Highlights
- Initiation of GLP-1 receptor agonists (GLP-1RAs) is associated with a 23% lower risk of hospitalization for heart failure (HHF) compared to DPP-4 inhibitors.
- In a direct real-world comparison, GLP-1RAs demonstrated comparable effectiveness to SGLT-2 inhibitors in preventing HHF, with a weighted hazard ratio of 1.02.
- The absolute risk reduction for HHF is most significant among patients with the highest baseline cardiovascular risk, suggesting a targeted approach for therapy escalation.
- Target trial emulation methodology provides robust evidence that bridges the gap between restrictive randomized controlled trials and heterogeneous observational data.
Background
Patients with type 2 diabetes (T2D) face a significantly elevated risk of developing heart failure (HF), a condition that contributes heavily to morbidity, mortality, and healthcare costs. While sodium-glucose cotransporter-2 inhibitors (SGLT-2is) have established themselves as the cornerstone for heart failure hospitalization (HHF) prevention, the role of glucagon-like peptide-1 receptor agonists (GLP-1RAs) in this specific outcome has remained controversial. Early cardiovascular outcome trials (CVOTs) for GLP-1RAs primarily focused on major adverse cardiovascular events (MACE), with HHF often appearing as a secondary or exploratory endpoint with varying results.
There is a critical clinical need to understand whether the heart failure benefits of GLP-1RAs are a class effect and how they perform relative to other glucose-lowering therapies in routine clinical care. Current guidelines are increasingly prioritizing cardiorenal protection over glycemic control alone, yet the comparative effectiveness of GLP-1RAs versus SGLT-2is—the current gold standard for HF—remains largely unexplored in head-to-head clinical trials. This review synthesizes recent evidence from a massive target trial emulation to clarify these therapeutic relationships.
Key Content
Target Trial Emulation: Methodological Rigor in Real-World Evidence
The core evidence discussed stems from a sophisticated target trial emulation conducted using population-based healthcare data from Stockholm, Sweden (2010–2021). This methodology is designed to mimic the design of a randomized controlled trial (RCT) using observational data, thereby reducing the common biases associated with real-world evidence (RWE), such as immortal time bias and selection bias. The study emulated two distinct trials: one comparing GLP-1RA to dipeptidyl peptidase-4 inhibitors (DPP-4is) and another comparing GLP-1RA to SGLT-2is.
To ensure balance across groups, researchers utilized inverse probability of treatment weighting (IPTW) to adjust for 72 different confounders, including baseline comorbidities, laboratory values (HbA1c, renal function), and concurrent medications. The use of MACE as a positive control outcome further validated the model, as the observed reduction in MACE with GLP-1RA use aligned perfectly with established RCT findings.
GLP-1RA vs. DPP-4 Inhibitors: Confirming Cardiovascular Superiority
In the first emulated trial involving 32,979 patients, GLP-1RAs demonstrated a clear clinical advantage over DPP-4is. Over a 3-year period, the absolute risk of HHF was 3.4% for GLP-1RA initiators versus 4.3% for DPP-4i initiators. This translated to a weighted hazard ratio (HR) of 0.77 (95% CI, 0.66–0.91). This finding is significant because it suggests that GLP-1RAs provide substantial protection against heart failure exacerbations compared to a technically “neutral” class like DPP-4is. Subgroup analyses confirmed these results across various ages, sexes, and even among those with a pre-existing history of heart failure.
The Comparison of Giants: GLP-1RA vs. SGLT-2 Inhibitors
The most provocative finding of the Stockholm study (Target Trial 2) was the comparison between GLP-1RAs and SGLT-2is. Including 30,104 patients, the analysis showed that the risk of HHF was nearly identical between the two classes (3.6% for GLP-1RA vs. 3.3% for SGLT-2i; HR 1.02 [95% CI, 0.85–1.18]). For years, clinical consensus has held SGLT-2is as uniquely potent for heart failure. However, this data suggests that in routine clinical practice, GLP-1RAs may offer comparable HHF protection. This parity was consistent for individual agents such as liraglutide and semaglutide, suggesting a robust class effect for the modern GLP-1RAs.
Risk Stratification and Clinical Prediction
The study highlighted that the absolute risk difference—the number of patients who actually avoid a hospital visit—was largest for those with higher predicted risk of heart failure at baseline. This aligns with broader shifts in diabetes care toward personalized risk-based therapy. In a related context, recent discussions in the literature (e.g., Pop-Busui et al., Diabetes Care 2025) emphasize the importance of screening natriuretic peptide (NP) levels to predict HF and death in individuals with T2D. Integrating NP screening with the high-potency cardiovascular benefits of GLP-1RAs or SGLT-2is could represent the next frontier in preventive cardiology for diabetic patients.
Expert Commentary
The results of this target trial emulation challenge the traditional hierarchy of diabetes medications. Historically, SGLT-2is were for heart failure and GLP-1RAs were for atherosclerotic MACE. This new evidence suggests the boundaries are blurring. Mechanistically, GLP-1RAs may reduce HHF through indirect pathways: improved endothelial function, reduction in systemic inflammation, and significant weight loss, which reduces the hemodynamic burden on the heart. Unlike SGLT-2is, which have a direct diuretic and sodium-modulating effect on the kidney, GLP-1RAs likely achieve cardiovascular stability through a broader metabolic and anti-inflammatory profile.
However, clinicians must remain cautious. While target trial emulations are robust, they are not a substitute for double-blind RCTs. One controversy remains: the study showed similar risks between GLP-1RA and SGLT-2i, but it does not suggest that GLP-1RAs should replace SGLT-2is in patients with established heart failure with reduced ejection fraction (HFrEF), where SGLT-2i evidence is insurmountable. Instead, this evidence supports the early use of GLP-1RAs as a potent tool for the *primary* prevention of HF in high-risk T2D patients.
Conclusion
The landscape of T2D management has evolved from glucose-centric to organ-protective. GLP-1RAs have now been shown in real-world, emulated trial settings to provide substantial protection against heart failure hospitalization—superior to DPP-4is and comparable to SGLT-2is. These findings reinforce the importance of initiating high-benefit therapies early in the disease course, particularly in patients identified as high-risk through biomarkers like natriuretic peptides. Future research should focus on the synergistic effects of combining GLP-1RAs and SGLT-2is, potentially offering a “dual-inhibitor” strategy to nearly eliminate the excess risk of heart failure in the diabetic population.
References
- Xu Y, Huang T, Zhang Y, Ji D, Tuttle KR, Carrero JJ, Fu EL. Risk of Heart Failure Hospitalization for GLP-1 Receptor Agonists Versus DPP-4 Inhibitors or SGLT-2 Inhibitors in Patients With Type 2 Diabetes: A Target Trial Emulation. Circulation. 2026 Feb 24. PMID: 41732861.
- Pop-Busui et al. Screening Natriuretic Peptide Levels Predicts Heart Failure and Death in Individuals With Type 1 and Type 2 Diabetes Without Known Heart Failure. Diabetes Care. 2025;48:2145-2153. PMID: 41719468.
- Sun et al. SGLT2i Versus Metformin for Delirium Prevention in Type 2 Diabetes: A Real-World, Head-to-Head Comparative Study. Diabetes Care. 2025;48:1361-1369. PMID: 41719462.
