Maternal BMI Associations with Maternal and Cord Blood Vitamin D Levels in a North American Subset of Hyperglycemia and Adverse Pregnancy Outcome (HAPO) Study Participants

Abstract

Objective: Obesity in pregnancy may be associated with reduced placental transfer of 25-hydroxyvitamin D (25-OHD). The objective of this study was to examine associations between maternal BMI and maternal and cord blood levels of 25-OHD in full term neonates born to a single racial cohort residing at similar latitude. Secondary objectives were to examine associations between maternal glucose tolerance with maternal levels of 25-OHD and the relationship between cord blood 25-OHD levels and neonatal size.

Methods: This study was conducted among participants of the Hyperglycemia and Adverse Pregnancy Outcomes (HAPO) Study meeting the following criteria: residing at latitudes 41-43°, maternal white race, and gestational age 39-41 weeks. Healthy pregnant women underwent measures of height, weight, and a 75-g fasting oral glucose tolerance test (OGTT) at approximately 28 weeks gestation. Maternal and cord blood sera were analyzed for total 25-OHD by HPLC tandem mass spectrometry. Statistical analyses included ANOVA and linear regression models.

Results: Maternal and cord blood (N = 360) mean levels (sd) of 25-OHD were 37.2 (11.2) and 23.4 (9.2) ng/ml, respectively, and these levels were significantly different among the 3 field centers (ANOVA p< 0.001). Maternal serum 25-OHD was lower by 0.40 ng/ml for BMI higher by 1 kg/m2 (p<0.001) in an adjusted model. Maternal fasting plasma glucose, insulin sensitivity, and presence of GDM were not associated with maternal serum 25-OHD level when adjusted for maternal BMI. Cord blood 25-OHD was lower by 0.26 ng/ml for maternal BMI higher by 1 kg/m2 (p<0.004). With adjustment for maternal age, field center, birth season and maternal serum 25-OHD, the association of cord blood 25-OHD with maternal BMI was attenuated. Neither birth weight nor neonatal adiposity was significantly associated with cord blood 25-OHD levels.

Conclusion: These results suggest that maternal levels of 25-OHD are associated with maternal BMI. The results also suggest that interpretation of neonatal 25-OHD levels may need to incorporate specific maternal factors in addition to season of birth and latitude.

Fig 1. Vitamin D levels by field center.

Bars display mean levels +/- sd, solid gray for maternal serum and hatched light grey for cord blood. Cleveland, OH: n = 118; Maternal 25-OHD = 36.1 ± 10.6, cord blood = 25.0 ± 10.0 ng/ml; Chicago, IL: n = 112; Maternal 25-OHD = 41.7 ± 10.7, cord blood = 24.7 ± 9.0 ng/ml; Toronto, ON: n = 130; Maternal 25-OHD = 34.1 ± 11.0, cord blood = 20.6 ± 8.0 ng/ml. Field center levels of 25-hydroxyvitamin D were significantly different for both maternal and cord blood, (ANOVA p< 0.001).

Fig 2. Maternal BMI versus Cord Blood 25-OHD Levels.

Scatterplot of cord blood 25-hydroxyvitamin D versus maternal BMI measured at OGTT, N = 360. Both the regression line (dotted line, β estimate = -0.26) and loess curve demonstrate the inverse association between these two variables.