Mothers have protected their babies from rickets by spending time outdoors and taking cod liver oil for almost 100 years, from before vitamin D was discovered. Now that avoidance of mid-day sunlight is advised, and cod liver oil is no longer used in pregnancy, we have seen rickets and other features of vitamin D deficiency re-emerge(Reference Holick1).
What is the scale of the problem in the UK?
Hypovitaminosis D affects adults in epidemic proportions in Western societies(Reference Calvo, Whiting and Barton2, Reference Prentice3), which is a problem aggravated by Western lifestyles with long hours of indoor work and by avoidance of sunshine aimed at reduction in skin cancer risks (Table 1). Vitamin D deficiency ( < 25 nmol/l) is more common in women than in men (for example, 9·2 v. 6·6 %, respectively, in British 45-year-olds)(Reference Hyppönen and Power4), and pregnancy is known to represent a particularly high-risk situation. However, as we will discuss in the following section, in the UK, there is currently no consensus on advice on vitamin D supplementation to pregnant women or generally agreed guidelines for relevant health care providers. Lack of consistent guidance leads to mixed messages, which make it difficult to effectively implement strategies for the primary prevention of vitamin D deficiency.
Barriers to adequate vitamin D intake and sunlight-induced skin synthesis in pregnancy
SACN, Scientific Advisory Committee on Nutrition.
The return of rickets in the UK
There is a long history of population-based approaches for the prevention of vitamin D deficiency in Britain(Reference Bivins5). By the mid-19th century, it was appreciated that rickets was prevented and cured by summer sunshine or by taking cod liver oil(Reference Rajakumar6). The discovery of vitamin D in the 1920s led to regular outdoor ‘airing’ of infants and routine cod liver oil consumption by many mothers and children(Reference Bivins5, Reference Rajakumar6). Indeed, neonatal/infantile hypocalcaemia and rickets had virtually disappeared by the 1930s. As a result of this knowledge, ergocalciferol was added to National Dried Milk during World War II, and cod liver oil was included as one of the five welfare foods distributed by the Ministry of Food to expectant/nursing mothers and young children(Reference Berry7). Since World War II, vitamin D supplementation has been consistently recommended during pregnancy(Reference Bivins5, 8, Reference Shaw and David9). However, after the war, uncontrolled vitamin D fortification of baby milks and baby foods provided intakes of up to 100 μg/d (4000 IU/d), which caused many cases of infantile hypercalcaemia(Reference Bivins5, Reference Arneil10, Reference Berry11). Uncontrolled vitamin D fortification was then banned(Reference Berry11), followed by resurgence of rickets in immigrant communities during the 1960s(Reference Bivins5, Reference Arneil12, Reference Stewart, Mitchell and Morgan13). Cod liver oil remained available at antenatal clinics, but uptake was poor in immigrant populations(Reference Datta, Alfaham and Davies14). Routine offers of cod liver oil (rich in vitamin A) to pregnant mothers ceased in the 1990s following formal advice from the Chief Medical Officer when the teratogenicity of excessive vitamin A intake was appreciated(15). No provision was made for continued vitamin D supplementation of pregnant mothers. This left women dependent on the combination of a national diet known to be poor in vitamin D and sun-induced skin synthesis that is effective in increasing serum 25-hydroxyvitamin D (25(OH)D) concentrations only for 5–6 months of the year(Reference Hyppönen and Power4, 8). The benefits of modest supplementation of immigrant mothers had been demonstrated in the UK by the 1990s(Reference Bivins5, Reference Arora and Arora16), suggesting that the resurgence of rickets was predictable and largely preventable had routine vitamin D supplementation been continued(Reference Bivins5).
Current conflicts in advice
The UK Department of Health (DoH) re-enforced their advice(17) for pregnant and breast-feeding mothers to ensure that they achieved intakes of vitamin D (10 μg/d (400 IU/d)) in 2007. This followed recommendations of the Scientific Advisory Committee on Nutrition(8) (Table 2) and a resurgence of rickets, especially in immigrant communities with high prevalence of vitamin D deficiency(Reference Prentice3, Reference Wharton and Bishop18–Reference Odeka and Tan21). DoH advice effectively endorses supplementation as diet provides little vitamin D, and sunlight-induced synthesis is limited in Britain(Reference Hyppönen and Power4, 8).
