Author + information
- Received March 19, 2014
- Revision received August 27, 2014
- Accepted August 29, 2014
- Published online February 1, 2015.
- Samir Kapadia, MD∗,
- William J. Stewart, MD∗,
- William N. Anderson, PhD†,
- Vasilis Babaliaros, MD‡,
- Ted Feldman, MD§,
- David J. Cohen, MD, MSc‖,
- Pamela S. Douglas, MD¶,
- Raj R. Makkar, MD#,
- Lars G. Svensson, MD, PhD∗,
- John G. Webb, MD∗∗,
- S. Chiu Wong, MD††,
- David L. Brown, MD‡‡,
- D. Craig Miller, MD§§,
- Jeffrey W. Moses, MD‖‖,
- Craig R. Smith, MD‖‖,
- Martin B. Leon, MD‖‖ and
- E. Murat Tuzcu, MD∗∗ ()
- ∗Cleveland Clinic Foundation, Cleveland, Ohio
- †Irvine, California
- ‡Emory University School of Medicine, Atlanta, Georgia
- §NorthShore University Health System, Evanston, Illinois
- ‖Saint Luke’s Mid America Heart Institute, Kansas City, Missouri
- ¶Duke Clinical Research Institute, Durham, North Carolina
- #Cedars-Sinai Medical Center, Los Angeles, California
- ∗∗St. Paul’s Hospital/University of British Columbia, Vancouver, British Columbia, Canada
- ††New York–Presbyterian Hospital/Weill Cornell Medical Center, New York, New York
- ‡‡Medical City Dallas Hospital, Dallas, Texas
- §§Stanford University, Stanford, California
- ‖‖New York–Presbyterian Hospital/Columbia University Medical Center, New York, New York
- ↵∗Reprint requests and correspondence:
Dr. E. Murat Tuzcu, Department of Cardiovascular Medicine, Cleveland Clinic, 9500 Euclid Avenue, Mail Code J2-3, Cleveland, Ohio 44195.
Objectives The aim of this report is to characterize the impact of balloon aortic valvuloplasty (BAV) in patients not undergoing aortic valve replacement in the PARTNER (Placement of AoRtic TraNscathetER Valves) trial.
Background The PARTNER trial is the only randomized trial with independently adjudicated data of inoperable severe symptomatic aortic stenosis patients, allowing outcome analysis of unoperated-on patients.
Methods The design and initial results of the PARTNER trial (Cohort B) were reported previously. After excluding patients with pre-randomization BAV, we compared patients undergoing BAV within 30 days of randomization (BAV group) with those not having BAV within 30 days of randomization (no BAV group) to characterize the use and impact of BAV.
Results In the PARTNER Cohort B study, 179 inoperable patients were randomized to standard treatment including 39 patients (21.8%) who had undergone a BAV before randomization (previous BAV group). Of the 140 patients who did not have BAV before enrollment in the study, 102 patients (73%) had BAV within 30 days of study randomization (BAV group). Survival at 3 months was greater in the BAV group compared with the no BAV group (88.2%; 95% confidence interval [CI]: 82.0% to 94.5% vs. 73.0%; 95% CI: 58.8% to 87.4%). However, survival was similar at 6-month follow-up (74.5%; 95% CI: 66.1% to 83.0% vs. 73.1%; 58.8% to 87.4%). There was improvement in quality of life parameters when paired comparisons were made between baseline and 30 days and 6 months between the BAV and no BAV groups, but this effect was lost at 12-month follow-up.
Conclusions BAV improves functional status and survival in the short term, but these benefits are not sustained. BAV for aortic stenosis patients who cannot undergo aortic valve replacement is a useful palliative therapy. (THE PARTNER TRIAL: Placement of AoRTic TraNscathetER Valve Trial; NCT00530894)
Aortic stenosis (AS) is a common condition among the elderly and is associated with poor survival without surgery once symptoms develop (1,2). In addition, patients with severe AS experience progressive symptoms with reduced functional status and quality of life (QOL). Despite the success of surgical valve replacement in alleviating symptoms, improving functional status, and extending survival (3,4), a substantial minority of patients with severe AS remain untreated due to prohibitive surgical risk (5,6). One-year mortality rates may exceed 50% in these patients (7).
