KORNELIUS EDYYI-SUN YANGCHIEN-NING HUANG2026-10-162026-10-162025-01-01Pure ID: 15364335Pure UUID: d795a4af-3e4c-4ea6-a38a-9e9c383904af10.3389/fphar.2025.1686851https://www.scopus.com/pages/publications/105020705840https://scholars.csmu.edu.tw/handle/123456789/29900Aims: Dual inhibition of sodium-glucose cotransporter (SGLT) 1 and 2 with canagliflozin may offer additional metabolic effects beyond selective SGLT2 inhibition; however, its comparative cardiovascular associations remain uncertain. This study compared the risks of major adverse cardiovascular events (MACE) and all-cause mortality between canagliflozin and selective SGLT2 inhibitors in routine clinical practice. Methods and results: We conducted a retrospective cohort study using a multicenter electronic health record database including over 118 million patients. Adults with type 2 diabetes, no prior cardiovascular disease, and new use of an SGLT inhibitor between January 2016 and December 2023 were identified. After applying strict exclusion criteria and 1:1 propensity score matching, 24,078 patients (mean age, 57 years; 47% women) were included: 12,039 initiated canagliflozin and 12,039 initiated other SGLT2 inhibitors. The primary outcome was MACE (composite of myocardial infarction, stroke, or all-cause mortality). Compared with other SGLT2 inhibitors, canagliflozin was associated with higher risk of MACE (hazard ratio [HR], 1.23; 95% confidence interval [CI], 1.14–1.33) and all-cause mortality (HR, 1.49; 95% CI, 1.33–1.68). Hemorrhagic stroke risk was also elevated (HR, 1.35; 95% CI, 1.02–1.79), while risks of ischemic stroke and myocardial infarction were similar. Conclusion: In this large real-world cohort, patients initiating canagliflozin had higher observed event rates for a composite of myocardial infarction, stroke, or all-cause mortality compared with those initiating selective SGLT2 inhibitors. These associations should be interpreted as exploratory and hypothesis-generating, given the observational design and differences from randomized trial evidence. Further research is needed to clarify potential differences among SGLT2 inhibitors in routine practice.enjournal article