Key result
Pillar-GDMT score moderately predicts all-cause mortality, stratifying high-risk patients to a ~41% event rate.
Why the study?
Achieved doses of quadruple guideline-directed medical therapy in HFrEF often remain low, prompting the design of a score incorporating all four pillars and doses to predict all-cause mortality after hospitalisation.
Does the novel Pillar-GDMT score improve the prediction of all-cause mortality in patients with HFrEF requiring hospitalization compared to previously published score systems?
Cohort (n=441)
No
Does the novel Pillar-GDMT score improve the prediction of all-cause mortality in patients with HFrEF requiring hospitalization compared to previously published score systems?
A novel Pillar-GDMT score that incorporates the achieved discharge doses of quadruple medical therapy provides superior predictive accuracy for all-cause mortality in HFrEF patients compared to existing scoring systems.
May aid post-hospitalization mortality stratification in HFrEF; leaves open prospective validation against established scores.
Introduction The initiation and optimisation of first-line guideline-directed medical therapy (GDMT) of heart failure (HF) with reduced ejection fraction (HFrEF) is of strategic importance. However, the achieved doses of the quadruple therapy (QT: RASi [ACEi/ARB/ARNI] + βB + MRA + SGLT2i) of HFrEF often remain low. Aims To design a novel, universally applicable "Pillar-GDMT" score considering the 4 pillars of HFrEF pharmacotherapy and their doses applied for prediction of all-cause mortality (ACM) after a HF hospitalisation (HFH) and to assess its predictive accuracy in comparison with previously published score systems. Methods Based on the retrospective data analysis of a consecutive, real-world patient (pt) cohort requiring HFH between 01.09.2021 and 01.09.2024 at a tertiary referral cardiology center, the "Pillar-GDMT" score model was designed including all 4 pillars of QT, considering the target doses (TDs) achieved at discharge (in case of RASi/βB/MRA rating the doses applied as follows: not receiving medication: 0 point, <50% of TDs: 1 point, ≥50% but <100% of TDs: 2 points, 100% of TDs: 3 points; in case of SGLT2i: not receiving medication: 0 point, not at TDs: 1 point, 100% of TDs: 3 points), with a maximal score of 12 points. The established model was tested with receiver operating characteristic (ROC) analysis and its predictive accuracy was compared with "simple GDMT score" and "modified HF Collaboratory (HFC) Treatment Score". Pts were divided into 3 risk categories based on their "Pillar-GDMT" score: low-risk (LR): 7-12 points, intermediate-risk (IMR): 4-6 points, high-risk (HR): 0-3 points. ACM was analysed with Kaplan-Meier method among the risk categories. The independent predictors of ACM were tested with multivariate Cox-regression analysis. Results The total cohort of 441 pts (male: 75%, age: 61 [51-71] years, "de novo" HFrEF: 38%, diabetes: 36%, eGFR <60 mL/min/1.73m²: 51%, LVEF: 25 [20-30] %, NT-proBNP at discharge: 2176 [963-4204] pg/mL) were evaluated. At discharge, RASis in 92%, βBs in 85%, MRAs in 95%, SGLT2is in 62%, while QT in 56% was applied. Median "Pillar-GDMT" score was 8 [5-9]. The ROC analysis showed moderate predictive accuracy (AUC: 0.661, 95% CI: 0.601-0.722, p<0.001), which exceeded that of "simple GDMT score" and "modified HFC Treatment Score". ACM of the risk categories showed significant differences (9% vs 21% vs 41%, LR vs IMR vs HR, p<0.001). Of the independent predictors of ACM lower "Pillar-GDMT" score, higher NT-proBNP and creatinine levels at discharge, diabetes mellitus, previously confirmed diagnosis of HF increased, while the use of CRT at discharge decreased the risk of ACM. Conclusions According to our analysis "Pillar-GDMT" score considering the first-line QT of HFrEF showed moderate predictive accuracy but exceeded previous score models. Low "Pillar-GDMT" score was also an independent predictor of mortality.
No takes yet. Share an insight, caveat, or question.
Bódi et al. (2025) conducted a cohort in Heart failure with reduced ejection fraction (HFrEF) requiring hospitalization (n=441). Pillar-GDMT score vs. Simple GDMT score and modified HFC Treatment Score was evaluated on All-cause mortality (AUC 0.661, 95% CI 0.601-0.722, p=<0.001). The novel Pillar-GDMT score demonstrated moderate predictive accuracy for all-cause mortality (AUC 0.661; 95% CI 0.601-0.722; P<0.001) in patients with HFrEF requiring hospitalization.
Synapse has enriched 5 closely related papers on similar clinical questions. Consider them for comparative context: