https://doi.org/10.1007/s00392-025-02625-4
1Universitätsspital Basel Abt. für Kardiologie Basel, Schweiz; 2Department of Experimental and Clinical Toxicology, Institute of Experimental and Clinical Pharmacology and Toxicology, Center for Molecular Signaling (PZMS) Homburg, Deutschland; 3Universitätsklinikum des Saarlandes Innere Medizin III - Kardiologie, Angiologie und internistische Intensivmedizin Homburg/Saar, Deutschland; 4ABDA-Bundesvereinigung Deutscher Apothekerverbände e. V. GB Arzneimittel Berlin, Deutschland
Background: Acutely decompensated heart failure (ADHF) is associated with high rates of rehospitalization and in-hospital mortality. Non-adherence to guideline-directed medical therapy (GDMT) may cause worsening of chronic heart failure (HF). This study aims to investigate adherence to HF medication in patients with ADHF, directly assessed through analysis of drugs and/or their metabolites in urine, and to identify patient-related factors influencing adherence.
Methods: This cross-sectional study included patients with ADHF, regardless of ejection fraction (EF), who presented to the emergency department (ED). The main inclusion criteria were a history of chronic HF, elevated NT-proBNP levels, and the requirement for intravenous diuretics. Medication adherence was assessed using indirect and direct methods. The indirect assessment involved questionnaires (RAI, BMQ), while the direct assessment was conducted through qualitative analysis of drugs and/or their metabolites in urine samples, utilizing liquid chromatography coupled with high-resolution mass spectrometry. The trial is registered under clinicaltrials.gov (NCT06459115).
Results: A total of 100 patients were included in the final analysis, of which 61% were men, with a mean (± standard deviation) age of 78±10 years, mean EF of 43.6±13.3 %, and median NT-proBNP of 4,846 pg/ml (interquartile range 2,091-9,863). The cohort comprised 35% with HF with reduced EF (HFrEF), 22% with mildly reduced EF (HFmrEF), and 43% with preserved EF (HFpEF). Notably, only 29% of HFrEF patients were on GDMT with all five drugs with a Class I recommendation, while 45% of the HFmrEF patients and 23% of the HFpEF patients were on a combination of a diuretic and a sodium-glucose co-transporter 2 (SGLT2) inhibitor. According to the adherence definition used in this study, 39% of patients were adherent (all substances/metabolites detectable in urine), 39% were partially adherent (1 substance/metabolite not detectable), and 22% were non-adherent (≥2 substances/metabolites not detectable). Interestingly, 41% of patients had drugs detected that were not on their medication plans (mainly cardiovascular drugs and painkillers). Non-adherent patients had significantly more substances prescribed for the treatment of chronic HF than partially (p=0.0047) and adherent (p<0.001) patients. Non-adherent patients were significantly younger than partially adherent (p=0.0144) and adherent (p=0.0054) patients. Adherent patients were significantly more likely to have an abnormal DemTect test result (p=0.049), an indicator for early dementia. In-hospital mortality was not associated with adherence status (p=0.63). Having medication prepared by a third party (family members or healthcare staff) reduced the risk of non- and partially adherence (p<0.001). Factors such as living alone (p<0.001), and low income (p<0.001) were associated with non-adherence. The prescription of mineralocorticoid receptor antagonists and thiazide diuretics was associated with poor adherence. Indirect adherence monitoring through questionnaires and patient self-reporting on medication intake was inaccurate compared to urine analysis.
Conclusions: Adherence rates in patients with ADHF presenting to the ED were notably low, and nearly half were found to be taking non-prescribed medications. Non-adherence to medical therapy may be an overlooked contributor to the worsening of chronic HF, potentially leading to hospitalization due to ADHF.
Created with BioRender.com