After the first human report in 2002 (8), transcatheter aortic valve replacement (TAVR) has emerged as a less invasive treatment option for patients with AS and a high or unacceptable surgical risk (9–11). The PARTNER (Placement of AoRtic TraNscathetER Valves) trial demonstrated that for patients who are not suitable candidates for surgery, TAVR led to a 20% absolute reduction in all-cause mortality at 1 year compared with standard therapy, and this benefit was sustained and actually more pronounced when patients were followed for 2 years (12). Beyond its mortality benefit, TAVR led to improvement in symptoms, functional status, and QOL, which may be more important than the survival benefit for these elderly patients (13).
The PARTNER trial was the first randomized trial with a collection of outcome adjudicated data on inoperable patients, allowing one to study the outcomes of unoperated-on patients with severe symptomatic AS. Although balloon aortic valvuloplasty (BAV) has been used for palliation as well as a bridge to surgical aortic valve replacement (AVR), the impact of BAV has not been studied with independent adjudication compared with standard medical therapy (14–16). The standard therapy arm of the PARTNER trial provides an opportunity to better understand the role of BAV in inoperable patients. In this report, we attempt to characterize the outcomes of standard therapy in patients not undergoing TAVR with a special focus on the role of BAV.
The design and initial results of the PARTNER trial (Cohort B) were published previously (17). In brief, the PARTNER program enrolled patients with severe AS, New York Heart Association functional class II, III, or IV heart failure symptoms, and prohibitively high surgical risk based on the Society for Thoracic Surgeons (STS) risk score and qualifying assessments by the heart team. Patients included in the present study were not considered to be suitable candidates for cardiac surgery because of coexisting medical conditions associated with a predicted probability of death or permanent disability ≥50%, as determined by at least 2 surgical investigators and reaffirmed by the study’s executive committee. These patients were then randomized to TAVR, using the Edwards SAPIEN heart valve system (Edwards Lifesciences, Irvine, California), or standard medical care, which often included BAV at the discretion of the investigators. There was no specific time stipulated in the protocol for TAVR after randomization, although treatment within 2 weeks of randomization was encouraged.
The PARTNER trial was funded by Edwards Lifesciences and designed collaboratively by the steering committee and the sponsor. The study was approved by each participating site’s Institutional Review Board, and all patients provided written informed consent. All events were independently adjudicated, and echocardiograms were interpreted by a core laboratory. The current analysis was carried out by academic investigators at the study sites and by the Health Economics and Technology Assessment Group at Saint Luke’s Mid America Heart Institute (Kansas City, Missouri). The authors had unrestricted access to the study data, drafted the manuscript, and vouch for the veracity and completeness of its content.
BAV use and impact on outcomes
We compared the characteristics of patients who had BAV before randomization (previous BAV group) with those who never had BAV. After excluding patients with pre-randomization BAV, we compared patients undergoing BAV within 30 days (BAV group) with those not having BAV within 30 days of randomization (no BAV group) to characterize the use and impact of BAV on these inoperable patients in the study.
A randomized sample size of 350 patients was computed to obtain appropriate power for analyzing the primary trial endpoints; the actual randomized enrollment was 179 in each trial arm. The trial was not powered for the various separate standard therapy analyses described in this paper. BAV was among the options that could be chosen for treatment in the standard therapy arm and was performed at various post-randomization time points. For purposes of presenting the effect of BAV, patients were placed into 3 groups: BAV before the trial, BAV during the trial, and no BAV. Because assignment to these 3 groups was not randomized, any comparisons are purely descriptive, and no formal statistical comparisons among the 3 groups are presented. The analysis was based on the October 9, 2012 data extract with analysis close date as of March 16, 2012. This close date was 3 years after the last randomization. The time of each event was calculated from the date of randomization. Deaths and other adverse events occurring after the analysis close date are not considered in the analysis. Categorical variables are presented as frequencies and compared with the use of the Fisher exact test. Continuous variables are presented as mean ± SD and compared using the t test or paired t tests as appropriate. Survival curves for time-to-event variables were constructed on the basis of all available follow-up data with the use of Kaplan-Meier estimates; patients were censored at the last date known alive or the analysis close date, whichever came first. To identify predictors of mortality, covariate analysis was performed using the Cox proportional hazards algorithm. All statistical analyses were performed using SAS software, version 9.3 (SAS Institute, Cary, North Carolina).
