR patients with ESRD and chronic kidney disease (9,10). The aim of this study was to assess the effect of ESRD on early and 1-year outcomes of TAVR in patients enrolled in the Society of Thoracic Surgeons (STS)/American College of Cardiology (ACC) TVT (Transcatheter Valve Therapies) Registry.
The TVT Registry was established as a data repository to track patient safety and real-world outcomes related to TAVR procedures performed in the United States. Participating centers collect patient demographics; comorbidities; functional status; patient-reported quality of life; hemodynamics; procedural details; and post-operative, 30-day, and 1-year outcomes, as described previously (11). A total of 72,631 patients undergoing TAVR from the participating 457 sites entered into the TVT Registry from November 2011 to June 2016 were included in our study for procedural and in-hospital outcomes. During this time, a total of 59,655 TVT Registry clinical records of Medicare patients age >65 years were linked to Medicare administrative claims using direct patient identifiers (name and social security numbers) by the U.S. Centers for Medicare & Medicaid Services. Claims filed through June 2016 were included in the analysis set, allowing a minimum of 10 weeks for claim processing for all patients in the cohort. Of the Medicare-linked TVT records, 15,429 records were excluded from this study cohort due to patient nonparticipation in the Medicare Parts A and B Fee for Service program at the time of the index procedure or an inability to link the index admission to a Medicare inpatient claim. After keeping only the first TVT procedure for each patient, 44,226 TVT records were included in the study cohort. These patients had similar baseline characteristics and in-hospital mortality and stroke as the excluded patients (Online Table 1).
Data elements definitions
Demographic data, pre-operative risk factors, and postoperative outcomes were identified and collected using standardized definitions that have been harmonized with the STS National Database wherever possible (12). The clinical indications for TAVR was based on the clinical judgment of 2 independent cardiac surgeons and risk calculation of procedural operative mortality (PROM) by the STS-PROM calculator, as described previously (10). Patients were considered inoperable if their risk of combined mortality or irreversible morbidity was >50% or there was an anatomical reason that would preclude chest procedures or cross-clamping the aorta, such as: 1) heavily calcified ascending aorta (porcelain aorta); 2) congenital or acquired chest wall deformity; 3) extensive mediastinal radiation; and 4) absence of sternal reconstructive options based on plastic surgery consultation or other anatomical reason to consider repeat sternotomy prohibitively hazardous. Procedural route was defined as transfemoral (TF) or alternative access (AA), which included transapical, transaortic, subclavian, or other routes. Gait speed was defined as slow if the 5 m walk time was >6 s (13). The Kansas City Cardiomyopathy Questionnaire was utilized to evaluate baseline quality of life (14).
Patients were initially stratified into 2 groups depending on the presence or absence of pre-procedural dialysis. In hospital, 30-day outcomes were reported to the TVT registry using standardized definitions including harmonization with the Valve Academic Research Consortium (VARC) and VARC-2 definitions for stroke, transient ischemic attacks, aortic valve interventions, major bleeding, and major vascular complications (15,16). All site-reported stroke, transient ischemic attack, and valve reintervention were adjudicated by a board-certified cardiologist using a combination of site-reported clinical information and targeted chart reviews. Device implant success was defined as successful vascular access, delivery, and deployment of a single device in the appropriate anatomic location, appropriate performance of the bioprosthetic valve (final aortic valve area >1.2 cm2 and a mean gradient <20 mm Hg or a peak velocity <3 m/s, without moderate or severe valve regurgitation), and successful retrieval of the delivery system. Aborted procedures were defined as those that were canceled or terminated after the patient entered the procedure room.
Baseline patient characteristics were summarized by percentages or the median (interquartile range [IQR] [25th to 75th percentile]) as appropriate and compared across subgroups using chi-square and Wilcoxon tests. For analyzing mortality, the Kaplan-Meier method was used for unadjusted 1-year event rates overall and within subgroups based on the following patient pre-procedural covariates: age, sex, renal function, ejection fraction, access site, and the STS-PROM. Differences in mortality risk across covariate subgroups were assessed using the Cox proportional hazards models. For unadjusted analyses, we fit a separate univariable Cox model for each covariate and included a set of binary subgroup indicators for modeling subgroups. For adjusted analyses, all subgroup indicator variables were included as covariates in a single multivariable Cox model. The variables included in this analysis were: age, sex, Chronic obstructive pulmonary disease, renal function, left ventricular ejection fraction, access site, prior SAVR, valve morphology, STS-PROM, and pre-operative atrial fibrillation or flutter. For these analyses, missing data were imputed to the most common category. Unadjusted and adjusted hazard ratios comparing mortality risks across subgroups were estimated along with 95% approximate confidence intervals and Wald-type p values.
