Vitamin D in pregnancy-time for action: a paediatric audit


Dr S Alexander, Department of Paediatrics, Chelsea and Westminster Hospital, 369 Fulham Road, London SW10 9NH, UK. Email


Maternal vitamin D deficiency is a significant public health issue. This retrospective multicentre audit was undertaken to identify the scale of maternal vitamin D deficiency in London, using infant vitamin D deficiency as a surrogate marker. During January 2006 to June 2008, 74 infants presented with symptomatic vitamin D deficiency, a prevalence of 1.6 per 1000 deliveries in London. Of these, 49% were hypocalcaemic at presentation and 27% had hypocalcaemic seizures. A telephone survey in June 2008 showed that no London National Health Service antenatal units had departmental guidelines on vitamin D. This audit is presented to raise awareness of current recommendations and the sequelae of maternal vitamin D deficiency.


Maternal vitamin D status in late pregnancy and during lactation influence the newborn infant’s vitamin D status and subsequent bone development. Therefore, babies born to women at high risk of vitamin D deficiency have a higher prevalence of vitamin D deficiency. Recent studies have shown that low maternal vitamin D levels in late pregnancy were associated with higher incidence of hypocalcaemia in the neonate, a decreased knee–heel length at birth and reduced whole body and lumbar spine bone mineral content in the child/offspring at age of 9 years.1 In addition, lack of vitamin D in children and adults has been associated with a higher incidence of non-skeletal pathology including autoimmune diseases like diabetes, cancers (colorectal and breast), infectious diseases, allergy and asthma.

There is an increase in the prevalence of vitamin D deficiency among children and women of childbearing age across the world.2 Recent reports from United Kingdom (UK) and Europe confirm that vitamin D deficiency during pregnancy is a significant public health issue, especially in the at-risk immigrant population.2 Pigmented skin and excessive body cover are specific risk factors.

The authors had noted an increase in the number of neonates and infants (<1 year old) presenting with vitamin D deficiency-related symptoms to their respective hospitals and believed that poor maternal vitamin D status during pregnancy (because of inadequate antenatal screening and supplementation) and lactation were major contributing factors. This retrospective audit was therefore undertaken to determine the extent of biochemical vitamin D deficiency in infants (<1 year old) as a surrogate marker of maternal vitamin D deficiency. In addition, we wanted to find out the number of London antenatal units with established guidelines regarding vitamin D in pregnancy.


A retrospective audit in four London hospitals was undertaken for the period of January 2006 to June 2008. The aim was to identify the number of children under the age of 1 year diagnosed with 25-hydroxy vitamin D deficiency (<50 nmol/l) through the biochemistry laboratory database of the respective hospitals. The clinic letters/hospital notes, blood results and radiological investigations of the eligible children were reviewed by the individual hospital teams and the information collected using an audit proforma. Babies that were under 36 weeks’ gestation or had an underlying systemic illness pre-disposing them to vitamin D deficiency were excluded. In one of the hospitals, information from the biochemistry database was not available to the clinicians and therefore the clinical database of children admitted with hypocalcaemic seizures secondary to vitamin D deficiency during the study period were included.

During the month of June 2008, the midwife coordinator or a senior midwife in the antenatal clinic of 24 out of 28 National Health Service maternity units providing antenatal care in London were contacted by telephone to enquire if their department had guidelines on vitamin D in pregnancy. Despite several attempts, antenatal clinic in 4 units could not be contacted.


In this multicentre London audit, 74 children under the age of 1 year with vitamin D deficiency were identified. Of these, 61% were under the age of 6 months (mean = 5 months) and 57% were males. Fifty-eight percent of children with a documented birth weight (n = 52), had a birth weight of <3 kg. Almost all infants (92%) belonged to the ethnic groups with pigmented skin, with 62% being of Asian origin. The ethnic group was not known for five children and one child was Caucasian. In those where feeding information were available (n = 59), the vast majority were exclusively breastfed at the time of the diagnosis (44) while 13 were on mixed breast and bottle feeds and two were on exclusive formula feeds.