Data were collected from all 17 clinical sites. Various poolability analyses were presented to the U.S. Food and Drug Administration, and after extensive review, the U.S. Food and Drug Administration accepted poolability of site results for the primary endpoint and all the secondary endpoints. The Pre-Market Approval was presented in this manner, and all subsequent publications have also used pooled site data; that practice has been followed in this manuscript.
In the PARTNER Cohort B study, 179 inoperable patients were randomized to standard treatment without TAVR. At the time of enrollment, the majority of the patients were on contemporary treatment for heart failure: diuretics, 80%; beta-blockers, 62%; statins, 61%; angiotensin-converting enzyme inhibitor or angiotensin receptor blocker, 41%; aspirin, 59%; and warfarin, 24%. Despite aggressive medical treatment, 94% of patients were in New York Heart Association (NYHA) functional class III or IV heart failure. Of these 179 patients, 39 (21.7%) had undergone BAV weeks to months before enrolling in the PARTNER trial (median, 121 days). The previous BAV group of patients had similar clinical and echocardiographic characteristics compared with patients who did not have BAV before enrollment (Table 1).
In the PARTNER trial, many centers used BAV as the next treatment tool in severely symptomatic patients who were already receiving optimal medical treatment. Of the 22 centers that participated in the PARTNER trial, 17 performed BAV in more than 50% of their patients who were randomized to standard therapy. Ten centers used it in more than 90% of their patients.
Of the 140 patients who did not have BAV before enrolling in the study, 102 patients were included in the BAV group, with BAV performed within 30 days of study randomization (median, 3 days). To understand the impact of BAV, characteristics of the patients undergoing their first BAV within 30 days of randomization were compared with those of the 38 patients who did not undergo BAV within 30 days of randomization (no BAV group) (Table 2). Patients undergoing BAV had more severe congestive heart failure symptoms and a higher logistic EuroSCORE. However, other characteristics including STS score, echocardiographic characteristics, and QOL parameters were not different between these 2 groups. There were 53 patients who had BAV past 30 days, and 12 of these had more than 1 such BAV.
Hemodynamic response to BAV
The BAV group of patients demonstrated improvement in gradients and aortic valve area. Echocardiographic information was available for 96 patients before and after BAV. After the first BAV, the mean aortic valve gradient determined by transthoracic echocardiography decreased significantly from 44.8 ± 15.9 mm Hg to 31.5 ± 12.1 mm Hg, and the aortic valve area increased significantly from 0.6 ± 0.2 cm2 to 0.8 ± 0.2 cm2.
In the BAV group of patients, a second BAV was performed in 29 patients 282 ± 158 days after the first BAV. Seven patients underwent a third BAV after 201 ± 144 days of the second BAV, 2 patients had a fourth BAV after an interval of 163 days after the third BAV. In contrast to the first BAV, hemodynamic improvement after the second BAV was negligible; the change in mean gradient from 46.4 ± 18.7 mm Hg to 44.5 ± 16.8 mm Hg was not significant, and aortic valve area was unchanged at 0.7 ± 0.2 (Table 3).
Complications of BAV
After 189 BAV procedures during the trial, 8 deaths occurred within 30 days after the BAV (4.2%). Stroke occurred in 2 patients (1.1%) during the same period. There was no change in magnitude of aortic regurgitation (1.6 ± 0.9+ vs 1.7 ± 0.9+). Aortic regurgitation was none, trace, mild, moderate, and severe in 14%, 31%, 57%, 11%, and 1% of patients, respectively, and after the first BAV, the counts were 9%, 32%, 55%, 16%, and 2%, respectively. No crossovers to TAVR were allowed after BAV for severe aortic regurgitation. All outcomes of the 3 groups of patients are reported in Table 4.
There was meaningful improvement in heart failure symptoms in the BAV group compared with the no BAV group, as demonstrated by improvement in NYHA functional class (Figure 1). There was improvement in QOL parameters when paired comparisons were made between baseline and 30 days and 6 months in the BAV group, but this effect was lost at 12-month follow-up (Table 5).
