Vitamin D screening and supplementation in pregnancy in Australia; is it safe to say one size fits all?

Vitamin D and its physiological role in pregnancy have been topics of much interest in recent years. Insufficient levels of this vitamin have been implicated in numerous obstetric and perinatal complications. The optimal level of serum vitamin D in pregnancy and the value and appropriate dosage of supplementation are subjects of controversy in current literature. Stemming from evidence gained in recent research, there have been proposals that the introduction of routine screening for vitamin D deficiency and/or routine supplementation should be undertaken nationwide in Australia.

Vitamin D is a fat-soluble steroid hormone that has long been known to play an integral role in bone metabolism and calcium homeostasis. In recent times, however, there has been emerging evidence that vitamin D has important non-skeletal functions, including in reproductive health.2,4,5,6,9,10

Deficiency/insufficiency has been associated with an increased risk in adults of malignancy, autoimmune disorders, certain infections and poor glycaemic control.2,4,5,7,9 Suboptimal levels have also been linked to a variety of adverse obstetric and perinatal outcomes, including gestational diabetes, pre-eclampsia, gestational hypertension, small for gestational age (SGA), caesarean section and preterm labour.2,4,5,7 As a logical progression from the discovery of these associations, it has been suggested that screening for suboptimal vitamin D levels in pregnancy and supplementation of women found to be deficient/insufficient may improve pregnancy and neonatal outcomes.

Vitamin D exists in the human body in two physiologically active forms, ergocalciferol (vitamin D2) and cholecalciferol (vitamin D3). It can be ingested in its natural form in the diet in foods such as fatty fish (mackerel, salmon, tuna), mushroom and egg yolk or in supplemented foods such as margarine, milk and yoghurt. The more important source of vitamin D for humans, however, is the conversion of 7-dehydrocholecalciferol, a cholesterol derivative, to vitamin D3 in the skin, a reaction produced by ultraviolet (UVB) light.2,5,7,9 Subsequently, anything that reduces the penetration of UVB into the skin will predispose to suboptimal vitamin D levels. For example, a high density of melanin (such as in dark-skinned people), a more obtuse incident angle of the sun, such as that found at extreme latitudes or in winter, sun protection as promoted in recent decades by sun-safe campaigns and practices of reducing skin exposure for cultural or religious reasons have all been shown to increase the prevalence of vitamin D insufficiency/deficiency.2,4,7

Supplementation of vitamin D can be in the form of vitamin D2 or D3. Both vitamin D2 and D3 are hydroxylated in the liver to form 25 hydroxy vitamin D (25(OH)D) and then hydroxylated in the kidney to 1,25 dihydroxyl vitamin D (1,25(OH)2D or calcitriol), which is the active form of the vitamin. Calcitriol (1,25(OH)2D) travels in the bloodstream to target receptors throughout the body where it carries out its various physiological functions, including regulation of up to 900 genes.9

Vitamin D has a short half-life (24 hours for vitamin D, two to three weeks for 25(OH)D and four to six hours for 1,25(OH)2D) and thus ongoing intake or in vivo production is necessary to ensure sustained circulating levels.2,7

Vitamin D toxicity leads to hypercalcaemia and its attendant consequences. This generally does not occur until levels of 220nmol/l (88ng/ml) are reached.4

Vitamin D levels are estimated by measuring 25(OH)D levels in serum. As demonstrated by a 2012 Cochrane review, supplementation of vitamin D during pregnancy leads to increased vitamin D levels at term2, however, there exist widely disparate opinions about the optimal serum level in pregnancy and appropriate dosage of supplementation during pregnancy. The US Institute of Medicine considers levels of 25(OH)D of 20ng/ml (50nmol) to be the lower limit of normal.5 They recommend intake of 400–600IU per day to achieve these levels, which they claim can be obtained solely from the diet without supplementation. The Endocrine Society suggests, however, that normal levels in pregnancy are above 30ng/ml (75nmol) and recommend an intake of 1500–2000IU day.4 A recent randomised controlled trial (RCT) by a group from the Medical University of South Carolina suggests higher levels still are optimal. This group undertook a multi-year RCT which showed that the optimal level of 25(OH)D in pregnancy is 32ng/ml (80nmol) and that this can be achieved with a dosing regimen of 4000IU per day.5,9

According to the recent Cochrane review on the subject2, there has been only one RCT analysing the link between Vitamin D deficiency and pre-eclampsia, which did not support this association. Three trials analysed in the review demonstrated that women who received supplementation had a reduced risk of having a low birth weight baby compared to those who received placebo or no treatment; however, the statistical significance was low. (It is suspected that vitamin D status may influence placentation and thus fetal growth.) The Cochrane review demonstrated no statistically significant increase in stillbirths or neonatal deaths in women who did not receive vitamin D supplementation compared to those who did. There have been no randomised control studies on preterm birth, maternal mortality, admission to NICU or APGAR scores and the review did not look at studies on gestational diabetes.

