Author + information
- Received December 27, 2016
- Revision received April 18, 2017
- Accepted May 4, 2017
- Published online August 7, 2017.
- Anja Stundl, MDa,
- Regina Schulte, MDa,
- Hannah Lucht, MDa,
- Marcel Weber, MDa,
- Alexander Sedaghat, MDa,
- Jasmin Shamekhi, MDa,
- Berndt Zur, MDb,
- Eberhard Grube, MDa,
- Fritz Mellert, MDc,
- Armin Welz, MDc,
- Rolf Fimmers, MDd,
- Georg Nickenig, MDa,
- Nikos Werner, MDa and
- Jan-Malte Sinning, MDa,∗ ()
- aDepartment of Medicine II, Heart Center Bonn, Bonn, Germany
- bInstitute of Clinical Chemistry and Clinical Pharmacology, Bonn, Germany
- cDepartment of Cardiothoracic Surgery, Heart Center Bonn, Bonn, Germany
- dInstitute of Medical Biometry, Informatics and Epidemiology, University Hospital Bonn, Bonn, Germany
- ↵∗Address for correspondence:
Dr. Jan-Malte Sinning, Heart Center Bonn, Department of Medicine II, University Hospital Bonn, Sigmund-Freud-Strasse 25, 53105 Bonn, Germany.
Objectives The aims of this study were to determine plasma elevations of biomarkers of myocardial injury associated with transfemoral (TF) transcatheter aortic valve replacement (TAVR) and to evaluate their prognostic value.
Background Increases in biomarkers of myocardial injury are a common finding after TAVR, but their clinical significance is unclear.
Methods In 756 consecutive TF TAVR patients, cardiac high-sensitivity troponin I (hsTnI) and creatine kinase MB (CK-MB) levels were measured at pre-defined time points to assess the occurrence of myocardial injury (defined as 15 times the upper reference limit for hsTnI [≥1.5 ng/ml] or 5 times the upper reference limit for CK-MB [≥18 μg/l]) during the first 72 h. The primary endpoint was all-cause mortality at 1 year.
Results After uneventful TF TAVR, hsTnI was elevated in 51.6% and CK-MB in 7.4% of patients, respectively. Myocardial injury was associated with transcatheter heart valve (THV) type: patients who received the LOTUS THV more frequently had myocardial injury compared with those who received other THVs (LOTUS, 81.6%; Direct Flow Medical, 56.4%; CoreValve, 51.2%; Evolut R, 42.7%; SAPIEN XT, 40.4%; SAPIEN 3, 36.6%; p < 0.001). Myocardial injury defined by hsTnI was not associated with adverse outcomes at 30 days (3.1% vs. 2.7%; p = 0.778) or 1 year (16.7% vs. 17.2%; p = 0.841). Likewise, a CK-MB increase was not associated with 30-day mortality (5.5% vs. 2.8%; p = 0.258) or 1-year mortality (16.4% vs. 17.3%; p = 0.856).
Conclusions Myocardial injury is common following TF TAVR. The extent of cardiac biomarker elevation depends on THV type but is not associated with adverse short- and long-term outcomes after uneventful TAVR.
Transcatheter aortic valve replacement (TAVR) has emerged as a therapeutic alternative to surgical aortic valve replacement for elderly patients with intermediate, high, and prohibitive surgical risk and has been shown to lead to substantial reductions in mortality and morbidity (1–3). Although considerable progress in research and technique has been made during the past decade in this field, continued efforts are required to further improve procedural safety and to minimize periprocedural complications. In line with this fact, debate is ongoing about the definition of periprocedural myocardial injury and the impact on outcomes. Periprocedural myocardial infarction is a very rare but life-threatening complication caused predominantly by coronary ostia occlusion after TAVR or the rare occurrence of coronary embolism, whereas a mild degree of myocardial injury frequently occurs after the procedure and might be a result of mechanical trauma due to myocardial tissue compression caused by balloon valvuloplasty and the deployment of the transcatheter heart valve (THV) itself, hypotension during rapid pacing, and especially by the anchoring of the THV (4–10). Although a distinct association with poor outcome and prognosis of myocardial injury or myocardial infarction occurring during cardiac surgery and percutaneous coronary intervention has been shown so far (11,12), the clinical significance of myocardial injury after TAVR in a transfemoral (TF), uneventful procedure is still not fully elucidated (5,13–17).
