目的：本研究旨在探究2型糖尿病患者并发远端对称性多发性神经病变（distal symmetric polyneuropathy，DSPN）的危险因素，并构建关于DSPN风险的预测列线图模型。方法：选取2021年7月至2024年7月 在武汉科技大学附属孝感医院收治的2型糖尿病患者1658例作为研究对象，收集一般资料和临床数据，按 7∶3比例将全部数据划分为训练集和验证集。训练集数据采用LASSO回归和二元Logistic回归建立列线图 模型，验证集数据进行模型验证。通过校准曲线、受试者工作特征曲线（receiver operator characteristic curve，ROC）下面积、决策曲线（decision curve analysis，DCA）评价模型的一致性、区分度和临床应用价值。 结果：研究发现，糖尿病病程（OR=1.195, 95%CI: 1.116-1.280）、空腹血糖（fasting plasma glucose，FPG）（OR= 1.614, 95%CI: 1.435-1.816）、中性粒细胞与淋巴细胞比值（neutrophil to lymphocyte ratio，NLR）（OR=1.388, 95%CI: 1.042-1.849）、尿微量白蛋白（microalbuminuria，mALB）（OR=1.536, 95%CI: 1.113-2.120）是DSPN的 独立危险因素，而高密度脂蛋白胆固醇（high-density lipoprotein cholesterol，HDL-C）（OR=0.252, 95%CI: 0.160-0.397）、25-羟基维生素D（25-Hydroxyvitamin D，25(OH)D）（OR=0.845, 95%CI: 0.825-0.864）则是独立 保护因素。校准曲线示列线图模型预测DSPN风险概率与实际概率的一致性较好。训练组中列线图预测 模型预测DSPN发生的AUC为0.896（95%CI: 0.878-0.913），验证组中列线图预测模型预测 DSPN 发生的 AUC 为0.888（95%CI: 0.860-0.917）。DCA图显示模型在较大的阈值范围内具有临床应用价值。结论：本 研究成功开发了一种基于糖尿病病程、FPG、NLR、mALB、HDL-C和25(OH)D等关键预测因子的高精度列 线图预测模型。

This study aims to identify the risk factors associated with distal symmetric polyneuropathy (DSPN) in patients with type 2 diabetes mellitus (T2DM) and to develop a predictive model for assessing the risk of DSPN. Methods: A total of 1,658 eligible T2DM patients were selected from Xiaogan Hospital, affiliated with Wuhan University of Science and Technology, between July 2021 and July 2024. General and clinical data were collected and then split into a training set and a validation set in a 7 ∶ 3 ratio. LASSO regression and binary logistic regression were employed to develop a nomogram model using the training set data, which was subsequently validated with the validation set data. The model’s accuracy, discrimination, and clinical applicability were assessed using calibration curves, the area under the receiver operating characteristic (ROC) curve, and decision curve analysis (DCA). Results: The analysis identified duration of diabetes (OR=1.195, 95% CI: 1.116-1.280), fasting blood glucose (FPG) (OR=1.614, 95% CI: 1.435-1.816), neutrophil-to-lymphocyte ratio (NLR) (OR=1.388, 95%CI: 1.042-1.849) and urinary microalbumin (mALB) (OR=1.536, 95% CI: 1.113-2.120) as independent risk factors for DSPN. Conversely, high-density lipoprotein cholesterol (HDL-C) (OR=0.252, 95%CI: 0.160-0.397) and 25-hydroxyvitamin D[25(OH)D] (OR= 0.845, 95% CI: 0.825-0.864) were identified as independent protective factors. The nomogram model’s predicted DSPN risk probability closely aligned with the actual probability, as demonstrated by the calibration curve. The area under the curve (AUC) for DSPN prediction was 0.896 (95% CI: 0.878-0.913) in the training group and 0.888 (95%CI: 0.860-0.917) in the validation group. The decision curve analysis (DCA) indicated that the model holds significant clinical value across a wide range of thresholds. Conclusion: This study successfully developed a highly accurate nomogram prediction model based on key predictors, including diabetes duration, FPG, NLR, mALB, HDL-C, and 25(OH)D.