Prognostic association of medication trajectories with 3-year mortality in heart failure and preserved ejection fraction: findings from the EPICAL2 cohort study

The aims of this study were to describe combinations of beta-blockers (BB), renin-angiotensin system (RAS) blockers, and mineralocorticoid receptor antagonist (MRA) prescriptions and their trajectories in heart failure with preserved ejection fraction (HFpEF) patients, and to assess their effect on the three-year all-cause and cardiovascular (CV)-mortality. We used data from the EPICAL2 cohort of 689 hospitalized HFpEF patients. Medication prescriptions were collected at hospital discharge and at 6, 12, and 24 months after discharge. A multi-trajectory approach was used to conjointly model groups of individuals following similar trajectories over medications prescriptions. We used Cox and Fine‐Gray models, to evaluate respectively the associations between 3-year all‐cause mortality and CV-mortality and the trajectory groups. Multi-trajectory modelling revealed five distinct trajectory groups: group1 (N = 232, 33.6%) stable ACEI/ARB and BB prescriptions, group 2 (N = 199, 28.8%) stable ACEI/ARB prescription, group 3 (N = 133, 19.3%) stable BB prescriptions, group 4 (N = 78, 11.3%) stable prescriptions of none of the medications, and group 5 (N = 47, 6.8%) stable ACEI/ARB, BB, and MRA prescriptions. As compared to the group 4 of patients receiving none of the three medications, patients receiving a stable prescription of one or a combination of two or the three medications over 2 years) had a lower overall mortality over 3-year follow-up, i.e., group 1 (HR = 0.5, 95% CI 0.4–0.8), group 2 (HR = 0.6, 95% CI:0.4–0.8), group 3 (HR = 0.5, 95% CI:0.4–0.7), and group 5 (HR = 0.5, 95% CI:0.3–0.9). However, none of these trajectory groups was associated with a lower CV-mortality over 3 years. In an unselected population-based sample of HFpEF patients, the long-term stable use of the combination ACEI/ARB and BB, BB exclusively, ACEI/ARB exclusively, or the combination ACEI/ARB and BB and MRAs was associated with reduced three-year all-cause mortality.