AUTHOR=Pan Yu-qing , Huang Xin-xin , Jiang Xiu-min TITLE=Risk factors and prediction model for low-birth-weight infants born to women with gestational diabetes mellitus JOURNAL=Frontiers in Public Health VOLUME=12 YEAR=2024 URL=https://www.frontiersin.org/journals/public-health/articles/10.3389/fpubh.2024.1432033 DOI=10.3389/fpubh.2024.1432033 ISSN=2296-2565 ABSTRACT=Aim

To explore the risk factors for low-birth-weight infants born to pregnant women with GDM to develop a prediction model and to construct a prediction nomogram for the risk of low birth weight infants born to pregnant women with GDM.

Methods

The clinical data of singleton infants diagnosed with gestational diabetes mellitus in Southeast China in 2019 were retrospectively reviewed. Gestational conditions and neonatal weight were assessed.

Results

A total of 386 diabetic mothers and infants were enrolled, including 193 in the low birth weight group and 193 in the normal birth weight group. The statistically significant factors were age over 36 years (OR = 1.916, 95% CI 1.048–3.505), junior high school education (OR = 4.454, 95% CI 1.882–10.543), history of fetal distress (OR = 0.120, 95% CI 0.016–0.925), gestational hypertension (OR = 3.681, 95% CI 1.357–9.986), preeclampsia (OR = 24.652, 95% CI 5.956–102.036), threatened preterm birth (OR = 18.393, 95% CI 8.457–39.999), triglycerides (OR = 0.642, 95% CI 0.485–0.850), and inadequate gestational weight gain (OR = 1.997, 95% CI 1.162–3.432). The area under the receiver operating characteristic curve (AUC) was 0.834 (95% CI: 0.794–0.874, p < 0.001), and the sensitivity and specificity were 82.38 and 87.56%, respectively. The goodness-of-fit test likelihood ratio 2 was 2.089 (p = 0.978). The comprehensive nomogram model showed that the discrimination and mean absolute error were 0.834 and 0.015, respectively. The calibration curves showed acceptable agreement between the predictions of the column line plots and the observations. The DCA curves showed good positive net yields in the prediction model.

Discussion

This study established a prediction model and risk score for low birth weight in pregnant women with GDM. It helps pregnancy clinics to identify the risk of low birth weight in newborns promptly, in addition to glycemic control and weight management for pregnant women with GDM, and should improve the appropriate treatment plan for pregnant women with higher risk, to provide personalized and precise treatment for pregnant women with GDM and improve infant outcomes.