UK sources of current advice on vitamin D supplementation in pregnancy and conflicts on who should be supplemented
* All links are as accessed on 9 July 2009.
The recent DoH re-enforcement of vitamin D supplementation in pregnancy coincided with a conflict in advice being developed by the National Institute for Health and Clinical Excellence on Antenatal Care (‘healthy pregnant mothers should not be routinely supplemented with vitamin D’, September 2007) and on Maternal and Child Nutrition (endorsing existing DoH guidelines, July 2007). The final recommendation was made by the National Institute of Health and Clinical Excellence Guideline Review Panel, and, in contrast to current DoH guidance, it did not endorse vitamin D supplementation for all pregnant women. Instead, it stated that ‘all women should be informed… about the importance for their own and their baby's health of maintaining adequate vitamin D stores during pregnancy and whilst breastfeeding’, and that ‘in order to achieve this, women may choose to take 400 IU/d as in the Healthy Start Multivitamin supplement’(22, 23). Continuing confusion in the UK recommendations is highlighted by the inclusion of vitamin D as one of the vitamins to be provided free of charge to low-income mothers as part of the Healthy Start Scheme (previously known as the Welfare Food Scheme), despite the relatively weak social differences in vitamin D status seen in the UK(Reference Hyppönen and Power4) and the lack of systematic implementation of vitamin D supplementation recommendations for other pregnant mothers. There have also been severe problems in the distribution of Healthy Start vitamins, with manufacturing being stopped due to ‘lack of demand’(Reference McGee24). Currently, the UK is the only one of the thirty-one European countries with a recommended daily vitamin D intake of 0 for women of child-bearing age(Reference Doets, de Wit and Dhonukshe-Rutten25), and therefore, the UK is doing nothing to reduce the risk of women becoming vitamin D deficient before they become pregnant.
Conflicting advice on maternal vitamin D intake is not unique to the UK. The North American health authorities are debating, not about whether to supplement vitamin D, but only about the dosage of vitamin D that should be provided. The Canadian Paediatric Society recently recommended that ‘all pregnant mothers should take vitamin D 50 μg/d (2000 IU/d) throughout pregnancy’(26), while shortly afterwards, the Federal Department of Health Canada re-enforced its recommendation of 5 μg/d (200 IU/d) for pregnant and breast-feeding women, publicly discounting the recommendations of the Paediatric Society. Currently, the European Commission recommends 10 μg/d (400 IU/d) vitamin D during pregnancy(27), while the WHO recommendation is set at 5 μg/d (200 IU/d)(28).
Gaps in the evidence and differences in its interpretation
Inconsistency in advice given by health authorities can arise from differences in the interpretation of available evidence and in the perceived importance of vitamin D deficiency in pregnancy. Furthermore, opinions about the safety of dosages>10 μg/d (400 IU/d) in pregnancy differ(8). As reviewed above, historical data suggest that the recommended dosage of 10 μg/d (400 IU/d) is safe and effective in preventing severe clinical vitamin D deficiency, rickets in children and osteomalacia in adults(Reference Bivins5, Reference Shaw and David9). However, the physiological effects of 1,25-dihydroxyvitamin D (the active hormonal metabolite) are known to extend beyond Ca metabolism and bone health, and evidence is accumulating to suggest that vitamin D intakes required to achieve optimal benefits are likely to be much higher(Reference Hollis and Wagner29, Reference Holick30).
