Effects of sodium/glucose cotransporter 2 inhibitors on the coagulation profile in patients with coronary-artery disease and type 2 diabetes mellitus: a retrospective cohort study

Figures

• 

  Figure 1. Participant selection flowchart.

• 

  Figure 2. Paired line plots for stratified analysis comparing coagulation profiles before and after treatment between the SGLT2i and non-SGLT2i groups of patients stratified into the normal reference range subgroup. (a), aPTT; (b), AT-III; (c), D-dimer; (d), INR; (e), PTA; (f), TT; (g), FDP; (h), PT. (b) Unlike the non-SGLT2i group, the SGLT2i group exhibits an upward shift in the box plot, indicating a higher median after treatment. (c) Unlike the non-SGLT2i group, the SGLT2i group exhibits a downward shift in the box plot, suggesting a lower median after treatment. (a, d–h) Similar to the non-SGLT2i group, the SGLT2i group exhibits no significant change in the median or trend of the connected lines after treatment. (b, c) AT-III and PTA follow a normal distribution, with the mean represented by a triangle.

• 

  Figure 3. Scatter plots illustrating the correlation between the predicted and observed values of (a) antithrombin III (AT-III), (b) prothrombin time (PT), and (c) D-dimer. Each black dot represents an individual observation. The red line indicates the fitted regression line, with the shaded area representing the confidence interval. (a, b) The strong alignment along the diagonal suggests a good model fit for AT-III and PT predictions. (c) While most data points closely align with the regression line, a few extreme values suggest potential outliers or high variability in D-dimer predictions.

• 

  Figure 4. Scatter plots illustrating the correlation between predicted and observed values for (a) antithrombin III (AT-III), (b) prothrombin time (PT), and (c) D-dimer after adjusting for diuretic use (yes/no) in the regression models. Each black dot represents an individual observation. The red line denotes the fitted regression line, and the shaded area represents the 95% confidence interval. (a, b) The close alignment of data points along the diagonal line indicates a good model fit for AT-III and PT. (c) While most data points cluster around the regression line, a few extreme values suggest potential outliers or increased variability in D-dimer predictions.

• 

  Figure 5. Proposed mechanisms by which SGLT2is modulated coagulation function in patients with CAD and T2DM via endothelial-function enhancement and anti-inflammatory effects. SGLT2is promoted nitric oxide production, suppressed the expression of inflammatory mediators, and reduced inflammatory-cell activation. These effects collectively mitigated the hypercoagulable state, decreased the risk of thrombosis, and could potentially reduce CV events in this high-risk patient population.