Background
Diabetic nephropathy (DN) has emerged as the leading cause of chronic kidney disease, with a significant proportion of DN patients progressing to end-stage kidney disease (ESKD), profoundly affecting their quality of life. Currently, no single clinical marker reliably predicts the likelihood and timing of progression to ESKD in DN patients. This study aims to develop a non-invasive
... [Show full abstract] predictive model to evaluate the risk and timing of ESKD onset in this population.
Methods
This study retrospectively analyzed data from 140 biopsy-confirmed DN patients. Key predictive variables were identified using multivariate Cox regression analysis, and a visual predictive nomogram was developed. The model was subsequently evaluated for its predictive performance.
Results
Of the 140 DN patients, 81 progressed to ESKD. Multivariate analysis identified estimated glomerular filtration rate, common logarithm of albumin-creatinine ratio, cystatin C, hemoglobin, and fibrinogen as independent predictors of progression to ESKD. Based on these significant factors, a nomogram was constructed. The area under the time-dependent receiver operating characteristic curve at 1, 2, 3, and 5 years were 0.898 (95% CI: 0.839–0.958), 0.889 (95% CI: 0.818–0.959), 0.876 (95% CI: 0.785–0.968), and 0.893 (95% CI: 0.796–0.990), respectively. Calibration curves demonstrated strong concordance between predicted and observed outcomes, while decision curve analysis indicated substantial net clinical benefit for practical application.
Conclusions
This study developed a predictive model to assess the risk and timing of ESKD progression in DN patients. As a quantitative tool, this model enables clinicians to estimate the 5-year risk of ESKD, facilitating timely interventions to improve patient outcomes.