Survival at 3 months was higher in the BAV group compared with the no BAV group (88.2%; 95% confidence interval [CI]: 82.0% to 94.5% vs. 73.0%; 95% CI: 58.8% to 87.4%). However, survival was similar at 6 months follow up (74.5%; 95% CI: 66.1% to 83.0% vs. 73.1%; 58.8% to 87.4%) (Figure 2). Of the 179 patients randomized to standard treatment, 135 patients died during the trial; the cause of death was cardiovascular in 69, noncardiovascular in 21, and unknown in 45 patients. At 3 months, there were fewer cardiovascular deaths in the BAV group (Figure 2B⇓).
Table 6 lists the univariate predictors of mortality in the unoperated-on severe symptomatic AS patients. Previous stroke or transient ischemic attack and a history of pacemaker implantation were the most important predictors of mortality. Interestingly, patients who underwent BAV before enrolling in the PARTNER trial had a higher mortality rate. The STS risk score was a significant predictor of mortality. Patients with an STS score <10 had better survival than those with an STS score of 10 to 15 or more than 15 (Figure 3).
In 11 of the 179 patients (6.1%), overall clinical status improved to the point that the decision of inoperability was reconsidered, and they underwent surgical AVR. Of the 39 patients who did not undergo BAV, only 1 (2.5%) underwent AVR. Fifty percent (5 of 10 with 1 unknown) of the patients undergoing surgery survived to 1 year after surgery.
There are several major observations that can be made from the detailed analysis of these prospectively collected data. First, the outcome of patients with severe symptomatic AS who do not undergo AVR is poor. Second, BAV improves outcomes in these inoperable patients for a few months. Third, repeat valvuloplasties are less effective than the first one. Fourth, the risk of BAV is not trivial but is in an acceptable range. Fifth, major comorbidities responsible for a high STS score predict mortality, even in patients not undergoing AVR.
In 1933, Grant (18) reported 10-year follow-up of 43 male war prisoners with AS with or without symptoms for 10 years. At the end of this follow-up, 28 had died; 10 died of cardiac failure, 7 of sudden cardiac death, 2 of subacute bacterial endocarditis, and 8 of unrelated causes. Harken et al. (19) reported that 49 of 54 patients with AS (91%) who were advised to have cardiac surgery and refused were dead within 6 months. Ross and Braunwald (1) reported in their classic paper “recently, we have obtained follow-up information on patients with isolated aortic valve stenosis who were studied hemodynamically at the National Heart Institute and in whom operation was refused or not carried out for a variety of reasons. Ten of 12 patients with severe aortic stenosis (transvalvular pressure gradients ≥ 50 mm Hg, or effective aortic valve orifices ≤0.70 cm2/m2 BSA) who were followed for at least five years after catheterization are now dead.” Subsequently, they reported data from 15 patients with severe well-characterized AS and without aortic surgery with a 52% mortality rate at 5 years, but these patients were young (32 to 59 years of age) (20,21). In 1991, O’Neill (22) reported data from Mansfield Scientific Aortic Balloon Valvuloplasty Registry for 492 patients with a 1-year survival rate of 64%, which is somewhat better than the standard therapy arm of the PARTNER study (22,23). However, the patient population was quite different, with only 30% of patients refused by consulting cardiovascular surgeons. The PARTNER trial population is older (84 years of age vs. 79 years of age in the Mansfield Registry). The inoperability in the PARTNER trial was determined by local surgeons and confirmed in a conference call by nationally recognized expert aortic surgeons. This process ensured a selection of inoperable patients who had more comorbidities compared with the Mansfield Registry. Vascular complication rates were higher in the Mansfield Registry, probably related to the unavailability of vascular closure devices. Very similar data were reported from the National Heart, Lung, and Blood Institute balloon valvuloplasty registry around a similar time frame from a patient population comparable to the Mansfield Registry (24). A well-characterized patient population, long-term follow-up, and adjudicated event monitoring from a contemporary era make the present paper unique.
Our study may be the first, and possibly the last, study in which inoperable high-risk medically managed patients with AS were carefully studied without aortic replacement in a randomized fashion. There are many similarities of these data with those of the studies done 80 and 45 years ago. However, Ross and Braunwald studied the natural history of AS in operable and young patients. It appears that there has been no significant improvement in the medical treatment of severe AS in a century.
The prospectively collected data from unoperated-on symptomatic AS patients in the PARTNER trial will serve as a benchmark for future trials testing new interventions. Although BAV was studied carefully in a multicenter registry in the 1990s and in many other reports since then, these reports did not use core laboratories for data collection and analysis, and patient selection was not defined. Our study offers an opportunity to evaluate the contemporary state of this procedure in a well-characterized and carefully selected group of patients. The rigor of data collection, uniformity of definitions, and the presence of an independent clinical events committee and an independent echocardiographic core laboratory make this analysis unique.
