摘要
目的评估孕8~16周孕妇空腹血脂与血浆葡萄糖(FPG)水平预测妊娠期糖尿病(GDM)的价值。方法选取行常规产前检查的孕妇878例,根据糖耐量情况将其分为GDM组(n=99)与糖耐量正常组(n=779),收集孕妇的一般资料,检查孕8~16周的血脂指标与FPG指标。采用Logistics回归分析评估GDM的独立危险因素,建立发生GDM风险的预测模型,采用Hosmer-Lemeshow拟合优度检验评估预测模型拟合能力。结果孕8~16周孕妇FPG水平与葡萄糖耐量试验(OGTT)检查空腹(r=0.447)、服糖后1 h(r=0.178)、服糖后2 h(r=0.143)的血糖值呈正相关(均P〈0.05)。孕前体质量指数(BMI)(OR=1.115,95%CI:1.052~1.172)、孕8~16周FPG(OR=8.148,95%CI:5.423~12.247)、年龄(OR=1.057,95%CI:1.014~1.103)、三酰甘油(TG)水平(OR=1.471,95%CI:1.159~1.869)与孕次〉3次(OR=1.779,95%CI:1.082~2.931)是孕妇发生GDM的独立危险因素(P〈0.05)。将孕前BMI、孕8~16周、年龄、TG与孕次〉3次等危险因素纳入Logistics回归分析,建立回归方程模型,Hosmer-Lemeshow拟合优度检验显示P=0.955,该模型预测GDM发生风险的ROC曲线下表面积(AUC)为0.762(95%CI=0.729~0.793,P=0.028),cut-off值为0.12,敏感度为68.6%,1-特异度为71.6%。结论将孕8~16周孕妇的实验室指标联合临床指标建立预测模型,可以更好地预测孕妇发生妊娠期糖尿病的风险。
Objective To evaluate the value of fasting plasma lipid levets and fasting plasma glucose (FPG) levels in predicting gestational diabetes mellitus (GDM) in pregnant women between 8 and 16 weeks. Methods A total of 878 pregnant women with routine prenatal examination were selected. According to the glucose tolerance of pregnant women, they were divided into GDM group ( n = 99) and normal glucose tolerance group (n = 779 ). General data of pregnant women were collected, and the blood lipid index and FPG index were examined at 8 - 16 weeks of gestation. The independent risk factors of GDM were evaluated by Logistic regression analysis, and the prediction model of GDM risk was established. The Hosmer-Lemeshow goodness-of-fit test was used to evaluate the fitting capacity of GDM model. Results FPG levels were positively correlated with blood glucose of OGTF at fasting ( r = 0. 447 ), after taking glucose for 1 h (r =0. 178) and after taking glucose for 2 h (r =0. 143), respectively (P 〈 0. 05). Body mass index (BMI) before pregnancy ( OR = 1. 115, 95% CI: 1. 052 - 1. 172), FPG in 8-16weeks of gestation (OR =8.148, 95%CI: 5.423 - 12. 247), age (OR = 1.057, 95%CI: 1. 014 - 1. 103), triglyceride (TG) ( OR = 1. 471, 95% CI: 1. 159 - 1. 869) and pregnancies 〉 3 times ( OR = 1. 779, 95 % CI: 1. 082 - 2.931 ) were independent risk factors of GDM in pregnant women ( P 〈 0. 05 ). The regression model was established by BMI, 8 - 16 weeks of gestation, age, TG and pregnancy times 〉 3. The Hosmer-Lemeshow goodness-of-fit test showed P = 0. 955, which predicted the ROC curve under risk of GDM occurrence had a surface area AUC of 0. 762 ( 95% CI = 0. 729 - 0. 793, P = 0.028), a cut-off value of 0. 12, a sensitivity of 68.6% and a 1-specificity of 71.6%.Conclusions Gestational diabetes mellitus in pregnanc women (GDM) can be predicted by establishing a predictive model combining laboratory indexes and clinical indexes of pregnant women with 8 - 16 weeks gestational diabetes.
作者
栾福娟
Luan Fujuan(Department of Obstetrics and Gynecology, Jiaozhou People's Hospital, Jiaozhou 266300, Chin)
出处
《中国实用医刊》
2017年第19期50-53,共4页
Chinese Journal of Practical Medicine
关键词
妊娠期糖尿病
血糖
血脂
预测模型
Gestational diabetes mellitus
Blood glucose
Blood lipid
Prediction model