In the present study, we sought to: 1) determine the incidence, degree, and timing of increases in cardiac biomarkers of myocardial injury (cardiac high-sensitivity troponin I [hsTnI] and creatine kinase MB [CK-MB]) associated with TF TAVR; and 2) evaluate the 30-day and 1-year prognostic value of myocardial injury associated with TF TAVR.
From January 2010 to August 2016, 756 consecutive patients with severe, symptomatic aortic stenosis underwent TF TAVR at our institution using the third-generation CoreValve and Evolut R (Medtronic, Minneapolis, Minnesota), the SAPIEN XT and SAPIEN 3 (Edwards Lifesciences, Irvine, California), the Direct Flow Medical (Direct Flow Medical, Santa Rosa, California), or the LOTUS (Boston Scientific, Natick, Massachusetts) prosthesis and were included in this observational study. The study was approved by the local ethics committee of the University of Bonn, and all patients provided written informed consent.
Before TAVR, invasive coronary angiography was performed in all patients. After 3-dimensional transesophageal echocardiography, angiography of the aortic root, and multislice computed tomography for evaluation of aortic annular dimension, the decision for TAVR was made by the local heart team. In all patients, diagnostic coronary angiography was performed with percutaneous revascularization when appropriate (stenosis of left main coronary artery ≥50% or stenosis of proximal vessel ≥70%). Final choice of prosthesis was left to the discretion of the operator on the basis of computed tomographic sizing algorithm according to the manufacturer’s recommendations for access vessels and aortic annular size. All TAVR procedures were performed with biplane fluoroscopy under local anesthesia in combination with a sedative or analgesic treatment. Intraprocedural transesophageal echocardiography was not routinely performed, and the procedure was guided predominantly by angiographic control.
In the present analysis, 180 patients in total were excluded because of missing pre-procedural hsTnI (n = 5), transapical approach (n = 160), need for conversion to open heart surgery (because of pericardial tamponade [n = 3], annular rupture [n = 2], device embolization [n = 3], high-grade aortic regurgitation [n = 2], or access-site complication [n = 1]), intraprocedural death [n = 1], and the use of extracorporeal circulation (n = 3), which all lead to a certain extent of myocardial injury.
The primary endpoint of this study was all-cause mortality at 1 year. Secondary endpoints were assessed according to Valve Academic Research Consortium (VARC) 2 criteria. Follow-up data were collected during routine outpatient clinic visits, from hospital discharge letters, or via telephone interviews with the referring cardiologists or primary care physicians.
In 756 patients, hsTnI and CK-MB levels were measured at baseline (before the index procedure), immediately after the procedure, and at 4 h, 24 h, 72 h, and 7 days after TAVR using a cardiac hsTnI immunoassay (Dimension Vista CTNI Flex reagent cartridge, Siemens Healthcare Diagnostics, Munich, Germany) and an immunoassay for CK-MB (Dimension Vista Mass creatine kinase MB isoenzyme Flex reagent cartridge, Siemens Healthcare Diagnostics). According to the updated VARC-2 criteria, myocardial injury was defined as a peak hsTnI value exceeding 15 times the upper reference limit (URL) for hsTnI (≥1.5 ng/ml) (or 5 times the URL for CK-MB [≥18 μg/l]) during the first 72 h or as a further increase of at least 50% of cardiac biomarkers that were increased at baseline (>99th percentile) and in which the peak value had to exceed the previously stated limit (18). Based on the 99th percentile in a healthy population and the requirement of ≤10% coefficient variation, the URLs at our institution were 0.10 ng/ml for hsTnI levels and 3.6 μg/l for CK-MB levels.
