The landscape of heart failure (HF) management is evolving alongside a deeper understanding of how metabolic dysfunction drives cardiac pathology and the progression of HF, particularly in phenotypes like HF with preserved ejection fraction (HFpEF) associated with obesity. Glucagon-like peptide-1 receptor agonists (GLP-1RAs), which were initially approved for the treatment of type 2 diabetes (T2D) and obesity, have cardiovascular benefits that extend beyond glycaemic control. Their effects in modulating weight, haemodynamic load, endothelial function and systemic inflammation position them as potentially valuable therapeutic interventions for HF.1,2
This review evaluates the proposed mechanisms of action of GLP-1RAs, critically analyses recent evidence from clinical trials and considers the optimal clinical use of GLP-1RAs and potential future directions.
Mechanisms of Action of GLP-1RAs: Beyond Glycaemic Control
Endogenous glucagon-like peptide-1 (GLP-1) is a peptide hormone secreted by intestinal L cells and brainstem neurons in response to nutrient intake. GLP-1 functions as an incretin, enhancing glucose-dependent insulin secretion, suppressing glucagon release and delaying gastric emptying. However, endogenous GLP-1 has an extremely short half-life due to rapid degradation by dipeptidyl peptidase-4 and renal clearance.2 Synthetic GLP-1RAs are structurally modified analogues engineered to resist dipeptidyl peptidase-4 cleavage, prolonging their half-life.2
Although the pancreatic effects of GLP-1 (glucose-dependent insulin secretion and glucagon suppression) are well-established, its extrapancreatic actions are increasingly being recognised as critical to the therapeutic effects of GLP-1RAs. The cardiovascular benefits of GLP-1-RAs arise primarily through indirect systemic mechanisms, with limited evidence supporting direct effects on cardiomyocytes (Figure 1). Of note, dual receptor agonists have significantly advanced incretin-based therapy by concurrently engaging glucose-dependent insulinotropic polypeptide (GIP) and GLP-1 receptors within a single peptide molecule. Although activation of the GLP-1 receptor primarily facilitates weight reduction through appetite suppression, the inclusion of GIP receptor agonism offers supplementary and synergistic advantages, including improved insulin sensitivity and overall metabolic health. Tirzepatide functions as an imbalanced dual agonist, characterised by a molecular structure exhibiting a markedly higher affinity for the GIP receptor than the GLP-1 receptor.3,4 This specific configuration is believed to generate a synergistic effect that exceeds the efficacy of merely combining two separate agents.3,4
Indirect Effects of GLP-1RAs
The evidence points to the indirect effects of weight loss and systemic metabolic improvement as being the cornerstones of GLP-1RA efficacy in HF. These effects are described in detail below.
Weight Loss
GLP-1RAs induce substantial weight loss by affecting appetite and delaying gastric emptying.5 Weight loss plays a significant role in improving symptoms among HFpEF patients with obesity-related HF. Improved diastolic function, reduced ventricular wall stress and decreased accumulation of epicardial fat may contribute to the expected favourable outcomes.6 Clinical trials have demonstrated significant weight loss with various GLP-1RAs, ranging from 3–5 kg with shorter-acting agents to >10 kg with high doses of longer-acting agents, such as semaglutide and tirzepatide.5,7,8 The magnitude of weight loss is a significant factor in the potential benefit of GLP-1RAs in HFpEF patients with obesity.7–9
Blood Pressure Reduction
GLP-1RAs may lower systolic blood pressure by approximately 2–5 mmHg.9,10 This could be attributed to natriuresis and the improved function of the vascular endothelium facilitated by the release of nitric oxide. Although individual studies have reported varying outcomes, the cumulative evidence suggests a consistent, modest and yet clinically important reduction in blood pressure associated with GLP-1RAs.10,11
Improved Glycaemic Control
In addition to their direct effects on the pancreas, GLP-1RAs improve insulin sensitivity in peripheral tissues, like skeletal muscle and the liver.2 This effect may arise from mechanisms involving reduced ectopic fat accumulation and better cellular glucose uptake. GLP-1RAs may help alleviate myocardial damage linked to chronic hyperglycaemia.2 Enhanced glycaemic control is crucial for HF patients with T2D, who face increased cardiovascular risks.
