妊娠期糖尿病患者启动药物干预风险模型的建立

Construction of a risk model for initiating pharmacological intervention in gestational diabetes mellitus

目的构建妊娠期糖尿病(GDM)启动药物干预的风险模型。方法回顾性分析2015年3月至2017年4月香港大学深圳医院GDM患者669例,比较药物干预组(n=110)和生活方式干预组(n=559)各临床指标的差异,利用logistic回归建立风险模型,应用受试者工作特征(ROC)曲线及Hosmer-Lemeshow拟合优度检验(HL检验)进行评价,并选取75例GDM进行内部验证。结果本研究纳入孕28周前的孕期体重增加(ΔW)、孕前体重指数(BMI)、口服糖耐量试验(OGTT)-0 h、OGTT-2 h共4个危险因素,建立风险模型:Z=0.270×ΔW+0.167×孕前BMI+1.768×OGTT-0 h+0.371×OGTT-2 h-18.787。模型ROC曲线下面积为0.809,HL检验P>0.05。内部验证结果:模型灵敏度64.7%,特异度79.3%,准确率76.0%。结论本模型有一定临床参考价值,但其泛化能力及模型的稳定性和适用性有待进一步验证。

Objective To establish a risk model of pharmacological intervention for gestational diabetes mellitus(GDM).Methods We retrospectively analyzed the data of 669 GDM in the University of HongKong-Shenzhen Hospital from Mar.2015 to Apr.2017.And indicators of pharmacological intervention group(n=110)and lifestyle intervention group(n=559)were compared.Logistic regression was used to establish a risk model.The ROC curve and Hosmer-Lemeshow goodness of fit test were used to test the model′s efficiency.A total of 75 GDM patients were selected to evaluate the model′s effect as internal validation.Results The four risk factors including weight gain during the first and second trimester(Δweight),pre-pregnancy BMI,OGTT-0 h and OGTT-2 h were included in the risk model.The model formula was:Z=0.270×Δweight+0.167×BMI+1.768×OGTT-0 h+0.371×OGTT-2 h-18.787.The AUC was 0.809,with Hosmer-Lemeshow goodness of fit test P>0.05.The internal validation showed the sensitivity,specificity,and the accuracy of this model were 64.7%,79.3%and 76.0%,respectively.Conclusion The risk model has certain clinical reference value,but its generalization ability and the stability and applicability of the model need further verification.