Abstract
PURPOSE: To investigate the associations between maternal glucose and lipid metabolism and adverse pregnancy outcomes in women with type 2 diabetes mellitus (T2DM). METHODS: A retrospective cohort of pregnant women with T2DM who delivered at Peking University First Hospital was included. Univariate and multivariate logistic regression were used to identify risk factors. ROC curves were applied to determine optimal cut-off points for predicting adverse pregnancy outcomes. RESULTS: A total of 398 pairs of pregnant women and newborns from 2019 to 2023 were included. The key complications included large-for-gestational-age (LGA, 21.6%), preeclampsia (PE, 14.3%), preterm birth (14.3%) and birth defects (11.6%). The achievement rates for both glycated hemoglobin A1c (HbA1c) and lipid levels in early pregnancy were approximately 50%. HbA1c in the third trimester was significantly associated with increased birth weight (for LGA, aOR = 3.819, 95% CI [1.534-9.509]). Elevated triglyceride (TG) in early pregnancy were critical risk factors for LGA, PE, and preterm birth (aOR = 2.627, 2.478, 2.233; all p < 0.05). The optimal cut-off point for third-trimester HbA1c predicting LGA was ≤5.95%. The optimal cut-off points for first-trimester TG predicting LGA, PE, and preterm birth were separately ≤1.51, ≤1.67, and ≤1.46 mmol/L. CONCLUSIONS: In Chinese pregnant women with T2DM, elevated plasma glucose and lipid levels at different pregnancy stages were independently linked to adverse outcomes, especially third-trimester HbA1c and first-trimester TG. TG level in early pregnancy lower than currently recommended may be beneficial to improve adverse outcomes.