Long-Term, Real-World Kidney Outcomes with SGLT2i versus DPP4i in Type 2 Diabetes without Cardiovascular or Kidney Disease

Background Contemporary guidelines recommend the use of sodium-glucose cotransporter 2 inhibitors (SGLT2is) independently of glycemic control in patients with type 2 diabetes and those with kidney disease, with heart failure, or at high risk of cardiovascular disease. Using a large Israeli database, we assessed whether long-term use of SGLT2is versus dipeptidyl peptidase 4 inhibitors (DPP4is) is associated with kidney benefits in patients with type 2 diabetes overall and in those without evidence of cardiovascular or kidney disease. Methods Patients with type 2 diabetes who initiated SGLT2is or DPP4is between 2015 and 2021 were propensity score-matched (1:1) according to 90 parameters. The kidney-specific composite outcome included confirmed ≥40% decline in eGFR or kidney failure. The kidney-or-death outcome included also all-cause mortality. Risks of outcomes were assessed using Cox proportional hazard regression models. The between-group difference in eGFR slope was also assessed. Analyses were repeated in patients’ subgroup lacking evidence of cardiovascular or kidney disease. Results Overall, 19,648 propensity score-matched patients were included; 10,467 (53%) did not have evidence of cardiovascular or kidney disease. Median follow-up was 38 months (interquartile range, 22– 55). The composite kidney-specific outcome occurred at an event rate of 6.9 versus 9.5 events per 1000 patient-years with SGLT2i versus DPP4i. The respective event rates of the kidney-or-death outcome were 17.7 versus 22.1. Compared with DPP4is, initiation of SGLT2is was associated with a lower risk for the kidney-specific (hazard ratio [HR], 0.72; 95% confidence interval [CI], 0.61 to 0.86; P < 0.001) and kidney-or-death (HR, 0.80; 95% CI, 0.71 to 0.89; P < 0.001) outcomes. The respective HRs (95% CI) in those lacking evidence of cardiovascular or kidney disease were 0.67 (0.44 to 1.02) and 0.77 (0.61 to 0.97). Initiation of SGLT2is versus DPP4is was associated with mitigation of the eGFR slope overall and in those lacking evidence of cardiovascular or kidney disease (mean between-group differences 0.49 [95% CI, 0.35 to 0.62] and 0.48 [95% CI, 0.32 to 0.64] ml/min per 1.73 m² per year, respectively). Conclusions Long-term use of SGLT2is versus DPP4is in a real-world setting was associated with mitigation of eGFR loss in patients with type 2 diabetes, even in those lacking evidence of cardiovascular or kidney disease at baseline.