预测妊娠期糖尿病发生风险的列线图的建立和评估  

Establishment and evaluation of a nomogram for predicting the risk of gestational diabetes mellitus

作　　者：陈秋景 冉爱冬[2] 黄春燕[1] Chen Qiujing;Ran Aidong;Huang Chunyan(Department of Endocrinology,Guangxi Hechi Peoples Hospital,Hechi Guangxi 547000,P.R.China;Department of Obstetrics,Guangxi Hechi Peoples Hospital,Hechi Guangxi 547000,P.R.China)

机构地区：[1]广西河池市人民医院内分泌科,广西河池547000 [2]广西河池市人民医院产科,广西河池547000

出　　处：《中国计划生育和妇产科》2022年第3期58-64,共7页Chinese Journal of Family Planning & Gynecotokology

摘　　要：目的 建立预测妊娠期糖尿病(gestational diabetes mellitus, GDM)发生风险的列线图模型,为更好防治GDM的发生提供指导。方法 选取2019年1月至12月广西河池市人民医院定期产检的孕妇作为调查对象,根据孕妇是否患有GDM,将其分为GDM组和非GDM组,采用自制调查表对孕妇进行调查。调查问卷内容包括基本信息、既往史、孕前情况、孕早期情况等内容。排除信息严重缺失及数据可疑的问卷。采用单因素分析和多因素Logistic回归分析筛选GDM的独立危险因素。利用R软件建立风险列线图模型,并对风险列线图模型的准确度和区分度进行验证。结果 本研究共收回问卷618份,有效问卷562份,有效率为90.94%。其中发生GDM的有131例,发生率为23.31%。经多因素Logistic回归分析,高龄、孕前高体质量指数、有糖尿病家族史、有GDM史、有巨大儿史、缺乏运动、有高血压、平均睡眠时间<7 h/d、孕期经常吃高热量食物、血红蛋白、甘油三酯、血清铁蛋白、孕早期空腹血糖升高为GDM发生的独立危险因素(P<0.05)。依此建立的列线图模型具有较好的区分度(AUC=0.988,95%CI:0.979-0.995)和准确度(H-L检验:χ^(2)=3.669,P=0.886)。结论 本研究依据GDM发生的重要风险因素构建了直观、个性化的GDM发生风险预测模型,经验证该预测模型具有较好的区分度和准确度,有助于临床医护人员或孕妇判断GDM发生风险,并制定针对性的预防措施,降低孕妇GDM的发生率。Objective A nomogram model for predicting the risk of gestational diabetes mellitus(GDM) was established to provide guidance for better prevention and treatment of GDM.Methods Pregnant women who received regular obstetric examinations in Guangxi Hechi People’s Hospital from January to December 2019 were selected as the survey subjects.According to whether the pregnant women suffered from GDM,the pregnant women were divided into GDM group and non-GDM group, and a self-made questionnaire was used to investigate the pregnant women.The contents of the questionnaire included basic information, past history, pre-pregnancy, and early pregnancy condition.Questionnaires with serious missing information and suspicious data were excluded.Univariate analysis and multivariate Logistic regression analysis were used to screen independent risk factors for GDM.The risk nomogram model was established by R software, and the accuracy and discrimination of the risk nomogram model were verified.Results A total of 618 questionnaires were recovered in this study, and 562 questionnaires were valid, with an effective rate of 90.94%.Among them, GDM occurred in 131 cases, the incidence rate was 23.31%.After multivariate Logistic regression analysis, advanced age, higher pre-pregnancy body mass index, family history of diabetes, history of GDM,history of large children, lack of exercise, hypertension, average sleep time <7 h/d, frequent eating high calorie food during pregnancy, hemoglobin, triglyceride, serum ferritin, and elevated fasting blood glucose in early pregnancy were independent risk factors for GDM(P<0.05).The nomogram model established on this basis had good discrimination(AUC=0.988,95% CI:0.979-0.995) and accuracy(H-L test: χ;=3.669,P=0.886).Conclusion In this study, an intuitive and personalized GDM risk prediction model was constructed based on the important risk factors of GDM.It has been verified that the prediction model has good discrimination and accuracy, and can assist clinical medical staff or pregnant women in jud

关 键 词：妊娠期糖尿病 风险评估 列线图 

分 类 号：R714.256[医药卫生—妇产科学]