The stroke analysis focused on estimating the cumulative incidence function (CIF); the cumulative probability of an endpoint occurring over time in a patient’s lifetime. The CIF is the appropriate parameter for describing nonfatal events from the patient’s perspective in a setting of high competing mortality risk (17). Unlike standard time-to-event methods, which describe the probability of a nonfatal endpoint occurring in a hypothetical death-free environment, the CIF models the probability that an endpoint will actually occur, given that death may preclude an event from occurring. All analyses were performed using R version 2.15.1 (R Foundation, Vienna, Austria) and SAS version 9.3 (SAS Institute, Cary, North Carolina).
Demographic and pre-procedural characteristics
A total of 72,631 patients were enrolled in the TVT registry from November 2011 to June 2016, 69,578 were nondialysis (95.7%) and 3,053 were ESRD patients (4.3%). Demographic and baseline characteristics are presented in Table 1. Patients with ESRD were younger (78 years vs. 84 years; p < 0.01), predominantly male (58.3% vs. 48.5%; p < 0.01), and more commonly African American (11.4% vs. 3.3%; p < 0.01). The presence of peripheral arterial disease, hypertension, diabetes mellitus, previous myocardial infarction, and worsened heart failure symptoms were significantly more prevalent in patients with ESRD and contributed to their higher STS-PROM (14.4% vs. 6.8%; p < 0.01). There was no difference in the prevalence of prior coronary revascularization, stroke, or atrial fibrillation between groups.
Concomitant moderate to severe mitral regurgitation (p < 0.01), tricuspid regurgitation (p < 0.01), and a lower ejection fraction (p < 0.01) were more common in the patients with ESRD (Table 1). There was no difference in the severity of the AS or the mean transvalvular gradients between groups.
Procedural and in-hospital outcomes
Procedures were elective in the majority of the patients; however, those in the ESRD group had a higher prevalence of urgent procedures (17.2% vs. 8.9%; p < 0.01). Reasons for inoperability varied between groups: debilitated state or deconditioning was more common in the nondialysis patient group, whereas numerous comorbid conditions were more commonly encountered in dialysis patients. A TF approach was utilized less frequently in the ESRD group (76.4% vs. 78.4%; p = 0.01) with similar device success (92.4% vs. 93.0%; p = 0.17).
Procedural complications were more commonly seen in the ESRD cohort leading to a longer hospitalization and intensive care unit stay (6 days vs. 5 days; p < 0.001). VARC major bleeding events (1.4% vs. 1.0%; p = 0.03) were more commonly seen in the ESRD group. There was no difference in vascular site complications, unplanned vascular procedures, and stroke rate between groups. Although in-hospital mortality was significantly higher in the ESRD group (5.1% vs. 3.4%; p < 0.001) the observed/expected mortality rate (O:E ratio) was lower in the ESRD group (0.32 vs. 0.44; p < 0.01) (Table 2). Online Table 2 represents the hazard ratios of the different characteristics of those dialysis patients versus nondialysis patients who were alive at 1-year post-TAVR.
Among the ESRD group, patients were stratified according to their STS-PROM: STS <8%, 8% to 15%, and >15%. All groups had a favorable O:E ratio; however, there was an incremental increase in the in-hospital mortality rate among the groups: 2.5% vs. 4.4% vs. 7.0%, respectively. Of the ESRD patients that were alive at discharge, the discharge locations depend on the STS-PROM: the higher the STS-PROM, the fewer were discharged to home and more to extended care/transitional care unit/Rehab (Table 3).
TF versus AA
Of the patients in the ESRD group, 2,332 and 721 patients underwent TAVR via TF and AA, respectively. AA was performed in patients with a higher STS-PROM score (15.6% vs 13.0%; p < 0.01). There was a statistically significant difference in the observed in-hospital mortality for TF (vs. AA) patients (4.1% vs. 8.5%; p < 0.001), with a smaller O:E ratio (0.26 vs. 0.46; p < 0.001). Major VARC bleeding were similar in 2 groups, and vascular access complications were less common in the AA approach when compared with the TF approach, but TF patients had a shorter hospitalization (Table 4).