Lipid Modulation
GLP-1RAs exert favourable effects on lipid profiles, primarily reducing triglycerides and, to a lesser extent, LDL cholesterol.12 These effects could be secondary to reduced hepatic lipogenesis, increased fatty acid oxidation and modulation of lipoprotein metabolism.13
Improved Energy Expenditure
GLP-1RAs may increase energy expenditure; however, the precise mechanisms are still under investigation. This effect of GLP-1RAs may involve increased thermogenesis and modulation of mitochondrial function in various tissues. The impact of increased energy expenditure on cardiac function and energetics requires further investigation.14
Anti-inflammatory Effects
GLP-1RAs reduce proinflammatory cytokines, including tumour necrosis factor-α and interleukin-6. By alleviating chronic systemic inflammation, GLP-1RAs may help prevent the progression of HF and adverse cardiovascular events.14,15
Potential Direct Effects on the Heart
Although the systemic indirect effects of GLP-1RAs are well established as the primary drivers of their clinical benefit in HF, emerging preclinical and mechanistic evidence suggests GLP-1RAs may also exert some direct cardioprotective effects, as discussed below. However, it is critical to recognise that these direct mechanisms are considered secondary, and in some cases remain controversial, compared with the proven impact of indirect pathways.
Direct Myocardial Effects
Although the presence of functional GLP-1 receptors on human ventricular cardiomyocytes remains a matter of debate, preclinical studies suggest potential direct cardioprotective effects. In animal models, GLP-1 enhances myocardial glucose usage, shifting metabolism from fatty acid to carbohydrate oxidation.16–19 However, the clinical relevance of these direct myocardial effects in humans requires further clarification. GLP-1 receptors are expressed in the atrial myocardium and the sinoatrial node. Activation of these receptors increases cAMP levels, which may explain the direct chronotropic effects of GLP-1 and its potential effect on the electrophysiological properties of atrial tissue.16
Endothelial Modulation
GLP-1 receptors are expressed in vascular endothelial cells. Activation of GLP-1 receptors improves endothelial function by enhancing nitric oxide bioavailability and subsequent vasodilation. This could improve myocardial perfusion.2,20
Antifibrotic Effects
GLP-1RAs downregulate transforming growth factor-β signalling pathways in myocardial fibrosis. In animal models, GLP-1RAs inhibited myocardial fibrosis and reduced ventricular hypertrophy.21
Antiarrhythmic Properties
GLP-1RAs may lower the risk of AF, a common comorbidity in HF that worsens prognosis, by reducing the volume of epicardial fat, modulating calcium handling within cardiomyocytes and improving left atrial remodelling.22–24
Natriuretic and Diuretic Effects
GLP-1RAs promote natriuresis and diuresis; these contribute to blood pressure reduction and volume homeostasis. These effects of GLP-1RAs are mediated through direct actions on the kidney. Subsequently, fluid overload, a key feature of HF, is reduced.25,26
Effects on Epicardial Adipose Tissue
GLP1-RAs decrease the volume of epicardial adipose tissue, which is often elevated in HFpEF and is a source of proinflammatory cytokines and adipokines.6
Neurohormonal Mechanisms
GLP-1RAs may modulate the activity of the autonomic nervous system by shifting the balance away from sympathetic overactivation towards parasympathetic tone.27–30 Studies have shown that GLP-1RAs can inhibit carotid body chemoreceptor activity, reducing sympathetic nervous system outflow. This effect may lower blood pressure and reduce excessive sympathetic drive in HF.27
Evidence From Clinical Trials
The clinical evidence supporting the use of GLP-1RAs in HF has primarily emerged from cardiovascular outcome trials (CVOTs) in patients with T2D and dedicated HF trials (Table 1).
Cardiovascular Outcome Trials in Type 2 Diabetes
CVOTs in T2D were designed primarily to establish cardiovascular safety in patients diagnosed with T2D who also have established cardiovascular disease (CVD) or have a high cardiovascular risk.8,9,31,32 These trials provided the first signs of the benefit of GLP-1RAs in HF. It is crucial to note that in most CVOTs the primary endpoint was a composite endpoint of major adverse cardiovascular events (MACE), with HF hospitalisation typically a prespecified secondary or exploratory outcome.