Majority (81%) of the patients were referred either through the Accident and Emergency department or General Practitioners with acute clinical presentations. Rest of the referrals source included neonatal teams, paediatric surgeons, orthopaedic team, gastroenterologist or health visitor. Twenty children (27%) presented with seizures that were secondary to hypocalcaemia. Other presentations were varied (lethargic/floppy child, respiratory infections, febrile illness, weight loss, skeletal deformity and ‘incidental finding’) but also included significant morbidities like failure-to-thrive and cardiomyopathy.

Biochemical investigations revealed that 54% of the children included in the audit were severely vitamin D deficient (<25 nmol/l) and 49% were hypocalcaemic on presentation. Information on whether maternal and sibling screening was undertaken or recommended couldn’t be obtained for most of the children.

Since the analysis of the data, all four units are in the process of evaluating and implementing NICE guidelines on vitamin D for pregnant and breastfeeding women. Telephone survey showed that none of the 24 National Health Service antenatal care providers in London contacted had a department guideline on vitamin D in pregnancy in June 2008. One department was undertaking an audit on selective vitamin D screening in pregnancy and another had recently had a department presentation to highlight the scale of the problem.


We report our findings to raise awareness among obstetricians and midwives regarding the re-emergence of neonatal and infant vitamin D deficiency and its contributing factors.

Previous studies have shown that more than 90% mothers of rachitic infants are deficient themselves. There is a positive correlation between the serum 25-OHD concentrations of mothers and their children. This association persisted when the correlation was performed according to age: age <12 months (r = 0.39, P = 0.012) and age >12 months (r = 0.45, P < 0.001).3 We believe that these mothers would almost certainly have been deficient during their antenatal period. Other studies have also shown adverse effects of antenatal vitamin D deficiency even as late as 9 years of age.1 Given the above observations, we chose infant vitamin D deficiency as a surrogate marker of antenatal vitamin D deficiency.

This audit shows that each year a number of infants present to London hospitals with significant morbidity secondary to nutritional vitamin D deficiency. The true prevalence of vitamin D deficiency in infants is difficult to determine unless a national newborn/infant screening programme is undertaken or a national surveillance by active reporting is set up. Under the current setting, using the total number of deliveries during the audit period (46 790 deliveries), gives an estimated prevalence of 1.6 symptomatic vitamin D deficiency per 1000 deliveries. This prevalence is much higher than the conditions which are currently subject of a national screening programme e.g. Down’s syndrome (risk of 1 in 1500 for a 20 year pregnant woman) and cystic fibrosis (risk of 1 in 2500 live-birth in population of western European origin). Moreover, vitamin D deficiency is preventable if appropriately addressed.

For the purpose of this audit, all infants with any possible known causes for vitamin D deficiency (prematurity, malabsorption disorders, renal tubular disorders, antiepileptic medication and genetic disorders) were excluded to ensure that only nutritional vitamin D deficiency was included. Most infants presented with symptoms related to vitamin D deficiency (Table 1). We believe that there are a large number of asymptomatic infants with vitamin D deficiency in the community. Therefore, the prevalence shown in this audit is almost certainly a major underestimate and the true prevalence of infant vitamin D deficiency is far higher.

Table 1.   Summary of clinical presentation and findings
  1. In 15 infants, the initial clinical presentation could not be determined.

Presenting symptomNumbersAge (months) at presentation, range (mean)Vitamin D levels (nmol/l), range (mean)Calcium level (mmol/l), range (mean)
Specific presentations (neuro-muscular or skeletal)
Seizures20 0.25–10.6 (4.69)  2–38 (15.8)1.07–1.68 (1.54)
Floppy baby4 1–12 (7.5)  2–25 (13)1.28–2.5 (1.8)
Skeletal deformity3 6–9 (7.7) 16–45 (32)2.14–2.27 (2.2)
Cardiomyopathy1 0.4 181.5
Non specific presentations
Others/Incidental (‘unwell’, prolonged jaundice, febrile illness, poor feeding, gastroenteritis, poor weight gain, sibling screening, constipation, etc.)29 0.2–12 (4.85)  6–48 (25.6)1.24–2.67 (2.0)

