A practical tool for risk stratification: eGDR as a predictor of incident diabetic kidney disease

Diabetic kidney disease (DKD) is a common complication of type 2 diabetes and a major cause of end-stage kidney failure. Insulin resistance contributes to its development, but most insulin resistance measures are not practical in daily care. The estimated glucose disposal rate (eGDR) uses common clinical data to estimate how the body processes glucose without relying on insulin levels. Previous studies found that lower eGDR values were linked to worse kidney and cardiovascular outcomes. This study looked at how eGDR was related to the 5-year risk of DKD and developed a simple model to predict that risk.

We analyzed data from adults with type 2 diabetes who were followed for five years and had no DKD at baseline. eGDR was calculated from waist circumference, HbA1c, and hypertension status. DKD was defined as new albuminuria or a decline in eGFR below 60 mL/min/1.73 m². Logistic regression tested the link between eGDR and DKD, and model accuracy and usefulness were assessed through discrimination, calibration, and decision-curve analysis.

During the 5-year follow-up, 27.1% of participants developed DKD. People with lower eGDR were more likely to develop DKD (adjusted OR = 0.86, 95% CI 0.76-0.97, p = 0.016). Those in the highest eGDR group had roughly half the risk compared with the lowest group (OR = 0.47, p = 0.017; p for trend = 0.021). When eGDR was combined with age, the model predicted DKD fairly well (C-index = 0.659) and matched observed results closely. Decision-curve analysis showed meaningful clinical benefit when predicted risk ranged from 0.10 to 0.35.

Lower eGDR was independently linked to a higher 5-year DKD risk in type 2 diabetes. A model based on eGDR and age may help identify moderate-risk patients and guide early kidney-protective care in everyday practice.

Not applicable.