Some older studies showed that vitamin D deficient rats had reduced fertility and litter size, associated with impaired ovarian function and spermatogenesis. The latter appeared to be calcium modulated and could be reversed with administration of calcium [46]. Sperm count and motility were both shown to be reduced and histological abnormalities of the testis were observed in another animal model, VDR null mutant mice [47]. However, the association between vitamin D deficiency and defective reproduction may be the result of hypocalcaemia rather than a vitamin D effect per se, as the infertile female VDR null mice developed normal fertility when fed a calcium rich diet [48].

In human in vitro fertilization (IVF) studies, women who achieved a clinical pregnancy had significantly higher follicular fluid (Fol F) levels of 25(OH)D than women with unsuccessful cycles [49]. Multiple regression analysis confirmed that Fol F 25(OH)D was associated with a successful IVF cycle, after adjusting for maternal age, body mass index, ethnicity and number of embryos transferred (p = 0.01); for each 2.5 nmol/L increase of Fol F 25(OH)D the likelihood of achieving a clinical pregnancy increased by 7% [49]. The authors suggested that higher Fol F 25(OH)D levels may enhance endometrial receptivity and implantation [49].

1,25(OH)₂D has effects on immune function within both the innate and adaptive immune systems. For example, 1,25(OH)₂D regulates the production of the antimicrobial protein cathelicidin, CAMP (LL37), an intracellular antimicrobial protein found within lysosomes of neutrophils, macrophages and also within trophoblasts [14,50]. Cathelicidin is a key element of innate immunity protecting against viral and bacterial infection. During pregnancy, a key requirement for the maternal immune system is the development of immune tolerance to the fetus. Recent work further suggests that 1,25(OH)₂D may be important to specific aspects of placental function [51]. Within the placenta, 1,25(OH)₂D can function as an intracrine regulator of CAMP (LL37) in trophoblasts, and may thus provide a novel mechanism for activation of innate immune responses in the placenta, assisting the pregnancy. 1,25(OH)₂D down-regulates IL-6 production [52] and there is some evidence that adverse neurodevelopmental outcomes following maternal infections are mediated by this cytokine [53]. Ongoing research is examining the role of 1,25(OH)₂D in decreasing the risk of infections during pregnancy and whether vitamin D status alters the risk of developing adverse outcomes following such infections.

In animal models (mainly rodents), vitamin D–deficient and –insufficient animals appear to have normal skeletal mineral content [54]. However, it is not clear that the vitamin D requirements of these largely nocturnal animals reflect human needs. Human studies are needed as inter-species variation in vitamin D requirements and metabolism may occur [55]. Among vitamin D deficient pregnant women, fetal ultrasound revealed splaying of the distal metaphysis of the fetal femur as early as 19 weeks, with the appearance being very similar to that seen in childhood (vitamin D-deficient) rickets [56]. There was a positive correlation between the maternal 25(OH)D level and the fetal metaphyseal cross-sectional area, with the latter 5% and 14% greater in fetuses of mothers who were vitamin D insufficient (serum 25(OH)D 25-50 nmol/L) and vitamin D deficient (serum 25(OH)D < 25 nmol/L) compared to vitamin D sufficient mothers (serum 25(OH)D > 50 nmol/L) [56]. There was, however, no association between maternal 25(OH)D level and fetal femur length at 19 weeks or 34 weeks gestation [56].

Early life immunomodulation is likely to also be affected by vitamin D status. At birth, cord serum 25(OH)D levels are directly correlated with levels of IL-10, an immunosuppressive cytokine [57]. In vitro studies show that 1,25(OH)₂D acts on naïve T cells to retard their development into either T helper 1 or 2 cells [58].

Recent data suggest an association between vitamin D deficiency and caesarean section. In a cross sectional study, caesarean section was almost four-times more common in women with vitamin D insufficiency (serum 25(OH)D < 37.5 nmol/L) compared to vitamin D-sufficient women (serum 25(OH)D ≥ 37.5 nmol/L), after adjustment for race, age, education level, insurance status, and alcohol use (Adjusted OR 3.84; 95% CI 1.71–8.62) [59].

