Effects of vitamin D in pregnancy on maternal and offspring health-related outcomes: An umbrella review of systematic review and meta-analyses

Vitamin D is a fat-soluble steroid hormone with roles in calcium-phosphate homeostasis and extraskeletal systems (cardiovascular, metabolic, respiratory, immune). Deficiency is common worldwide among groups with low sun exposure, darker skin, and in pregnant women, with prevalence varying widely (5%–90% by country). Pregnancy alters vitamin D requirements and metabolism, and low levels have been linked to hypertensive disorders, gestational diabetes mellitus (GDM), preterm birth, and low birth weight. Numerous meta-analyses have examined associations between maternal vitamin D status or supplementation and maternal/offspring outcomes, but findings had not been synthesized in an umbrella review. This study aimed to systematically collate and appraise meta-analytical evidence on vitamin D deficiency and supplementation in pregnancy in relation to maternal and offspring health outcomes, to inform clinicians, researchers, and policymakers.

Prior meta-analyses have assessed associations between low maternal 25(OH)D and adverse pregnancy outcomes (e.g., preterm birth, SGA/LBW, GDM, pre-eclampsia) and examined supplementation effects on maternal-neonatal outcomes. However, evidence was dispersed across studies with variable quality and heterogeneity, and no umbrella review had comprehensively organized these data. Umbrella reviews synthesize findings from multiple meta-analyses to provide high-level overviews and evaluate the strength and certainty of evidence.

Protocol and registration: Registered in PROSPERO (CRD42022368003). Search strategy: PubMed, Embase, and Cochrane Library were searched from inception to October 2, 2023, for systematic reviews and meta-analyses on vitamin D in pregnancy. Deduplication via EndNote 20, followed by title/abstract and full-text screening by two independent reviewers; reference lists were hand-searched. Eligibility (PECOS): P: pregnant women; E: vitamin D intake (supplementation) or measured vitamin D levels; C: lower dose/placebo/no supplementation or lower vitamin D levels as defined by authors; O: any health-related outcomes; S: meta-analyses of randomized controlled trials (RCTs) or observational studies. For overlapping meta-analyses on similar outcomes, the most recent was included, or the one with the largest number of studies if published the same year. Exclusions: conference abstracts, animal studies, meta-analyses not from systematic reviews, and non-English/Chinese publications. Data extraction: First author, year, outcome(s), study design, number of studies and participants, effect sizes (OR, RR, MD, SMD) with 95% CIs, heterogeneity (I²), and publication bias (Egger’s/Begg’s) where available. One author extracted data and another verified; disagreements resolved by consensus or a third reviewer. Quality assessment: AMSTAR 2 used to appraise methodological quality of included meta-analyses, categorizing as high, moderate, low, or critically low. Evidence grading: A modified framework (considering number of studies, heterogeneity, and concordance between RCTs and observational studies) categorized evidence for significant outcomes as level 1 (convincing) to level 4 (limited/contrasting).

