【摘要】 目的：构建并评估2型糖尿病患者并发高尿酸血症的临床预测模型，探究其独立危险因素。方法：回顾性分析2020年1月至2021年12月昭通市中医医院内分泌科收治的553例2型糖尿病患者的病历资料，按是否并发高尿酸血症将患者分为高尿酸血症组和非高尿酸血症组。采用LASSO回归筛选预测变量并构建Logistic回归模型，筛选出2型糖尿病患者并发高尿酸血症的独立危险因素，绘制列线图对模型进行可视化展示，并通过区分度、校准度、临床适用度对模型的预测效能进行评估。结果：2组患者在体重指数、并发糖尿病肾病、规范降糖治疗、总胆固醇、甘油三酯、血清肌酐、血尿素氮、胰岛素抵抗指数、湿热困脾证、气阴两虚证等方面差异有统计学意义（P<0.05）；对LASSO回归筛选出8个预测变量进行多重共线性诊断，删除气阴两虚证后构建预测模型并进行多因素Logistic回归分析。结果显示7个预测变量均为2型糖尿病患者并发高尿酸血症的独立危险因素（P<0.05），该模型的区分度、校准度、临床适用度均较好。结论：体重指数、并发糖尿病肾病、规范降糖治疗、总胆固醇、甘油三酯、胰岛素抵抗指数、湿热困脾证是2型糖尿病患者并发高尿酸血症的独立危险因素，以此为基础构建的临床预测模型能够为临床防治2型糖尿病并发高尿酸血症提供可靠依据。更多还原

【Abstract】 Objective: To construct and evaluate a clinical prediction model for type 2 diabetes patients with hyperuricemia and explore its independent risk factors. Methods: Retrospective analysis was conducted on the medical records of 553 type 2 diabetes patients admitted to the department of endocrinology of Zhaotong Hospital of Traditional Chinese Medicine from January 2020 to December 2021. The patients were divided into hyperuricemia group and non-hyperuricemia group according to whether hyperuricemia was complicated or not. LASSO regression was used to select predictive variables and construct a Logistic regression model to identify independent risk factors for type 2 diabetes patients with hyperuricemia. The model was visualized by drawing a nomogram, and the predictive efficacy of the model was evaluated by the degree of differentiation, calibration and clinical applicability. Results: There were statistically significant differences(P<0.05) between the two groups of patients in body mass index, concurrent diabetic nephropathy, standardized hypoglycemic therapy, total cholesterol, triglycerides, serum creatinine, blood urea nitrogen, insulin resistance index, dampness-heat obstructing the spleen syndrome, and qi-yin deficiency syndrome. Eight predictive variables were selected by LASSO regression for multicollinearity diagnosis, and the prediction model was constructed after the qi-yin deficiency syndrome was deleted and multi-factor Logistic regression was performed. The results showed that seven predictive variables were independent risk factors for type 2 diabetes patients with hyperuricemia(P<0.05), and the model had good differentiation, calibration and clinical applicability. Conclusion: Body mass index, concurrent diabetic nephropathy, standardized hypoglycemic therapy, total cholesterol, triglycerides, insulin resistance index, and dampness-heat obstructing the spleen syndrome are independent risk factors for type 2 diabetes patients with hyperuricemia. The clinical prediction model based on these factors can provide a reliable basis for the clinical prevention and treatment of hyperuricemia in patients with type 2 diabetes.更多还原