Background: Low vitamin D levels during pregnancy have been associated with adverse pregnancy outcomes by few studies, and not by others.
Methods: We measured maternal 25-hydroxyvitamin D [25(OH)D] in first trimester serum samples from 5,109 women with singleton pregnancies. Information on maternal and infant outcomes was obtained through record linkage of laboratory data to birth and hospital data. Pregnancy outcomes included small for gestational age (SGA), preterm birth, preeclampsia, gestational diabetes mellitus, miscarriage and stillbirth. Multivariate logistic regression was conducted to assess the association between low 25(OH)D (<25, <37.5 and <50 nmol/L) with each pregnancy outcome and a composite of any severe pregnancy outcomes (SGA<3rd centile, preterm birth<34 weeks, early-onset preeclampsia or stillbirth). Predictive accuracy was assessed.
Results: Median (interquartile range) 25(OH)D for the total population was 56.4 nmol/L (43.3-69.8). 25(OH)D levels showed significant variation by parity, smoking, weight, season of sampling, country of birth and socio-economic disadvantage. After adjusting for maternal and clinical risk factors, low 25(OH)D levels were not associated with any pregnancy outcome. The area under the Receiver Operating Characteristics curve (AUC) and likelihood ratio (LR) for the composite of severe pregnancy outcomes of 25(OH)D <25 nmol/L were 0.51 and 1.44; and for risk factors alone were 0.64 and 2.87, respectively. Adding 25(OH)D information to maternal and clinical risk factors did not improve the ability to predict severe adverse pregnancy outcomes (AUC=0.64; LR=2.32; P=0.39).
Conclusions: Low25(OH)D levels in first trimester are not associated with adverse pregnancy outcomes and do not predict complications any better than maternal and clinical risk factor information.