Objective To investigate the association of triglyceride to high density lipoprotein-cholesterol ratio (TG/HDL-C) in early pregnancy with the risk of gestational diabetes mellitus (GDM).

Methods Retrospectively collected clinical data of singleton pregnant women who received regular antenatal care and delivered at Peking Union Medical College Hospital from July 2020 to June 2022. Based on the results of the 75 g oral glucose tolerance test (OGTT) from 24 to 28 weeks, pregnant women were classified into GDM group and normal glucose tolerance (NGT) group. Multiple Logistic regression was used to evaluate the correlation between TG /HDL-C in early pregnancy (8-12⁺⁶weeks) and GDM, and triglyceride-glucose (TyG) index was used as a reference to assess the value of TG/HDL-C in early pregnancy in predicting GDM.

Results A total of 1617 singleton pregnant women who met the inclusion and exclusion criteria were enrolled, with 372 (23.01%) in the GDM group and 1245 (76.99%) in the NGT group. After adjusting for confounding factors, such as maternal age, ethnicity, pre-pregnancy BMI, GDM history and family history of diabetes, pregnant women in the highest TG/HDL-C quartile had a 2.46-fold higher risk of developing GDM than those in the lowest TG/HDL-C quartile (OR=2.46, 95% CI: 1.73-3.51). Pregnant women in the highest TyG index quartile had a 2.36-fold higher risk of developing GDM than those in the lowest TyG index quartile (OR=2.36, 95% CI: 1.67-3.37). The efficacy of TG/HDL-C in early pregnancy in predicting GDM (area under the curve: 0.607 vs. 0.608) and the degree of improvement in the basic risk model of GDM (net reclassification improvement: 0.240 vs. 0.270; integrated discrimination improvement: 0.022 vs. 0.024) were both close to the TyG index.

Conclusions Higher TG/HDL-C in early pregnancy was independently associated with higher risk of GDM. Its predictable value was comparable to that of TyG index.