Abstract
OBJECTIVE: This study aimed to investigate the independent association between the early pregnancy triglyceride to high density lipoprotein cholesterol (TG/HDL-C) ratio and the risks of gestational diabetes mellitus (GDM) and neonatal hypoglycemia (NH), with a focus on its dose-response relationship and mediating pathways. METHODS: This retrospective cohort study included 17,056 women with singleton pregnancies who were registered and delivered at our hospital. Early pregnancy fasting lipid profiles were collected, and the TG/HDL-C ratio was calculated. Propensity score matching (PSM) was applied to balance baseline characteristics. Logistic regression, restricted cubic spline models, and mediation analysis were employed to assess associations, nonlinear relationships, and potential mediating pathways. Subgroup analyses were conducted to explore effect heterogeneity across populations with different clinical features. RESULTS: After PSM, balanced cohorts were obtained (10,131 cases for GDM analysis and 2,123 cases for NH analysis). A significant dose-response relationship was observed between increasing TG/HDL-C ratio quartiles and GDM risk. In the fully adjusted model after matching, the adjusted odds ratios (aORs) for Q2-Q4 were 1.20, 1.44, and 1.69, respectively (all P < 0.05). A nonlinear relationship was identified, with a threshold inflection point at 2.91; below this value, the risk increased more steeply. In contrast, the TG/HDL-C ratio was associated with NH only at the highest quartile (Q4: aOR = 1.39, P < 0.05), with no clear dose-response trend. Mediation analysis revealed that 11.14% of the effect of the TG/HDL-C ratio on NH was mediated through GDM, while the mediating effect of GH was minimal (1.41%). CONCLUSION: An elevated TG/HDL-C ratio in early pregnancy is an independent risk factor for GDM, demonstrating a significant dose-response relationship with a nonlinear threshold. The inflection point of 2.91 may serve as a practical reference for early risk stratification. The association with NH is modest and partially mediated through GDM. Monitoring the TG/HDL-C ratio at the initial prenatal visit may facilitate early identification of high-risk individuals, enabling targeted interventions to reduce GDM incidence and potentially lower the risk of neonatal hypoglycemia through improved GDM management.