Author + information
- Received February 16, 2018
- Revision received February 28, 2018
- Accepted February 28, 2018
- Published online May 14, 2018.
- Mônica Samuel Avila, MDa,∗,
- Silvia Moreira Ayub-Ferreira, MD, PhDa,∗,
- Mauro Rogerio de Barros Wanderley Jr., MDa,
- Fatima das Dores Cruz, RNa,
- Sara Michelly Gonçalves Brandão, RNa,
- Vagner Oliveira Carvalho Rigaud, PhDa,
- Marília Harumi Higuchi-dos-Santos, MD, PhDc,
- Ludhmila Abrahão Hajjar, MD, PhDb,c,
- Roberto Kalil Filho, MD, PhDb,c,
- Paulo Marcelo Hoff, MD, PhDc,
- Marina Sahade, MDc,
- Marcela S.M. Ferrari, MDc,
- Romulo Leopoldo de Paula Costa, MDc,
- Max Senna Mano, MD, PhDc,
- Cecilia Beatriz Bittencourt Viana Cruz, MDb,c,
- Maria Cristina Abduch, VMDb,
- Marco Stephan Lofrano Alves, MD, PhDb,
- Guilherme Veiga Guimaraes, PhDa,
- Victor Sarli Issa, MD, PhDa,
- Marcio Sommer Bittencourt, MD, MPH, PhDb,c,d and
- Edimar Alcides Bocchi, MD, PhDa,∗ ()
- aHeart Failure Department, Heart Institute (InCor) do Hospital das Clínicas da Faculdade de Medicina da Universidade de São Paulo, São Paulo, Brazil
- bHeart Institute (InCor) do Hospital das Clínicas da Faculdade de Medicina da Universidade de São Paulo, São Paulo, Brazil
- cInstituto do Câncer do Estado de São Paulo-Universidade de São Paulo, São Paulo, Brazil
- dCenter for Clinical and Epidemiological Research, University Hospital, University of São Paulo, São Paulo, Brazil
- ↵∗Address for correspondence:
Dr. Edimar Alcides Bocchi, Heart Failure Department, Heart Institute (InCor) do Hospital das Clínicas da Faculdade de Medicina da Universidade de São Paulo, R. Dr. Eneas de Carvalho Aguiar 44, São Paulo 05403900, Brazil.
Background Anthracycline (ANT) chemotherapy is associated with cardiotoxicity. Prevention with β-blockers remains controversial.
Objectives This prospective, randomized, double-blind, placebo-controlled study sought to evaluate the role of carvedilol in preventing ANT cardiotoxicity.
Methods The authors randomized 200 patients with HER2-negative breast cancer tumor status and normal left ventricular ejection fraction (LVEF) referred for ANT (240 mg/m2) to receive carvedilol or placebo until chemotherapy completion. The primary endpoint was prevention of a ≥10% reduction in LVEF at 6 months. Secondary outcomes were effects of carvedilol on troponin I, B-type natriuretic peptide, and diastolic dysfunction.
Results Primary endpoint occurred in 14 patients (14.5%) in the carvedilol group and 13 patients (13.5%) in the placebo group (p = 1.0). No differences in changes of LVEF or B-type natriuretic peptide were noted between groups. A significant difference existed between groups in troponin I levels over time, with lower levels in the carvedilol group (p = 0.003). Additionally, a lower incidence of diastolic dysfunction was noted in the carvedilol group (p = 0.039). A nonsignificant trend toward a less-pronounced increase in LV end-diastolic diameter during the follow-up was noted in the carvedilol group (44.1 ± 3.64 mm to 45.2 ± 3.2 mm vs. 44.9 ± 3.6 mm to 46.4 ± 4.0 mm; p = 0.057).