Both the LEADER (liraglutide) and SUSTAIN-6 (subcutaneous semaglutide) trials reported significant reductions in the primary outcome (incidence of MACE).10,33 With regard to the secondary outcome of HF hospitalisation, the results were neutral, with LEADER showing a non-significant relative risk reduction (RRR) of 13% with liraglutide compared with placebo and SUSTAIN-6 showing no significant difference between the semaglutide and placebo treatment groups.10,33 Similarly, the REWIND trial, in patients with T2D and established CVD or cardiovascular risk factors, reported that dulaglutide significantly reduced the incidence of MACE (primary outcome); however, dulaglutide had a neutral effect on HF hospitalisation.34
The AMPLITUDE-O trial demonstrated significant dose-dependent effects of efpeglenatide on HF hospitalisation in patients with T2D and CVD or renal disease, along with one risk factor.11 In that trial, hospitalisation for HF was a prespecified secondary endpoint. The pooled analysis of both the 4 mg and 6 mg doses of efpeglenatide showed a 37% RRR in HF hospitalisation (HR 0.63).11 However, a dose-specific analysis revealed that the beneficial effect of efpeglenatide on HF hospitalisation was primarily seen in the 6 mg group, in which there was a 50% RRR in HF hospitalisation (HR 0.50); in contrast, the 4 mg dose of efpeglenatide had no statistically significant effect on HF hospitalisation.11
Although not a T2D trial, SELECT (semaglutide) enrolled patients with overweight or obesity (BMI ≥27 kg/m2) and established CVD. Semaglutide 2.4 mg weekly significantly reduced the primary endpoint (incidence of MACE).32 It also significantly reduced the secondary composite HF endpoint (cardiovascular death, HF hospitalisation or urgent HF hospital visit), with an RRR of 18%.32
The SOUL trial evaluated oral semaglutide in patients with T2D who had established atherosclerotic cardiovascular disease, chronic kidney disease (CKD), or both. The semaglutide group were significantly less likely to experience MACE. In SOUL, composite HF endpoints were analysed as a secondary outcome, with oral semaglutide resulting in a non-significant reduction in HF events.35
Although these CVOTs were not specifically powered to detect differences in HF outcomes, they provided the first valuable signs of the benefit of GLP-1RAs in patients with HF. The results were heterogeneous: some trials, such as AMPLITUDE-O and SELECT, demonstrated significant reductions in HF-related endpoints, whereas others, like LEADER, showed only non-significant favourable trends. Crucially, despite this variability, subsequent meta-analyses pooling the data from injectable GLP-1RAs have demonstrated a significant overall reduction in the risk of hospitalisation for HF with GLP-1RAs.36
Dedicated Heart Failure Trials
The STEP-HFpEF, STEP-HFpEF DM and SUMMIT trials represent key advances in the evaluation of GLP-1RAs in the treatment of HF. These studies enhance our understanding of the therapeutic potential and efficacy of GLP-1RAs in the treatment of HF.
The STEP-HFpEF trial randomised 529 patients with obesity (BMI ≥30 kg/m2), HFpEF (left ventricular ejection fraction [LVEF] ≥45%) and New York Heart Association (NYHA) Class ≥II symptoms to once-weekly subcutaneous semaglutide 2.4 mg or placebo.9 The semaglutide group exhibited significant improvements in symptoms and physical limitations, assessed using the Kansas City Cardiomyopathy Questionnaire Clinical Summary Score (KCCQ-CSS), with the change in KCCQ-CSS being the primary trial endpoint.9 Semaglutide also resulted in significant improvements in secondary endpoints, including the 6-minute walk distance and weight loss. However, the reduction in HF hospitalisations with semaglutide did not reach statistical significance.9
The STEP-HFpEF DM randomised controlled trial enrolled 616 patients with obesity (BMI ≥30 kg/m2) and both HFpEF (LVEF ≥40%) and T2D.37 STEP-HFpEF DM specifically investigated the effect of once-weekly subcutaneous semaglutide versus placebo on the primary endpoint of change in KCCQ-CSS at 52 weeks. A statistically significant improvement in KCCQ-CSS was found in the semaglutide group compared with placebo, although the magnitude of improvement was numerically smaller than in STEP-HFpEF.8,37 Secondary endpoints in STEP-HFpEF DM, including changes in 6-minute walk distance and body weight, also favoured semaglutide.37 A key strength of STEP-HFpEF DM is its focus on the prevalent comorbidities of HFpEF and diabetes. However, the lower dose of semaglutide used and a focus on patient-reported outcomes rather than on cardiovascular events are limitations of STEP-HFpEF DM.37