Vitamin D in pregnancy
The importance of vitamin D during pregnancy is suggested by the presence of nuclear vitamin D receptors and of the vitamin D-activating 1-α-hydroxylase enzyme in pregnancy-specific tissues such as the decidua and placenta(Reference Evans, Bulmer and Kilby31). Circulating maternal concentrations of 1,25-dihydroxyvitamin D rise from early in the first trimester and increase progressively during gestation, being twice as high in late pregnancy than postpartum or in non-pregnant controls(Reference Evans, Bulmer and Kilby31). These physiological changes are accepted as important for ensuring fetal Ca supplies and for inducing immunological adaptations required for successful maintenance of pregnancy(Reference Evans, Bulmer and Kilby31). There is evidence for alterations in vitamin D metabolism in women with pre-eclampsia(Reference August, Marcaccio and Gertner32–Reference Hypponen36), with recent studies suggesting reductions in the incidence with higher maternal vitamin D status(Reference Bodnar, Catov and Simhan37) and intake(Reference Haugen, Brantsaeter and Trogstad38). The immunomodulatory effects of 1,25-dihydroxyvitamin D, which have been proposed to explain the associations with pre-eclampsia(Reference Hypponen36), would also be biologically relevant for reduction in the risk of miscarriage(Reference Bubanovic39), and could explain the recent observations for higher success rates for in vitro fertilisation for women with higher compared to lower 25(OH)D concentrations(Reference Ozkan, Jindal and Greenseid40). Preliminary data from a randomised placebo controlled trial of high-dose vitamin D supplementation in pregnancy (100 μg/d, presented in the Vitamin D Workshop in Bruges, October 2009) suggested that supplementation at these dosages was safe and did not lead to elevations in maternal Ca levels(Reference Hollis and Wagner41). They also reported reductions in the rate of preterm births and pregnancy-related complications. However, full evaluation of these data will need to wait for the formal publication of the findings.
Maternal vitamin D deficiency and offspring health
Vitamin D status in neonates is related to maternal vitamin D status (serum 25(OH)D)(Reference Shenoy, Swift and Cody20, Reference Cockburn, Belton and Purvis42, Reference Brooke, Brown and Bone43). Randomised trials show maternal and cord 25(OH)D to increase after maternal vitamin D supplementation(Reference Brooke, Brown and Bone43–Reference Mallet, Gugi and Brunelle45). Neonatal complications of extreme maternal vitamin D deficiency are life threatening (e.g. severe hypocalcaemic fits with high risks of resultant brain damage(Reference Shenoy, Swift and Cody20) and neonatal heart failure(Reference Maiya, Sullivan and Allgrove46)) in addition to the well-recognised risks of fractures and rickets. The evidence for the severe complications of vitamin D deficiency comes from an expanding series of case reports. For obvious ethical reasons, these rare complications have not been, and they are unlikely ever to be, examined by randomised trials of supplementation. Indeed, experience has already shown that trials proposing to randomise mothers with low serum 25(OH)D concentrations are deemed unacceptable by most ethics committees, and approvals have been granted only to investigate the influences of vitamin D supplementation in women who are not severely deficient (serum 25(OH)D>25 nmol/l). Vitamin D insufficiency has been associated with dose-wise reductions in bone mineral content in offspring(Reference Javaid, Crozier and Harvey47) and perinatal growth restriction(Reference Brooke, Brown and Bone43), and also with increased risk of immunological disorders such as type 1 diabetes(Reference Brekke and Ludvigsson48, Reference Elenkov, Wilder and Bakalov49) and acute respiratory infection(Reference Karatekin, Kaya and Salihoglu50). The range of maternal and offspring health outcomes associated with maternal vitamin D status has been the subject of a number of recent reviews(Reference Lucas, Ponsonby and Pasco51–Reference Perez-Lopez54). The only end-point for possible adverse effects of maternal vitamin D ‘repletion’ is the association of high maternal 25(OH)D concentrations with increased atopy and asthma risks in the offspring(Reference Gale, Robinson and Harvey55), which is a finding challenged by reports suggesting beneficial effects of higher v. lower maternal vitamin D intakes on early childhood wheezing(Reference Devereux, Litonjua and Turner56). Overall, the available evidence suggests unequivocal benefits for avoidance of vitamin D deficiency during pregnancy, while the possible risks of milder forms of hypovitaminosis D provide promising scope for the prevention of a number of disorders. However, vitamin D dosages required to reach serum concentrations currently considered ‘optimal’ for health are likely to be higher than 10 μg/d(Reference Yu, Sykes and Sethi57, Reference Hollis58) and, as Scientific Advisory Committee on Nutrition and National Institute of Health and Clinical Excellence state, larger dosages in pregnancy require formal safety assessment(8, 22, 23).
The public health problem, what is being done and what more should be done
As stated in the Scientific Advisory Committee on Nutrition Update on Vitamin D(8), there is concern that vitamin D ‘recommendations are overlooked by health professionals, as well as by the general public’, and further that the ‘uptake of vitamin drops in the UK is very low even amongst those entitled to receive free supplies’. The recent provision of supplements containing 10 μg (400 IU) of vitamin D to pregnant/breast-feeding women and their offspring should lead to improvements in the situation, provided that the supply chain problems can be overcome(Reference McGee24).
