Season and vitamin D status are independently associated with glucose homeostasis in pregnancy

Background: Vitamin D status and season are intrinsically linked, and both have been proposed to be associated with glucose homeostasis in pregnancy, with conflicting results. We aimed to determine if exposure to winter and low maternal 25 hydroxyvitamin D (25OHD) in early pregnancy were associated with maternal glucose metabolism. Methods: This is a secondary data analysis of 334 pregnant women enrolled in the ROLO study, Dublin. Serum 25OHD, fasting glucose, insulin and insulin resistance (HOMA-IR) were measured in early (12 weeks' gestation) and late pregnancy (28 weeks' gestation). Season of first antenatal visit was categorised as extended winter (November-April) or extended summer (May-October). Multiple linear regression models, adjusted for confounders, were used for analysis. Results: Those who attended their first antenatal visit in extended winter had lower 25OHD compared to extended summer (32.9 nmol/L vs. 44.1 nmol/L, P < 0.001). Compared to those who attended their first antenatal visit during extended summer, extended winter was associated with increased HOMA-IR in early-pregnancy (46.7%) and late pregnancy (53.7%), independent of 25OHD < 30 nmol/L and confounders. Early pregnancy 25OHD < 30 nmol/L and extended winter were independently associated with significantly higher fasting glucose in late pregnancy (B = 0.15 and 0.13, respectively). Conclusions: Women who attended their first antenatal visit during the months of extended winter were more likely to have raised insulin resistance in early pregnancy, which had a lasting association to 28 weeks, and was independent of 25OHD. Our novel findings imply that seasonal variation in insulin resistance may not be fully explained by differences in vitamin D status. This could reflect circannual rhythm or seasonal lifestyle behaviours, and requires further exploration.