Data are expressed as mean ± SD if normally distributed or as median and interquartile range (IQR) if not normally distributed.
For comparisons of continuous variables between 2 groups, the Student t test or Mann-Whitney U test was performed, depending on the distribution of the variable. Data resulting from repeated measurements over time were analyzed using linear mixed-effects models to compare the variation in time between subgroups of patients. In case of group-by-time interaction, the measurements at different time points were compared between groups separately. When comparing more than 2 groups, analysis of variance or the Kruskal-Wallis test was used. Categorical variables are expressed as frequencies and percentages. For categorical variables, the chi-square test was used for further analysis.
hsTnI levels were categorized according to the definition of myocardial injury of the updated VARC-2 criteria: <15 times the URL (<1.5 ng/ml) versus ≥15 times the URL (≥1.5 ng/ml). CK-MB levels were categorized as follows: <5 times the URL (<18 μg/l) versus ≥5 times the URL (≥18 μg/l).
The unadjusted cumulative event rates were estimated using Kaplan-Meier methods, and statistical assessment was performed using the log-rank test. To identify predictors of myocardial injury after TAVR, a binary logistic regression analysis was applied. After adjusting for patients’ baseline and procedural characteristics, results are reported as odds ratios with 95% confidence intervals (CIs). Statistical significance was assumed when the null hypothesis could be rejected at p < 0.05. Statistical analyses were conducted with SPSS Statistics Version 22.0 (IBM Corporation, Somers, New York), and MedCalc version 18.104.22.168 (MedCalc Software, Mariakerke, Belgium).
The investigators initiated the study, had full access to the data, and wrote the manuscript. All authors vouch for the accuracy and completeness of the data and all analyses and confirm that the study was conducted according to the protocol.
Baseline characteristics according to the occurrence of myocardial injury (defined as ≥15 times the URL [≥1.5 ng/ml] of hsTnI) are summarized in Table 1. A total of 756 patients with a median logistic EuroSCORE (European System for Cardiac Operative Risk Evaluation) of 17.0% (IQR: 10.7% to 28.6%) underwent TF TAVR. Patients with myocardial injury according to VARC-2 criteria (n = 390 [51.6%]) were older (81.4 ± 6.1 years vs. 80.3 ± 6.1 years; p = 0.013), less often male (45.3% vs. 54.7%; p < 0.001), and had a lower logistic EuroSCORE (16.0% [IQR: 10.1% to 26.0] vs. 17.8% [IQR: 11.3% to 31.1%]; p = 0.008).
On the basis of echocardiography, patients with the occurrence of myocardial injury had significantly higher left ventricular ejection fractions (LVEFs) (57.1 ± 11.2% vs. 49.0 ± 14.9%; p < 0.001) and higher mean pressure gradients (43.2 ± 16.0 mm Hg vs. 40.2 ± 15.6 mm Hg; p = 0.011). Furthermore, patients with periprocedural myocardial injury had significantly lower baseline levels of N-terminal pro–brain natriuretic peptide (2,060.0 pg/ml [IQR: 807.3 to 5,480.0 pg/ml] vs. 3,581.5 pg/ml [IQR: 1,587.5 to 9,448.0 pg/ml]; p < 0.001).
When using CK-MB for the definition of myocardial injury, patients with the occurrence of myocardial injury had a lower logistic EuroSCORE (13.5% [IQR: 8.7% to 22.6%] vs. 17.4% [IQR: 11.0% to 29.0%]; p = 0.033), higher LVEFs (57.3 ± 12.0% vs. 52.8 ± 13.9%; p = 0.022), and lower N-terminal pro–brain natriuretic peptide levels (1,237.0 pg/ml [IQR: 673.0 to 3,950.0 pg/ml] vs. 2,950.0 pg/ml [IQR: 1,159.0 to 7,594.0 pg/ml]; p = 0.001) (Online Table 1).