Observational studies of vitamin D status during pregnancy and physical characteristics of the offspring are few and provide conflicting results. One study found pregnancies of mothers with low vitamin D status were shorter (by 0.7 week; 95% CI −1.3, −0.1) [60], and the babies had poorer intrauterine long bone growth [60], while in other studies, offspring had lower birth weight [61,62,63]. However another study found the opposite, with babies of vitamin D deficient mothers being both heavier and longer, compared to the offspring of vitamin D-sufficient mothers [64]. Further examination of these findings suggests that the effects may vary according to the VDR genotype of the offspring. Thus within the offspring of vitamin D deficient mothers, there was lower birth weight in those having the FF or Ff genotype (where the F allele is associated with increased vitamin D receptor activity [65]), but not in offspring with the ff genotype (P-value for interaction after adjustment for potential confounding factors = 0.02) [66]. A large randomized controlled trial on this issue has been completed in South Carolina, USA, with recent presentation at the 14th International Vitamin D meeting, Brugges, Belgium, in October 2009 [67]. The media reported on a reduction in both preterm birth and small-for-dates babies, with vitamin D supplementation [41], but formal publication of the findings had not occurred at the time of writing this review, thus these reports cannot be formally evaluated.

At this stage, all findings arise from animal studies, primarily in rats, and it is not clear whether there is a direct translation to human fetal outcomes. However, there is evidence that 1,25(OH)₂D regulates cell differentiation and proliferation so that effects on organ development during fetal life are plausible and warrant further investigation. In rats, maternal vitamin D deficiency was associated with an increase in nephron number, but lower renal corpuscle size [68]. Low maternal vitamin D status may slow neonatal cardiac development [69] and alter brain morphology [70,71], with changes in the latter persisting into adulthood [72].

Excess vitamin D administration during pregnancy could also potentially be of concern. Rats receiving very high-dose vitamin D during gestation and early development (considerably higher than would be administered to humans) had adverse changes in elastin content and organization in the aorta consistent with increased later risk of hypertension or aneurysm [73]. Nevertheless, in two studies in pigs, vitamin D administration to two sows at doses that achieved serum 25(OH)D levels, approximating recommended human levels, was associated with coronary lesions in offspring at six weeks [74,75]. The relevance of these findings to humans remains to be determined. In 2008, a cohort study investigating these issues reported no statistically significant associations between antenatal maternal 25(OH)D levels and cardiac measures in offspring at nine years, including in blood pressure, carotid intima-media thickness, arterial compliance and cardiac structure [62].

Maternal vitamin D status influences bone health in the offspring [76]. In a UK cohort study, winter-born babies had lower bone mineral content (BMC) than those born in summer [77] and maternal serum 25(OH)D in late pregnancy was directly associated with total and lumbar spine BMC in offspring at nine years of age [77,78].

An increased risk of acute lower respiratory tract infection (ALRI) in newborn infants was associated with lower maternal 25(OH)D levels in a case-control study in Turkey [79], but there was no association between postnatal serum 25(OH)D levels and infant ALRI in a case-control study in Canadian children aged 1–25 months [80]. In a prospective study in Tanzania, children born to mothers with lower 25(OH)D levels at 12–27 weeks gestation, had a 46% (95% CI 11%–91%) increased risk of contracting HIV infection by 24 months of age, including postnatally during breastfeeding, and a 61% higher risk of dying (95% CI 25%–107%) [81].

Increased risk for a winter season-of-birth is seen in some autoimmune disorders, such as Crohn’s disease [82] and type 1 diabetes [83,84,85], although the latter finding may be limited to comparisons within homogeneous populations [86], or in those with specific susceptibility genotypes [87].

