Longitudinal changes in vitamin D during twin pregnancy: association with maternal-neonatal outcomes

Existing studies suggest vitamin D (VD) deficiency links to adverse pregnancy outcomes in singletons; however, its association with specific adverse outcomes and neonatal health in twin pregnancies remains unclear. This study aimed to explore the relationship between maternal VD levels and maternal and neonatal outcomes in twin pregnancies.

This study collected VD levels, pregnancy conditions, and neonatal anthropometry from 324 twin pregnancies. Peripheral blood serum was collected from mothers in mid- and late pregnancy to measure VD concentrations. Logistic models assessed VD's association with pregnancy complications and neonatal anthropometry. The restricted cubic splines (RCS) method was used to estimate the risk threshold for spontaneous preterm birth (sPTB). Accounting for sample size and based on the Akaike (AIC) and Bayesian (BIC) information criteria, three knots were placed at the 10th, 50th, and 90th percentiles of the VD distribution. Missing data were handled using sensitivity analyses-including both optimistic and conservative assumptions-and multiple imputation methods.

Among all the participants, the mean mid-pregnancy VD concentration was 25.08 ± 8.31 ng/mL, with 25.6% having sufficient levels, 46.9% insufficient, and 27.5% deficient. Mid-pregnancy VD deficiency independently predicted sPTB (OR: 2.19; 95% CI: 1.07-4.50; p = 0.033). The RCS revealed an L-shaped VD-sPTB risk relationship. Each 1 ng/mL VD decrease increased sPTB risk by 14.94% (OR = 0.87). No statistically significant evidence was found between rising VD levels from mid- to late pregnancy and reduced sPTB risk vs. persistent low gestational VD levels. The analysis did not provide evidence for a statistically significant association between maternal mid-pregnancy VD levels and the incidence of hypertensive disorders of pregnancy (HDP), gestational diabetes mellitus (GDM), iatrogenic preterm birth (iPTB), or neonatal birth weight (all p > 0.05). In contrast, chorionicity (p = 0.004), HDP (p = 0.046), and neonatal sex (p = 0.021) were significant determinants of birth weight.

In twin pregnancies, maintaining adequate VD levels during mid-pregnancy represents a critical preventive window for reducing sPTB. For women with deficiency at this stage, the timing of supplementation may be more critical than dose. Although maternal VD status shows no significant association with neonatal anthropometry, the continued monitoring and optimization of VD levels throughout pregnancy remain essential for maternal-neonatal health.