Establishment of a nomogram model for predicting the risk of diabetic nephropathy in diabetic patients

Objective To analyze the risk factors of diabetic nephropathy in patients with type 2 diabetes,and to develop and verify a visidualized evaluation tool to assist in clinical prediction of diabetic nephropathy.Methods A total of 559 patients with type II diabetes mellitus (T2DM) who met the criteria were selected, including 280 patients in the alone T2DM group (control group) and 279 patients in the complicated microalbuminuria group (DN group). Clinical data were collected, and factors related to diabetic nephropathy (DN) were screened by univariate analysis. Statistically significant variables were included in multivariate logistic regression model to analyze DN risk factors. R-software was used to construct a nomogram model for predicting DN risk. Bootstrap was used to verify the model, to plot the ROC curve, and to calculate the predictive performance of the C-index evaluation model. Conformance testing is performed by plotting the calibration curves of the predicted and actual results. The Hosmer-Lemeshow test was used to judge the goodness of fit of the model. P>0.05 indicates that the model has good goodness of fit. Results Age, course of diabetes, neutrophil count, anemia, triglycerides, body mass index, diabetic peripheral neuropathy, and thyroid stimulating hormone(TSH) were associated with the development of diabetic nephropathy(both P<0.05), and aging, TSH>4.6 mU/L, triglyceride ≥ 1.7 mmol/L, diabetic peripheral neuropathy, and diabetes course >1 year were independent risk factors for T2DM complicated with DN(both P<0.05). These factors were included and a nomogram model was successfully constructed. The nomogram model had good prediction performance, with an area under the ROC curve of 0.852 (95% CI=0.822 to 0.882) and an internal verification C-index of 0.846.The calibration curve showed a good correlation between the predicted results and the actual results (P=0.178).Conclusions The individualized nomogram model for predicting the early risk of DN patients constructed in this study has a high resolution and high clinical application value. It has guiding significance for screening high-risk population with DN and formulating intervention strategies.