妊娠期糖尿病患者产后2年内发生2型糖尿病的影响因素及预测模型的构建

Influencing factors of occurrence of type 2 diabetes in women with gestational diabetes mellitus within 2 years after delivery and prediction model construction

目的探讨妊娠期糖尿病(GDM)患者产后2年内发生2型糖尿病(T2DM)的影响因素及预测模型的构建。方法采用病例对照设计法对苏州大学附属苏州九院产检诊断为GDM并分娩的359例患者进行2年的随访。根据随访期间是否发生T2DM分为T2DM组和非T2DM组。对2组资料进行单因素分析,再行多因素Logistic回归分析并建立预测模型。采用拟合优度检验和受试者工作特征(ROC)曲线评价模型效能。结果产后2年随访中,53例失访,最终306例完成随访。随访期间未被诊断为T2DM的患者266例(非T2DM组),被确诊为T2DM的患者40例(T2DM组)。T2DM组与非T2DM组糖尿病家族史、孕前体质量指数(BMI)、诊断GDM时的口服葡萄糖耐量试验(OGTT)-2 h血糖(OGTT-2hPG)、孕期参加孕妇学校次数、产后BMI、产后内脏脂肪面积(VFA)等指标比较,差异有统计学意义(P<0.05)。多因素Logistic回归分析结果显示,糖尿病家族史、OGTT-2hPG、产后BMI、产后VFA均是GDM患者产后2年内发生T2DM的影响因素(P<0.05),孕期参加孕妇学校是保护因素(P<0.05)。Hosmer-Lemeshow拟合优度检验结果表明预测模型拟合度良好(χ^(2)=2.076、P=0.665)。模型ROC的曲线下面积为0.891(95%CI:0.828~0.954),最大约登指数对应的截断值为0.795,灵敏度为0.890,特异度为0.847。结论基于孕期OGTT-2hPG指标、孕期参加孕妇学校次数、糖尿病家族史、产后BMI、VFA指标建立的风险预测模型对GDM患者产后2年内的T2DM风险具有一定预测价值。

Objective To explore the influencing factors and prediction model construction of type 2 diabetes(T2DM)in women with gestational diabetes mellitus(GDM)within 2 years after delivery.Methods A total of 359 patients who diagnosed as GDM in prenatal examination and delivered in Suzhou Ninth Hospital Affiliated to Soochow University were selected for a 2-year follow-up.According to whether T2DM occurred during the follow-up period,the patients were divided into T2DM group and non-T2DM group.Univariate analysis was performed on data of the two groups,and multivariate Logistic regression analysis was performed to establish the prediction model.The goodness of fit test and receiver operating characteristic(ROC)curve were used to analyze and evaluate the effectiveness of the model.Results During the 2-year postpartum follow-up,53 cases fell off,and 306 patients completed the follow-up.Among the 306 patients who completed the 2-year follow-up,266 were not diagnosed with T2DM during the follow-up period(non-T2DM group),while 40 were diagnosed with T2DM(T2DM group).Statistically significant differences were observed between the T2DM andnon-T2DM groups in family history of diabetes,pre-pregnancy body mass index(BMI),2-hour postprandial glucose level(OGTT-2hPG)for GDM diagnosis,the number of visits to prenatal classes during pregnancy,postpartum BMI,and postpartum visceral fat area(VFA)(P<0.05).Multivariate Logistic regression analysis revealed that family history of diabetes,OGTT-2hPG,postpartum BMI,and postpartum VFA were influencing factors in the development of T2DM within 2 years postpartum among GDM patients(P<0.05),while attending prenatal classes during pregnancy emerged as a protective factor(P<0.05).The Hosmer-Lemeshow goodness-of-fit test indicated good fit of the prediction model(χ^(2)=2.076,P=0.665).The area under the ROC curve for the model was 0.891(95%CI,0.828 to 0.954),with a cutoff value of 0.795 corresponding to the maximum Youden index,a sensitivity of 0.890 and a specificity of 0.847.Conclusion The risk prediction model based on OGT-2hPG index during pregnancy,the number of pregnant women attending school during pregnancy,family history of diabetes,postpartum BMI,VFA index has a certain predictive value for the risk of T2DM in GDM patients within 2 years after delivery.