Prediction of gestational diabetes mellitus using early-pregnancy data: a secondary analysis from a prospective cohort study in Iran

Abstract

Background: Early identification of gestational diabetes mellitus is essential for improving maternal and neonatal outcomes. While risk factors such as advanced maternal age, elevated pre-pregnancy body mass index, multiparity, and a history of gestational diabetes have been recognized, the role of serum biomarkers remains uncertain. This study explores the predictive value of early-pregnancy laboratory findings in conjunction with maternal demographic and clinical characteristics for gestational diabetes mellitus.

Methods: Early-pregnancy data from the first pregnancy visits at 6-12 weeks of gestation from women in the Mothers and Children's Health cohort were collected. Comprehensive maternal demographic data (e.g., age and body mass index) and obstetrics history (e.g., gravidity, parity, miscarriage, intrauterine growth retardation, gestational diabetes mellitus, and preeclampsia) were recorded. Maternal blood samples were analyzed for complete blood count and biochemistry parameters. Gestational diabetes mellitus was diagnosed based on 75-g oral glucose tolerance test results between 24 and 28 weeks of gestation, following the International Association of Diabetes and Pregnancy Study Groups criteria. Multivariate logistic regression analysis assessed the predictive capacity of various variables. Receiver operating curve analysis was conducted to identify optimal predictive cut-offs for continuous variables.

Results: 1,565 pregnant women with a mean age of 32.6 ± 5.7 years, mean body mass index of 25.5 ± 4.9 kg/m², mean gravidity of 1.1 ± 1.1, and mean parity of 0.8 ± 0.8 were included. 297 pregnancies (19.0%) were complicated by gestational diabetes mellitus. In the multivariate analysis, higher maternal age (p < 0.001, odds ratio = 1.076 [1.035-1.118]), a history of gestational diabetes mellitus (p < 0.001, odds ratio = 3.007 [1.787-5.060]) and preeclampsia (p = 0.007, odds ratio = 2.710 [1.310-5.604]), and elevated early-pregnancy fasting blood sugar (p < 0.001, odds ratio = 1.062 [1.042-1.083]) emerged as independent predictors of gestational diabetes mellitus. Moreover, the receiver operating curve yielded an optimal cut-off of 89.5 mg/dL for early-pregnancy fasting blood sugar in predicting gestational diabetes mellitus.

Conclusions: Our findings demonstrated that, in addition to established risk factors, a history of preeclampsia and elevated early-pregnancy fasting blood glucose are independent predictors of gestational diabetes mellitus. Therefore, close monitoring of pregnant women with these risk factors in early pregnancy is warranted to facilitate timely diagnostic and therapeutic interventions, reducing the burden of gestational diabetes.

Trial registration: Not applicable.

Keywords: Fasting blood sugar; Gestational diabetes mellitus; Prediction; Pregnancy complications.

Fig. 1

The flow diagram for the participant enrollment process

Fig. 2

Illustration of ROC curve analysis on the predictability of GDM based on the maternal FBS level in the first visit during the early pregnancy