Implementation of guideline-directed medical therapy among patients with acute decompensated heart failure admitted to cardiac rehabilitation: the PROMETEO registry
A Villaschi, M L Oreni, A Toccafondi, I Cusmano, S Gonella, R Gonella, G Coni, A Torri, P Grati, M Ambrosetti, D Cianflone, N MoriciAbstract
Introduction
Cardiac rehabilitation (CR) should be considered in patients after a recent hospitalization for acute heart failure (HF) to improve clinical outcomes. However recent epidemiological data are scant, referral rates are low and its role in optimization of guideline-directed medical therapy (GDMT) remains to be elucidated.
Aims
To provide an updated epidemiological description of patients referred to CR in Italy after hospitalization for acute HF, focusing on the use and optimization of GDMT across left-ventricular ejection fraction (LVEF) spectrum.
Methods
PROMETEO is a prospective, multicenter Italian registry enrolling consecutive adult patients admitted to CR after acute HF between 05/2023 and 06/2025. Patients were stratified as HFrEF (LVEF <40%) or HFmrEF/HFpEF (LVEF ≥40%). Baseline characteristics were collected at admission. Data on GDMT and loop diuretics prescription were also recorded at admission and discharge. GDMT was defined as ACEi/ARB/ARNI, beta-blockers, MRAs and SGLT2i for HFrEF and SGLT2i for HFmrEF and HFpEF. A modified GDMT score was calculated for HFrEF patients. The Wilcoxon signed-rank test and the McNemar's test were used for within-group comparisons of changes in medical therapy from baseline to discharge, as appropriate. A multivariable general linear regression model was built to evaluate the association between variables deemed clinically relevant and increase in the modified GDMT score from admission to discharge.
Results
Among 263 patients with available LVEF, median age was 68.7 years, 21.7% were female, and 74.9% had HFrEF. Compared with HFmrEF/HFpEF, HFrEF were younger, less frequently female and with higher NT-proBNP levels. At CR admission, 72.1% of HFrEF patients received ACEi/ARB/ARNI, 91.9% beta-blockers, 79.2% MRAs, and 66.0% SGLT2 inhibitors, with a median GDMT score of 6 (IQR 4–8). At discharge, ARNI use increased significantly (58.9%) and the median GDMT score increased to 7 (IQR 4–8; p=0.013), alongside higher beta-blocker dosing and reduced loop diuretic use. Chronic kidney disease and higher baseline GDMT score were negatively associated with GDMT optimization, while higher BMI showed a positive association. In HFmrEF/HFpEF patients, SGLT2 inhibitor use increased from 28.8% to 47.0% (p=0.010), with a concomitant reduction in loop diuretic prescription and dose.
Conclusion
Patients referred to CR after acute HF are mainly male with HFrEF and show substantial comorbidity and treatment complexity. Residential CR represents a key opportunity for GDMT optimization across the HF spectrum.For image description, please refer to the figure legend and surrounding text.