BACKGROUND The aim of this study is to identify the association between the lipid profile of healthy nondiabetic, nonobese pregnant women in the first weeks of the third trimester of pregnancy and macrosomia or large-for-gestational-age (LGA) neonates with normal pregnancies. MATERIALS AND METHODS In this cohort study, 200 pregnant healthy women without gestational diabetes mellitus (GDM), obesity, or hypertension and carrying a single fetus in a prenatal clinic of a referral hospital were included based on a convenience sampling. Then, we took a blood sample to assess fasting blood sugar (FBS), triglyceride (TG), total cholesterol, low-density lipoprotein (LDL), and high-density lipoprotein (HDL). GDM was assessed after administering 50 g of oral glucose. All cases were followed until the end of pregnancy. The main outcome measurement was neonatal birth weight. RESULTS Only 154 mothers met eligibility criteria. There were eight cases (5.2%) with macrosomia (birth weight ≥ 4000 g) and 35 cases (22.7%) with LGA. Linear regression showed that mothers' TG and neonates' gender were independent predictors of the birth weight of the children (R-square = 0.52, p < 0.001). Logistic regression analysis showed that maternal FBS and TG are the most independent variables which can predict the presence of macrosomia (Nagelkerke R-square = 0.53, p < 0.001) and maternal TG and child gender are the most independent variables that can predict the presence of LGA in neonates of a healthy mother (Nagelkerke R-square = 0.49, p < 0.001). CONCLUSION Maternal triglyceride levels may be a significant predictor of fetal size in late pregnancy but not in early pregnancy. Our study reinforces that this is true not only in the case of macrosomia (birth weight > 4500 g), but also for LGA.