Author + information
- Received June 19, 2017
- Revision received August 24, 2017
- Accepted September 19, 2017
- Published online December 4, 2017.
- Sammy Elmariah, MD, MPHa,b,∗ (, )
- William F. Fearon, MDc,
- Ignacio Inglessis, MDa,
- Gus J. Vlahakes, MDd,
- Brian R. Lindman, MDe,
- Maria C. Alu, MSf,
- Aaron Crowley, MAf,
- Susheel Kodali, MDf,g,
- Martin B. Leon, MDf,g,
- Lars Svensson, MDh,
- Philippe Pibarot, DVM, PhDi,
- Rebecca T. Hahn, MDf,g,
- Vinod H. Thourani, MDj,
- Igor F. Palacios, MDa,
- D. Craig Miller, MDc,
- Pamela S. Douglas, MDk,
- Jonathan J. Passeri, MDa,
- on behalf of the PARTNER Trial Investigators and PARTNER Publications Office
- aCardiology Division, Department of Medicine, Massachusetts General Hospital, Harvard Medical School, Boston, Massachusetts
- bBaim Institute for Clinical Research, Boston, Massachusetts
- cDivision of Cardiovascular Medicine, Department of Medicine, Stanford University School of Medicine, Stanford, California
- dDivision of Cardiac Surgery, Department of Surgery, Massachusetts General Hospital, Harvard Medical School, Boston, Massachusetts
- eCardiovascular Medicine Division, Department of Medicine, Vanderbilt University Medical Center, Nashville, Tennessee
- fCardiovascular Research Foundation, New York, New York
- gDivision of Cardiology, Department of Medicine, Columbia University Medical Center/New York–Presbyterian Hospital, New York, New York
- hDepartment of Cardiothoracic Surgery, Cleveland Clinic Foundation, Cleveland, Ohio
- iDepartment of Medicine, Québec Heart and Lung Institute, Laval University, Quebec City, Québec, Canada
- jDepartment of Cardiac Surgery, Medstar Washington Hospital Center, Washington, DC
- kCardiology Division, Department of Medicine, Duke Clinical Research Institute/Duke University Medical Center, Durham, North Carolina
- ↵∗Address for correspondence:
Dr. Sammy Elmariah, Massachusetts General Hospital, 55 Fruit Street, GRB 815, Boston, Massachusetts 02114-2696.
Objectives The authors sought to evaluate the impact of transapical (TA) transcatheter aortic valve replacement (TAVR) on mortality, left ventricular (LV) ejection fraction (LVEF) improvement, and functional recovery in patients with LV dysfunction.
Background LV injury inherent to TA access for structural heart disease interventions may be particularly detrimental to the LV, functional recovery, and survival in patients with LV dysfunction.
Methods The study included patients enrolled within the PARTNER I (Placement of Aortic Transcatheter Valves) trial that underwent transfemoral (TF) or TA TAVR. Analyses of clinical outcomes were stratified by the presence of baseline LV dysfunction (LVEF<50%) and adjusted for the propensity of receiving TA TAVR.
Results Of 2,084 subjects, 1,057 underwent TA TAVR. TA access was associated with increased 2-year all-cause mortality in those with (adjusted hazard ratio [HRadjusted]: 1.52; 95% confidence interval [CI]: 1.12 to 2.07; p = 0.008) and without (HRadjusted: 1.38; 95% CI: 1.10 to 1.74; p = 0.006) LV dysfunction. TA TAVR portended increased 2-year cardiac mortality in subjects with LVEF<50% (HRadjusted: 1.92; 95% CI: 1.21 to 3.05; p = 0.006), but not with LVEF≥50% (HRadjusted: 1.29; 95% CI: 0.87 to 1.90; p = 0.21). In those with LVEF<50%, greater improvements in LVEF (TF–TA difference +4.04%, 95% CI: 2.39% to 5.69%; p < 0.0001) and 6-min walk distance (TF–TA difference +45.1 m, 95% CI: 18.4 to 71.9 m; p = 0.001) occurred within 30 days after TF versus TA TAVR.