Myocardial injury following TAVR
On the basis of serial measurements of hsTnI (1, 4, 24, and 72 h and 7 days after TAVR), myocardial injury (Δ hsTnI ≥15 times the URL) occurred in 390 of 756 patients (51.6%) (Figure 1A). When using CK-MB for the definition of myocardial injury (Δ CK-MB ≥5 times the URL), 55 of 742 patients (7.4%) had periprocedural myocardial injury (Figure 1B). Following TF TAVR, both hsTnI and CK-MB levels showed marked increases, with a maximum peak at 4 h post-TAVR (hsTnI: 2.36 ng/ml [IQR: 1.70 to 3.39 ng/ml] vs. 0.80 ng/ml [IQR: 0.48 to 1.14 ng/ml]; p < 0.001; CK-MB: 22.90 μg/l [IQR: 19.15 to 28.58 μg/l) vs. 6.50 μg/l [IQR: 4.10 to 9.50 μg/l]; p < 0.001) (Figure 2, Online Figure 1).
Periprocedural characteristics according to the VARC-2 definition of myocardial injury for hsTnI are shown in Table 2. Most patients (97.1%) underwent TAVR via the TF approach. Patients who received either the Direct Flow Medical or LOTUS THV more often had myocardial injury (Figures 3A and 4) (Direct Flow Medical, 56.4% vs. 43.6%; LOTUS: 81.6% vs. 18.4%) compared with those who received other THVs (CoreValve, 51.2% vs. 48.8%; SAPIEN XT, 40.4% vs. 59.6%; Evolut R, 42.7% vs. 57.3%; SAPIEN 3, 36.6% vs. 63.4%; p < 0.001) and showed a significant increase in baseline hsTnI (a 57-fold increase for the LOTUS THV and a 33-fold increase for the Direct Flow Medical THV) (Figure 3B). When divided according to the median, the first half of patients receiving the LOTUS THV (n = 51) more frequently experienced myocardial injury than the second half (n = 52) (90.2% vs. 73.1%; p = 0.025) (Online Figure 2). This might be at least in part explained by the fact that THV deployment took significantly longer (16.0 min [IQR: 12.0 to 25.0 min] vs. 12.0 min [IQR: 9.0 to 19.3 min]; p = 0.001) and that the THV had to be resheathed more often for accurate positioning (27.5% vs. 3.8%; p = 0.001). For all other THVs, this association was not found.
Furthermore, in patients who underwent TAVR with the LOTUS valve system (n = 103) compared with those who received 1 of the other THVs, cardiac hsTnI elevation 4 h after the procedure was significantly associated with the need for new pacemaker implantation because of new-onset conduction disturbances (2.47 ng/ml [IQR: 1.69 to 3.76 ng/ml] vs. 1.25 ng/ml [IQR: 0.71 to 2.13 ng/ml]; p < 0.001). Patients undergoing TAVR with the SAPIEN 3 valve had the lowest increase of cardiac hsTnI. Furthermore, annular diameter (23.5 ± 2.3 mm vs. 24.3 ± 2.5 mm; p < 0.001) but not the degree of oversizing reflected by the cover index area was related to the occurrence of myocardial injury. In addition, procedure time was significantly longer (70.0 min [IQR: 54.0 to 90.0 min] vs. 59.0 min [IQR: 47.0 to 80.0 min]; p < 0.001) in patients who developed myocardial injury.