BAV is performed mainly to bridge high-risk patients to surgical AVR or transcatheter AVR in clinical practice. Reasons for bridging include temporary contraindications to valve replacement (e.g., sepsis, severe debilitation, acute neurological event, coagulopathy), other significant cardiac lesions (coronary, mitral valve, tricuspid valve, myocardial disease) where the relative contribution of AS to heart failure remains questionable, or in patients with severe noncardiac comorbidities in whom the role of AS in presenting symptoms is difficult to determine (e.g., severe lung disease, cirrhosis, severe debilitation). Infrequently, BAV can be used for patients with symptomatic severe AS who require urgent major noncardiac surgery. It can be helpful for palliation in adult patients with AS in whom AVR cannot be performed because of serious comorbid conditions, although short-lived improvement makes such an effort only a temporary success. This application of BAV may be helpful to relieve extreme symptoms or facilitate discharge of patients who are hospital bound. Studies in the past have shown that although BAV may result in temporary relief of symptoms, restenosis is virtually certain within 6 to 12 months (22,25–27). This led to the guidelines recommending BAV as a reasonable bridge to surgery in hemodynamically unstable patients at high risk of AVR or as a palliative procedure in patients with AS at high risk of surgical AVR (Class IIb indication) (28). This analysis from the PARTNER trial highlights important facts that are very relevant in clinical practice. There appears to be survival benefit in patients undergoing BAV in the first 3 months. Therefore, one could argue that if patients are not able to undergo AVR due to temporary contraindications such as infection and debility, there may be a survival benefit of a strategy with BAV as a bridge to definitive therapy. There is evidence of a high mortality rate in patients being screened for TAVR, representing a potential need for such considerations (29,30). Because the mortality matches patients not undergoing BAV at 6 months, the data confirm the need for definitive therapy (i.e., AVR or TAVR sooner rather than later after BAV). Importantly, this analysis shows improvement in QOL that persists for ∼6 months after BAV. Although improvement in heart failure symptoms has been reported with BAV, QOL improvement is important additional data. The distinction between previously reported functional improvement data and health status measures reported here is an important one. In this population, patient-reported QOL (based on the Kansas City Cardiomyopathy Questionnaire and SF-12) are used, whereas most other studies have relied on external assessment (NYHA functional class). This further substantiates the use of BAV as a palliative therapy.
These data also provide some insight beyond balloon valvuloplasty in these inoperable patients. A few patients randomized to standard therapy did undergo surgery, and they had poor outcomes, especially compared with those randomized to undergo TAVR in whom the 1-year mortality rate was 31% (17). Patients who underwent apico-descending aorta conduit or surgical AVR had poor outcomes, with 33% and 27% mortality rates in the first 30 days, respectively.
There are several implications of this analysis for clinical practice. BAV can improve symptoms and QOL, although early bridging to AVR is essential for a better outcome. QOL can be improved to some extent with BAV for ∼6-month duration.
There are some limitations of this analysis. Although the clinical characteristics of patients who did and did not undergo BAV are similar, selection bias in the decision regarding when to perform BAV cannot be characterized. The patients in this cohort were all candidates for TAVR, so the results cannot be extrapolated to patients who were excluded from the trial or so-called Cohort C patients.
Detailed procedural data on BAV were not collected in these patients, and therefore procedural variables could not be studied. Catheterization data are limited, including pre- and post-valvuloplasty data. Predictors of mortality could not be studied in a time variant model due to the small number of patients. Further, clinical characteristics of patients before repeat BAV were not collected in the trial. Because the crossovers were not allowed until the primary endpoint (1 year), patients who needed relief of AS for survival underwent repeat BAV at the discretion of the treating physician.
At 1 year, the survival of patients with inoperable severe AS was 49.3%, with only 28.4% patients without hospitalization and only 21% of the patients in NYHA functional class I or II. Although BAV was used in 84% of these patients, it was performed at the discretion of physicians taking care of the patients, and there was no survival benefit at 1 year, patients undergoing BAV in the trial had better 3-month survival and better 6-month QOL compared with patients not undergoing BAV in the trial. Standard therapy for AS in patients who cannot undergo surgery is inadequate. BAV for some patients who cannot undergo AVR is a useful palliative therapy.