The SUMMIT trial provides further evidence of the usefulness of GLP-1RAs in HF. SUMMIT included 731 patients with HFpEF and obesity with NYHA Class ≥II symptoms, and evaluated the dual GLP-1/GIP receptor agonist tirzepatide.8 Compared with placebo, once weekly tirzepatide significantly reduced the combined risk of cardiovascular death or worsening HF events (hospitalisation or an urgent hospital visit requiring intravenous therapy). There was a statistically significant reduction in the primary endpoint, which was a composite of cardiovascular death or worsening HF events, with an RRR of 33%.8 SUMMIT also demonstrated significant improvements in patient-reported outcomes: at 52 weeks, tirzepatide increased KCCQ-CSS, with a between-group mean difference of 6.9 points, and increased 6-minute walk distance, with a between-group mean difference of 18.3 m.8
Notably, in patients with obesity and HFpEF, these dedicated HFpEF trials showed a significant improvement in patient-reported outcomes measured by the KCCQ-CSS. Although SUMMIT also demonstrated a reduction in the composite endpoint of cardiovascular death or worsening HF events, the magnitude of the improvement, specifically in the KCCQ-CSS observed with higher-dose semaglutide (STEP-HFpEF; with a mean increase of 8 points) and tirzepatide (SUMMIT), appears to be greater than what has been reported for other HFpEF trials investigating the benefits of angiotensin receptor–neprilysin inhibitor (Paragon-HF), sodium–glucose cotransporter 2 inhibitors (EMPEROR-Preserved, DELIVER) and spironolactone (TOPCAT).38–41
Unresolved Role of GLP-1RAs in Heart Failure with Reduced Ejection Fraction
Unlike the positive results reported for GLP-1RAs in HFpEF, findings from dedicated trials in heart failure with reduced ejection fraction (HFrEF) report neutral findings. The LIVE trial, conducted in stable HFrEF patients (with or without T2D), found no significant effect of liraglutide on the change in LVEF with increased arrhythmic events.42 The FIGHT trial, which enrolled patients hospitalised for acute decompensated HFrEF, showed no improvement with liraglutide on the primary composite endpoint of death, HF rehospitalisation or change in N-terminal pro B-type natriuretic peptide levels.43
Notably, patients receiving liraglutide were reported to have a numerically higher risk of the composite outcome of death and HF hospitalisation overall, although this was not statistically significant (HR 1.30; 95% CI [0.92–1.83]; log-rank p=0.14), along with a potential increased risk of arrhythmic events.43,44 Given these neutral-to-concerning results, there is currently no evidence to support initiating GLP-1RAs for the primary treatment of HFrEF. If their use is considered for a compelling comorbidity like obesity, caution is advised.45
Identifying Optimal Candidates for GLP-1RA Therapy in Heart Failure
The selection of patients with HF for GLP-1RA therapy requires careful consideration of the HF phenotype, comorbidities and available evidence. Detailed considerations based on current knowledge are presented below.
Heart Failure with Preserved Ejection Fraction and Metabolic Comorbidities
The evidence most strongly supports the use of GLP-1RAs in the metabolic phenotype of HFpEF with NYHA Class II–IV symptoms, defined by obesity (BMI ≥30 kg/m2) with or without T2D. This is the patient group in which the STEP-HFpEF, STEP-HFpEF DM and SUMMIT trials demonstrated robust benefits.8,9,30 The agents used in these trials, including higher-dose semaglutide (2.4 mg) and the dual GLP-1/GIP receptor agonist tirzepatide, have demonstrated compelling benefits. All the dedicated HF GLP-1RA trials convincingly demonstrated significant improvements in HF-related symptoms (KCCQ-CSS), physical function (6-minute walk distance) and weight loss. Furthermore, the SUMMIT trial demonstrated a significant decrease in the hard composite endpoint of cardiovascular death or worsening HF.8,9,37 These data provide robust evidence for the benefit of tirzepatide in patients with obesity and HFpEF. The CVOTs provide additional evidence for cardiovascular risk reduction with GLP-1RAs in patients with T2D.
HFpEF and Obesity and Recent or Recurrent Hospitalisation
Patients with HFpEF, obesity and a history of recent HF decompensation (within the preceding 12 months) constituted a significant proportion of the populations in the STEP-HFpEF and SUMMIT trials. The SUMMIT trial demonstrated that tirzepatide significantly reduced the risk of HF hospitalisation or an urgent visit requiring intravenous diuretics.8 This suggests that GLP-1RAs can be beneficial in stabilising high-risk HFpEF patients and reducing subsequent exacerbations.
Integration of GLP-1RAs with Guideline-directed Medical Therapy and the Management of Other Comorbidities
A key practical question is how to integrate GLP-1RAs with existing therapies for HFpEF and its common comorbidities. Based on the current evidence, GLP-1RAs should be considered alongside other guideline-directed medical therapies for HFpEF when indicated for T2D or obesity.
