In patients with heart failure and improved ejection fraction (HFimpEF), withdrawing vs. continuing guideline-directed medical therapy (GDMT) with renin-angiotensin inhibitors (RASi)/angiotensin receptor–neprilysin inhibitors (ARNi) and mineralocorticoid receptor antagonists (MRA), but not beta-blockers (BB), is significantly associated with worse survival and morbidity at one year, according to a registry-based observational study published March 17 in Circulation.

In what they believe is the first registry-based study to evaluate this association, Christine Basile, MD, et al., examined data from the Swedish HF registry on the discontinuation of RASi/ARNi, MRAs and BB in patients registered between June 11, 2000 and Dec. 31, 2023 with a first recorded LVEF <40% and a later LVEF ≥40%. Withdrawal was defined as being on treatment at the first (reduced) but not the second (improved) registration.

A total of 8,728 patients were included in the analysis (median age, 70 years; 30% women). The proportion receiving each class of drug when the LVEF was <40% was 96% for RASi/ARNi, 46% for MRA and 94% for BB.

Results showed that withdrawal of GDMT with improvement in LVEF was rare: 4.4% for RASi/ARNi, 17.2% for MRA and 3.3% for BB. Predictors of withdrawal included lower use of other HF medications, higher LVEF at the later LVEF registration and a longer time between the two EF assessments (median, 12 months). Of note, 34.1% had an LVEF ≥50% at the time of improved EF.

The primary outcome, a composite of cardiovascular mortality and HF hospitalization, occurred in 9% (n=786) of the overall cohort, with a crude event rate of 9.9 per 100 patient-years.

After weighting, withdrawal was independently associated with a higher risk of the primary outcome, by 38% for RASi/ARNi and 36% for MRA. Notably, no association was found for BB withdrawal. At one year, the absolute risk difference was 6.0% and 3.9%, and the number needed to harm was 17 and 25 patients. However, in the subgroup of patients with an improved LVEF of 40-49% vs. ≥50%, BB withdrawal was associated with a higher risk of the primary outcome (p for interaction = 0.03).

The authors write that the findings support continuing RASi/ARNi and MRA in patients with HFimpEF. However, the results "highlight the need for randomized controlled trials testing BB withdrawal" in this population.