Conclusions Compared with TF TAVR, TA TAVR is associated with a disproportionate risk of cardiac mortality in patients with LV dysfunction and with delayed and less robust improvement in LV function and overall functional status. Caution is warranted when considering TA access for structural heart disease interventions, particularly in patients with LV dysfunction. (Placement of Aortic Transcatheter Valves [PARTNER]; NCT00530894)
- left ventricular ejection fraction
- transcatheter aortic valve replacement
Transcatheter aortic valve replacement (TAVR) has emerged as an effective therapy for intermediate- and high-risk patients with symptomatic severe aortic stenosis (AS) (1–3). Given this success, a host of new interventional approaches to structural heart disease are in clinical trials, many of which require a transapical (TA) approach. The TA approach involves direct left ventricular puncture and sheath insertion via a left lateral thoracotomy and ventriculotomy, and has consequently been associated with more frequent and severe myocardial injury (4,5). Transapical access has been associated with increased early mortality and slowed recovery (6), but whether ventriculotomy is particularly detrimental in patients with left ventricular (LV) dysfunction is unknown.
Although transfemoral (TF) transcatheter heart valve implantation represents the default and most commonly performed approach for TAVR, TA access is performed when peripheral vascular disease or tortuosity preclude the TF or alternative vascular approaches. Conflicting data exist regarding the impact of TA TAVR on focal and overall LV functional recovery after TAVR (7–10), and limited data suggest that TA TAVR may be associated with increased mortality in subjects with severely depressed LV systolic function (11,12). Despite the diminishing application of TA TAVR that has occurred with smaller profile transcatheter heart valve delivery systems, the impact of TA access on LV functional recovery and on mortality in patients with LV dysfunction is of tremendous clinical importance in helping to prioritize alternative access options, not only for TAVR in patients with LV dysfunction, but also, given the anticipated resurgence of the TA approach, for emerging transcatheter mitral valve replacement (TMVR) technologies and other valvular interventions. We therefore sought to determine the impact of the TA approach on mortality, LV systolic function improvement, and symptomatic and functional recovery in patients with LV dysfunction undergoing TA TAVR within the randomized PARTNER I (Placement of Aortic Transcatheter Valves) trial.
We included high-risk patients with AS enrolled within the PARTNER trial (cohort A), as well as those included within the continued access registry receiving either TF or TA TAVR. The design of the PARTNER trial has been reported previously (1,2). Briefly, the randomized portion of the trial included patients with symptomatic severe AS. After assessment of vascular anatomy, patients were allocated to either the TF or TA group and then randomized to either TAVR using the SAPIEN heart valve system (Edwards Lifesciences, Irvine, California) or surgical aortic valve replacement. After completion of the randomized portion of the trial, continued access registries allowed for the continued use of TAVR in the treatment of high-risk patients meeting the inclusion and exclusion criteria of the trial.
Inclusion criteria included severe AS, defined as a site-measured echocardiographic aortic valve area ≤0.8 cm2 plus either a peak velocity ≥4 m/s or a mean valve gradient ≥40 mm Hg (at rest or stress), New York Heart Association (NYHA) functional class II or greater, and high-risk status for surgical aortic valve replacement as determined by experienced surgeons. Patients were considered to be at high surgical risk if their predicted risk of 30-day perioperative mortality was ≥15%. The Society of Thoracic Surgery-Predicted Risk of Mortality (STS-PROM) risk score was used as the criterion for subject eligibility for patients with no other operative contraindications. Relevant exclusion criteria included a bicuspid or noncalcified aortic valve, coronary artery disease requiring revascularization, LV ejection fraction (LVEF) of <20%, and severe (4+) mitral or aortic regurgitation. The full exclusion criteria have been previously reported (1,2). Our analysis included all subjects who had baseline echocardiographic LVEF available.
