The Danish-Norwegian randomized trial on beta-blocker therapy after myocardial infarction: Design, rationale, and baseline characteristics

Abstract

Graphical Abstract

Open in new tabDownload slide

Introduction

Beta-blockers have been widely used and recommended after myocardial infarction (MI) based on landmark trials conducted in the 1980s.^1–3 Since then, improvements in the secondary preventive therapies and introduction of coronary revascularization have improved the prognosis following MI, and the continued efficacy of beta-blocker therapy has been questioned.^4,5 Beta-blockers have an undisputed role in post-MI patients with reduced left ventricular ejection fraction (LVEF < 40%),⁶ but for patients without reduced LVEF, no adequately powered randomized controlled trial has been performed in the reperfusion era. The European Society of Cardiology has a class IIA recommendation, level of evidence (LOE) B, for long-term beta-blocker therapy in patients with acute coronary syndromes,⁷ whereas the American College of Cardiology/American Heart Association has a class IIB, LOE B recommendation.⁸

Despite wide-spread use and tolerability, adherence to beta-blocker therapy is low, and side effects are reported.^1,3,9 However, beta-blockers do reduce symptoms of angina, which has profound effects on quality of life and may decrease the need for recurrent coronary revascularization.¹⁰ Thus, the benefit of beta-blocker therapy must be weighed against the impact on side effects and quality of life.

The ‘Danish Trial of Betablocker therapy after myocardial infarction without heart failure’ (DANBLOCK)¹¹ and the Norwegian ‘BEtablocker Treatment after Acute Myocardial Infarction in revascularized patients without reduced left ventricular ejection fraction’ (BETAMI)¹² are ongoing randomized trials designed to test the effectiveness and risks of beta-blocker therapy after MI in patients with LVEF ≥ 40%. The trials have similar designs and were, from the very beginning, coordinated with the aim of conducting sub-studies on pooled data.¹¹ However, the inclusion- and event rates have been lower than expected. To enhance feasibility, the decision was made to combine the trials and publish the initial results jointly. This paper describes the design, rationale, and baseline characteristics of the first 4000 randomized patients in the combined trial.

Study design and methods

Study design and objectives

BETAMI and DANBLOCK are randomized, open-labelled, multicenter clinical trials conducted at 25 sites in Denmark and 20 sites in Norway, covering the vast majority of the countries. Both trials are investigator-initiated and funded by grants that did not influence the design or conduction of the study. The primary objective is to test whether long-term beta-blocker therapy is superior to no such therapy in patients after MI without heart failure (HF). The trials are performed in accordance with the ethical principles of the ICH-GCP Guidelines, have been approved by the relevant authorities and ethics committees, and are registered at http://register.clinicaltrials.gov. BETAMI and DANBLOCK will remain separate trials until the end of follow-up, where data from the trials will be combined and the main results published together.

Study population

The flowchart of the combined trial is presented in Figure 1. In brief, the trial will include patients with a MI within 14 days, LVEF ≥ 40%, no HF, and no absolute indication/contraindication to beta-blockers. Differences in inclusion criteria are shown in Table 1. Patients receiving beta-blocker therapy prior to randomization will not be excluded.

Figure 1

Study flow chart of the combined DANBLOCK–BETAMI trial.

Open in new tabDownload slide

Table 1

Open in new tab

Study entry criteria

Table 1

Open in new tab

Study entry criteria

Randomization and medical therapy

Patients can be randomized during their hospital stay for the index MI, at ambulatory visits, or remotely through telephone (DANBLOCK only) within 14 days. This approach was selected in order to minimize bias, and it aligns with the pragmatic nature of the study. All patients will be provided with careful oral and written information about the trial from study personnel. They will be given ample time to consider their participation. Patients will be randomized 1:1 to beta-blocker therapy or no beta-blocker therapy. The type and dosage of the beta-blocker are at the discretion of the treating cardiologist and will be initiated within 14 days from the index MI. When patients on beta-blocker therapy are randomized to no therapy, the treating cardiologist is recommended to gradually discontinue the beta-blocker as soon as clinically possible and consider alternative therapy whenever needed. Information on adherence, dosage, and crossover rates will be monitored through linkage to national prescription registries at the study end.

Endpoints and follow-up

Table 2 displays the primary and secondary endpoints obtained from national registries and self-reported questionnaires. All first events of the primary endpoint will be identified through the national registries and adjudicated by an independent clinical endpoint adjudicating committee (CEAC).¹³ Patient-related outcomes will be assessed at baseline and after 12 months in both trials. Supplementary material includes information on questionnaires, registries, diseases, and surgical codes.

