Abstract
INTRODUCTION: Gestational diabetes mellitus's (GDM's) prevalence in Sri Lanka ranges from 5.5% to 11.5%. It is associated with maternal and perinatal complications, emphasizing the need for early screening and intervention. This study aims to determine the predictive effect of early pregnancy lipid profile and fasting plasma glucose for GDM. METHODS: It is a prospective cohort study of 172 pregnant women attending antenatal clinics at a tertiary hospital in Jaffna, Sri Lanka. Prediction was derived by calculating odds ratios (ORs) and 95% confidence intervals (CIs) in multivariable logistic regression, assessing lipid and glucose effects on GDM risk. RESULTS: The study included 172 participants (mean age: 29.84±5.38). GDM's prevalence was 16.9%, and 57.14% of these mothers were obese. Significant differences in fasting plasma glucose (FPG) values were observed between the first visit and at 24-28 weeks. GDM mothers showed elevated total cholesterol and low-density lipoprotein (LDL) levels. Triglyceride (TG) levels correlated significantly with FPG at the Point of Assessment (POA), identifying a 0.945 mmol/L cutoff with 75% sensitivity and 77.1% specificity. Logistic regression confirmed a significant TG-GDM relationship. There is an association between FPG levels measured in early pregnancy and the likelihood of developing GDM later on. Specifically, when FPG levels in early pregnancy surpass a cutoff value of 3.94 mmol/L, there is an increased risk of GDM, indicated by an OR of 3.81 Conclusion: Early pregnancy FPG and TG levels are potential markers for predicting GDM. FPG shows higher predictive efficacy than TG. Total cholesterol, LDL, and high-density lipoprotein (HDL) lack predictive ability.