【摘要】 目的 探讨糖尿病肾病(DN)患者并发高尿酸血症(HUA)的影响因素，并构建logistic回归预测模型。方法 选取2021年7月至2023年7月信阳市中心医院收治的325例DN患者为研究对象，根据是否并发HUA分为并发组(67例)、未并发组(258例)。采用单因素、多因素LASSO-logistic回归分析DN患者并发HUA的影响因素，基于多因素分析结果构建logistic回归预测模型，并检验该模型预测效能。结果 与未并发组相比，并发组体重指数(BMI)、糖化血红蛋白(Hb A1c)水平较高，DN病程较长，高血压、吸烟史、饮酒史占比较大(P<0.05); LASSO-logistic回归分析显示，BMI、高血压、DN病程、吸烟史、饮酒史、Hb A1c是DN患者并发HUA的独立影响因素(P<0.05);基于多因素结果，构建logistic回归风险预测模型，Logit(P)在0～1，检验结果显示，该模型预测DN患者并发HUA的曲线下面积为0.852(95%CI:0.749～0.957),Hosmer-Lemeshowχ~2=7.356,P=0.082。结论 BMI、高血压、DN病程、吸烟史、饮酒史、Hb A1c是DN患者并发HUA的独立影响因素，基于上述因素构建logistic回归模型具有较高预测价值，可作为临床预测并发HUA的有效模型，并对后续临床决策具有一定指导意义。更多还原

【Abstract】 Objective To explore the influencing factors of hyperuricemia( HUA) in patients with diabetic nephropathy( DN),and to construct a logistic regression prediction model. Methods A total of 325 patients with DN in Xinyang Central Hospital from July 2021 to July 2023 were selected as the research objects,and they were divided into concurrent group( 67 cases) and non concurrent group( 258 cases) according to whether they were complicated with HUA. Univariate and multivariate LASSOlogistic regression were used to analyze the influencing factors of patients with DN complicated with HUA. The logistic regression prediction model was constructed based on the results of multivariate analysis,and the prediction efficiency of the model was tested. Results Compared with the non complicated group,the concurrent group had higher body mass index( BMI) and glycosylated hemoglobin( Hb A1c) level,longer duration of DN,and a larger proportion of hypertension,smoking history and drinking history( P < 0. 05). LASSO-logistic regression analysis showed that BMI,hypertension,duration of DN,smoking history,drinking history and Hb A1c were the independent influencing factors of HUA in patients with DN( P < 0. 05). Based on the results of multiple factors,the logistic regression risk prediction model was constructed,and the Logit( P) was 0-1. The test results showed that the area under the curve of the model for predicting patients with DN complicated with HUA was 0. 852( 95% CI: 0. 749-0. 957),Hosmer-Lemeshow χ~2= 7. 356,P = 0. 082. Conclusion BMI,hypertension,DN duration,smoking history,drinking history and Hb A1c are the independent influencing factors of patients with DN complicated with HUA.The logistic regression prediction model based on the above factors has high predictive value,which can be used as an effective model for clinical prediction of complicated HUA,and has certain guiding significance for subsequent clinical decision-making.更多还原