The trial was approved by the institutional review board at each site. All patients provided written informed consent.
Transthoracic or transesophageal echocardiography was performed at baseline to assess eligibility for enrollment in the PARTNER I trial. Follow-up transthoracic echocardiography was performed before discharge, at 1- and 6-month visits, and annually thereafter. The Duke Clinical Research Institute (Durham, North Carolina) served as the echocardiographic core laboratory and independently analyzed all echocardiograms (13). All chamber parameters were measured according to the recommendations of the American Society of Echocardiography (14). LVEF was measured using the biplane Simpson’s volumetric method combining apical 4-chamber and 2-chamber views. The LV endocardial border was traced contiguously from one side of the mitral annulus to the other, excluding the papillary muscles and trabeculations, and any apical tethering of the mitral leaflets. In the very small number of images (<1%) with microbubble contrast, borders were traced similarly. In addition to the Simpson's method, LVEF was also determined by visual estimation (in 5-point increments), and when the definition of the LV endocardial border was not adequate for biplane tracing, was substituted to provide a single “combined LVEF” determination in all patients. Echocardiographic data reported here were obtained from rest studies.
Categorical baseline patient characteristics were reported as percentages and were compared by the chi-square or Fisher test, as appropriate. Continuous baseline patient characteristics were reported as mean ± SD and were compared by Student t test. Clinical outcomes were summarized as time-to-event variables using the Kaplan-Meier method to estimate failure rates and were compared by the log-rank test.
Propensity scores were developed because clinical risk factors differed between patients receiving TF and TA TAVR. The propensity score was defined as the posterior probability of receiving TA TAVR for each patient conditional on a set of covariates. Logistic regression was used to model the propensity score, which included the following parameters: age, sex, body surface area, STS-PROM score, diabetes mellitus, hyperlipidemia, hypertension, smoking status, NYHA functional class, oxygen-dependent chronic obstructive pulmonary disease, pulmonary hypertension, prior percutaneous coronary intervention, prior coronary artery bypass graft surgery, cerebrovascular disease, peripheral arterial disease, frailty, major arrhythmia, and permanent pacemaker.
Adjusted comparisons were performed stratifying by the propensity score quintile. The Cochran-Mantel-Haenszel test was used to compare categorical variables, the analysis of variance model was used to compare continuous variables, and a stratified Cox proportional hazards model was used to compare time-to-event variables. A multiplicative interaction term was added to the Cox proportional hazards models to test for differential associations between LV dysfunction and treatment approach. The time course of LVEF and 6-min walk distance (6MWD) were modeled using a propensity score–adjusted linear mixed model based on data from the baseline, 30 day, and 180 day visits.
All tests were 2-sided, and a p value <0.05 was considered significant. Statistical analyses were performed using SAS version 9.4 (SAS Institute, Cary, North Carolina).
Our analysis included 2,084 high-risk patients with symptomatic severe AS from the PARTNER I trial randomized cohort and continued access registry, of which 1,042 received TF TAVR, and 1,078, TA TAVR. Patients receiving TA TAVR were younger, more often female, and more likely to have a history of hyperlipidemia, hypertension, smoking, and established coronary artery, cerebrovascular, and peripheral vascular disease when compared with TF TAVR patients (Online Table 1). In addition, STS-PROM and logistic EuroSCORE scores were greater in TA patients than in TF patients.
Similar patterns were noted when further stratifying by the presence of LV dysfunction (Table 1). Patients receiving TA TAVR were younger and more likely to be female and to have a history of hypertension, smoking, coronary artery disease, cerebrovascular disease, peripheral arterial disease, and history of coronary artery bypass grafting surgery, regardless of baseline LV systolic function (Table 1). In patients with LVEF<50%, prior myocardial infarction was more prevalent among TA patients; whereas in patients with preserved LVEF, prior percutaneous coronary intervention was more prevalent among TA patients. After propensity score adjustment, measured baseline patient characteristics were well-balanced between TA and TF groups both with and without LV dysfunction.
