Mediating role and discriminative performance of early‑pregnancy triglyceride‑to‑adiponectin ratio in the relationship between pre-pregnancy overweight (BMI ≥ 25 kg/m²) and gestational diabetes mellitus in a Korean cohort

Although pre-pregnancy overweight status is a well-established risk factor for gestational diabetes mellitus (GDM), the mechanisms through which excess adiposity contributes to impaired glucose tolerance during pregnancy remain incompletely understood. This study aimed to examine whether the early-pregnancy serum triglyceride‑to‑adiponectin ratio (TAR) may partially explain the association between pre-pregnancy overweight status and GDM and to evaluate its exploratory discriminative performance for GDM. This secondary analysis included 589 pregnant women from a prospective cohort study conducted in South Korea. Participants were enrolled at 10–14 weeks of gestation (2014–2016), and GDM was diagnosed at 24–28 weeks of gestation using a two-step protocol. TAR was calculated from serum samples obtained at 10–14 weeks (triglycerides [mmol/L]/adiponectin [mg/L]). Pre-pregnancy BMI was categorized as overweight (≥ 25.0 kg/m²) or non-overweight (< 25.0 kg/m²). Logistic and linear regression models were used to assess the associations, and mediation analysis was performed to evaluate the indirect effect of TAR. The discriminative performance was assessed using receiver operating characteristic (ROC) curve analysis. Pre-pregnancy overweight (adjusted OR = 4.08; 95% CI: 1.73–9.63) and each 1 kg/m² increase in pre-pregnancy BMI (adjusted OR = 1.21; 95% CI: 1.09–1.34) were independently associated with an increased GDM risk. Each SD increase in TAR was also significantly associated with elevated GDM risk (adjusted OR = 1.89; 95% CI: 1.17–3.05). In exploratory tertile analysis, the highest tertile of TAR showed an elevated adjusted OR (10.54; 95% CI: 1.31–84.75). Linear regression showed a positive association between pre-pregnancy BMI and TAR (adjusted β = 0.03 per unit increase in BMI; 95% CI: 0.01–0.05). Mediation analysis suggested that approximately 20% of the association between pre-pregnancy overweight status and GDM may be statistically explained through TAR. ROC analysis demonstrated that TAR exhibited promising exploratory discrimination for GDM (AUC = 0.861; 95% CI: 0.801–0.922), with an optimal cutoff of 0.51 mmol/mg, yielding 86.5% sensitivity and 80.1% specificity in this cohort. When compared with a model containing both triglycerides and adiponectin as separate covariates, TAR performed comparably (AUC 0.861 vs. 0.859, P = 0.736), suggesting that its discriminative performance is derived largely from integrating the complementary information of its two components rather than adding new discriminative information. This study provides exploratory evidence that elevated early-pregnancy TAR is independently associated with an increased risk of GDM, and that the association between pre-pregnancy overweight status and GDM may be partially statistically explained through TAR. In this cohort, TAR showed promising exploratory discrimination, but its clinical utility, including any proposed threshold or discriminatory role, requires confirmation in larger and more diverse populations. ( https://clinicaltrials.gov/ ) Trial Number: NCT02276144, on October 10, 2014.