Conclusions In this largest clinical trial of β-blockers for prevention of cardiotoxicity under contemporary ANT dosage, the authors noted a 13.5% to 14.5% incidence of cardiotoxicity. In this scenario, carvedilol had no impact on the incidence of early onset of LVEF reduction. However, the use of carvedilol resulted in a significant reduction in troponin levels and diastolic dysfunction. (Carvedilol Effect in Preventing Chemotherapy-Induced Cardiotoxicity [CECCY]; NCT01724450)
Despite advances in the survival of patients with cancer, their prognosis remains limited by complications that are frequently treatment related (1). Cardiovascular effects of chemotherapeutic agents are responsible for a significant proportion of severe complications, particularly among female patients with breast cancer (2). One of the most widely used agents (3,4), anthracyclines (ANT), is responsible for early and late dose-related cardiotoxicity, particularly overt heart failure (HF) (5–7).
Several primary and secondary prevention strategies have been proposed to reduce ANT-induced cardiotoxicity (8). These strategies extrapolate guideline-oriented therapies for HF to patients with chemotherapy-induced cardiomyopathy. This includes the use of β-blockers as a mainstay HF treatment based on their beneficial effects on neurohumoral activation, symptoms, and prognosis (9,10). In nonrandomized studies of secondary prevention of cardiotoxicity with enalapril and carvedilol, recovery of left ventricular ejection fraction (LVEF) was reported with early treatment after initial clinical presentation (11,12). Therefore, an earlier approach could determine greater success in the management of cardiotoxicity.
However, the use of β-blockers for primary prevention of cardiotoxicity remains controversial. Previous trials supporting β-blocker use have many design limitations, such as being observational, open-label, single-blind, combined cardiovascular drugs, or limited number of patients (13–16). On the contrary, in the recent PRADA (Prevention of Cardiac Dysfunction During Adjuvant Breast Cancer Therapy) trial, metoprolol succinate was not associated with a reduction in the incidence of cardiotoxicity (17). However, carvedilol and metoprolol succinate have distinct pharmacologic properties (18,19).
In the face of controversial evidence supporting the use of β-blockers for primary prevention of anthracycline-induced cardiotoxicity (7,8,20), we conducted the CECCY (Carvedilol Effect in Preventing Chemotherapy-Induced Cardiotoxicity) trial, the aim of which was to test the effects of carvedilol for primary prevention of cardiotoxicity.
The CECCY trial was a prospective, double-blind, randomized, placebo-controlled study conducted in the Heart Failure Department of Heart Institute (InCor) and the Cancer Institute, São Paulo, Brazil. Patients were referred from the Cancer Institute, and the Heart Failure Team of the Heart Institute was responsible for allocation, randomization, and optimization of the dose of carvedilol/placebo. Data were collected, managed, and analyzed by the Heart Failure Team after the end of the study.
The institutional review board at both institutions approved the trial protocol (Figure 1). All participants were informed about the research objectives, research protocol, and treatment alternatives involved in the study, and all participants provided written informed consent to participate in the study. The trial was registered at the ClinicalTrials.gov (NCT01724450) before study initiation.
We included all consecutive patients with HER2-negative breast cancer tumor status and therapy that included anthracycline, cyclophosphamide, and taxane from April 23, 2013 to January 3, 2017. The standard chemotherapy protocol comprised 4 cycles of cyclophosphamide 600 mg/m2 and doxorubicin 60 mg/m2 every 21 days (with a total cumulative dose of 240 mg/m2), followed by paclitaxel 80 mg/m2 weekly for 8 weeks.
Eligibility requirements included an age of at least 18 years, diagnosis of invasive breast adenocarcinoma, with an indication for adjuvant or neoadjuvant therapy.
Exclusion criteria included the impossibility of left ventricular (LV) function evaluation; prior history of chemotherapy or radiation; HF symptoms; prior diagnosis of cardiomyopathy, coronary artery disease, moderate to severe mitral and aortic disease; use of angiotensin-converting enzyme inhibitors, angiotensin receptor blockers, or β-blockers; contraindication to the use of β-blockers; and patients with HER2 expression.