British immigrants are recognised as at ‘high risk’ of hypovitaminosis D, but the problem is also common in pregnant White women, even when living in the Southern England(Reference Javaid, Crozier and Harvey47). As can be seen in Fig. 1, 90 % of white pregnant mothers in the Avon Longitudinal Study of Parents and Children(Reference Golding, Pembrey and Jones59) had 25(OH)D concentrations < 50 nmol/l during winter and spring; 28 % were seriously deficient ( < 25 nmol/l), and virtually no one reached 75 nmol/l (currently considered optimal). Over the year, 60 % of expectant mothers (approximately 403 000 English/Welsh women) are likely to require vitamin D supplementation for avoidance of serum 25(OH)D < 50 nmol/l (90 % being White). Over 150 000 mothers will have deficiency ( < 25 nmol/l), and 59 % of these being White (estimated from data presented in Fig. 1 and published prevalence rates for non-White British immigrants(Reference Yu, Sykes and Sethi57)). As seen in Fig. 1, hypovitaminosis D in White women in the UK is largely a problem during winter and spring, suggesting that treatment during these seasons would benefit most women.
Prevalence of hypovitaminosis D in pregnant White women living in Southern England (Avon Longitudinal Study of Parents and Children study(Reference Golding, Pembrey and Jones59), pilot sample, n 354). Error bars represent 95 % CI for prevalence. ■, < 25 nmol/l; 
, < 50 nmol/l; 
, < 75 nmol/l.
The UK Health Minister emphasised (December 2007) that ‘women should contact their GP for a blood test if they think they may be lacking the vitamin’ (Table 2). However, this approach does not allow for the high prevalence of hypovitaminosis D, for the high cost of serum 25(OH)D assays (currently approximately £10·50–£25 depending on the assay and laboratory) or for the undesirability of delays in starting supplementation while awaiting results. Indeed, given the data that have been discussed already, it can be argued that relatively little is gained by measuring individual 25(OH)D concentrations since routine supplementation of a pregnant/breast-feeding woman at 10 μg/d (400 IU/d) can currently be provided for £3·64/year(Reference McGee24).
Conclusions
The prevalence of hypovitaminosis D in expectant mothers in Britain is unacceptably high. We, therefore, suggest that all pregnant mothers should be offered vitamin D supplementation throughout the pregnancy to provide cheap, safe and effective prevention of overt vitamin D deficiency. As shown by previous UK experience(Reference Bivins5, Reference Arora and Arora16), 10 μg/d (400 IU/d) of vitamin D should be enough to prevent the major bony and life-threatening complications of severe clinical deficiency. Higher doses may well be needed to achieve adequate neonatal vitamin D repletion(Reference Doets, de Wit and Dhonukshe-Rutten25, Reference Yu, Sykes and Sethi57), but well-designed randomised controlled trials are urgently needed to establish the potential benefits (and safety) of higher maternal vitamin D intakes.
Acknowledgements
We thank Professor George Davey Smith, Dr Jon Tobias, Professor Deborah Lawlor and the Avon Longitudinal Study of Parents and Children Study Group for allowing pilot data on vitamin D status in pregnant women in the Avon Longitudinal Study of Parents and Children study to be presented, and Professor Carol Dezateux for constructive comments on an earlier version. Ms Kate Noonan (Clinical Chemistry, Barts and The London NHS Trust) is gratefully acknowledged for carrying out the pilot 25(OH)D assays, which were funded in part by North (originally North East) Thames NHS R&D Directorate. E. H. is funded by the DoH (UK) Public Health Career Scientist Award. The present work was undertaken at the Great Ormond Street Hospital/University College London, Institute of Child Health, which received a proportion of funding from the DoH's National Institute of Health Research (‘Biomedical Research Centres’ funding). The Medical Research Council provides funds for the MRC Centre of Epidemiology for Child Health. The authors declare no conflicts of interest. Sources of funding had no influence on the writing of the paper or on the decision to submit for publication. Contributions: E. H. initiated the study; both authors jointly developed concepts and wrote the paper. B. J. B. contributed to the development of the Avon Longitudinal Study of Parents and Children pilot study of maternal vitamin D status and provided part funding for the serum 25(OH)D assays.
