Antenatal Vitamin D Status Is Not Associated with Standard Neurodevelopmental Assessments at Age 5 Years in a Well-Characterized Prospective Maternal-Infant Cohort

Abstract

Background: Although animal studies show evidence for a role of vitamin D during brain development, data from human studies show conflicting signals.

Objective: We aimed to explore associations between maternal and neonatal vitamin D status with childhood neurodevelopmental outcomes.

Methods: Comprehensive clinical, demographic, and lifestyle data were collected prospectively in 734 maternal-infant dyads from the Cork BASELINE Birth Cohort Study. Serum 25-hydroxyvitamin D [25(OH)D] concentrations were quantified at 15 weeks of gestation and in umbilical cord sera at birth via a CDC-accredited liquid chromatography-tandem mass spectrometry method. Children were assessed at age 5 y through the use of the Kaufman Brief Intelligence Test (2nd Edition, KBIT-2) and the Child Behaviour Checklist (CBCL). Linear regression was used to explore associations between 25(OH)D and neurodevelopmental outcomes.

Results: 25(OH)D concentrations were <30 nmol/L in 15% of maternal and 45% of umbilical cord sera and <50 nmol/L in 42% of mothers and 80% of cords. At age 5 y, the mean ± SD KBIT-2 intelligence quotient (IQ) composite score was 104.6 ± 8.6; scores were 107.2 ± 10.0 in verbal and 99.8 ± 8.8 in nonverbal tasks. Developmental delay (scores <85) was seen in <3% of children across all domains. The mean ± SD CBCL total problem score was 21.3 ± 17.5; scores in the abnormal/clinical range for internal, external, and total problem scales were present in 12%, 4%, and 6% of participants, respectively. KBIT-2 and CBCL subscale scores at 5 y were not different between children exposed to low antenatal vitamin D status, either at 30 or 50 nmol/L 25(OH)D thresholds. Neither maternal nor cord 25(OH)D (per 10 nmol/L) were associated with KBIT-2 IQ composite scores [adjusted β (95% CI): maternal -0.01 (-0.03, 0.02); cord 0.01 (-0.03, 0.04] or CBCL total problem scores [maternal 0.01 (-0.04, 0.05); cord 0.01 (-0.07, 0.09)].

Conclusion: In this well-characterized prospective maternal-infant cohort, we found no evidence that antenatal 25(OH)D concentrations are associated with neurodevelopmental outcomes at 5 y. The BASELINE Study was registered at www.clinicaltrials.gov as NCT01498965; the SCOPE Study was registered at http://www.anzctr.org.au as ACTRN12607000551493.