A nomogram prediction model incorporating noninvasive lens AGEs and conventional biochemical indicators for assessing and predicting diabetic kidney disease

Diabetic kidney disease (DKD) is a major microvascular complication of diabetes, associated with high morbidity, mortality, and healthcare costs. Early detection is challenging due to the lack of highly specific diagnostic tools. This study aimed to develop and validate a nomogram for predicting DKD risk in patients with type 2 diabetes mellitus (T2DM) by integrating conventional biochemical indicators with noninvasive lens advanced glycation end product (AGE) measurements. A total of 868 patients with T2DM from Shanghai Fifth People's Hospital (November 2019 to February 2024) were enrolled. Independent predictors of DKD were identified using logistic regression and incorporated into a predictive nomogram. Model performance was evaluated using receiver operating characteristic curves, calibration plots, and decision curve analysis. The cutoff value for lens AGE fluorescence (0.306) was derived from our previous study using ROC analysis with Youden's index. Seven variables were independently associated with DKD: systolic blood pressure, glycated hemoglobin, triglycerides, serum cystatin C, 25-hydroxyvitamin D, red blood cell count, and lens AGE values. The nomogram showed good discrimination, with an AUC of 0.809 in the training cohort and 0.806 in the validation cohort. Calibration demonstrated close agreement between predicted and observed risk, and clinical utility was confirmed. This novel nomogram provides a practical and highly specific tool for early DKD screening and individualized risk assessment in T2DM patients.