The records of 44,226 Medicare patients were available for longitudinal follow-up to assess readmissions and 1-year mortality. Of these patients, 1,733 were on dialysis, 2,634 were nondialysis with serum creatinine (Cr) ≥2 mg/dl, and 39,782 were nondialysis with serum Cr <2 mg/dl. For the characteristics of these patients, see Online Table 3. The presence of renal dysfunction had incremental risk for 30-day, 6-month, and 1-year mortality (Central Illustration). When patients were stratified by Cr <2 mg/dl, ≥2 mg/dl, or requiring dialysis, there was an incremental mortality risk observed to 1 year: patients on dialysis had a 36.8% 1-year mortality rate compared with 30.6% in those whose creatinine was ≥2 mg/dl and 17.9% in those whose creatinine was <2 mg/dl (p < 0.01). In patients who survived TAVR, those who were not dialysis-dependent had a significantly larger number of days alive and out of the hospital when compared with the ESRD group (173 days vs. 167 days; p < 0.01). Worse renal function was not associated with an increased risk of stroke to 1 year (Figure 1). Among a list of common comorbidities, dialysis use was one of the most strongly associated with 1-year mortality (adjusted hazard ratio: 1.28; 95% confidence interval: 1.17 to 1.41; p < 0.001).
TAVR is an effective treatment option for high-risk patients with severe AS. This patient cohort represents the largest published group of high-risk/inoperable patients with severe AS and ESRD who were treated with TAVR. The major findings of our study are:
1. Patients with ESRD are younger and are more commonly diabetic and hypertensive;
2. Despite younger age, these patients are at significantly higher risk based on STS-PROM due to comorbid conditions;
3. Vascular complications and major bleeding were more commonly seen in their procedures; and
4. In-hospital and 1-year mortality after TAVR were significantly higher than comparable high-risk nondialysis patients; but
5. There was no increased risk of stroke in patients with ESRD.
Collectively, these findings demonstrate the important effect of severe renal disease in this patient population.
Chronic kidney disease and TAVR
Chronic kidney disease (CKD) in the TAVR population is quite prevalent and there is a progressive risk of mortality as renal function worsens. Yamamoto et al. (18) noted that patients with CKD stages 3b and 4 were associated with a 1.7 risk of death at 1 year after TAVR when compared with patients with CKD stage 1 to 2. Only patients with stage 4 CKD had a higher 30-day mortality (9). Our study further advances this association, as there was increased mortality risk in patients with serum creatinine >2 mg/dl, and ESRD patients had a 21% increased risk of mortality, despite adjustment for other comorbidities. During 1-year follow-up, it is apparent that patients on dialysis who survive the procedure have a higher rate of rehospitalization, limiting their overall improvement in quality of life.
A recently published series (19) showed no difference in 30-day mortality in patients with ESRD, contrasting the findings in our study. It is likely that this study was not powered to show an association between ESRD and mortality due to their small sample size. An international registry of 2,075 patients undergoing TAVR that included 56 individuals on dialysis found a similar 50% increased risk of 1-year mortality after TAVR with ESRD (10). The overall mortality in our population and the one presented by Allende et al. (10) note a comparable survival similar to those seen in the inoperable control arm of the PARTNER (Placement of Aortic Transcatheter Valves) trial (4). When compared with previously published surgical data (3) it appears that patients on dialysis are at high risk of developing complications during all aortic valve-related procedures.
Dialysis has been proven to be a life-saving therapy for patients with ESRD. The survival of patients on dialysis with AS who are not treated with surgery is unknown. The prognosis of patients with ESRD is strongly influenced by the number of comorbidities and their functional status (20). The presence of increasing comorbid conditions (i.e., peripheral vascular disease, coronary artery disease, congestive heart failure, diabetes, chronic obstructive pulmonary disease) is associated with approximately 50% 1-year survival in older adults on dialysis (21). Having severe mitral regurgitation or severe tricuspid regurgitation prior to TAVR has been shown in some studies to be an independent predictor of mortality (22). In this study, the patients with ESRD were more likely to have severe mitral regurgitation or severe tricuspid regurgitation. The majority of the patients in our cohort with ESRD have multiple comorbidities in addition to experiencing the effects of severe AS, and as a result, it is not surprising that their survival is limited during midterm follow-up.