Table 2

Open in new tab

Primary and secondary endpoints

Table 2

Open in new tab

Primary and secondary endpoints

Safety monitoring

The primary safety endpoint includes a composite of all-cause mortality, recurrent MI, incident HF, malignant ventricular arrhythmia, or resuscitated cardiac arrest at 30 days following randomization. The 30-day safety endpoint was chosen as results from a previous observational study have indicated an increased risk with beta-blocker discontinuation within this time frame.¹⁴ In addition, all serious adverse reactions will be reported in the study databases. Two independent Data Safety Monitoring Boards monitor safety. Trial discontinuation will be recommended in case of an imbalance in event rates or if the risk to current and future trial patients outweighs the potential impact of continuing. No interim analyses are planned.

Statistical considerations

The combined trial will be analysed as a superiority trial and aims to achieve a pre-specified number of events under modelled event rates. A power calculation has been performed in which 950 events will provide a power of 80% to detect a true treatment effect equal to a hazard ratio of 1.2 for no beta-blocker therapy. It is expected that ∼5700 randomized patients will yield the 950 desired primary endpoints based on the current event rate for patients randomized in the trial. All major treatment comparisons will be performed according to the intention-to-treat principle.

The primary endpoint is time to the first event of the primary endpoint, assessed after the last patient randomized has completed a minimum of 6 months of follow-up and after the accrual of 950 endpoints. The null hypothesis is that the rate of the composite primary endpoint in the allocated groups is equal. The primary analysis will be a Cox proportional hazards regression model with a randomization group as the main covariate. The analysis will be adjusted for inclusion site and LVEF. A hazard ratio for no beta-blocker vs. beta-blocker therapy with a 95% confidence interval will be estimated, and a P-value for the null hypothesis of a hazard ratio equal to 1 will be computed. The survival curves for the two groups will be estimated and plotted with the Kaplan–Meier estimator, and the equality of the survival curves will be tested with the log-rank test. The primary endpoint will be analysed in the following subgroups: sex, tertiles of age, country, type of MI, LVEF-values, and beta-blocker dosage tertiles.

Trial status

The first patient was randomized in October 2018 in BETAMI and in December 2018 in DANBLOCK. The inclusion rate declined during the COVID-19 pandemic; however, a total of 5228 patients were included by July 2023. Accrual of events will be monitored by linkage to the registries, and study inclusion stopped when it is anticipated that the required number of endpoints will be reached within a minimum of a 6-month follow-up period. End of inclusion is expected in January 2024, and end of follow-up in December 2024.

Baseline characteristics

The baseline characteristics of the first 4000 randomized patients are presented in Table 3. The median age (62 years) and distribution of sex (79% men) are equal. The majority had a normal LVEF at randomization (DANBLOCK:83.2%, BETAMI:84.1%). There was a difference in the rates of coronary revascularization procedures, with 99.8% of patients in BETAMI receiving the procedure compared to 88.9% in DANBLOCK.

Table 3

Open in new tab

Baseline characteristics

Table 3

Open in new tab

Baseline characteristics

Discussion

DANBLOCK and BETAMI have been designed and planned in close collaboration.^11,12 Due to lower-than-expected event and inclusion rates, in part due to the COVID-19 pandemic, the Danish and Norwegian steering committees recommended to join the trials and publish main results together, thereby increasing both the feasibility of the trials and the generalizability of the results. Management of atherosclerotic cardiovascular disease and the organization of the healthcare systems in Denmark and Norway are, overall, similar; every person is given a unique personal identity number, which allows for linkage between nationwide health registries with virtually complete follow-up and exact censoring information.¹⁵ The differences in inclusion criteria are minor and not expected to impact the results. BETAMI exclusively includes patients who are revascularized, whereas DANBLOCK does not have this criterion. However, 89% of patients in DANBLOCK underwent coronary revascularization, thereby mitigating this difference. Both trials include patients with mildly reduced LVEF. The proportion of patients in this subgroup was relatively low but similar in the two trials. Information on the distribution of LVEF in MI patients is not available in the registries. It is possible that recruitment of patients in this subgroup may be lower as some physicians hesitate to abstain from beta-blocker therapy in patients with mildly reduced LVEF due to contradictory findings from previous studies.^16,17

Justification of the primary endpoint

The primary endpoint is presented in Table 2. The original endpoints of BETAMI and DANBLOCK are displayed in the Supplementary Material. The components of the primary endpoint are chosen for the clinical relevance and plausibility of the effect of beta-blocker therapy. Beta-blockers exert negative chronotropic and ionotropic effects, possess anti-arrhythmic qualities, and reduce heart rate, blood pressure, cardiac contractility, oxygen demand, and angina symptoms.^18,19 The primary composite endpoint provides a comprehensive picture of the effect of beta-blockers for future recommendations, increases the overall eventrate, and enhances statistical efficiency, leading to increased feasibility.

Unstable angina and coronary revascularization

Unstable angina was a component of the original primary endpoint in DANBLOCK. It was decided to replace unstable angina with coronary revascularization for several reasons. The introduction of highly sensitive troponins has affected the incidence of unstable angina, and its continued relevance in clinical trials has been questioned.²⁰ Even though adjudication criteria exist, the diagnosis comes with a risk of ascertainment and reporting bias. In contrast, coronary revascularization offers advantages such as ease of evaluation, clinical feasibility, and less information bias. Coronary revascularization is closely linked to MI and angina, consequently, it may be affected by beta-blocker therapy.

