基于中医体质及危险因素的妊娠期糖尿病预测模型的研究

Study on the Prediction Model of Gestational Diabetes Based on TCM Constitution and Risk Factors

[目的]探讨妊娠期糖尿病(GDM)的易感中医体质类型及危险因素,构建预测模型并进行初步验证。[方法]选取2020年8月至2021年7月建档并定期产检的孕妇作为研究对象,收集一般临床资料、中医体质类型及妊娠8~14周实验室检测指标,根据妊娠24~28周糖耐量检查确诊为GDM的孕妇153例作为GDM组,抽取同期糖耐量正常的孕妇150例作为对照组,应用Logistic回归分析GDM的独立危险因素,建立GDM发病风险预测模型,采用受试者工作特征曲线(ROC)评估模型预测效能。[结果]孕妇年龄、孕前BMI(体质量指数)、妊娠8~14周FPG(血浆空腹血糖)、痰湿质是GDM的独立危险因素(P<0.05),将这些相关危险因素建立GDM预测模型,ROC曲线分析显示:孕妇年龄、孕前BMI、妊娠8~14周FPG用于预测GDM的AUC(曲线下面积)分别为0.585、0.687、0.666;模型的AUC为0.817,敏感度为66.7%,特异度为86.0%,Hosmer-Lemeshow拟合优度检验,χ2=5.869,P=0.662,显示模型预测价值较高。[结论]年龄、孕前BMI、妊娠8~14周FPG、痰湿质是GDM独立危险因素,对GDM预测有一定价值,联合危险因素建立的GDM预测模型可提高预测效能。

[Objective]To explore the susceptible TCM constitution types and risk factors of gestational diabetes mellitus(GDM),and to construct a predictive model for preliminary verification.[Methods]In this study the pregnant women who took regular examination after filing from August 2020 to July 2021 were selected as the research objects.The general clinical data,TCM constitution types and laboratory test indexes at 8~14 weeks of pregnancy were collect⁃ed.According to the glucose tolerance test at 24~28 weeks of pregnancy,153 pregnant women diagnosed with GDM were selected as the GDM group,and 150 pregnant women with normal glucose tolerance during the same period were selected as the control group.Logistic regression was used to analyze the independent risk factors of GDM,and the risk prediction model of GDM was established.The receiver operating characteristic curve(ROC)was used to evaluate the prediction efficiency of the model.[Results]Age,pre-pregnancy BMI(body mass index),FPG(plasma fasting blood glucose)at 8~14 weeks of pregnancy,and phlegm-dampness constitution were independent risk factors for GDM(P<0.05).These related risk factors were used to establish a GDM prediction model.The ROC curve analysis showed that AUC(area under the curve)of pregnant women’s age,pre-pregnancy BMI,and FPG at 8~14 weeks of pregnancy for predicting GDM were 0.585,0.687,and 0.666,respectively.The AUC of the model was 0.817,the sensitivity was 66.7%,and the specificity was 86.0%.Hosmer-Lemeshow goodness of fit test,χ2=5.869,P=0.662.It showed that the model had high predictive value.[Conclusion]Age,pre-pregnancy BMI,FPG at 8~14 weeks of pregnancy,and phlegm-dampness constitution are independent risk factors for GDM,which have certain value in predicting GDM.GDM prediction model combined with the risk factors were established to promote the prediction efficiency.