Although there is a subset of patients who survive more than 1 year, careful selection of the ESRD patient is required to focus the therapy on patients who will survive long enough to receive benefit. Concomitant atrial fibrillation, tricuspid regurgitation, the duration of dialysis, type of dialysis, degree of comorbidities, presence of frailty, and degree of functional impairment (assessed by the Kansas City Cardiomyopathy Questionnaire), are some factors that may influence the heart team’s decision in treating these patients. Careful adjudication of the technology should be done to ensure that TAVR is offered to patients who will benefit the most from it.
AA in dialysis patients
Our results demonstrate a high incidence of vascular complications and major bleeding events in ESRD patients undergoing TF TAVR. These complications are likely due to the prevalence of vascular calcification and platelet dysfunction seen in patients on dialysis, and the use of a commercially available first-generation TAVR system. Although present, these complications did not translate into prolonged hospitalization or worsening mortality. ESRD patients who were treated with AA did not see an increase risk of access site bleeding, but were associated with a numerically higher operative mortality, albeit with similar O:E mortality ratio. Although all TAVR access routes are complimentary, this population remains at high risk of complications, emphasizing the need for careful procedural planning.
Although younger, patients with ESRD had a higher prevalence of diabetes, hypertension, and peripheral vascular disease, which increase their operative risk. Patients undergoing TAVR had a lower than expected procedural mortality calculated by the STS-PROM irrespective of their renal function. However, the observed mortality was in-hospital rather than 30-day, which is the expected mortality, likely leading to a lower than expected O:E ratio. Although the STS-PROM has been validated in high-risk patients undergoing surgical AVR, it was not specifically intended as a TAVR risk calculator. We now have an STS/ACC in-hospital TAVR mortality score, which has been validated in the U.S. and European TAVR population. This score has performed as well as the STS-PROM in determining in-hospital as well as 30-day mortality TAVR risk. It also performs better than EuroSCORE I and II and the German AV Score (23). The STS/ACC in-hospital mortality score still has its shortcomings, and a true 30-day mortality TAVR score is currently being developed based on the STS/ACC TVT Registry (24,25).
Renal disease and stroke
In the high-risk population with severe AS, the incidence of a cerebrovascular event after TAVR varies from 1.5% to 6% (26). Factors associated with stroke risk after TAVR include previous history of cerebrovascular event, non-TF access, and higher functional disability (27). Although patients with ESRD have a higher prevalence of peripheral vascular disease (as a marker of atherosclerotic burden) and disability, we did not find an association between worsening renal function and increased risk of stroke after TAVR. Although unexpected, the lack of association may be explained by the younger age in our cohort or the absence of formalized post-procedural neurocognitive assessment.
As TAVR continues to be adopted for the treatment of severe AS, patient selection remains critical to maintain optimal outcomes. Patients with ESRD who have severe AS still remain a high-risk group for TAVR. Even though patients with ESRD survive the procedure and initial hospitalization after TAVR, they remain at high risk for early mortality and rehospitalization. Future efforts should concentrate on identifying the factors associated with survival after TAVR in patients with ESRD to improve patient selection.
If ESRD patients are treated with TAVR, realistic expectations in survival and rehospitalization should be identified prior to the procedure. Although procedural outcomes appear acceptable, a 1-year survival in only one-half of the treated patients raises concerns regarding the benefit of treatment in this patient population. Further analysis is necessary to delineate the patients with ESRD who are likely to have favorable procedural and functional outcomes. In the meantime, careful consideration should be made before any decision to treat patients with ESRD with very high STS-PROM requiring an AA approach.
Our findings only apply to high-risk and inoperable severe aortic stenosis patients with ESRD. Treatment options for patients with severe AS and lower surgical risk need to be assessed in clinical trials. The data included in the TVT registry is self-reported, voluntary, and not measured by a central core laboratory. Early mortality is reported as in-hospital rather than the more standard 30-day mortality due to incomplete 30-day outcomes data. This may have led to an artificially low O:E ratio, because expected mortality is at 30 days. The fields assessed in the registry did not include the duration of dialysis, the type of dialysis (peritoneal vs. hemodialysis), and the etiology of the renal disease, which may have influenced the results.
