Aims To test the hypothesis that an earlier post-acute myocardial infarction (AMI) eplerenone initiation in patients with left ventricular systolic dysfunction (LVSD) and heart failure (HF) is associated with better long-term outcomes.Methods and resultsThe 6632 patients of the EPHESUS study were randomized from day 3 to 14 after the index AMI (median = 7 days), of these 3319 were assigned to eplerenone. We analysed the differential effects of time-to-eplerenone initiation vs. placebo, based on the median time to initiation of treatment (<7 days-'earlier', ≥7days-'later'). Effects on outcomes were evaluated over a mean 16-month follow-up, using Cox proportional hazards regression analysis. The earlier eplerenone initiation (<7 days) reduced the risk of all-cause mortality by 31 (P = 0.001) when compared with the 'earlier' placebo' and also reduced the risks of cardiovascular (CV) hospitalization/CV mortality by 24 (P < 0.0001) and sudden cardiac death (SCD) by 34 (P < 0.0001). In contrast, later eplerenone initiation (≥7 days) had no significant effect on outcomes. Interactions between time-to-randomization and treatment were significant. These associations remained substantially unchanged after risk adjustment in multivariable models.ConclusionAn earlier eplerenone administration (3-7days) post-AMI improved outcomes in patients with LVSD and HF. This benefit was not observed when eplerenone was initiated later (≥7days).