Feed

To investigate the long-term renal effects of tofogliflozin, with the shortest half-life among sodium-glucose cotransporter 2 inhibitors, in real-world clinical practice.

We retrospectively enrolled patients with type 2 diabetes mellitus treated at three affiliated hospitals between January 2015 and May 2024. The primary outcome was the change in estimated glomerular filtration rate from baseline to 72 months. After inverse probability of treatment weighting, linear mixed-effects models were used to compare estimated glomerular filtration rate changes between the tofogliflozin and non-tofogliflozin groups. As a sensitivity analysis, multiple imputation was performed to address missing baseline urinary albumin-to-creatinine ratio data. Secondary outcomes included estimated glomerular filtration rate slopes, changes in glycated hemoglobin and body weight at 72 months, and changes in urinary albumin-to-creatinine ratio at 48 months.

Among 598 participants (tofogliflozin, n = 245; non-tofogliflozin, n = 353), estimated glomerular filtration rate decline was significantly slower in the tofogliflozin group than in the non-tofogliflozin group (P < 0.001). At 72 months, the least squares mean estimated glomerular filtration rate change was -4.51 and -10.64 mL/min/1.73 m², respectively (between-group difference: 6.14 mL/min/1.73 m²). Significant between-group differences were observed in estimated glomerular filtration rate slope, glycated hemoglobin, and body weight, but not in urinary albumin-to-creatinine ratio. Renal findings were consistent in the multiple imputation sensitivity analysis.

Real-world clinical data over 72 months suggested that tofogliflozin may slow estimated glomerular filtration rate decline.