Research Article: The Association between Maternal 25-Hydroxyvitamin D Concentration during Gestation and Early Childhood Cardio-metabolic Outcomes: Is There Interaction with Pre-Pregnancy BMI?

Date Published: August 5, 2015

Publisher: Public Library of Science

Author(s): E. Jessica Hrudey, Rebecca M. Reynolds, Adriëtte J. J. M. Oostvogels, Ingeborg A. Brouwer, Tanja G. M. Vrijkotte, Cheryl S. Rosenfeld.


Both maternal 25-hydroxyvitamin D(25OHD) status and pre-pregnancy BMI(pBMI) may influence offspring cardio-metabolic outcomes. Lower 25OHD concentrations have been observed in women with both low and high pBMIs, but the combined influence of pBMI and 25OHD on offspring cardio-metabolic outcomes is unknown. Therefore, this study investigated the role of pBMI in the association between maternal 25OHD concentration and cardio-metabolic outcomes in 5-6 year old children. Data were obtained from the ABCD cohort study and 1882 mother-child pairs were included. The offspring outcomes investigated were systolic and diastolic blood pressure, heart rate, BMI, body fat percentage(%BF), waist-to-height ratio, total cholesterol, LDL cholesterol, HDL cholesterol, triglycerides, glucose, C-peptide, and insulin resistance(HOMA2-IR). 62% of the C-peptide samples were below the detection limit and were thus imputed using survival analysis. Models were corrected for maternal and offspring covariates and tested for interaction with pBMI. Interaction with pBMI was observed in the associations with insulin resistance markers: in offspring of overweight mothers(≥25.0kg/m2), a 10 nmol/L increase in maternal 25OHD was associated with a 0.007(99%CI:-0.01,-0.001) nmol/L decrease in C-peptide and a 0.02(99%CI:-0.03,-0.004) decrease in HOMA2-IR. When only non-imputed data were analyzed, there was a trend for interaction in the relationship but the results lost significance. Interaction with pBMI was not observed for the other outcomes. A 10 nmol/L increase in maternal 25OHD was significantly associated with a 0.13%(99%CI:-0.3,-0.003) decrease in %BF after correction for maternal and child covariates. Thus, intrauterine exposure to both low 25OHD and maternal overweight may be associated with increased insulin resistance in offspring, while exposure to low 25OHD in utero may be associated with increased offspring %BF with no interactive effects from pBMI. Due to the limitations of this study, these results are not conclusive, however the observations of this study pose important research questions for future studies to investigate.

Partial Text

Exposure to certain nutritional factors in utero, such as insufficient maternal 25-hydroxyvitamin D (25OHD), may be related to adverse cardio-metabolic outcomes in offspring [1, 2]. Maternal 25OHD deficiency can occur during pregnancy, in part as a result of fetal demand, and the prevalence varies from 5 to 67% depending on location, ethnicity, and definition of deficiency [2–4]. Laboratory and observational evidence suggest that an individual’s own 25OHD status may influence the risk for developing chronic diseases, such as type 2 diabetes mellitus and cardiovascular disease [3, 5] and some studies suggest that maternal 25OHD status during gestation also influences this risk in offspring [2, 6, 7]. Therefore it is possible that maternal 25OHD deficiency during gestation contributes to cardio-metabolic abnormalities in offspring, which then track into adulthood and increase the risk of future chronic disease [2, 3, 8].

2150 eligible mother-child pairs did not complete follow-up as a result of non-response, leaving Amsterdam or death. The mothers in this non-response group were younger, less educated, more often non-Dutch, and were more likely to have smoked during pregnancy (Table 1). Of those who did complete follow-up, 17.7% of the women had deficient 25OHD levels and 20.7% had insufficient levels (Table 2). The mean maternal 25OHD concentration in the study sample was 60.4 nmol/L (interquartile range, 18.9–102.7 nmol/L), which was measured at a mean gestational age of 16.1 weeks. Maternal 25OHD deficiency was more likely if blood samples were drawn in winter, pBMI was higher, and women were younger, less educated, not Dutch, multiparous, smokers during pregnancy, and not users of vitamin D supplements. The children of deficient mothers were generally older and more sedentary. 25OHD deficiency was more prevalent in both underweight (26.4%) and overweight (31.5%) women when compared to normal weight women (13.6%), and the relationship between pBMI and 25OHD was significantly non-linear (p = 0.002) (Fig 2). In the study sample there were 1426 normal weight, 72 underweight and 384 overweight women.

This study observed inverse associations between maternal 25OHD and the markers of IR (C-peptide and HOMA2-IR) in children whose mothers were overweight during pregnancy. Additionally, an inverse relationship was observed between maternal 25OHD and offspring %BF, but no interaction with pBMI was observed in this association or for any other outcomes. A non-linear association between pBMI and maternal 25OHD was also observed.

The results of this study suggest that low maternal 25OHD is associated with increased IR in the children of overweight women, and that low maternal 25OHD is associated with increased %BF in children with no interactive effects from pBMI. Due to the limitations of this study, these results are not conclusive, however the observations of this study pose important research questions for future studies to investigate.