Yang Jing, Jiang Sheng
State Key Laboratory of Pathogenesis, Prevention and Treatment of High Incidence Diseases in Central Asia; Department of Endocrinology, The First Affiliated Hospital of Xinjiang Medical University, Urumqi, 830017, People's Republic of China.
Int J Gen Med. 2022 May 20;15:5089-5101. doi: 10.2147/IJGM.S363474. eCollection 2022.
To develop a nomogram model that predicts the risk of diabetic nephropathy (DN) incidence in type 2 diabetes mellitus (T2DM) patients.
We collect information from electronic medical record systems. The data were split into a training set (n=521) containing 73.8% of patients and a validation set (n=185) holding the remaining 26.2% of patients based on the date of data collection. Stepwise and multivariable logistic regression analyses were used to screen out DN risk factors. A predictive model including selected risk factors was developed by logistic regression analysis. The results of binary logistic regression are presented through forest plots and nomogram. Lastly, the c-index, calibration plots, and receiver operating characteristic (ROC) curves were used to assess the accuracy of the nomogram in internal and external validation. The clinical benefit of the model was evaluated by decision curve analysis.
Predictors included serum creatinine (Scr), hypertension, glycosylated hemoglobin A1c (HbA1c), blood urea nitrogen (BUN), body mass index (BMI), triglycerides (TG), and Diabetic peripheral neuropathy (DPN). Harrell's C-indexes were 0.773 (95% CI:0.726-0.821) and 0.758 (95% CI:0.679-0.837) in the training and validation sets, respectively. Decision curve analysis (DCA) demonstrated that the novel nomogram was clinically valuable.
Our simple nomogram with seven factors may help clinicians predict the risk of DN incidence in patients with T2DM.
建立一种预测2型糖尿病（T2DM）患者发生糖尿病肾病（DN）风险的列线图模型。
我们从电子病历系统中收集信息。根据数据收集日期，将数据分为一个包含73.8%患者的训练集（n = 521）和一个包含其余26.2%患者的验证集（n = 185）。采用逐步和多变量逻辑回归分析筛选出DN危险因素。通过逻辑回归分析建立包含所选危险因素的预测模型。二元逻辑回归的结果通过森林图和列线图呈现。最后，使用c指数、校准图和受试者工作特征（ROC）曲线评估列线图在内部和外部验证中的准确性。通过决策曲线分析评估模型的临床益处。
预测因素包括血清肌酐（Scr）、高血压、糖化血红蛋白A1c（HbA1c）、血尿素氮（BUN）、体重指数（BMI）、甘油三酯（TG）和糖尿病周围神经病变（DPN）。训练集和验证集的Harrell's C指数分别为0.773（95%CI：0.726 - 0.821）和0.758（95%CI：0.679 - 0.837）。决策曲线分析（DCA）表明新的列线图具有临床价值。
我们的包含七个因素的简单列线图可能有助于临床医生预测T2DM患者发生DN的风险。
