Vitamin D status in pregnancy and childhood associates with intelligence quotient at age 7 years: An Odense child cohort study.

OBJECTIVES: Animal studies indicate a key role for vitamin D in brain development and function, but observational studies in humans only suggests a borderline positive association between prenatal vitamin D exposure and cognitive development in the offspring. Knowledge gaps include insights in exposure time window and differences by sex for the association. We aimed to investigate the association between blood concentrations of serum 25-hydroxyvitamin D measured at four different time points and intelligence quotient score at the age of 7 years, including analyses spilt by child sex. METHODS: In Odense child cohort, we included 1404 mother-child pairs with serum 25-hydroxyvitamin D data from early pregnancy to age 7 years. Full-scale intelligence quotient was assessed with Wechsler Intelligence Scale for Children - fifth edition. Associations were adjusted for maternal education, pre-pregnancy body mass index, gestational age, sex and head circumference. Subanalyses stratified by sex were performed. RESULTS: The median (interquartile range) serum 25-hydroxyvitamin D in cord was 45.88 (31.15-61.08) nmol/L; early pregnancy, 66.45 (51.29-78.74); late pregnancy, 79.13 (59.69-97.31); 7 years, 66.29 (53.45-80.23) nmol/L. The mean (standard deviation) full-scale intelligence quotient was 99.44 (11.98). In adjusted analyses, cord serum 25-hydroxyvitamin D < 50 nmol/L was associated with 2.2 points lower full-scale intelligence quotient compared to the reference (50-75 nmol/L) in boys, ß = -2.2; 95% confidence interval = [-4.3, -0.1], p = 0.039. The same association with full-scale intelligence quotient was found for early pregnancy serum 25-hydroxyvitamin D, ß = -2.5 [-4.6, -0.3], p = 0.025, primarily driven by an association in boys, ß = -4.0 [-7.2, -0.8], p = 0.015; and for serum 25-hydroxyvitamin D at 7 years in girls, ß = -3.0 [-6.0, -0.1], p = 0.042. CONCLUSION: In this cohort, serum 25-hydroxyvitamin D < 50 nmol/L in both early gestation and cord blood in boys and current serum 25-hydroxyvitamin D < 50 nmol/L in girls were independent risk factors for two to four points lower full-scale intelligence quotient at the age of 7 years. Vulnerability to hypovitaminosis D, especially in pregnancy, may relate to child sex.

Vitamin D status and tooth enamel hypomineralization are not associated in 4-y-old children: An Odense Child Cohort study.

Early fetal stages of tooth development are vitamin D-dependent, suggesting an impact of vitamin D status in pregnancy on tooth mineralization in human populations. We examined the association between pregnancy and cord serum 25-hydroxyvitamin D (s-25(OH)D) and hypomineralization of the second primary molars (HSPM) in the 4-year-old children in the prospective, population-based Odense Child Cohort, Denmark. S-25(OH)D was measured in early pregnancy (<20 weeks, n = 753); late pregnancy (≥20 weeks, n = 841); and in umbilical cord blood (n = 1,241) using liquid chromatography-tandem mass spectrometry. HSPM was scored using modified European Academy of Paediatric Dentistry judgment criteria. The median [Q1;Q3] s-25(OH)D was 65.0 [49.4;78.0], 79.2 [60.4;95.8], and 45.1 [31.2;60.5] nmol/L in early pregnancy, late pregnancy, and cord blood, respectively. The prevalence of HSPM was 54.7%; creamy/white demarcated opacities 79.5%; yellowish/brownish demarcated opacities 14.9%; post-eruptive breakdown 5.2%; atypical restoration 0.4%. No univariate or adjusted associations with HSPM were detected for pregnancy or cord s-25(OH)D as a continuous variable or categorized into quartiles or routine clinical cut-offs, or when classifying HSPM by severity. In exploratory multiple regression analysis, HSPM was inversely associated with the length of gestation, adjusted odds ratio (aOR) 0.82 (95% C.I 0.74-0.92, p < 0.001), and directly associated with maternal education, aOR 1.57 (95% C.I 1.18-2.08, p = 0.002). In a population with relatively high s-25(OH)D concentrations and generally healthy mothers and children, pregnancy and cord blood vitamin D status was not associated with HSPM. The associations between HSPM and shorter gestational length and higher maternal education warrant further study.