The Diabetes Autoimmunity Study in the Young (DAISY) cohort has observed that maternal intake of vitamin D via food, recorded at birth, was significantly associated with a decreased risk of islet autoimmunity in offspring during the first four years of life [88]. A Norwegian case control study found cod liver oil use during pregnancy was associated with lower risk of type 1 diabetes in the offspring, indicating vitamin D or the n-3 fatty acids eicosapentaenoic acid and docosahexaenoic acid in the cod liver oil, or both, may have a protective effect against type I diabetes [89]. These findings are also consistent with the finding of an inverse association between postnatal vitamin D supplementation in early life and type 1 diabetes risk [90,91].

Low vitamin D intake during pregnancy has been associated with an increased risk of asthma, eczema or hay fever [92,93], while winter birth was associated with higher IgE levels and lower IL-10 than summer birth [57]. As IL-10 is important in the development of tolerance to exogenous antigens and inhibition of mast cell degranulation [57], these findings are consistent with the notion of low vitamin D levels at birth increasing the risk of an allergic propensity. However, the ecological patterns are not consistent with this because some studies show a higher prevalence of asthma in warmer climates with higher ambient UVR than other locations [94,95,96,97]. These regional patterns are not necessarily conflicting, because they may also reflect uncontrolled confounding, in that higher house dust mite allergen levels may occur in warmer climates. A birth cohort has reported an increase in eczema and asthma among children whose mothers had higher 25(OH)D levels during late pregnancy (>75 versus <30 nmol/L) [62]. Randomized controlled trial data are required but not yet available.

Vitamin D is a potent inducer of nerve growth factor synthesis [98]. Genes of the vitamin D pathway (25 hydroxylase, 1α hydroxylase and 24 hydroxylase) and VDR are expressed in rat brain cells [99] and the distribution of VDR and 1α hydroxylase is thought to be similar in rat and human brains [100]. In experimental studies, rats deprived of vitamin D prenatally had enlarged lateral ventricles, decreased cortical thickness and heavier and longer brains, and altered patterns of apoptosis [70,71].

Seasonal and population level patterns of schizophrenia incidence indicate the possible role of low intrauterine vitamin D in increasing later schizophrenia risk. This is a difficult but important area to research [101,102,103,104], with animal work indicting transient prenatal vitamin D deficiency is associated with subtle alterations in learning and memory functions in adult rats [105]. In a Finnish birth cohort study, regular vitamin D supplementation (maternal self-report) during the first year of offspring life was associated with a reduced risk of schizophrenia in males (but not females) (RR = 0.08, 95% CI 0.01–0.95) [106].

Observational studies provide indirect evidence that low vitamin D status during early life may increase the risk of multiple sclerosis [107,108]. These findings include a striking season-of-birth pattern in a large cohort of Northern Hemisphere MS cases [109], a congruent pattern in the Southern hemisphere [110], and a maternal parent-of-origin effect [111,112]. Using prenatal ambient UVR levels as an instrumental variable for vitamin D levels in pregnancy [113], an inverse association between maternal ambient UVR exposure during the late first trimester and MS risk was observed [110]. This inverse association was independent of region of birth, indicating that the first trimester ambient UVR level was not acting merely as a marker of longer term UVR exposure due to region of residence [110].

Seasonal patterns of birth in adults [114] and children [115] with brain tumors and epilepsy [116] have been noted and could be evidence of an effect of low vitamin D status, but this is far from conclusive. A tentative link between vitamin D status and autism has been postulated, arising mainly from the findings of population level studies [117,118]. Further, adult body mass index and obesity prevalence has been noted to vary as a function of month of birth [119,120]. In adult ecological and observational studies there is considerable evidence that vitamin D status may be related to risk or prognosis of a number of cancers, including of the prostate [121] and breast [122]. One animal study has shown that maternal antenatal vitamin D supplementation was associated with greater mean prostatic weight and a histologically more differentiated prostatic architecture in offspring in adulthood [123]. Whether such changes also occur in humans and have relevance for later prostate cancer risk is unknown.