Clinical Outcomes of GLP-1 Receptor Agonist and SGLT2 Inhibitor Combination Therapy in Heart Failure: A Real-World Propensity-Matched TriNetX Analysis

Background: Heart failure patients are frequently prescribed SGLT2 inhibitors, but the incremental real-world benefit of adding GLP-1 receptor agonists is uncertain. Methods: A retrospective propensity-matched cohort study was conducted using the TriNetX Global Collaborative Network (171 healthcare organizations). Adults aged ≥18 years with incident heart failure who initiated either dual SGLT2 inhibitor and GLP-1 receptor agonist therapy or SGLT2 inhibitor monotherapy within 1 month of the first heart failure diagnosis were compared. Outcomes over 365 days included all-cause mortality, all-cause hospitalization, acute myocardial infarction, atrial fibrillation/flutter, acute kidney failure, pulmonary edema, new-onset diuretic use, urinary tract infection, retinopathy, and laboratory hypoglycemia (glucose ≤ 70 mg/dL). Cox proportional hazards models were used; the proportional hazards assumption was formally tested. Bonferroni and Benjamini–Hochberg adjustments were applied for multiple comparisons. E-values quantified robustness to unmeasured confounding. Appendicitis was used as a negative control outcome. Results: After 1:1 propensity score matching, 3421 patients were included in each cohort. Cohorts were well-balanced (all standardized mean differences < 0.10). Over 365 days of follow-up, no significant differences were observed between dual therapy and monotherapy in all-cause mortality (HR 0.92, 95% CI 0.77–1.09), all-cause hospitalization (HR 1.16, 95% CI 0.96–1.40), acute myocardial infarction (HR 1.20, 95% CI 0.92–1.55), atrial fibrillation/flutter (HR 1.05, 95% CI 0.85–1.30), acute kidney failure (HR 1.05, 95% CI 0.89–1.26), new-onset diuretic use (HR 0.92, 95% CI 0.78–1.09), or urinary tract infection (HR 1.16, 95% CI 0.90–1.48). Dual therapy was associated with a significant increase in retinopathy (HR 2.66, 95% CI 1.81–3.93; Bonferroni p = 0.001; E-value 4.77, lower CI bound 3.01), and a modest increase in laboratory hypoglycemia (HR 1.22, 95% CI 1.04–1.44) and pulmonary edema (HR 1.35, 95% CI 1.06–1.70), both of which survived FDR but not Bonferroni adjustment. Prespecified subgroup analyses showed lower mortality with dual therapy in patients aged 18–64 (HR 0.64, 95% CI 0.43–0.93) and in women (HR 0.60, 95% CI 0.45–0.81). The negative control outcome (appendicitis) was non-significant. Conclusions: In a propensity-matched real-world cohort with a conservative cohort definition, dual SGLT2 inhibitor and GLP-1 receptor agonist therapy was not associated with significant differences in mortality or major cardiovascular outcomes compared with SGLT2 inhibitor monotherapy. Subgroup signals favoring dual therapy in younger and female patients, alongside safety signals for retinopathy and laboratory hypoglycemia, are hypothesis-generating and should be confirmed in prospective trials.