Endocrine Abstracts

Objectives: Maternal metabolism has a major impact on foetal growth, and the risk of developing obesity, cardiovascular disease and diabetes in later life. Identification of maternal metabolic parameters in early pregnancy that predict birth weight (BW), is pivotal in the prevention of these diseases. We evaluated whether maternal triglyceride (TG) or fructosamine levels in early pregnancy, as possible reflectors of maternal insulin resistance (IR), could predominantly contribute to BW and whether this is sex dependent.

Study design: The data were obtained from the Amsterdam Born Children and their Development cohort study. Non-fasting TG and fructosamine levels were determined in early gestation (median 13 weeks). Associations between maternal TG and fructosamine levels, and BW - small for gestational age (SGA) - large for gestational age (LGA), were analysed for each sex separately.

Results: In total 3514 pregnant women were included. With every increase of 1 mmol/l TG, the BW increased significantly by 81.7 g. This increase was larger with boys (107.3 g, 95% CI 66.0-148) compared to girls (60.5 g, 95% CI 23.6-97.4). However, no association was found with fructosamine. The results were adjusted for gestational age at blood sampling, total duration of pregnancy, maternal height, age, parity, ethnicity, educational level, smoking, alcohol and pre-pregnancy BMI. These covariates were also used in a different statistical test (R-squared), and explained 29.2% of the variance in BW. Adding fructosamine to this model, had no added value in predicting BW (R² stayed 0.292). Contrary, TG levels raised the R² from 0.292 to 0.299 (P<0.001). In total 8.3% children in our population were LGA. The odds of a new-born LGA with higher maternal TG levels were increased (OR 1.6, 95% CI 1.3-2.0). No increased odds were found for fructosamine levels (OR 1.0, 95% CI 1.0-1.0).

Conclusions: This study shows that fructosamine levels, measured in the first trimester of a physiological pregnancy, are not significantly associated with BW, in contrast to maternal TG levels. This association is more prominent with boys. Our data suggest that the lipid profile in early pregnancy is more predictive for pregnancy outcomes, like LGA, than glycaemic metabolic parameters, such as fructosamine. This may give a different focus on metabolic variables (TG vs fructosamine) involved in early patterns of maternal IR during a physiologic pregnancy. Additional studies could show whether maternal TG levels should be included in the screening or follow-up of pregnancies with a pronounced IR (e.g. in gestational diabetes).