The PARTNER trial was funded by Edwards Lifesciences and the protocol was designed collaboratively by the sponsor and the steering committee. The current analysis was carried out under the auspices of the PARTNER Publications Office by academic investigators at the study sites, by an independent biostatistician (Dr. Anderson), and by the Health Economics and Technology Assessment Group at Saint Luke’s Mid America Heart Institute, Kansas City, Missouri (Dr. Cohen). Dr. Anderson is a consultant for and holds equity in Edwards Lifesciences. Dr. Babaliaros is an investigator for the PARTNER trial (Edwards Lifesciences). Dr. Feldman has received grant support from and consulting fees/honoraria from Abbott Vascular, Boston Scientific Corporation, and Edwards Lifesciences. Dr. Cohen has received grant support from Abbott Vascular, Boston Scientific Corporation, Edwards Lifesciences, Eli Lilly/Daiichi-Sankyo, MedRad, Medtronic, and Merck/Schering-Plough; and consulting fees/honoraria from Cordis, Eli Lilly, Medtronic, Schering-Plough, St. Jude Medical, and The Medicines Company. Dr. Makkar has received grant support from Edwards Lifesciences and St. Jude Medical; consulting fees/honoraria from Abbott Vascular, Cordis, and Medtronic; and has other financial interest in Entourage Medical. Dr. Brown has received consulting fees/honoraria from Edward Lifesciences, Medtronic, St. Jude Medical, and Abbott Vascular. Dr. Svensson has received travel reimbursement from Edwards Lifesciences related to his work as an unpaid member of the PARTNER Trial Executive Committee; holds equity in Cardiosolutions and ValvXchange; and has Intellectual Property Rights/Royalties from Posthorax. Dr. Webb is a consultant for Edwards Lifesciences. Dr. Miller is supported by an R01 research grant from the NHLBI #HL67025; has received travel reimbursements from Edwards Lifesciences related to his work as an unpaid member of the PARTNER Trial Executive Committee; and has received consulting fees/honoraria from Abbott Vascular, St. Jude Medical, and Medtronic. Drs. Smith, Leon, and Tuzcu have received travel reimbursement from Edwards Lifesciences related to their work as unpaid members of the PARTNER Trial Executive Committee. All other authors have reported that they have no relationships relevant to the contents of this paper to disclose.
- Abbreviations and Acronyms
- aortic stenosis
- aortic valve replacement
- balloon aortic valvuloplasty
- confidence interval
- New York Heart Association
- quality of life
- Society of Thoracic Surgeons
- transcatheter aortic valve replacement
- Received March 19, 2014.
- Revision received August 27, 2014.
- Accepted August 29, 2014.
- 2015 American College of Cardiology Foundation
- Schwarz F.,
- Baumann P.,
- Manthey J.,
- et al.
- Bach D.S.,
- Siao D.,
- Girard S.E.,
- Duvernoy C.,
- McCallister B.D. Jr..,
- Gualano S.K.
- Ben-Dor I.,
- Pichard A.D.,
- Gonzalez M.A.,
- et al.
- Cribier A.,
- Eltchaninoff H.,
- Bash A.,
- et al.
- Thomas M.,
- Schymik G.,
- Walther T.,
- et al.
- Rodes-Cabau J.,
- Webb J.G.,
- Cheung A.,
- et al.
- Gurvitch R.,
- Wood D.A.,
- Tay E.L.,
- et al.
- Frank S.,
- Johnson A.,
- Ross J. Jr..
- O'Neill W.W.
- McKay R.G.
- ↵(1991) Percutaneous balloon aortic valvuloplasty. Acute and 30-day follow-up results in 674 patients from the NHLBI Balloon Valvuloplasty Registry. Circulation 84:2383–2397.
- Berland J.,
- Cribier A.,
- Savin T.,
- Lefebvre E.,
- Koning R.,
- Letac B.
- Otto C.M.,
- Mickel M.C.,
- Kennedy J.W.,
- et al.
- Lieberman E.B.,
- Bashore T.M.,
- Hermiller J.B.,
- et al.
- Bonow R.O.,
- Chatterjee K.,
- de Leon A.C. Jr..,
- et al.