Randomization, allocation, and intervention
Randomization was on a 1:1 ratio to receive carvedilol or placebo. The randomization included a pre-defined stratification according to menopause status considering the potential difference in cardiotoxicity risk between pre- and post-menopause (21). The randomization was performed with a computer-based randomization list in blocks of 4, including 2 carvedilol and 2 placebos. Data on randomization and allocation were maintained in the custody of an independent research pharmacy at the Cancer Institute.
Carvedilol and placebo were administered in a progressive manner with incremental dosing at 3-week intervals beginning with a dose of 3.125 mg twice a day, which was then increased to 6.25 mg, then to 12.5 mg, to a maximum dose of 25 mg every 12 h or until the appearance of intolerable symptoms or heart rate ≤60 beats/min or systolic blood pressure <110 mm Hg. Carvedilol and placebo were continued until completion of chemotherapy.
All eligible patients underwent a baseline transthoracic echocardiogram and routine laboratory tests including biomarkers before randomization. If the patient met the eligibility criteria, the randomization was performed. After randomization, the medication was initiated on the first day of chemotherapy. The following sequential measurements of biomarkers were performed in a median of 19 days after each ANT cycle. The echocardiography, troponin I (TnI), and B-type natriuretic peptide (BNP) follow-up evaluations are summarized in Figure 1.
The quantitative TnI determination was obtained by means of a 3-step sandwich immunoassay using direct chemiluminescent technology and constant amounts of 2 monoclonal antibodies. An auxiliary reagent is included to reduce nonspecific binding using the ADVIA Centaur TnI-Ultra commercial kit (Siemens Healthcare Diagnostics, Tarrytown, New York). The level of detection was 0.006 ng/ml. Levels <0.006 were reported as 0.005 ng/ml. The normal range of TnI was <0.04 ng/ml. Plasma concentration of BNP was obtained with a 2-step sandwich immunoassay using direct chemiluminescent technology and constant amounts of 2 monoclonal antibodies using the commercial kit ADVIA Centaur (Siemens Healthcare, Malvern, Pennsylvania). The results are presented in pg/ml. The level of detection was 2 pg/ml. Levels <2 are reported as 1 pg/ml.
Transthoracic echocardiography was performed with a commercially available system (Envisor Philips, Philips Healthcare, Andover, Massachusetts). All the measurements were performed and reported according to the recommendations of the American Society of Echocardiography (22). LVEF was measured by Simpson rule, throughout apical 4- and 2-chamber views. We also assessed the following echocardiographic parameters: left atrium diameter, interventricular septum diameter, posterior wall thickness, LV end-diastolic diameter, LV end-systolic diameter, and mitral inflow with the use of Doppler echocardiography (23). Diastolic function was evaluated by mitral inflow E/A pattern, E/A ratio, E velocity deceleration time, annular tissue Doppler curves (e′/a′), and E/e′ ratio (22). All echocardiography data was stored including the original DICOM (Digital Imaging and Communications in Medicine) images. These images were stored on a secure institutional drive. All scans were read by experienced board-certified echocardiographers who were blinded to all clinical characteristics.
The primary endpoint was an early onset drop in LVEF of at least 10% from baseline until the end of chemotherapy at 6 months (24). The secondary endpoints were changes in the levels of TnI (TnI >0.04 ng/ml), BNP, and in diastolic dysfunction recommended by guidelines at the time of the study design (23).
The sample size was calculated with an expected incidence of cardiotoxicity of 23% with the use of ANT and an expected reduction to 8% with the addition of β-blockers (15). We estimated a loss of follow-up of 5%, statistical significance level of 95%, and 80% power.
Quantitative variables are expressed as mean ± SD when normally distributed or as median (interquartile range) if the hypothesis of normality was rejected. For troponins and BNP levels, we performed a natural logarithm (ln) transformation to normalize the data. Categorical variables were compared using the Fisher exact test, including the comparison of the prevalence of LV dysfunction at the end of chemotherapy and occurrence of side effects. For the clinical outcomes, we performed a log-rank test to compare the incidence of events across groups.