Mortality and major adverse events
TA was associated with increased 2-year all-cause mortality compared with TF access in patients with (adjusted hazard ratio [HR]: 1.52; 95% confidence interval [CI]: 1.12 to 2.07; p = 0.008) and without (adjusted HR: 1.38; 95% CI: 1.10 to 1.74; p = 0.006) LV dysfunction after propensity score adjustment (Figure 1, Table 2). Cardiac mortality was increased at 2 years in patients undergoing TA TAVR with LVEF <50% (adjusted HR: 1.92; 95% CI: 1.21 to 3.05; p = 0.006), but not in patients with LVEF ≥50% (adjusted HR: 1.29; 95% CI: 0.87 to 1.90; p = 0.21). Moreover, rates of cardiovascular death related to arrhythmia or sudden death at 2 years were increased after TA TAVR in those with LV dysfunction (adjusted HR: 2.65; 95% CI: 1.14 to 6.15; p = 0.02), but not in those with preserved LV function (adjusted HR: 1.39: 95% CI: 0.72 to 2.70; p = 0.33). These differences in mortality between TF and TA approaches were observed despite higher rates of moderate or severe paravalvular aortic regurgitation with TF versus TA TAVR at hospital discharge through 6-month follow-up (discharge 11.0% vs. 5.4%; p < 0.0001, padjusted = 0.0007; 6-month 14.0% vs. 5.7%; p < 0.0001, padjusted <0.0001). Transapical TAVR was associated with an increased noncardiovascular mortality among the entire cohort (adjusted HR: 1.66; 95% CI: 1.21 to 2.28; p = 0.002) and in those with preserved LV function (adjusted HR: 1.70; 95% CI: 1.17 to 2.47; p = 0.006).
Rates of repeat hospitalization, major stroke, and myocardial infarction were comparable between TA and TF TAVR subjects, regardless of LV function (Table 2).
Symptom and functional status
Baseline functional status, as quantified by 6MWD, was similar in those undergoing TF TAVR (173.9 m, 95% CI: 164.0 to 183.8 m) and TA TAVR (176.4 m, 95% CI: 166.1 to 186.8 m; p = 0.73) (Figure 2), excluding patients not able to perform the test. Although improvement was significantly greater after TF TAVR within the first 30 days (+46.3 m, 95% CI: 37.1 to 55.5 m vs. +23.8 m, 95% CI: 14.1 to 33.4 m; p = 0.0009), the extent of improvement was comparable after 6 months (+59.0 m, 95% CI: 49.8 to 68.2 m vs. +53.0 m, 95% CI: 43.3 to 62.6 m; p = 0.37) in all TAVR patients.
In subjects with baseline LV dysfunction, the extent of improvement in 6MWD was greater after TF TAVR compared with TA TAVR at 30 days (+68.8 m, 95% CI: 50.3 to 87.3 m vs. +23.6 m, 95% CI: 4.24 to 43.04 m; p = 0.001) and statistically comparable at 6 months (TF +76.4 m, 95% CI: 58.0 to 94.9 m vs. TA +58.9 m, 95% CI: 39.5 to 78.3 m; p = 0.20), and the total distance walked was comparable across groups at 6-month follow-up (Figure 2). In subjects without baseline LV dysfunction, the extent of functional improvement was comparable after TF and TA TAVR at 30 days (+36.9 m, 95% CI: 26.3 to 47.5 m vs. +22.8 m, 95% CI: 11.8 to 33.9 m; p = 0.07) and at 6 months (+51.3 m, 95% CI: 40.7 to 61.9 m vs. +49.3 m, 95% CI: 38.2 to 60.3; p = 0.26).
