“Hold the drug, strength tomorrow”: The prognostic effect of therapy adherence after a hospitalisation for heart failure with reduced ejection fraction
P P Schaffer, F Banfi-Bacsardi, Zs Forrai, T G Gergely, E M Boldizsar, K Czurko, M Solazzo, D Pilecky, P Andreka, Zs Piroth, B MukAbstract
Introduction
Adherence to guideline-directed medical therapy (GDMT) in heart failure (HF) remains a significant challenge nowadays, which is emphasised by the 2025 scientific statement of the ESC HFA and the ESC Working Group on Cardiovascular Pharmacotherapy. Not only “door-to-GDMT” time, but also “time on GDMT” has a severe prognostic effect.
Aims
To assess the long-term application of quadruple therapy (QT: RASi + βB + MRA + SGLT2i) in a consecutive cohort of patients who were hospitalised for HF with reduced ejection fraction (HFrEF) and received QT upon discharge from hospital; and to evaluate its effect on prognosis.
Patients and methods
A retrospective observational study was conducted in a consecutive patient cohort (n=551) hospitalised for HF between 2021 and 2025 at our cardiology centre (male: 75%, age: 63 [52-71] years, de novo HF: 42%, LVEF: 25 [20-30]%, coronary artery disease: 45%, diabetes: 40%, eGFR < 60 mL/min/1.73m2: 52%, NT-proBNP at hospital admission: 5866 [2728-10616] pg/mL, GDMT at discharge: RASi: 94%, βB: 85%, MRA: 95%, SGLT2i: 68%, triple therapy [TT: RASi + βB + MRA]: 82%, QT: 61%). Among patients who received QT at discharge, the QT application was evaluated at 1 year of follow-up (FU). The prognostic effect (all-cause mortality [ACM]/HF hospitalisation [HFH]/composite endpoint of ACM/HFH) of adherence to QT was analysed using the Kaplan-Meier method and the log-rank test.
Results
At discharge, 61% of the total cohort (TC) were on QT; at 1 year of FU, 59% of the TC were on QT. Of the 257 patients on QT at discharge, 79% were still on QT at 1 year, while 21% had QT suspended. Among QT-adherent and non-QT-adherent patients, there were no differences in baseline characteristics except hypertension (66% vs 81%, p=0.021). In those patients, SGLT2i was discontinued in 56%, βB in 49%, MRA in 36%, and RASi in 20%; 34% of them had at least 2 medications discontinued. During the median 670 [339-1083] days of FU, HFH rate was significantly higher among patients with non-adherence to QT (17% vs 35%, p=0.004; QT at 1 year vs non-QT at 1 year, respectively). The composite endpoint of ACM/HFH also occurred more frequently in non-QT-adherent patients (19% vs 37%, p=0.008).
Conclusions
Our retrospective study revealed that non-QT-adherent patients had a significantly worse prognosis, with higher HFH and ACM/HFH rates. Our findings suggest that conscious efforts should be made to maintain medication adherence in daily clinical practice.Adherence1Adherence2