孕早期体脂分析对妊娠期糖尿病的预测价值研究

Predictive value of body fat analysis in early pregnancy for gestational diabetes mellitus

目的探讨妊娠妇女孕早期体脂分析对妊娠期糖尿病(gestational diabetes mellitus,GDM)的预测价值。方法纳入2018年8月至2020年1月于郑州大学第二附属医院妇产科门诊建卡符合纳入标准的所有初孕妇,收集其基本临床资料及体脂率(percentage of body fat,PBF)、内脏脂肪指数(visceral adiposity index,VAI)、总胆固醇(total cholesterol,TC)、甘油三酯(triglyceride,TG)、高密度脂蛋白(high density lipoprotein,HDL)及低密度脂蛋白(low density lipoprotein,LDL)等研究指标,随访至妊娠24~28周观察GDM的筛查结果。为控制体重指数(body mass index,BMI)这一混杂因素,将BMI水平进行分组,通过单因素分析及多因素logistic回归分析不同BMI分组下GDM的独立危险因素,进一步通过受试者工作曲线(receiver operating characteristic,ROC)计算曲线下面积(area under the curve,AUC)。结果收集有效样本量1130例,其中BMI<25的孕妇534例,有81例(15.2%)被诊断为GDM;25≤BMI<28的孕妇432例,有107例(24.8%)被诊断为GDM;BMI≥28的孕妇164例,有27例(16.5%)被诊断为GDM。通过多因素logistic回归分析及ROC曲线分析结果显示,在BMI<25孕妇组中PBF、VAI及TG预测GDM发生风险的AUC分别为0.839、0.838、0.829,最佳阈值点为28.25%、6.5、1.92 mmol/L,其敏感度为90.1%、81.5%、84.0%,特异度为69.8%、78.6%、71.3%;在25≤BMI<28孕妇组中PBF、VAI及TG预测GDM发生风险的AUC分别为0.761、0.769、0.773,最佳阈值点为28.8%、6.5、1.91mmol/L,其敏感度为86.9%、79.4%、70.1%,特异度为57.2%、66.2%、76.7%;在BMI≥28孕妇组中PBF、VAI及TG预测GDM发生风险的AUC分别为0.838、0.864、0.762,最佳阈值点为30.15%、6.5、2.18mmol/L,其敏感度为85.2%、83.2%、70.4%,特异度为73.7%、78.1%、76.6%。结论妊娠早期孕妇的PBF、VAI及TG水平均可作为GDM发生的独立预测指标。利用妊娠早期孕妇的PBF、VAI及TG水平评估低危孕妇GDM的发病风险,可及早预测低风险孕妇GDM的发生,并实施干预,以降低母儿不良妊娠结局的发生率。

Objective To study the predictive value of body fat analysis of pregnant women in early pregnancy for gestational diabetes mellitus(GDM).Methods Collect all primiparas,who met the inclusion criteria,at the outpatient clinic of the Second Affiliated Hospital of Zhengzhou University from August 2018 to January 2020,and collect basic clinical data and percentage of body fat(PBF),visceral adiposity index(VAI),total cholesterol(TC),triglycerides(TG),high density lipoprotein(HDL)and low density lipoprotein(LDL)and other research indicators,and they were followed up to 24-28 weeks of pregnancy to observe the screening results of GDM.In order to control the confounding factor of BMI,the pregnant women were grouped according to the BMI level,and the independent risk factors of GDM under different BMI groups were analyzed through single factor analysis and multivariate logistic regression analysis,and the receiver operating characteristic(ROC)was further used to calculate the area under the curve(AUC).Results This study collected 1130 effective samples,of which 534 pregnant women were with BMI<25,81 cases(15.2%)diagnosed as GDM;432 cases of pregnant women were with 25≤BMI<28,107 cases(24.8%)diagnosed as GDM;164 cases of pregnant women were with BMI≥28,27 cases(16.5%)diagnosed as GDM.The results of multivariate regression analysis and ROC curve analysis showed that in the pregnant women group with BMI<25,the AUC of PBF,VAI and TG in predicting GDM risk was 0.839,0.838 and 0.829,and the best threshold point was 28.25%,6.5 and 1.92 mmol/L;the sensitivities were 90.1%,81.5%and 84.0%,and the specificities were 69.8%,78.6%and 71.3%.In the pregnant group with 25≤BMI<28,the AUC of PBF,VAI and TG predicting GDM risk was 0.761,0.769 and 0.773,and the optimal threshold points were 28.8%,6.5,and 1.91 mmol/L;the sensitivity was 86.9%,79.4%,and 70.1%,and the specificity was 57.2%.66.2%and 76.7%.In pregnant women with BMI≥28,the AUC of PBF,VAI and TG predicting GDM risk was 0.838,0.864 and 0.762,and the best threshold points were 30.15%,6.5 and 2.18 mmol/L,the sensitivity was 85.2%,83.2%and 70.4%,and the specificity was 73.7%,78.1%and 76.6%.Conclusions The PBF,VAI and TG levels of pregnant women in early pregnancy can be used as independent predictors of the occurrence of GDM.Using the PBF,VAI and TG levels of pregnant women in early pregnancy to assess the risk of gestational diabetes in low-risk pregnant women can predict the occurrence of GDM in low-risk pregnant women as early as possible,and interventions can be implemented to reduce the incidence of adverse pregnancy outcomes.