At baseline, 95.8% and 94.1% of subjects receiving TF and TA TAVR, respectively, were classified as possessing NYHA functional class III or IV symptoms. By 30-day follow-up, only 16.7% of TF patients and 24.3% of TA patients remained in NYHA functional class III/IV (p < 0.0001). There was no difference in symptom classification between the 2 groups at 6-month follow-up (11.4% TF vs. 13.0% TA; p = 0.31). Rates of persistent NYHA functional class III/IV symptoms were similarly decreased at 30 days and 6 months in those with and without baseline LV dysfunction.
LV systolic function after TAVR
LVEF was 52.6% (95% CI: 51.8% to 53.4%) in patients undergoing TF TAVR and 52.6% (95% CI: 51.7% to 53.4%; p = 0.92) in those undergoing TA TAVR. Among patients with LVEF <50% that survived to 6 months and who had LVEF available at all time points (n = 225), baseline LVEF was 35.3% (95% CI: 34.0% to 36.5%) and 37.1% (95% CI: 35.8% to 38.4%) in TF and TA subjects, respectively. A significant increase in overall LVEF was observed within 30 days of both TF (difference 9.03%, 95% CI: 7.89% to 10.18%; p < 0.0001) and TA (difference 4.99%, 95% CI: 3.80% to 6.18%; p < 0.0001) TAVR (Figure 3); however, the magnitude of improvement was more robust after TF TAVR at both 30-day (TF–TA difference +4.04%, 95% CI: 2.39% to 5.69%; p < 0.0001) and 6-month (TF–TA difference +4.12%, 95% CI: 2.47% to 5.77%; p < 0.0001) time points. Among patients with baseline LVEF <50%, 34.7% of TF TAVR patients (100 of 288) experienced an absolute increase in LVEF >10% by 30-day follow-up compared with only 21.6% of TA TAVR patients (59 of 273) (p = 0.0006).
In this study of patients with severe symptomatic AS at high risk for surgical aortic valve replacement, we found that clinical outcomes after TA TAVR, relative to TF TAVR, are effected by the presence of baseline LV dysfunction (LVEF <50%). Specifically, TA TAVR is associated with a 2-fold increase in the risk of cardiovascular death in subjects with LVEF <50%, but no increase in the risk of cardiovascular death in subjects with preserved LVEF. In addition, whereas TA TAVR has previously been associated with delayed functional improvement (2), we found that improvement in 6MWD is particularly delayed in those subjects with LVEF <50%. In subjects with LV dysfunction, LV systolic function improved rapidly (<30 days) after TAVR, regardless of approach. However, the magnitude of LVEF improvement was greater after TF TAVR than with TA TAVR, both at 30 days and 6 months after TAVR. Additionally, we found that a greater proportion of subjects undergoing TF TAVR experience an absolute increase in LVEF >10%.
The observed differences in mortality between TF and TA TAVR among patients with LV dysfunction are notable, especially in light of reduced rates of paravalvular aortic regurgitation seen with TA TAVR using the first-generation SAPIEN device (6). Previous data from the PARTNER trial have found significantly increased mortality risk in patients with greater than trace paravalvular aortic regurgitation (15). Although further evidence is needed, we anticipate that substantial reductions in paravalvular aortic regurgitation, as seen with the newer generation SAPIEN 3 transcatheter heart valve, may therefore reveal more substantial risk differences between TF and TA TAVR in patients with LV dysfunction (16,17).