Infants identified in this audit above the age of 6 months may well have other factors like a poor diet (prolonged exclusive breastfeeding) contributing to the vitamin D deficiency. The vast majority (97%) of our patients were breastfed (either exclusively or combined with formula). Breast milk is a poor source of vitamin D especially in mothers who are themselves deficient, as was probably the case in our patients. Infants receiving fortified formula milk (at least 500 ml/day) or multivitamin supplements will receive the recommended daily intake of vitamin D. The ideal plasma level of 25-Hydroxy vitamin D required are unknown but most experts would agree that levels of up to 50–80 nmol/l are required to ensure good bone health.

In the last two decades, the number of studies reporting an increase in maternal vitamin D deficiency has risen. In a recent study in London, 47% of Asian, 64% of Middle Eastern, 58% of African/Caribbean pregnant women were found to have vitamin D levels below 25 nmol/l compared with 13% of Caucasian pregnant women.4 Vitamin D deficiency has been reported in 60–84% of non-western pregnant women living in the Netherlands, 61% in New Zealand, 80% in Iran, 42% in India and 25% in United Arab Emirates.2 These figures also help to demonstrate the potential scale of the problem in the immigrant population in the western world including UK. Exposure to adequate sunlight accounts for 90% of vitamin D production, with pigmented skins requiring up to ten times more exposure than lighter skin. Moreover, northerly latitudes reduce the efficiency of the sun’s uVB rays. As a result, supplementation with vitamin D is probably the only way of meeting the recommended nutritional dosage. At present, 30% of the London population belongs to ethnic minority communities compared with 6% in England as a whole. (London Maternity Services Review Project)

It is important to note that 13% of lighter skinned Caucasian pregnant women in London have also been found to have low vitamin D levels4. The UK National Diet and Nutrition Survey found that 28% of women aged 19–24 years and 13% of those aged 25–34 years had plasma vitamin D levels below 25 nmol/l.5 Similarly, a number of reports in the paediatric literature between 1999 and 2006 have documented an increasing prevalence of rickets in the UK.

Clear recommendations on vitamin D supplementation in pregnancy have been issued as far back as 1991, when Committee on Medical Aspects of Food and Nutrition Policy (COMA) recommended that all pregnant and breastfeeding women should receive vitamin D [10 microgram (μg)/400 IU/day), but this did not translate into clinical practice. More recently, the Department of Health (DoH) in 2005 and the scientific advisory committee on nutrition (SACN) in 2007 reiterated the above recommendation. The re-emergence of vitamin D deficiency in the UK has now made organisations like the National Institute of Clinical Governance (NICE) to modify their recommendations in 2008 (c.f. 2003). It now classifies vitamin D as a high priority recommendation, and states that—‘during the booking appointment at the beginning of pregnancy, midwives should offer every woman information and advice on the benefits of taking a vitamin D supplement (10 microgram/day) during pregnancy and while breastfeeding’. They have also specifically advised that health professionals should check that the at-risk women are following the advice while calling for research into the effectiveness of routine antenatal vitamin D supplementation. Women and children of South Asian, African, Caribbean and Middle Eastern descent, and those who remain covered when outside, those with diet low in vitamin D and those with pre-pregnancy body mass index of 30 or more are identified to be at greatest risk. (NICE clinical guideline 62: Antenatal Care, Routine care for the healthy pregnant women, NICE public health guidance 11: Improving the nutrition of pregnant and breastfeeding mothers and children in low-income households. March 2008).

The Department of Health’s Healthy Start Scheme which replaced the Welfare Food Scheme for women from low income group and disadvantaged households in the UK aims to address nutritional issues in this group by giving vouchers for free milk and fresh fruit and vegetables to children and mums-to-be in addition to vitamin supplements containing vitamin C, D (10 micrograms) and Folic Acid. For those not eligible for the healthy start scheme, supplements are available from the local pharmacy.