Many patients with HFpEF have underlying coronary artery disease; the major CVOTs included significant numbers of patients with coronary artery disease and provide compelling evidence for MACE reduction with GLP-1RAs. However, the dedicated HF trials were not designed to address atherosclerotic CVD outcomes.
Chronic kidney disease also frequently coexists with T2D and HFpEF. Several GLP-1RAs CVOTs included patients with chronic kidney disease and showed consistent cardiovascular and some renal benefits. The FLOW trial has recently provided clear evidence for kidney protection with semaglutide.46 That dedicated kidney outcomes study, involving over 3500 patients with both T2D and established chronic kidney disease, demonstrated that once-weekly subcutaneous semaglutide 1.0 mg significantly reduced the primary composite renal outcome (including kidney failure onset, a ≥50% persistent decline in estimated glomerular filtration rate (eGFR), or renal/cardiovascular death) compared with placebo.46
Adverse Effects and Tolerability of GLP-1RAs in Clinical Practice
Although the cardiovascular benefits of GLP-1RAs are significant, their use is commonly associated with well-characterised adverse effects that can affect tolerability and long-term medication adherence, as detailed below.
Gastrointestinal Effects
The most frequent adverse effects of GLP-1RAs are gastrointestinal, including nausea, vomiting, diarrhoea and a sensation of early satiety. These effects occurred in 40–70% of treated patients in clinical trials.47 These symptoms are due to the drug’s mechanisms of action, particularly delayed gastric emptying, and are the primary reason for treatment discontinuation, representing a major barrier to adherence. The incidence and severity of gastrointestinal side effects may increase with higher doses or more rapid dose escalation, as commonly employed for greater weight reduction. Such gastrointestinal effects are a major reason for treatment discontinuation and are intrinsically linked to the mechanisms responsible for the weight loss efficacy of GLP-1RAs.
The cornerstone of management is dose titration (start low, go slow). Patients should be counselled to expect these potential side-effects.48,49
Heart Rate Increase
GLP-1RAs are known to cause a modest, dose-dependent increase in resting heart rate. This is thought to be mediated by direct GLP-1 receptor stimulation in the sinoatrial node.50 This modest effect was not associated with adverse cardiovascular outcomes in any of the major clinical trials.
Pancreatitis
There is a known, although rare, risk of acute pancreatitis with GLP-1RAs. GLP-1RAs should be discontinued immediately if pancreatitis is suspected and are generally avoided in patients with a prior history of pancreatitis.51
Gallbladder Disease
Rapid weight loss can increase the risk of cholelithiasis and cholecystitis, and was observed in major GLP-1RA trials.52
Contraindications
GLP-1RAs are contraindicated in patients with a personal or family history of medullary thyroid carcinoma or in patients with multiple endocrine neoplasia syndrome type 2.53
Future Directions and Unanswered Questions
Despite recent evidence from clinical trials, key questions remain about the optimal use of GLP-1RAs for HF. Although there is strong evidence supporting the benefit of GLP-1RAs in populations with obesity, their efficacy in other populations of particular HF subtypes and specific cardiomyopathies is unclear. Therefore, additional studies on different phenotypes of HFpEF are essential. The identification of subgroups that are most likely to benefit from GLP-1RAs could be guided by biomarkers, and cardiac imaging may help determine whether the advantages of GLP-1RAs extend beyond the metabolic or obesity phenotypes. In addition, the long-term effects of GLP-1RA therapy on HF need further investigation. Longer studies and real-world evidence collected through registries are essential.
Conclusion
GLP-1RAs are gaining attention as a new option for treating the obesity-related phenotype of HFpEF. This represents a significant improvement in our approach to addressing the metabolic factors of this complex syndrome. Compelling evidence from landmark clinical trials, particularly STEP-HFpEF, STEP-HFpEF DM and SUMMIT, has established the efficacy of GLP-1RAs in improving HF-related symptoms and quality of life, as well as reducing worsening HF events, in the prevalent phenotype of HFpEF in the setting of obesity and T2D. Their multiple mechanisms of action, which include significant weight loss, improved glycaemic control, blood pressure reduction, anti-inflammatory effects and potential direct cardioprotective and vascular effects, target multiple pathophysiological pathways implicated in HF progression. However, it is crucial to note that current evidence does not support the use of GLP-1RAs for the primary treatment of HFrEF. If considered for the indication of obesity in HFrEF, caution and close monitoring are warranted.