When using the VARC-2 definition of myocardial injury for CK-MB, the analyses also showed significant differences regarding valve type (Δ CK-MB ≥5 times the URL vs. Δ CK-MB <5 times the URL: CoreValve, 7.9% vs. 92.1%; SAPIEN XT, 7.7% vs. 92.3%; Direct Flow Medical, 10.3% vs. 89.7%; Evolut R, 2.6% vs. 97.4%; LOTUS, 15.0% vs. 85.0%; SAPIEN 3, 2.2% vs. 97.8%) (Online Figure 3A), procedure time (80.5 min [IQR: 60.0 to 95.0 min] vs. 63.0 min [IQR: 50.0 to 87.0 min]; p = 0.001), and rate of pre-dilation (5.2% vs. 94.8%; p = 0.027). Patients with balloon aortic valvuloplasty for pre-dilation, which was performed according to the operator’s preference, less frequently developed CK-MB defined myocardial injury (Online Table 2). TAVR procedures with the Direct Flow Medical and the LOTUS prostheses also led to a significant increase in baseline CK-MB (2-fold increase for the Direct Flow Medical THV and 3-fold increase for the LOTUS THV) and consecutively to a higher occurrence of myocardial injury, whereas patients undergoing TAVR with the SAPIEN 3 THV had the lowest increases in CK-MB (Online Figures 3B and 4).
Clinical outcomes according to the occurrence of troponin-defined myocardial injury are summarized in Table 3. The analyses revealed that the occurrence of myocardial injury was not related to 30-day mortality (Δ hsTnI ≥15 times the URL vs. Δ hsTnI <15 times the URL: 3.1% vs. 2.7%; p = 0.778), 1-year mortality (16.7% vs. 17.2%; p = 0.841), 3-year mortality (25.9% vs. 25.4%; p = 0.878), or 5-year mortality (30.3% vs. 30.1%; p = 0.952) (Figures 5A to 5D). The same could be found using the VARC-2 definition of myocardial injury by CK-MB (Online Table 3, Online Figures 5A to 5D). Furthermore, post-procedural stroke (3.3% vs. 1.1%; p = 0.038), major bleeding (4.6% vs. 1.6%; p = 0.020), and acute kidney injury (18.7% vs. 11.2%; p = 0.004) occurred more frequently in patients with troponin-defined myocardial injury than in patients without. When using CK-MB, the occurrence of myocardial injury was associated with myocardial infarction (1.8% vs. 0.0%; p < 0.001) in the patient group with myocardial injury compared with the group without (Online Table 3).
Predictors of myocardial injury
After adjusting for patients’ baseline and procedural characteristics, univariate regression analysis revealed that age, male sex, logistic EuroSCORE, EuroSCORE II, atrial fibrillation, previous myocardial infarction, pulmonary hypertension, LVEF, pressure mean gradient, end-diastolic volume, end-systolic volume, valve type, annular diameter, and procedure time were associated with a higher risk for the occurrence of hsTnI-based myocardial injury (Table 4). After multivariate analysis, age (hazard ratio [HR]: 1.039; 95% CI: 1.005 to 1.074; p = 0.023), LVEF (HR: 1.816; 95% CI: 1.439 to 2.293; p < 0.001), end-diastolic volume (HR: 0.713; 95% CI: 0.569 to 0.893; p = 0.003), valve type compared with the SAPIEN 3 THV (Evolut R: HR: 1.438; 95% CI: 0.697 to 2.967; SAPIEN XT: HR: 1.333; 95% CI: 0.439 to 4.042; CoreValve: HR: 2.696; 95% CI: 1.440 to 5.048; Direct Flow Medical: HR: 2.667; 95% CI: 1.008 to 7.051; LOTUS: HR: 9.656; 95% CI: 4.358 to 21.393; p < 0.001), and procedure time (HR: 1.891; 95% CI: 1.464 to 2.443; p < 0.001) were the only independent variables predicting myocardial injury.
For CK-MB, multivariate regression analysis revealed that LVEF, estimated glomerular filtration rate, valve type compared with the SAPIEN 3 THV, pre-dilation, and procedure time were the only independent variables predicting myocardial injury according to the VARC-2 definition for CK-MB (Online Table 4).