In addition, recent initiatives aimed at the general public have also highlighted the need for vitamin D supplementation in pregnancy. ‘The Pregnancy Book 2007’ (DoH) recommends that high risk women with dark skin or those who always cover their skin should obtain vitamin D supplements while the newly introduced NHS online one-stop shop (pregnancy care planner) advises on the importance of vitamin D and recommends that pregnant women may choose to take 10 microgram of vitamin D per day. (

Despite all of the above government initiatives there seems to be a significant lack of awareness and implementation at the point of service delivery. In June 2008, 2 months after the NICE guidance publication on maternal nutrition, majority of the London antenatal units did not have any specific guidance on vitamin D. Although this situation may have now improved with units wanting to be compliant with NICE guidelines, we believe that the Royal College of Obstetricians and Gynaecologists (RCOG) and Royal College of midwifery should make unequivocal recommendations in this field for all units offering antenatal care.

While, we support the above recommendations which we believe apply to the women starting their pregnancy with vitamin D levels in the normal range, there is an urgent need to identify the ideal amount and timing of administration of vitamin D supplementation in the high-risk group, who are most probably already deficient at the start of the pregnancy. Supplementation of pregnant women, with 400 IU vitamin D/day during the last trimester of pregnancy did not significantly increase circulating 25(OH) D concentrations in the mothers or their infants at term.6 In the most recent study, by Yu et al.,4 180 pregnant women in a London hospital were randomised in to one of three groups: a daily 800 IU of vitamin D (ergocalciferol), a stat dose of 200 000 IU of calciferol or no treatment from 27 weeks until delivery. Even though the maternal and umbilical cord vitamin D levels were significantly higher in the vitamin D arms compared with the placebo arm, only 30% of treated women and 8% of the babies at term in these groups achieved vitamin D sufficiency (levels >50 nmol/l). Thus, women who are deficient at the start of the pregnancy are likely to remain deficient at the end of the pregnancy despite taking 400 IU of vitamin D dosage daily throughout their pregnancy. In the absence of evidence on how best to manage the high risk group, one management option would be to check vitamin D levels at booking and treat appropriately, possibly with higher doses. However, the least an antenatal care provider should do is to implement the NICE guidance of recommending 10 microgram (400 units) of vitamin D per day during pregnancy.

The other point worth highlighting here is that breast milk is a poor source of vitamin D, and more so if the mother is vitamin D deficient. Vitamin D supplements as part of a multivitamin preparation is recommended for breastfed infants from the age of 6 months (or 1 month if the vitamin D status is in question), if they are consuming <500 ml of formula milk. In the 1- to 4-year age group, DoH recommend universal vitamin D supplementation while NICE recommends supplementation in those eligible for the Healthy start programme. Importance of vitamin D supplements in infants and young children should therefore be emphasised to parents in antenatal classes, especially to those at high risk.

We recognise the limitations of this audit, which include an inability to exactly quantify the clinical problem in the population. Data like the proportion of exclusively breastfed babies at the time of the diagnosis, dietary intake in infants over 6 months and extent of maternal body cover could not be collected for all cases because of its retrospective methodology.

In conclusion, vitamin D deficiency among infants is re-emerging in the at-risk ethnic population. Ensuring that pregnant mothers are replete in vitamin D will have significant beneficial effects on the offspring and the general health of the community. While, current recommendations are clear there seems to be a clinical inertia to implement them. We believe there is an urgent need to raise awareness among all antenatal care service providers and pro-actively implement vitamin D supplementation in pregnant and breastfeeding women and infants.

Disclosure of interests

Nil to declare.

Contribution of authors

SS, MC, SA participated in the conception and design of the audit. All authors participated in data acquisition. SS and SA participated in the analysis of the data. All authors contributed to the drafting of the article. All authors have seen and approved the final version.


Nil to declare.


We thank the Biochemistry department staff that conducted the database search and allowed access to the data.