Although previous studies have associated TA TAVR with increased mortality, reduced quality of life, and slower recovery as compared with TF TAVR (6,18,19), this is the first report documenting disproportionately increased risk in subjects with LV dysfunction subjected to the transapical approach. The mechanism responsible for increased mortality in patients with LV dysfunction undergoing TA TAVR is unclear, but inspection of event curves suggests that the hazard of TA TAVR in those with LV dysfunction begins immediately post-TA TAVR and increases with time at least to 2-year follow-up. Therefore, although TA TAVR may be associated with increased procedural and in-hospital mortality, our observations suggest a persistent hazard with TA TAVR, rather than a purely procedure-related hazard. In subjects with baseline LV dysfunction, transapical TAVR also results in reduced improvement in LV systolic function as compared with TF TAVR. We previously found that the lack of LV systolic function improvement within 30 days of TAVR is associated with markedly reduced survival (10). The direct left ventricular trauma incurred during TA TAVR may hinder LV recovery, despite alleviation of the pressure-load of AS. We also found that cardiovascular deaths related to arrhythmia or sudden death were increased in patients with LV dysfunction, suggesting that scar-mediated arrhythmias may in part drive the increased mortality attributable to TA TAVR in patients with LV dysfunction, but whether progressive heart failure or alternative mechanisms additionally contribute to cardiovascular mortality in TA TAVR patients remains uncertain.
Our observations have substantial clinical implications. Avoidance of the transapical approach in patients with LV dysfunction in need of TAVR in favor of other alternative approaches appears prudent. In addition, as transcatheter therapies for valvular heart disease continue to evolve and grow, caution should be exercised when adopting a TA approach in patients with LV dysfunction. This concern is of particular relevance for emerging TMVR technologies, not only because many of these devices will be delivered via the LV apex, but also because of the high prevalence and long-term prognostic significance of LV dysfunction among those with severe mitral regurgitation (20–22). Furthermore, TMVR devices currently under development target patients with secondary (“functional”) mitral regurgitation (23), which by definition possess diseased LVs. Our findings, therefore, highlight a potential pitfall of emerging TA TMVR and other structural heart intervention technologies, and underscore the need for a fuller understanding of the consequences of TA procedures and the rapid development of alternative approaches.
The use of an echocardiographic core laboratory, clinical event adjudication, and large sample size are significant strengths of the present analysis, but several limitations must also be acknowledged. First, patients with severe LV dysfunction (LVEF <20%) and those with low-gradient AS without contractile reserve were excluded from the PARTNER trial. However, we anticipate that the use of TA TAVR in patients with more diseased LVs would likely magnify our observations. Second, comparison of TF and TA approaches was not randomized and is therefore prone to confounding; however, we used propensity score adjustment to minimize the risk of confounding. Third, our analysis was prone to survival selection bias, given that follow-up LVEFs were only available in those that survived. Focused analyses of LVEF solely in patients that survived to 6-month follow-up mitigate this concern. Finally, the PARTNER I trial included a highly selected cohort of patients at very high risk for surgical intervention. Whether results of this analysis are applicable to the larger population of patients with severe AS or with other valve lesions warrants further investigation.
In summary, compared with TF TAVR, TA TAVR is associated with disproportionate risk of 2-year cardiovascular mortality in patients with baseline LV dysfunction. This mortality hazard is supported by increased risk of death related to arrhythmia and sudden death, delayed and less robust improvement in LV systolic function, and delayed improvement in functional status after TA TAVR. These findings emphasize the need for caution when considering transapical access for any structural heart intervention, particularly in patients with LV dysfunction.
WHAT IS KNOWN? TA TAVR has been associated with more frequent and severe myocardial injury, increased mortality, and delayed recovery when compared to the transfemoral approach.
WHAT IS NEW? Using data from the PARTNER I trial, we identified a disproportionate increase in mortality in patients with left ventricular systolic dysfunction undergoing TA TAVR, emphasizing the need for caution when considering TA access for structural heart procedures in patients with left ventricular dysfunction.
WHAT IS NEXT? Further efforts are needed to identify methods of mitigating the risk of TA access in patients with left ventricular dysfunction and to identify alternative access approaches for emerging transcatheter mitral valve replacement technologies.
The authors thank Feifan Zhang, Girma M. Ayele, and Thomas McAndrew for assistance with statistical analyses.
