Fasting plasma glucose in the first trimester is related to gestational diabetes mellitus and adverse pregnancy outcomes

Abstract

Purpose: To investigate and identify first-trimester fasting plasma glucose (FPG) is related to gestational diabetes mellitus (GDM) and other adverse pregnancy outcomes in Shenzhen population. Methods: We used data of 48,444 pregnant women that had been retrospectively collected between 2017 and 2019. Logistic regression analysis was used to evaluated the associations between first-trimester FPG and GDM and adverse pregnancy outcomes, and used to construct a nomogram model for predicting the risk of GDM. The performance of the nomogram was evaluated by using ROC and calibration curves. Decision curve analysis (DCA) was used to determine the clinical usefulness of the first-trimester FPG by quantifying the net benefits at different threshold probabilities. Results: The mean first-trimester FPG was 4.62 ± 0.42 mmol/L. A total of 6998 (14.4%) pregnancies developed GDM.489(1.01%) pregnancies developed polyhydramnios, the prevalence rates of gestational hypertensive disorder (GHD), cesarean section, primary cesarean section, preterm delivery before 37 weeks (PD) and dystocia was 1130 (2.33%), 20,426 (42.16%), 7237 (14.94%), 2386 (4.93%), and 1865 (3.85%), respectively. 4233 (8.74%) of the newborns were LGA, and the number of macrosomia was 2272 (4.69%), LBW was 1701 (3.51%) and 5084 (10.49%) newborns had admission to the ICU, which all showed significances between GDM and non-GDM groups (all P < 0.05). The univariate analysis showed that first-trimester FPG was strongly associated with risks of outcomes including GDM, cesarean section, macrosomia, GHD, primary cesarean section, and LGA (all OR > 1, all P < 0.05), furthermore, the risks of GDM, primary cesarean section, and LGA was increasing with first-trimester FPG as early as it was at 4.19–4.63 mmol/L. The multivariable analysis showed that the risks of GDM (ORs for FPG 4.19–4.63, 4.63–5.11 and 5.11–7.0 mmol/L were 1.137, 1.592, and 4.031, respectively, all P < 0.05) increased as early as first-trimester FPG was at 4.19–4.63 mmol/L, and first-trimester FPG which was also associated with the risks of cesarean section, macrosomia and LGA (OR for FPG 5.11–7.0 mmol/L of cesarean section: 1.128; OR for FPG 5.11–7.0 mmol/L of macrosomia: 1.561; OR for FPG 4.63–5.11 and 5.11–7.0 mmol/L of LGA: 1.149 and 1.426, respectively, all P < 0.05) and with its increasing, the risks of LGA increased. Furthermore, the nomogram had a C-indices 0.771(95% CI: 0.763~0.779) and 0.770(95% CI:0.758~0.781) in training and testing validation respectively, which showed an acceptable consistency between the observed, validation and nomogram-predicted probabilities, the DAC curve analysis indicated that the nomogram had important clinical application value for GDM risk prediction. Conclusions: FPG in the first trimester was an independent risk factor for GDM which can be used as a screening test for identifying pregnancies at risk of GDM and adverse pregnancy outcomes.