Incident heart failure

In the combined trial, incident HF includes diagnoses from hospitalizations or outpatient clinics. Traditional definitions of cardiovascular outcomes in trials specify HF as hospitalization due to worsening HF.¹³ However, in this trial, no patients had HF at inclusion, and in recent years, HF is increasingly diagnosed and managed at outpatient clinics. In Denmark, incident HF diagnosed at outpatient clinics increased from 23% in 1997 to 61% in 2017.²¹ If only hospitalized HF were considered, the number of events would be considerably underestimated. Additionally, there has been an increased focus on using non-hospitalized HF events as endpoints in cardiovascular trials.^13,22,23 To ensure high validity of the endpoints, the HF endpoint and the additional components of the primary composite endpoint will undergo adjudication by a CEAC.

Dosage of beta-blocker therapy

The beta-blocker dosages used in the landmark trials were high.^1,2 The current trial follows a pragmatic approach without study visits to increase dosage but with a recommendation to increase dosage at the general practitioner. It is expected that a minority of participants will be uptitrated. The trial will thereby mimic current clinical practice, as patients are typically prescribed lower beta-blocker dosages in a contemporary setting.²⁴

Contemporary beta-blocker trials

A randomized trial published in 2018 allocated 801 post-MI patients to carvedilol or no such therapy.²⁵ They found no beneficial effects of beta-blockers after a median follow-up of 3.9 years. The trial was, however, not powered to detect a difference in the treatment effect for the primary endpoint.

In addition to the combined BETAMI–DANBLOCK trial, three randomized clinical trials are currently evaluating beta-blocker therapy after MI in patients without HF (REDUCE,²⁶ REBOOT,²⁷ and ABYSS²⁸). The trials are presented in Table 4. A meta-analysis combining REDUCE, REBOOT, and BETAMI-DANBLOCK is planned. Altogether, the trials will be able to perform analyses on subpopulations, and resolve the question of beta-blocker therapy after MI in patients without HF.

Strengths and limitations

The combined trial has several strengths, including its randomized, pragmatic design and extensive long-term follow-up through national registries lowering costs, minimizing attrition, and increasing feasibility. Even though the registries have high diagnostic accuracies, an independent CEAC will review and validate all components of the primary endpoint, adding to the robustness of the findings. The utilization of data from national registries also allows for the assessment of treatment adherence and crossover rates.

Nevertheless, it is important to acknowledge certain limitations. The primary endpoint was changed during the enrollment period. This modification was necessary to enable the combination of the trials and was done with no knowledge of the distribution of endpoints. The open-label nature of the trial introduces the possibility of bias as participants and study personnel are aware of the treatment arm. Participants' previous knowledge of and experiences with beta-blockers may influence their expectations to the therapy or lack thereof. In turn, this may influence both adherence to the allocated treatment as well as responses to the patient-reported outcome measures during follow-up. Furthermore, the lack of placebo in itself introduces a theoretic overestimation of the effect of beta-blocker therapy on the patient-reported outcomes. We aim to reduce these limitations by collecting all primary and key secondary outcomes from nationwide registries. In addition, all primary outcomes will be adjudicated by an independent CEAC adhering to strict and objective criteria. Nevertheless, sub-studies with an emphasis on patient-related outcomes should be interpreted with these limitations in mind.

Conclusion

The trial will provide evidence for the efficacy of long-term beta-blocker therapy after MI in patients with normal or mildly reduced LVEF and will potentially affect current clinical practice.

Acknowledgement

All primary investigators, study nurses, and members of the steering committees should be acknowledged for their contribution in the DANBLOCK and BETAMI trials. All patients who gave their informed consent to participate in the trials should, furthermore, be acknowledged.

Funding

The DANBLOCK trial is supported by grants from the Danish Heart Foundation [18-R124-A8323-22091] and the Novo Nordisk Foundation [NNF18OC0034582]. The BETAMI trial has received grants from the Health South-East research program in Norway [2 017 205] and the Research Council of Norway [302 454]. These funding organizations are not otherwise involved in the trial.

Authors contributions

All authors have made substantial contributions to the conception and design of the trial, have given their final approval of the manuscript to be published, and have agreed to be accountable for all aspects of the work in ensuring that questions related to the accuracy or integrity of any part of the work are appropriately investigated and resolved. A.M.K. and J.M. drafted the manuscript. E.P., D.A., S.H., M.H.O., T.S.G.S., A.B., V.R., M.W.F., and T.L. revised the manuscript critically for important intellectual content.

Conflict of interest: None of the authors report COIs of relevance for the present work. J.M. report lecture fees from Novartis, Boehringer Ingelheim, and Sanofi outside the submitted work. T.S.G.S. report lecture fees from AstraZeneca.

Data availability

Due to confidentiality agreements, supporting data cannot be made openly available.

References

Author notes