Long-term effect of beta-blockers after myocardial infarction with preserved left ventricular ejection fraction in a per protocol analysis of the REDUCE-AMI trial
A Humphreys, R Hofmann, C Olarte Parra, A Berglund, B Lindahl, M Hernan, A Matthews, T JernbergAbstract
Background
The multinational registry-based pragmatic randomised trial REDUCE-AMI showed that initiation of beta-blocker treatment in individuals with myocardial infarction and preserved left ventricular ejection fraction (LVEF ≥ 50%) did not reduce the risk of death or recurrent myocardial infarction compared with no beta-blockers in an intention-to-treat analysis (hazard ratio 0.96, 95% confidence interval 0.79, 1.16) (1). However, among participants in Sweden who attended follow-up visits, 18.1% of those assigned to beta-blockers were reported to have discontinued treatment by 11–13 months, while 14.3% of those assigned to no beta-blockers were reported to have initiated beta-blockers. A secondary analysis of the trial is required to assess the per-protocol effect of beta-blockers compared with no beta-blockers under full adherence to the treatment strategies.
Methods
We restricted the per protocol analysis to the Swedish participants of the REDUCE-AMI trial for whom linked longitudinal data were available. All dispensation data relating to the beta-blockers metoprolol succinate and bisoprolol (the treatment strategies defined in the trial protocol) were manually extracted. Individuals were censored if and when they deviated from their assigned treatment strategy, taking into consideration time varying indications and contraindications (2). Risks were estimated using pooled logistic regression models with adjustment via inverse probability weighting for baseline and time-varying covariates associated with non-adherence to account for selection bias introduced by informative censoring.
Results
The study population was composed of 4,786 individuals. The median follow up was 30 months in the beta-blocker group and 38 months in the no beta-blocker group. Adherence was 71.8% in the beta-blocker group, and 89.7% in the no beta-blocker group. The estimated 5-year adjusted risk of the composite outcome of death or recurrent myocardial infarction was 10.2% (8.5%, 12.0%) in the beta-blocker group and 11.7% (10.1%, 13.4%) in the no beta-blocker group, corresponding to a risk difference of -1.5% (-3.9%, 1.0%), a risk ratio of 0.87 (0.70, 1.10), and an average hazard ratio by 5 years of 0.87 (0.70, 1.10).
Discussion
This per protocol analysis found little evidence of a large effect of beta-blockers on the composite outcome of death or myocardial infarction in individuals with myocardial infarction and preserved left ventricular ejection fraction, consistent with intention-to-treat estimates.Estimated risk of death or MI