Prevention of cardiovascular events in heart failure with mildly reduced or preserved ejection fraction: a comprehensive network meta-analysis of eight randomized controlled trials using reconstructed individual patient's data.

Background: Cardiometabolic therapies such as mineralocorticoid receptor antagonists (MRAs), sodium glucose cotransporter 2 inhibitors (SGLT2i), and glucagon-like peptide 1 receptor agonists (GLP-1 RA) reduce clinical heart failure events in patients with heart failure with mildly reduced or preserved ejection fraction (HFpEF). However, evidence for the benefit of one strategy or their combination is not well established. To optimize guidelines in this population, evidence of the most effective therapeutic options or strategies is needed. Methods: The present network meta-analysis used reconstructed individual patient data from published Kaplan-Meier curves of cardiometabolic therapies, identified through MEDLINE, EMBASE, and Cochrane library (CENTRAL) up to March 31, 2025 (PROSPERO; CRD420251007431). The primary outcome was the composite of time to cardiovascular death or heart failure (HF) hospitalization. One-stage approach using stratified random-effect Cox regression model was used to re-estimate the treatment-effects for heart failure events. The relative effect of combination therapies was estimated using established methods for making indirect comparisons by an additive network meta-analysis model. Findings: Eight trials involving 25,440 patients with HFpEF were analyzed. The mean age of participants ranged between 61.7 and 72 years, 11,355 (45%) were women, 10,242 (40%) with type 2 diabetes, and 4841 (19%) with obesity. In these patients, the combination of MRA, SGLT2i, and GLP-1 RA was the most effective for all clinical HF event (P-score 1.00). Compared with placebo, this combination reduced by 58% (HR 0.42 [0.31-0.56]) the risk of combined endpoint of cardiovascular death or HF hospitalization. The corresponding estimated absolute risk reduction (ARR) over a median follow-up of 2.5 years was 6.9% (95% CI, 4.6%-8.7%]), with a number need to treat (NNT) of 14 (95% CI 12-22). There was also a risk reduction of 73% (HR 0.27 [0.18-0.42]; ARR 6.2% [4.2%-7.5%]; NNT 16 [13-24]) for HF hospitalization, and 43% (HR 0.57 [0.44-0.72]; ARR 3.0% [1.9%-4.0%]; NNT 34 [26-52]) for cardiovascular death. Among the double combinations, MRA and GLP-1 RA strategies were the most effective in the reduction in the risk of HF hospitalization (P-score 0.77) and cardiovascular death (P-score 0.75), while SGLT2 and GLP-1 RA strategies were the most effective in the reduction in the risk of the composite endpoint (P-score 0.77). Interpretation: In patients with HF and LVEF > 40% for whom few treatment options are available, adjunctive combination of MRA, SGLT2i, and GLP-1 RA to standard care has the potential to confer benefit in heart failure events (moderate quality of evidence). Furthermore, large-scale randomized combination therapies with extended follow-up are needed to confirm these findings, and to explore potential benefits in subgroups, optimized protocols, and inform clinical guidelines. Funding: None.