For the longitudinal analysis of the echocardiography parameters, ln-transformed troponins, and BNP, we plotted the mean and 95% confidence intervals stratified by treatment group on linear graphs. We then constructed longitudinal linear mixed-effects models to compare the 2 treatment groups. For the longitudinal mixed models, we constructed models including the interaction of time and treatment group with a fixed intercept, assuming baseline measurements of each predictor were similar across groups due to the randomized design of the study. The difference between groups was considered significant if the interaction term between group and time was significant. Due to the previously known nonlinearity of the evolution of troponin over time after ANT use, which was also seen in our data, we pre-specified an additional model for the evaluation of ln-transformed troponin levels across time. In this model, time was included as a quadratic function by the inclusion of first- and second-order time variables to accommodate the biphasic behavior of the troponin levels in each group. Then, we evaluated the interaction of time and time squared with the treatment group to assess the longitudinal differences in the distribution of the ln troponin over time. Additionally, to provide a single overall level of significance for the difference in the TnI levels between the 2 groups, we then constructed a model without interaction terms and compared it with a model including both interaction terms using a likelihood ratio test as 1 model was nested into the other. For the longitudinal analysis of diastolic dysfunction, we have used generalized estimating equations using the logit function, as the outcome was binary. For the analysis of the difference in prevalence of diastolic dysfunction, an interaction term of treatment group and medication was created, and the analysis was performed with the same strategy described for the longitudinal mixed-effects models used for the continuous variables. The statistical analysis was performed using Stata version 14.0 (StataCorp, College Station, Texas) and the level of significance was defined as p <0.05.
From April 2013 to January 2017, we screened 1,122 consecutive patients with breast cancer referred for ANT chemotherapy (Figure 2). We randomized 200 patients; 8 of them had no valid randomization. The reasons for an invalid randomization were change in chemotherapy scheme (n = 4) and error in the information about menopausal status (n = 4). Thus, 192 patients were randomly assigned to receive carvedilol or placebo for the intention-to-treat analysis. The baseline characteristics of the patients were statistically balanced across groups (Table 1). Online Table 1 demonstrates the percentage of patients at the different dose ranges of carvedilol or placebo. Also, the mean serum creatinine levels were also balance between groups, 0.70 ± 0.12 mg/dl in carvedilol group and 0.723 ± 0.13 mg/dl.
During the follow-up, 27 patients (14.0%) had a decrease of at least 10% in LVEF at 6 months after the initiation of chemotherapy. Of these patients, 14 (14.5%) were in the carvedilol group and 13 (13.5%) were in the placebo group (p = 1.0). In an exploratory manner, considering an alternative definition of cardiotoxicity as a decrease of 10 percentage points to a value below the lower normal value of 55% in our center (25), only 1 patient (1%) in the placebo group and none in the carvedilol group fulfilled criteria for cardiotoxicity and had a decrease of LVEF to 35% (Table 2).
The mean baseline LVEF was 65.2 ± 3.6% in the placebo group and 64.8 ± 4.7% in the carvedilol group. After 6 months of chemotherapy, LVEF was 63.9 ± 5.2% in the placebo group and 63.9 ± 3.8% in the carvedilol group (Table 2), a nonsignificant absolute LVEF reduction of 1.3% in the placebo group and 0.9% in the carvedilol group. No changes in LVEF from baseline throughout the 6 months were noted between groups (p = 0.84) (Figure 3A).
Taking into account menopausal status, no difference was seen between carvedilol and placebo in both subgroups pre- and post-menopause.
During the follow-up, a trend toward a less-pronounced increase in LV end-diastolic diameter was noted in the carvedilol group compared with placebo (from 44.1 ± 3.3 mm to 45.2 ± 3.2 mm vs. from 44.9 ± 3.6 mm to 46.4 ± 4.0 mm, respectively; p = 0.057) (Figure 3B). No differences across groups in left atrium septal or posterior wall or LV end-systolic diameter were noted over time (Online Tables 2 to 7).
TnI levels increased from baseline until the end of the study in both groups (Table 2). However, this increase of TnI levels over time was attenuated in the carvedilol arm (p = 0.003) (Central Illustration). Sixty-five patients (33.8%) had plasma levels of TnI >0.04 ng/ml. Of these, 25 (26.0%) were in the carvedilol and 40 (41.6%) were in the placebo group (p = 0.03) (Table 2, Online Figure 1). After the end of ANT cycles, a drop occurred in TnI levels in both groups (Central Illustration). For BNP, however, no difference occurred across groups over the course of follow-up (p = 0.85) (Table 2).
We found a lower incidence of diastolic dysfunction in the carvedilol compared with the placebo group (p = 0.039) (Table 2). The incidence of diastolic dysfunction in the placebo group increased progressively during ANT treatment, but not in the carvedilol group (Online Figure 2). The majority (91%) of diagnosed diastolic disjunction cases were classified as diastolic dysfunction grade 1, relaxation abnormality.
No differences were found in the incidence of clinical events across groups. Two deaths (2.1%) occurred in the placebo and 2 (2.1%) in the carvedilol group (p = 1.00), all due to cancer progression. In the placebo group, we observed 1 case of de novo HF and 1 case of asymptomatic atrial flutter with normal LVEF. A detailed description of all clinical events is presented in Online Table 7.
A significant difference occurred in both systolic and diastolic blood pressure in the carvedilol compared with the placebo group at 12 and 24 weeks (Online Figures 3 and 4). Similarly, heart rate in the carvedilol group was lower than that in the placebo group from 6 weeks up to 24 weeks (Online Figure 5).
No differences occurred in the incidence of side effects or carvedilol/placebo discontinuation across groups. The most common adverse event in the placebo group was dizziness and in the carvedilol group symptomatic hypotension (Online Table 8). No serious adverse events occurred, and carvedilol was well tolerated. Nine patients stopped the drug due to side effects, 6 (6.2%) in the placebo group and 3 (3.1%) in the carvedilol group (p = 0.30).
To the best of our knowledge, the CECCY trial is the largest prospective randomized double-blind trial evaluating the use of cardiovascular drugs for primary prevention of ANT cardiotoxicity. In patients receiving contemporary doses of ANT, a low prevalence of cardiovascular comorbidities and risk factors for cardiotoxicity, a small percentage of patients experienced an early onset reduction in LVEF, which was not influenced by the use of carvedilol. Nevertheless, carvedilol was associated with attenuated peak levels of TnI and a lower percentage of patients experiencing increases in serum TnI levels. Also, its use was associated with a trend toward a less pronounced increase in LV diastolic diameter and a reduction in the percentage of patients with diastolic dysfunction.
Our TnI results are in accordance with those of a small single-blind trial showing TnI reduction with carvedilol (26). The effect of carvedilol on attenuation of troponin levels is not well understood. Antioxidant pharmacological proprieties of carvedilol and its protection against free radicals could contribute to its beneficial effects (1). The attenuation of TnI increase could improve prognosis of patients receiving high-dose ANT treatment. High TnI predicts subsequent LVEF reduction and cardiac events (1,27–29). However, there are currently no data showing that troponin-based management improves cardiovascular outcomes in this population. In fact, the correlation between TnI levels and LVEF reduction has been reported in coronary artery disease but not in diffuse myocardial injury or other causes of cardiac troponin release not associated with myocardial ischemia (30). Troponin cardiac release may result from chronic cardiac damage without acute myocardial necrosis (30). In this scenario, evidence exists that any increased serum levels of troponin might be associated with an impaired outcome in various clinical settings (30). Accordingly, our observed discordance between the elevations of troponin without a change in LVEF is not well established. The elevation of troponin in our study was a mild elevation, barely exceeding the upper limit of mortality, demonstrating that chemotherapy with ANT at these doses may lead to myocardial injury, but this lesion may not be expressive enough to affect myocardial function.