In high-risk and inoperable patients with severe AS undergoing TAVR, ESRD was associated with a significantly increased risk of mortality at 30 days and 1 year, despite adjustment for other confounding factors. These findings have important potential implications in patient selection, and therefore, careful patient selection to optimize use of this procedural resource is necessary.
COMPETENCY IN MEDICAL KNOWLEDGE: Patients with ESRD and severe AS undergoing TAVR in the STS/ACC TVT Registry had a high mortality rate at 30 days and 1 year.
TRANSLATIONAL OUTLOOK: Randomized trials are needed to compare various management strategies for patients with ESRD and severe AS cognizant of their heterogeneous surgical risks.
↵∗ Drs. Szerlip and Zajarias contributed equally to this work and are co-first authors.
Dr. Szerlip has served as a speaker for Edwards Lifesciences and Medtronic. Dr. Zajarias has served as a consultant to Edwards, Abott, and Boston Scientific; and has received research support from Medtronic, Edwards, Boston Scientific, and Abbott. Dr. Vemulapalli has received grants from the American College of Cardiology (ACC), the Society of Thoracic Surgeons, Abbott Vascular, PCORI, and Boston Scientific; has served as a consultant and/or on the Advisory Board for Boston Scientific, Novella, and Janssen; and has served as a consultant for Zafgen and Premiere. Dr. Brennan has served as a consultant for Edwards Lifesciences. Dr. Lindman has received investigator-initiated research grants from Edwards Lifesciences and Roche Diagnostics; has received a research grant from Medtronic; and has served as a consultant for Roche Diagnostics. Dr. Carroll has served as an investigator for the Medtronic Low Risk TAVR Trial and Edwards PARTNER 2 Trial. Dr. Grover has served as a consultant for JenaValve. Dr. Shahian has served on ACC/STS TVT-related working groups (uncompensated). Dr. Peterson has served as co-investigator for the TAVR Registry analytic center at DCRI. Dr. Rumsfeld has served as Chief Innovative Officer for the American College of Cardiology. Dr. Brindis has served as Senior Medical Officer for the ACC-NCDR. Dr. Thourani has served as an advisor to Edwards Lifesciences, Boston Scientific, Abbott Vascular, Gore Vascular, and JenaValve. Dr. Mack has served as co-principal investigator for the Edwards Lifesciences PARTNER trial and Abbott COAPT Trial; and has served as study chair for the Medtronic APOLLO Trial. All other authors have reported that they have no relationships relevant to the contents of this paper to disclose.
Listen to this manuscript's audio summary by Editor-in-Chief Dr. Valentin Fuster on JACC.org.
- Abbreviations and Acronyms
- alternative access
- chronic kidney disease
- end-stage renal disease
- surgical aortic valve replacement
- Society of Thoracic Surgeons-Predicted Risk of Mortality
- transcatheter aortic valve replacement
- valve academic research consortium
- Received November 14, 2018.
- Revision received February 12, 2019.
- Accepted March 5, 2019.
- 2019 American College of Cardiology Foundation
- Straumann E.,
- Meyer B.,
- Blumberg A.,
- Jenzer H.R.
- Brennan J.M.,
- Edwards F.H.,
- Zhao Y.,
- et al.
- Popma J.J.,
- Adams D.H.,
- Reardon M.J.,
- et al.
- Rodes-Cabeu J.,
- Webb J.G.,
- Cheung A.,
- et al.
- American College of Cardiology; Society of Thoracic Surgeons
- Leon M.B.,
- Piazza N.,
- Nikolsky E.,
- et al.
- Kappetein A.P.,
- Head S.J.,
- Genereux P.,
- et al.
- Yamamoto M.,
- Hayashida K.,
- Mouillet G.,
- et al.
- Nombela-Franco L.,
- Eltchaninoff H.,
- Zahn R.,
- et al.
- Edwards F.H.,
- Cohen D.J.,
- O’Brien S.M.,
- et al.,
- for the Steering Committee of the Society of Thoracic Surgeons/American College of Cardiology Transcatheter Valve Therapy Registry
- Arsalan M.,
- Weferling M.,
- Hecker F.,
- et al.
- Deneault B.,
- Kirtane A.,
- Kodali S.,
- et al.