The PARTNER trial was funded by Edwards Lifesciences. This analysis was supported in part by the American Heart Association (14 FTF20440012 to Dr. Elmariah) and the MGH Heart Center Hassenfeld Cardiovascular Research Scholar Program (to Dr. Elmariah). Dr. Elmariah has received institutional research support from Siemens, Edwards Lifesciences, Boehringer Ingelheim, and the American Heart Association; and consulting fees from Medtronic. Dr. Fearon has received institutional research support from Abbott, Edwards Lifesciences, Medtronic, and ACIST Medical; and consulting fees from HeartFlow. Dr. Inglessis has received institutional research support from Medtronic, St. Jude Medical, and W.L. Gore and Associates; and is a proctor for Medtronic and Edwards Lifesciences. Dr. Lindman has received research institutional support from Roche Diagnostics and Edwards Lifesciences; and is a consultant for Roche and Medtronic. Ms. Alu is a consultant for Claret Medical. Dr. Kodali is on the steering committee for Edwards Lifesciences and is on the advisory board and holds equity in Thubrikar Aortic Valve Inc. and Dura Biotech. Dr. Svensson holds equity in Cardiosolution and Valvexchange; is chairman of the PARTNER Publishing Office; and has intellectual property with Postthorax. Dr. Pibarot has received institutional research support from Edwards Lifesciences and Medtronic. Dr. Hahn has received institutional research support from Edwards Lifesciences. Dr. Thourani has received institutional research support from Edwards Lifesciences. Dr. Miller has received institutional research support from Edwards Lifesciences, Medtronic, and Abbott; and consulting fees from Medtronic. Dr. Douglas has received institutional research support from Edwards Lifesciences. Dr. Passeri has received institutional research support from Edwards Lifesciences; has been a speaker at an educational symposium sponsored by Medtronic; and has received consulting fees from Medtronic. All other authors have reported that they have no relationships relevant to the contents of this paper to disclose.
- Abbreviations and Acronyms
- 6-min walk distance
- aortic stenosis
- confidence interval
- hazard ratio
- left ventricle/ventricular
- left ventricular ejection fraction
- New York Heart Association
- Society of Thoracic Surgery-Predicted Risk of Mortality
- transcatheter aortic valve replacement
- transcatheter mitral valve replacement
- Received June 19, 2017.
- Revision received August 24, 2017.
- Accepted September 19, 2017.
- 2017 American College of Cardiology Foundation
- Paradis J.M.,
- Maniar H.S.,
- Lasala J.M.,
- et al.
- Blackstone E.H.,
- Suri R.M.,
- Rajeswaran J.,
- et al.
- Ando T.,
- Holmes A.A.,
- Taub C.C.,
- DeRose J.J.,
- Slovut D.P.
- Unbehaun A.,
- Pasic M.,
- Buz S.,
- et al.
- Elmariah S.,
- Palacios I.F.,
- McAndrew T.,
- et al.
- Urena M.,
- Webb J.G.,
- Eltchaninoff H.,
- et al.
- Lang R.M.,
- Bierig M.,
- Devereux R.B.,
- et al.
- Yang T.H.,
- Webb J.G.,
- Blanke P.,
- et al.
- Binder R.K.,
- Stortecky S.,
- Heg D.,
- et al.
- Gada H.,
- Kirtane A.J.,
- Wang K.,
- et al.
- Reynolds M.R.,
- Magnuson E.A.,
- Wang K.,
- et al.
- Nishimura R.A.,
- Otto C.M.,
- Bonow R.O.,
- et al.
- Crawford M.H.,
- Souchek J.,
- Oprian C.A.,
- et al.
- Enriquez-Sarano M.,
- Tajik A.J.,
- Schaff H.V.,
- Orszulak T.A.,
- Bailey K.R.,
- Frye R.L.
- De Backer O.,
- Piazza N.,
- Banai S.,
- et al.