Our results concerning LVEF changes are closed to the recent PRADA trial results with contemporary ANT dosage (17). PRADA also did not demonstrate the protective effect of metoprolol succinate in the primary prevention of cardiotoxicity. This study showed a numerically modest absolute reduction in the LVEF in both groups (placebo group drop of 1.8% and 1.6% in metoprolol group). In our study, the decrease was even lower (drop of 1.3% in placebo group and 0.9% in carvedilol group), though the technique for measuring LVEF was different in the 2 studies, as cardiac magnetic resonance imaging was performed in the PRADA trial.
In contrast with our findings, several randomized trials reported a higher incidence of cardiotoxicity and a beneficial effect of carvedilol or nebivolol in prevention of cardiotoxicity (15,16,26,31,32). Potential explanations for these apparent discrepancies may include small sample sizes, previously reported higher doses of ANT, heterogeneity of included populations with differences in risk factors, cardiovascular comorbidities, types of cancer and chemotherapy protocols, and differences in follow-up duration and in study methodology. Patients receiving high cumulative doses of ANT and with a higher prevalence of risk factors have a high risk for cardiotoxicity (25).
Despite the lack of differences in the incidence of clinically defined cardiotoxicity across our groups, the use of carvedilol was associated with a tendency toward reduction in LV end-diastolic diameter and attenuation of worsening of diastolic LV function. It suggests that carvedilol could influence LV remodeling in this setting.
First, the clinical trial was conducted in a single center. Despite that, the sample was a representative population of breast cancer patients, and it is, to the best of our knowledge, the largest study using cardiovascular drugs for cardiotoxicity prevention. Second, the incidence of early onset cardiotoxicity was 13.5% to 14.5%, lower than we expected, which might reduce the statistical power to study carvedilol effects. Third, the dose of carvedilol was optimized during chemotherapeutic treatment and may indicate that the target dose was reached at a later stage of chemotherapy. However, the effect of carvedilol was already noted early at 6 weeks based on heart rate. Also the maximal dose of carvedilol was obtained before the time of the peak of TnI levels and the highest cumulative dose of ANT. Fourth, the maximum tolerated dose of carvedilol and placebo less than what was expected. This finding could be predictable in the carvedilol group but not in the placebo group, suggesting an unknown mechanism of chemotherapy treatment and cancer impairing the titration of the drug even in the placebo arm. Fifth, other parameters to assess diastolic dysfunction have been published more recently (33), but this does not invalidate the evaluation in the present study, because all parameters that were used are extremely reproducible and have had an important role in the evaluation of diastolic dysfunction and in the prognosis in predicting HF (34). Sixth, the interobserver variability might have influenced the repeated LVEF measurements, though this is unlikely to affect the point estimates of our findings, as this equally affects both groups across time. Seventh, as the primary endpoint was not met, any statements related to the secondary endpoints need to carefully interpreted, although the additional information provided by the changes in troponin levels and diastolic function over time adds meaningful information and may aid on the mechanistic understanding of the pathophysiology of LV dysfunction in this scenario. And finally, the short follow-up period (6 months) could be considered a limitation. Nevertheless, early reduction in LVEF of 10% has been previously reported at the end of chemotherapy in patients developing cardiotoxicity, and indeed the detection of the initial disease allows the possibility of prompt treatment and consequent recovery of LVEF (12).
The incidence of early onset cardiotoxicity with contemporary doses of ANT was lower than expected with moderate to high doses of ANT. In this scenario, the use of carvedilol did not result in significant changes in LV function in up to 6 months of follow-up. However, the TnI elevation was significant and attenuated by carvedilol use, suggesting a protective effect of carvedilol in myocardial injury.
