Abstract
Diabetic kidney disease (DKD) is the leading cause of end-stage renal disease in diabetes. Beyond chronic hyperglycemia, glycemic variability (GV) has emerged as a contributor to diabetic complications, but its prognostic role in DKD is unclear. This single-center retrospective cohort study enrolled 156 patients with DKD and followed them for a median of 36 months. GV was assessed using the coefficient of variation and patients were stratified into quartiles. The primary endpoint was a composite of ≥40% decline in estimated glomerular filtration rate (eGFR), end-stage renal disease, or kidney-related death. Cox regression and linear mixed-effects models were applied. Patients in the highest GV quartile had a significantly higher incidence of renal events than those in the lowest (43.6% vs 12.8%, P for trend = .015), corresponding to nearly a 3-fold increased risk (hazard ratio = 2.79, 95% confidence interval: 1.18-6.62). Each 1-standard deviation increase in coefficient of variation was associated with a 22% to 30% higher risk of adverse outcomes (all P < .05), and mixed-effects models confirmed a dose-response relationship with accelerated eGFR decline. Subgroup analyses showed stronger associations in patients with glycated hemoglobin A1c ≥7.0%, baseline eGFR <60 mL/min/1.73 m², and those not treated with SGLT2 inhibitors. In conclusion, higher GV is independently associated with adverse renal outcomes and faster eGFR decline in DKD. Incorporating GV into risk stratification may support individualized glycemic management and help delay disease progression.