In our study, we were able to show that myocardial injury according to the VARC-2 definition (Δ hsTnI ≥15 times the URL) occurred in more than one-half of patients (51.6%) undergoing TF TAVR during the first 72 h with a peak cardiac hsTnI level at 4 h post-TAVR, whereas myocardial injury using the peak CK-MB level (Δ CK-MB ≥5 times the URL) occurred in <8% of patients. The extent of cardiac biomarker elevation was associated with prosthesis type and showed a learning-curve effect for the LOTUS valve. However, the occurrence of myocardial injury in transvascular patients with uneventful post-procedural courses was not associated with adverse short- and long-term outcomes.
A certain degree of myocardial injury defined by cardiac biomarker elevation is a common finding after TAVR. The high incidence has been proved in numerous previous studies (5,6,13–18), depending on the definition of myocardial injury. Until now, data indicated that myocardial injury has several predominantly procedure-related origins, including factors such as mechanical trauma due to myocardial compromise caused by the balloon and THV itself, because of several short episodes of extreme hypotension, regional or global myocardial ischemia due to balloon valvuloplasty and/or valve implantation (4–10). In our previous study (15), we hypothesized that deployment and anchoring of the THV itself might be associated with the degree of myocardial injury and could be a relevant source for cardiac biomarker release. In this study, we were able to demonstrate a relationship between prosthesis type and the extent of myocardial biomarker increase and found that patients treated with the LOTUS valve showed the highest increases in cardiac biomarkers, although without any impact on outcomes. Part of this device-specific increase is explained by device-related attributes: the LOTUS THV is mechanically expanded in the aortic annulus and has an additional adaptive seal around the outer aspect of the lower valve frame that is useful to further minimize paravalvular leakage (19). The resulting myocardial tissue trauma caused by the anchoring of the prosthesis in the native aortic valve annulus could presumably contribute to significant increases in cardiac biomarkers reflecting greater myocardial tissue compression and mechanical trauma. Conversely, patients undergoing TAVR with use of the balloon-expandable SAPIEN 3 prosthesis had the lowest increases in cardiac biomarkers. The SAPIEN 3 THV is a balloon-expandable valve (19), and part of the increase might be explained by similar device-related properties (myocardial compromise caused by balloon dilation and the skirt at the outer part of the distal frame of the prosthesis). However, it could be assumed that balloon-expandable THVs are expected to require only a brief high-pressure application during the implantation procedure, whereas self-expanding THVs appear to need a continuous application of pressure, which in turn might result in substantially greater myocardial damage. Interestingly, we found a certain learning-curve effect for the mechanically expanded LOTUS THV, as deployment time was longer and resheathing had to be applied more frequently for accurate positioning. With growing experience, both deployment time and resheathing attempts could be decreased and myocardial injury according to the VARC-2 definition occurred less frequently with this valve type but still more often than with other THV types. Although the pathophysiologic explanation for this might be the anchoring of the valve frame in the aortic annulus, the reasoning behind this remains elusive and calls for further investigation.
In line with the aforementioned, we could further demonstrate that patients undergoing balloon aortic valvuloplasty for pre-dilation less frequently developed myocardial injury. Although it is unlikely that balloon aortic valvuloplasty is protective for myocardial injury, this finding can be explained by the fact that current practice in our center is to implant self-expanding valves such as the Evolut R and mechanically expanded valves such as the LOTUS without pre-dilation.
In previous studies, increased short- and long-term mortality was shown to be associated with greater increases in biomarkers of myocardial injury following the index procedure (5,13,14,16–18). This was most likely because the study protocol was defined to include all patients undergoing TAVR, irrespective of the route chosen and of the occurrence of complications such as conversion to open heart surgery, pericardiocentesis or pericardiotomy, or other serious events. Thus, in contrast to our study, several patients with poor outcomes per se, who had myocardial biomarker increases that were just an epiphenomenon, were included. To date, however, only a few studies have indicated that the occurrence of myocardial injury has no influence on short- and long-term mortality. Similar to our study, Barbash et al. (14) found evidence that myocardial injury, defined as a troponin increase according to the VARC-2 definition, had no profound impact on overall survival (15). A potential explanation for this inconsistency across studies might be the heterogeneous patient cohorts with respect to the access site, exclusion of periprocedural complications and, not the least, the use of different assays to determine cardiac biomarkers.
