Cardiovascular effectiveness and musculoskeletal safety of tirzepatide, GLP-1 receptor agonists, and SGLT2 inhibitors in normal-weight adults with type 2 diabetes and established atherosclerotic cardiovascular disease: a cohort study using a federated electronic health record network

Type 2 diabetes (T2D) is associated with atherosclerotic cardiovascular disease (ASCVD). We aimed to compare the cardiovascular effectiveness and musculoskeletal safety of tirzepatide, glucagon-like peptide−1 receptor agonists (GLP-1 RAs), and sodium–glucose cotransporter 2 (SGLT2) inhibitors among normal-weight adults with T2D and established ASCVD. We conducted an active-comparator, retrospective cohort study using the TriNetX Analytics Network (data extracted on February 16, 2026). Adults (≥ 18 years) with T2D, ASCVD, and a baseline body mass index between 18.5 and 24.9 kg/m 2 were included. Three prespecified pairwise comparisons were performed (tirzepatide vs. GLP-1 RAs, tirzepatide vs. SGLT2 inhibitors, and GLP-1 RAs vs. SGLT2 inhibitors). Cohorts were propensity score matched 1:1. The primary endpoint was three-point major adverse cardiovascular events (MACE) including all-cause mortality, acute myocardial infarction, and ischemic stroke. The secondary cardiovascular outcome was the occurrence of each MACE component. The safety outcomes included composites for sarcopenia, osteoporosis, and falls/fractures. After matching, the cohort sizes were 634 pairs (tirzepatide vs. GLP-1 RAs), 4,438 pairs (GLP-1 RAs vs. SGLT2 inhibitors), and 545 pairs (tirzepatide vs. SGLT2 inhibitors). The hazard of three-point MACE was lower with tirzepatide than with GLP-1 RAs (14.2% vs. 20.3%; hazard ratio [HR] = 0.63, 95% confidence interval [CI]: 0.48–0.83) and with SGLT2 inhibitors (13.0% vs. 22.6%; HR = 0.64, 95% CI 0.48–0.86); the reduction was driven by lower ischemic stroke risk. GLP-1 RAs and SGLT2 inhibitors had similar MACE risk (17.3% vs. 19.6%; HR = 0.99, 95% CI 0.90–1.09); acute myocardial infarction was less frequent with GLP-1 RAs, whereas all-cause mortality was higher in the GLP-1 RA group. Composites of osteoporosis diagnoses were more frequent with tirzepatide than with GLP-1 RAs or SGLT2 inhibitors, whereas falls/fractures did not differ significantly between groups. Composites of sarcopenia diagnoses were less frequent with tirzepatide and GLP-1 RAs. In normal-weight adults with T2D and established ASCVD, the hazard of three-point MACE was lower with tirzepatide than with GLP-1 RAs and SGLT2 inhibitors, driven by fewer ischemic strokes, but was accompanied by higher osteoporosis diagnoses.