Abstract Background Guideline-directed medical therapy reduces mortality/morbidity in heart failure (HF) with reduced ejection fraction (EF). However, evidence about its use and prognostic role once EF has improved is limited. Purpose To assess use and associations with mortality/morbidity of renin-angiotensin system inhibitors (RASi), angiotensin-receptor neprilysin inhibitors (ARNi), beta-blockers (BBL), and mineralocorticoid receptor antagonists (MRA) in patients with HF and improved EF (HFimpEF). Methods We analyzed patients with HFimpEF, defined as a first recorded EF <40% and a subsequent EF ≥40%, registered in the Swedish HF registry between 2004 and 2021. Index date was the SwedeHF registration reporting the second (i.e. improved) EF assessment. To reduce the bias related to the initial recommendation for MRA limited to more severe HF (NYHA III-IV), the assessment of MRA was restricted to patients with index date 2016 or later, when Swedish guidelines on HF included a recommendation for MRA in patients with NYHA class II–IV. The association between treatment use and the composite outcome cardiovascular mortality (CVM)/HF hospitalization (HHF) (with censoring >3 years) was assessed by Cox proportional hazard models in a 2:1 propensity score-matched cohort. Propensity scores for treatment use were calculated by a logistic regression model including 38 variables, achieving a standardized mean difference <0.1 for all evaluated variables. Since matching reduces the sample size and may limit generalizability, a Cox proportional hazard model was fitted in the overall cohort adjusting, rather than matching, for the propensity score (Overall HR). Results Of 4700 patients with HFimpEF (mean age 68±12 yrs, 70% male), 94%, 95%, and 50% received RASi/ARNi, BBL, and MRA, respectively. Regardless of the specific treatment, non-users had lower BMI, education level and use of other HF medications, but older age, higher EF at improvement, higher NT-proBNP, and more likely peripheral artery disease, anemia, and valvular disease. In the matched cohorts, RASi/ARNi use was associated with a statistically significant lower risk of CVM/HHF [hazard ratio (HR) 0.71, 95% confidence interval (CI) 0.53–0.95] (Figure). There was no statistically significant association between CVM/HHF and beta-blockers (HR 0.75, (95% CI 0.53-1.08), and MRA (HR 1.16, 95% CI 0.90-1.49). These results were consistent after adjustment for the propensity score in the overall cohort (Figure). Conclusion In patients with HFimpEF use of GDMT was high. Treatment with RASi/ARNi was associated with lower morbidity and mortality, whereas no statistically significant association was observed with beta-blockers which might be explained by limited statistical power.Composite outcome.
Read full abstract