COMPETENCY IN PATIENT CARE AND PROCEDURAL SKILLS: In patients undergoing ANT chemotherapy, carvedilol might prevent or delay the onset of cardiac injury and adverse myocardial remodeling.
TRANSLATIONAL OUTLOOK: Clinical trials with longer follow-up are required to evaluate the safety and efficacy of carvedilol for prevention of cardiotoxicity in patients receiving ANT chemotherapy, determine optimum timing and dosing regimens, and delineate the impact of treatment on both cardiac function and clinical outcomes.
The authors extend special thanks to Maria de Lourdes Ribeiro for supporting the administrative logistics of the trial.
↵∗ Drs. Avila and Ayub-Ferreira contributed equally to this article and are joint first authors.
This study was supported by Fundação de Amparo à Pesquisa do Estado de São Paulo (FAPESP 2010/18078-8) as part of Mônica Samuel Avila’s doctoral thesis under the guidance of Silvia Moreira Ayub-Ferreira. Baldacci Laboratories donated the carvedilol and placebo used in this study. The Baldacci Laboratories did not participate in any phase of the study. Dr. Issa has received speaking honoraria from Novartis. Dr. M. Bittencourt has received unrestricted funding support from Sanofi; and speaking honoraria from Boston Scientific. Dr. Bocchi has received consulting fees from Servier and AstraZeneca; subsidized travel/hotel/registration fees and other honoraria from Servier; serves on steering committees for Servier and Novartis; and is a contracted researcher for Jansen and Bayer/Merck. All other authors have reported that they have no relationships relevant to the contents of this paper to disclose.
- Abbreviations and Acronyms
- B-type natriuretic peptide
- heart failure
- natural logarithm
- left ventricular/ventricle
- left ventricular ejection fraction
- troponin I
- Received February 16, 2018.
- Revision received February 28, 2018.
- Accepted February 28, 2018.
- 2018 American College of Cardiology Foundation
- Bloom M.W.,
- Hamo C.E.,
- Cardinale D.,
- et al.
- Ewer M.S.,
- Ewer S.M.
- Valachis A.,
- Nilsson C.
- Henriksen P.A.
- Vejpongsa P.,
- Yeh E.T.
- Hamo C.E.,
- Bloom M.W.
- Cardinale D.,
- Colombo A.,
- Lamantia G.,
- et al.
- Cardinale D.,
- Colombo A.,
- Bacchiani G.,
- et al.
- Seicean S.,
- Seicean A.,
- Alan N.,
- Plana J.C.,
- Budd G.T.,
- Marwick T.H.
- Bosch X.,
- Rovira M.,
- Sitges M.,
- et al.
- Kalay N.,
- Basar E.,
- Ozdogru I.,
- et al.
- Yue T.L.,
- Cheng H.Y.,
- Lysko P.G.,
- et al.
- Khosrow-Khavar F.,
- Filion K.B.,
- Al-Qurashi S.,
- et al.
- Lang R.M.,
- Badano L.P.,
- Mor-Avi V.,
- et al.
- National Institutes of Health
- Zamorano J.L.,
- Lancellotti P.,
- Rodriguez Muñoz D.,
- et al.
- Nabati M.,
- Janbabai G.,
- Baghyari S.,
- Esmaili K.,
- Yazdani J.
- Sawaya H.,
- Sebag I.A.,
- Plana J.C.,
- et al.
- Cardinale D.,
- Colombo A.,
- Torrisi R.,
- et al.
- Cardinale D.,
- Sandri M.T.,
- Colombo A.,
- et al.
- Tashakori Beheshti A.,
- Mostafavi Toroghi H.,
- Hosseini G.,
- Zarifian A.,
- Homaei Shandiz F.,
- Fazlinezhad A.
- Nagueh S.F.,
- Smiseth O.A.,
- Appleton C.P.,
- et al.