- Included evidence: 16 meta-analyses (total participants within meta-analyses ranged 1,465–67,484; 9 observational meta-analyses on 16 outcomes; 7 RCT meta-analyses on 31 outcomes; 18 outcomes specific to women with GDM). - Methodological quality: All included meta-analyses were rated critically low by AMSTAR 2 due to multiple critical flaws. - Observational associations (maternal): - Preterm birth: vitamin D <50 nmol/L vs ≥50 associated with higher risk (OR 1.28, 95% CI 1.08–1.52; 21 studies). - Gestational diabetes mellitus: deficiency vs normal associated with higher risk (OR 1.38, 95% CI 1.22–1.57; 31 studies). - Preeclampsia: highest vs lowest 25(OH)D associated with lower risk (RR 0.68, 95% CI 0.55–0.85; 29 studies). - Miscarriage: lower vitamin D associated with higher risk (e.g., OR 1.60, 95% CI 1.11–2.30; 6 studies); recurrent miscarriage: OR 4.02 (95% CI 2.23–7.25; 12 studies). - Bacterial vaginosis: deficiency vs normal (OR 1.54, 95% CI 1.25–1.91; 14 studies). - Observational associations (offspring): - Birth weight: low vitamin D <30 nmol/L associated with −87.8 g (MD −87.82, 95% CI −119.73 to −55.91; 15 studies). - Low birth weight <2500 g: vitamin D <50 nmol/L associated with higher risk (OR 2.39, 95% CI 1.25–4.57; 8 studies). - Head circumference: low vitamin D <30 nmol/L associated with smaller head circumference (MD −0.19 cm, 95% CI −0.32 to −0.06; 7 studies). - Neurodevelopment/childhood disorders: highest vs lowest maternal vitamin D associated with lower risks of ADHD (OR 0.59, 95% CI 0.44–0.81; 5 studies) and ASD (OR 0.57, 95% CI 0.33–0.99; 5 studies); lower maternal vitamin D associated with lower child mental development scores (MD −1.12 points, 95% CI −1.82 to −0.42; 9 studies). Maternal 25(OH)D lower in mothers of children with type 1 diabetes (MD −2.54 ng/mL, 95% CI −4.65 to −0.44; 8 studies). - RCT evidence: vitamin D supplementation during pregnancy: - Maternal outcomes (general): - Preeclampsia: ≤2000 IU/day vs placebo reduced risk (OR 0.29, 95% CI 0.09–0.95; 3 studies); >2000 vs ≤2000 IU/day showed no significant difference (OR 0.80, 95% CI 0.51–1.24). - GDM: >2000 vs ≤2000 IU/day reduced risk (OR 0.70, 95% CI 0.51–0.95; 7 studies). - Biomarkers: supplementation >400 IU/day increased maternal 25(OH)D (SMD 2.07, 95% CI 1.51–2.63; 15 studies), total antioxidant capacity (SMD 2.13, 95% CI 1.04–3.23; 6–9 studies), glutathione (SMD 4.37, 95% CI 2.90–5.84; 9 studies), and decreased malondialdehyde (SMD −0.46, 95% CI −0.87 to −0.05; 6 studies). Lipids improved: total cholesterol SMD −0.67 (95% CI −1.19 to −0.14; 7 studies), LDL-C SMD −0.49 (95% CI −0.68 to −0.29; 7 studies), triglycerides SMD −0.59 (95% CI −1.01 to −0.17; 6 studies); HDL-C increased SMD 0.41 (95% CI 0.23–0.58; 8 studies). - Offspring outcomes (general): - Birth length increased (MD 0.269 cm, 95% CI 0.024–0.514; 20 studies); third-trimester humeral length increased (MD 0.13 cm, 95% CI 0.06–0.21; 2 studies). - Fetal or neonatal mortality reduced (RR 0.69, 95% CI 0.482–0.985; 13 studies). - Neonatal vitamin D status improved: higher concentration (MD 27.7, 95% CI 20.5–34.9; 25 studies) and lower insufficiency (RR 0.508, 95% CI 0.384–0.673; 12 studies). - No consistent effects on birth weight, LBW rate, head circumference, SGA, macrosomia (general populations in RCTs). - Women with GDM (supplementation vs placebo/no supplementation): - Reduced cesarean section (RR 0.75, 95% CI 0.63–0.89; 9 studies), maternal hospitalization (RR 0.13, 95% CI 0.02–0.98; 2 studies), and preterm birth (OR 0.37, 95% CI 0.22–0.62; 10 studies). - Reduced macrosomia, polyhydramnios, fetal distress, and neonatal hyperbilirubinemia in some analyses; improved fasting plasma glucose (MD −10.2 mg/dL), fasting insulin (MD −5.02 µIU/mL), and HOMA-IR (MD −1.06). - Dose considerations: Very high dose (>2000 IU/day) showed benefit for reducing GDM risk vs ≤2000 IU/day; several outcomes improved with >400 IU/day compared with ≤400 IU/day or placebo.

The umbrella review consolidates evidence that low maternal vitamin D status is associated with multiple adverse outcomes (preterm birth, SGA/LBW, miscarriage, bacterial vaginosis, GDM) and poorer offspring neurodevelopment-related outcomes. However, RCTs show mixed effects of supplementation: while supplementation improves maternal and neonatal vitamin D status and some clinical outcomes (e.g., reduced pre-eclampsia with lower-dose vs placebo, reduced fetal/neonatal mortality, increased birth length), it does not consistently reduce risks like preterm birth or SGA in unselected populations. This discrepancy may reflect inclusion of participants without baseline deficiency in RCTs, attenuating potential benefits. Biological plausibility exists for effects on inflammation (labor and pre-eclampsia pathways) and neurodevelopment (vitamin D influences dopaminergic neuron differentiation). For GDM, convergent evidence from observational studies and RCTs suggests vitamin D plays a role in prevention and in improving maternal-neonatal outcomes among affected women, although some updates contest certain benefits and are limited by small study numbers. Overall, monitoring vitamin D status and targeted supplementation, particularly in deficient or high-risk groups, appears clinically relevant.

Vitamin D deficiency is common in pregnancy and is associated with adverse maternal and offspring outcomes. Supplementation during pregnancy, particularly at doses >400 IU/day, can improve maternal/neonatal vitamin D status and some clinical outcomes (e.g., reduced pre-eclampsia vs placebo, lower fetal/neonatal mortality, increased birth length), and in women with GDM, can improve metabolic markers and several maternal-neonatal outcomes. Clinicians should monitor vitamin D status during pregnancy and consider supplementation, especially in deficient or high-risk women. Future research should include well-designed RCTs targeting vitamin D–deficient populations, evaluate effects on recurrent miscarriage and bacterial vaginosis, and standardize dosing and outcome measures.

- All included meta-analyses were appraised as critically low quality on AMSTAR 2, limiting confidence in pooled estimates. - Many associations are derived from observational studies with substantial heterogeneity (often I² >50%), introducing residual confounding and inconsistency. - Some RCT meta-analyses included few trials (n=2–3) for certain outcomes, reducing reliability. - Participants in RCTs were not uniformly vitamin D deficient, which may dilute detectable effects of supplementation on outcomes like preterm birth and SGA. - Vitamin D status was often measured at a single time point, reflecting short-term levels subject to variability (diet, sun exposure); longer-term measures (e.g., hair) may be more informative. - Conflicting findings exist for some GDM-related outcomes in updated analyses with small study numbers.