Abstract Aims/hypothesis Sodium-glucose cotransporter 2 inhibitors (SGLT2i) are indicated for the treatment of type 2 diabetes, and evidence shows that they may confer a cardioprotective effect to reduce the incidence of cardiovascular conditions in obese type 2 diabetes patients. The mechanisms underlying this effect remain unclear, but an SGLT2i, canagliflozin (CANA), has been reported to increase levels of fibroblast growth factor-21 (FGF21) and enhance activation of its downstream target proteins. We thus aimed to study whether the effect on FGF21 by CANA can reduce obesity-related cardiac injury. Methods 1) Animal study: Sixteen-week-old normal diet and high-fat diet male mice were treated with or without 30 mg/kg CANA once a day for 24 weeks, and were subsequently examined for body weight, blood glucose levels, adipose tissue changes, lipid accumulation in the liver, and cardiac function. 2) Cell study: An in vitro study that treated H9c2 cardiomyocyte cells cultured in a simulated high-lipid environment containing palmitic acid (PA) with FGF21 recombinant protein was also conducted to examine protein expression profiles. Furthermore, the effect of CANA on FGF21 expression. 3) Human clinical trial: A randomized double-blinded control trial (NCT 05764811) was performed to test effect of CANA on MRI liver fibrosis scores of obese type 2 diabetes patients (n = 40). Results 1) High-fat diet mice treated with CANA for 24 weeks had lower average body weight than untreated controls (34 g vs. 40 g, p < 0.001). In addition, CANA treatment stimulated FGF21 protein production in liver tissue and plasma at higher levels for both normal diet and high-fat diet mice, and reduced expression levels of proteins associated with fibrosis and apoptosis. 2) H9c2 cardiomyocyte culture studies suggested that a simulated high-lipid environment with PA caused higher expression of apoptotic proteins, but the addition of FGF21 recombinant protein significantly reduced apoptosis signalling. 3) MRI before and after 4 weeks of CANA treatment in obese type 2 diabetes patients showed a significant reduction in liver fibrosis scores for those receiving CANA treatment. Conclusion CANA treatment stimulated FGF21 protein expression and reduced fibrosis and apoptosis signalling protein expression. Moreover, reductions in liver fibrosis scores were observed in obese type 2 diabetes patients treated with CANA. Taken together, this suggests that CANA-induced FGF21 expression may exert a cardioprotective effect against fibrosis and apoptosis and can potentially reduce the risk of obesity-related cardiac injury.