Because this is an observational clinical study, the exact mechanism, by which myocardial injury occurs, remains speculative and cannot be answered in detail. Apart from this, the single-center nature of this study is a further limitation. For further verification and generalization of our results, larger studies are needed.
Depending on the definition, myocardial injury is common following TF TAVR. The extent of cardiac biomarker elevation depends on THV type but is not associated with adverse short- and long-term outcome after uneventful TAVR.
WHAT IS KNOWN? A certain degree of myocardial injury defined by increases in cardiac biomarkers is a common finding after uneventful TAVR. In previous studies, an increased short- and long-term mortality was shown to be associated with greater increases in biomarkers of myocardial injury following the index procedure. However, in these analyses, patients with serious complications such as conversion to open heart surgery and also transapical patients have been included, so that the clinical significance for patients after uneventful TF TAVI ultimately remains unclear.
WHAT IS NEW? The extent of cardiac biomarker elevation defining myocardial injury does not predict short- and long-term mortality following TAVR, but it is associated with THV type, in particular with next-generation devices such as the Direct Flow Medical and LOTUS prostheses. Patients who received the LOTUS THV more frequently had myocardial injury compared with those who received other prostheses (LOTUS, 81.6%; Direct Flow Medical, 56.4%; CoreValve, 51.2%; Evolut R, 42.7%; SAPIEN XT, 40.4%; SAPIEN 3, 36.6%; p < 0.001).
WHAT IS NEXT? The reasoning behind the increased release of cardiac biomarkers defining myocardial injury remains elusive and will have to be assessed in larger clinical and partially experimental studies.
For supplemental tables and figures, please see the online version of this article.
Drs. Sinning, Grube, Nickenig, and Werner have received research grants and speaking honoraria from Medtronic, Edwards Lifesciences, and Boston Scientific. Dr. Grube works as proctor for Medtronic and Boston Scientific. All other authors have reported that they have no relationships relevant to the contents of this paper to disclose.
- Abbreviations and Acronyms
- confidence interval
- creatine kinase MB
- European System for Cardiac Operative Risk Evaluation
- hazard ratio
- high-sensitivity troponin I
- interquartile range
- left ventricular ejection fraction
- transcatheter aortic valve replacement
- transcatheter heart valve
- upper reference limit
- Valve Academic Research Consortium
- Received December 27, 2016.
- Revision received April 18, 2017.
- Accepted May 4, 2017.
- 2017 American College of Cardiology Foundation
- Kim W.-K.,
- Liebetrau C.,
- Linden A.,
- et al.
- Rodés-Cabau J.,
- Gutiérrez M.,
- Bagur R.,
- et al.
- Piazza N.,
- Grube E.,
- Gerckens U.,
- et al.
- Rodés-Cabau J.,
- Webb J.G.,
- Cheung A.,
- et al.
- Thomas M.,
- Schymik G.,
- Walther T.,
- et al.
- Ribeiro H.B.,
- Webb J.G.,
- Makkar R.R.,
- et al.
- Thygesen K.,
- Alpert J.S.,
- White H.D.
- Yong Z.Y.,
- Wiegerinck E.M.,
- Boerlage-van Dijk K.,
- et al.
- Sinning J.M.,
- Hsmmerstingl C.,
- Schueler R.,
- et al.
- Paradis J.M.,
- Maniar H.S.,
- Lasala J.M.,
- et al.
- Ribeiro H.,
- Nombela-Franco L.,
- Muñoz-García A.J.,
- et al.