摘要
目的研究2型糖尿病(type 2 diabetes mellitus,T2DM)患者生化指标预后预警筛查模型。方法研究对象为2017年9月至2019年8月收治的274例T2DM患者,其中49例合并T2DM并发症,另225例无并发症;采集两组一般临床资料、血尿生化指标、代谢组学指标及眼科检查结果,将单因素分析有差异纳入Logistic回归模型,逐步模式筛选建立T2DM生化指标预后预警筛查模型,并绘制受试者工作特征(ROC)曲线分析模型对T2DM并发症的预测价值。结果年龄(X1)、饮酒(X2)、病程(X3)、BMI(X4)、CHO(X5)、hs-CRP(X6)、Cys C(X7)、ADPN(X8)、β-假尿苷(X9)、D-氨基葡萄糖酸(X10)、棕榈酸(X11)、CARE(X12)成功进入模型;Logistic回归风险筛查模型预测T2DM并发症的ROC曲线下面积值为0.780(0.712~0.849),约登指数0.554,截断值0.955,对应灵敏度、特异性分别为67.30%、87.10%。结论T2DM生化指标预后预警筛查模型对T2DM并发症具有良好的预测价值,值得临床重视。
Objective To study the early warning and screening model using biochemical indicators forprognosis of patients with type 2 diabetes mellitus(T2DM).Methods 274 patients with T2DM admitted from September,2017 to August,2019 were selected as study subjects,including 49 patients with T2DM complications and 225 patients without.The general clinical data,biochemical indicators of hematuria,metabolomics indicators and ophthalmic test results were collected for all patients.Factors with significant impact from the univariate analysis were included in the logistic regression model,and the early warning and screening model using biochemical indicators for prognosis of T2DM was established by a stepwise pattern.The receiver operating characteristic(ROC)curve was plotted to analyze the predictive value for T2DM complications.Results Age(X1),drinking(X2),course of disease(X3),BMI(X4),CHO(X5),hs-CRP(X6),Cys C(X7),ADPN(X8),β-pseudouridine(X9),D-glucosamine(X10),palmitic acid(X11)and CARE(X12)were successfully enrolled in the model.The area under the ROC curve of logistic regression risk screening model for predicting complications of T2DM was 0.780(0.712-0.849).The Yoden index was 0.554,and the cutoff value was 0.955.The corresponding sensitivity and specificity were 67.30%and 87.10%,respectively.Conclusion The early warning and screening using biochemical indicators for prognosis of T2DM can be used to predict complications of T2DM and it is worthy of clinical attention.
作者
王星星
WANG Xingxing(Department of Laboratory,Shenzhen Longhua District People′s Hospital,Shenzhen 437600,China)
出处
《标记免疫分析与临床》
CAS
2020年第4期603-608,共6页
Labeled Immunoassays and Clinical Medicine
基金
深圳市龙华区科技创新资金项目(编号:2017088)。
关键词
2型糖尿病
生化指标
预后
预警筛查
Type 2 diabetes mellitus
Biochemical indicators
Prognosis
Early warning and screening