Effects of dapagliflozin on inflammatory biomarkers and cardiac function in patients with acute myocardial infarction after percutaneous coronary intervention and type 2 diabetes: an exploratory observational study with propensity score matching

Abstract Background Acute myocardial infarction (AMI) triggers a robust inflammatory response that contributes to adverse ventricular remodeling. Sodium-glucose cotransporter-2 (SGLT2) inhibitors have demonstrated anti-inflammatory properties beyond glycemic control, but their effects on inflammatory biomarkers in AMI patients after percutaneous coronary intervention (PCI) with comorbid type 2 diabetes mellitus (T2DM) remain poorly characterized. Patients with comorbid type 2 diabetes mellitus (T2DM) are at particularly high risk due to amplified inflammation, coronary microvascular dysfunction, and elevated restenosis rates following PCI. Methods This single-center, prospective exploratory observational cohort study enrolled 60 AMI patients with comorbid T2DM after successful PCI at Baoji Central Hospital between January 2023 and June 2025 (Ethics: BZYL2022-26-1). No single primary endpoint was pre-specified because of the exploratory design. Patients were categorized into dapagliflozin ( n = 41) and control ( n = 19) groups based on clinical treatment decisions; 4 patients (all control) were excluded, yielding 56 for analysis. Three inflammatory biomarkers (IL-10, TNF-α, MMP-9), cardiac function parameters (LVEF, SV, LVEDD, LVESD), and metabolic indices were assessed at baseline, 1 month, 6 months, and 1 year. Bonferroni correction (α′ = 0.0167) and propensity score matching (PSM) were applied. Results The principal exploratory finding was significantly elevated IL-10 at 1 year in the dapagliflozin group vs. controls (3.22 ± 3.51 vs. 1.36 ± 0.67 pg/mL, P = 0.001), surviving Bonferroni correction (α′ = 0.0167). MMP-9 declined in both groups (dapagliflozin: -66.2%, P < 0.001; control: -48.7%), consistent with natural post-AMI decline, with no significant between-group difference at 1 year ( P = 0.323). Within the dapagliflozin group, LDL-C decreased by 20.8% ( P < 0.001) and LVESD by 8.8% ( P = 0.002). After PSM (14 pairs), IL-10 between-group differences remained significant ( P = 0.028). Conclusions In this exploratory study, dapagliflozin was associated with significantly elevated IL-10 at 1 year—the only inflammatory biomarker showing a significant between-group difference surviving Bonferroni correction and PSM. MMP-9 declined in both groups, consistent with natural post-AMI inflammation, without significant between-group difference. These hypothesis-generating findings are consistent with a possible anti-inflammatory effect of dapagliflozin post-AMI and warrant confirmation in larger randomized trials.