摘要
目的探讨2型糖尿病肾病患者高凝血状态的风险因素并构建列线图模型。方法回顾性分析2020年1月—2024年6月本院收治的200例2型糖尿病肾病患者的临床资料,根据血浆纤维蛋白原(FIB)水平将患者分为高凝血组(FIB≥4 g/L,n=82)与非高凝血组(FIB<4 g/L,n=118)。比较两组患者的临床资料;lasso回归初筛2型糖尿病肾病患者高凝血状态的潜在预警指标;多因素二元logistic回归分析患者高凝血状态的风险因素;构建列线图模型;Hosmer-Lemeshow检验、校准曲线、决策曲线评价列线图模型的拟合优度、校准度及临床适用性;绘制模型的受试者工作特征(ROC)曲线图。结果两组患者在透析时间、TC、TG、SCr、CRP方面的差异具有统计学意义(P<0.05)。透析时间、TC、TG、SCr、CRP是2型糖尿病肾病患者高凝血状态的潜在预警指标。透析时间、TC、TG、SCr、CRP是2型糖尿病肾病患者高凝血状态的风险因素。Hosmer-Lemeshow检验结果(χ^(2)=5.539,P=0.699)提示列线图模型的拟合优度较高。列线图模型对2型糖尿病肾病患者高凝血状态的预测效果与实际结果一致性高。当风险阈值>0.10时,与任何单一预测因子相比,该列线图模型预测2型糖尿病肾病患者高凝血状态的临床净获益更高。绘制列线图模型的ROC曲线,曲线下面积(AUC)为0.962,95%CI为0.925~0.984,灵敏度82.93%,特异度99.15%,提示该列线图模型具有较高的预测效能。结论本研究基于透析时间、TC、TG、SCr、CRP构建的列线图模型能够有效预测患者的高凝血风险,为临床决策提供了有力的工具,帮助制定个性化的治疗方案,提高糖尿病肾病患者的管理水平。
Objective To explore the risk factors of hypercoagulability in type 2 diabetes nephropathy patients and construct a nomogram model.Methods The clinical data of 200 patients with type 2 diabetes nephropathy who were admitted to our hospital from January 2020 to June 2024 were retrospectively analyzed.According to the level of plasma fibrinogen(FIB),the patients were divided into hyperthrombosis group(FIB≥4 g/L,n=82)and non hyperthrombosis group(FIB<4 g/L,n=118).Compare the clinical data of two groups of patients.Lasso regression was used to screen the potential early warning indicators of hypercoagulability in type 2 diabetes nephropathy patients.Multivariate binary logistic regression was used to analyze the risk factors for high coagulation status in patients.Build a column chart model;Hosmer Lemeshow test,calibration curve,and decision curve are used to evaluate the goodness of fit,calibration,and clinical applicability of the column chart model.Draw the Receiver Operating Characteristic(ROC)curve of the model.Results The differences in dialysis time,TC,TG,SCr,and CRP between the two groups of patients were statistically significant(P<0.05).Dialysis time,TC,TG,SCR,CRP are potential early warning indicators of hypercoagulability in patients with type 2 diabetes nephropathy.Dialysis time,TC,TG,SCR,CRP are risk factors of hypercoagulability in type 2 diabetes nephropathy patients.The Hosmer Lemeshow test results(χ^(2)=5.539,P=0.699)indicate a high degree of goodness of fit for the column chart model.The prediction effect of nomogram model on hypercoagulant state in type 2 diabetes nephropathy patients is highly consistent with the actual results.When the risk threshold is>0.10,compared with any single predictor,this nomogram model has a higher clinical net benefit in predicting hypercoagulability in patients with type 2 diabetes nephropathy.Draw the ROC curve of the column chart model,with an area under the curve(AUC)of 0.962,a 95%confidence interval of 0.925-0.984,a sensitivity of 82.93%,and a specificity of 99.15%,indicating that the column chart model has high predictive performance.Conclusions The nomogram model based on dialysis time,TC,TG,SCR,CRP in this study can effectively predict the high coagulation risk of patients,provide a powerful tool for clinical decision-making,help to develop personalized treatment programs,and improve the management level of patients with diabetes nephropathy.
作者
陈浩华
聂寒
吴奇
Chen Haohua;Nie Han;Wu Qi(Affiliated Hospital of Jiujiang University,Jiujiang,Jiangxi 332000,China)
出处
《齐齐哈尔医学院学报》
2025年第16期1560-1564,共5页
Journal of Qiqihar Medical University
基金
江西省卫生健康委科技计划项目(202131092)
江西省中医药管理局科技计划项目(2022B384)。
关键词
2型糖尿病肾病
高凝血状态
风险因素
列线图
Type 2 diabetes nephropathy
Hypercoagulable state
Risk factors
Nomogram