There has been a recent retrospective cross-sectional study, however, demonstrating that vitamin D levels were inversely proportional to blood glucose levels and HbA1c.6 There have also been observational studies demonstrating an association between vitamin D deficiency and preterm labour and caesarean section.

Overall, it appears that while there is biological plausibility to the hypothesis that vitamin D deficiency in pregnancy can increase the risk of a variety of pregnancy and neonatal complications – and there is lower level evidence in the form of observational, cross-sectional and case-control studies – this has yet to be proven with high-level evidence. This paucity of good-quality, high-level evidence highlights a need for further RCTs in this field.

The prevalence of vitamin D deficiency in pregnancy obviously depends on the definition used. According to the IOM definition of <20ng/ml (<50nmol/l), 100 per cent of Somali immigrants living in Sweden, 89 per cent of urban Japanese women, 54.7 per cent of UK women in the first trimester and 46 per cent of labouring women in Pakistan are vitamin D deficient.2 Using a definition of deficiency <37.5nmol/l, one study found that 29 per cent of black pregnant women and five per cent of white pregnant women in the US were deficient.2

Three recent Australian studies done in Victoria and NSW demonstrated high levels of vitamin D deficiency in the populations tested. In a study of 147 women attending the GDM Clinic in Westmead hospital, 41 per cent of 147 women had vitamin D insufficiency or deficiency (as defined by levels <25nmol/l and 25–50nmol/l, respectively).6 In a 2010 cross-sectional study, it was found that the rate of suboptimal vitamin D levels in pregnant women was 35 per cent in Canberra and 25.7 per cent in Campbelltown, Sydney. Definitions for deficiency and insufficiency were <25nmol/l and 26–50nmol/l, respectively. In this study it was found that skin exposure (attire), ethnicity, season and body mass index were determinants of vitamin D levels.7 In a third study, done in rural Victoria in 2008–09, 65.5 per cent of subjects had suboptimal levels of vitamin D (<75nmol/l). This study also found that ethnicity and season affected rates of deficiency.9 The authors of all three studies suggest that these rates of vitamin D inadequacy warrant routine screening for vitamin D levels antenatally and/or routine supplementation.

The proposal that universal screening is warranted throughout Australia is questioned by the findings of a small 2011 study from Northern Queensland. The results of this study were that none
of the subjects in a study of 116 women booking into the Cairns Base Hospital Antenatal Clinic were vitamin D insufficient, using a definition of insufficiency of <50nmol/l.1 If a threshold of 75nmol/l was used to define insufficiency, 6.9 per cent had insufficient levels. The results of this study suggest that nationwide universal screening may not be indicated. While the incidence of vitamin D deficiency in southern states has been shown to be significant, the incidence in more northerly populations is likely to be much lower. Therefore, it may be more prudent in these areas to use targeted screening in groups known to be at risk such as women with skin that has a high melanin content, those with gestational diabetes and those who have limited skin exposure due to cultural/religious reasons. Further research is required in other regions of northern Australia before we are able to confidently conclude that universal screening is not needed in each of these areas.

In addition, it is evident from the aforementioned study that routine supplementation of pregnant women, prior to screening for vitamin D levels, as suggested in some previous studies, could potentially lead to toxic levels in pregnant women in the northern regions. Of women in this study, ten per cent had serum vitamin D levels of 151–200nmol/l. Given that the toxic level of vitamin D is over 220nmol/l, routine supplementation could easily lead to hypercalcaemia in these women.

A standard definition of vitamin D deficiency in pregnancy and appropriate dose for supplementation is needed to guide clinical practice. While it does appear that vitamin D deficiency may be associated with adverse pregnancy and neonatal outcomes, this has not been proven by RCTs, with the exception of the outcome of low birth weight. That is not to say that vitamin D deficiency is not an issue of public health significance. On the contrary, the results of trials done to date suggest that pregnancy and neonatal outcomes are likely to be significantly improved if the population is vitamin D replete. Supplementation of women with proven deficiency is unlikely to cause harm and is likely to lead to benefit; however, the introduction of universal screening may not be either cost effective or necessary. An argument may be made that it is more prudent to bring in screening for high-risk populations, such as those in colder climates (the southern states where the prevalence of vitamin D deficiency has been shown to be significantly higher than in northern Australia), with dark skin and little sun exposure for cultural/societal reasons.