OUP user menu

Vitamin D and gestational diabetes mellitus

Maysa Alzaim , Richard J Wood
DOI: http://dx.doi.org/10.1111/nure.12018 158-167 First published online: 1 March 2013


The incidence of gestational diabetes mellitus (GDM) is increasing worldwide. GDM can be responsible for an important proportion of adverse fetal and maternal outcomes during pregnancy, and it is associated with long-term health deterioration for both mother and child. Therefore, it is important to identify potentially modifiable risk factors for GDM. Accumulating evidence links vitamin D deficiency with abnormal glucose metabolism, and epidemiological studies have shown that women who develop GDM are more likely to be vitamin D deficient. This review discusses the prevalence, risk factors, and outcomes of GDM and vitamin D deficiency in pregnant women, outlines the possible mechanism of action of vitamin D in glucose homeostasis, and summarizes emerging evidence that associates vitamin D deficiency with the risk of developing GDM. This critical review of the literature indicates there is a need for intervention trials to test the possible beneficial effect of vitamin D supplementation in pregnant women with low vitamin D status to reduce the risk of developing GDM.

  • gestational diabetes
  • glucose metabolism
  • insulin resistance
  • vitamin D


The American Diabetes Association (ADA) defines gestational diabetes mellitus (GDM) as carbohydrate intolerance of variable severity, with onset or first recognition during pregnancy.1 Pregnancy is a diabetogenic condition in which gestational steroid hormones induce peripheral insulin resistance; thus, the stress on pancreatic β-cell function increases to levels typically associated with type 2 diabetes.2 Compared with normal pregnant women, women with GDM have impaired β-cell function and reduced β-cell adaptation, resulting in insufficient insulin secretion to maintain normal glycemia.3 Usually, insulin resistance increases in mid-pregnancy and through the third trimester. For this reason, the ADA recommends that all pregnant women be screened for GDM between weeks 24 and 28 of gestation, with the exceptions of women with low risk status, including maternal age <25 years, normal body weight, not being members of a racial/ethnic group with a high prevalence of diabetes, no history of poor obstetric outcome, and no family history of diabetes.1 Though varying criteria can be used, a diagnosis of GDM is characterized by a fasting plasma glucose level of =126 mg/dl (7.0 mmol/L) or a casual plasma glucose level of =200 mg/dl (11.1 mmol/L), if confirmed on a subsequent day, and precludes the need for any oral glucose tolerance test.1

The prevalence of GDM is increasing in parallel with the rising prevalence of type 2 diabetes (T2D) and the growing global obesity epidemic.4,5 This increase in GDM could be related to the obesity-associated increase in T2D, which, if undiagnosed before pregnancy,5 might explain the observed increasing GDM prevalence trends. Women at risk of T2D are also at risk of GDM.4 The prevalence of GDM around the world is from 1% to 14% of all pregnancies, depending on the ethnic mix of the population and the GDM diagnostic tests administered.6 In the United States, there is an increasing occurrence of GDM. One recent study reported that, between the time periods 1989–1990 and 2003–2004, the prevalence of GDM doubled from 1.9% to 4.2%.7 Annually, 135,000 cases of GDM, representing, on average, 3–8% of pregnancies, are diagnosed in the United States with a higher prevalence seen among Native-American, Asian, African-American, and Hispanic populations than among non-Hispanic whites.5 A higher incidence of GDM has also been reported in Australia (6–10%)8 as well as in developing countries such as India (14.3%), China (13.9%),9 Saudi Arabia (12.5%),10 and Iran (7%).11 GDM imposes a significant economic burden on society. In 2007, the short-term annual medical costs associated with GDM in the United States were estimated at $636 million–$596 million for maternal costs and $40 million for neonatal costs, excluding the long-term costs of adverse sequelae.12

Gestational Diabetes and Adverse Health Outcomes

Maternal health effects

GDM can also be responsible for an important proportion of fetal and maternal morbidity and mortality.7 During pregnancy, the short-term maternal health consequences of GDM include an increased risk of hypertension, preeclampsia, urinary tract infections, and caesarean delivery.7 Longitudinal studies of the long-term adverse health effects of GDM found that as many as 70% of women who experience GDM will develop T2D within 10 years after delivery.2 In addition to the increased risk of developing T2D later in life, women who have had GDM with one pregnancy are 35–80% more likely to develop GDM again in future pregnancies.9

Offspring health effects

Untreated hyperglycemia allows glucose to travel freely from the mother to the fetus, forcing the fetus to increase its own production of insulin.3 Excess insulin production by the fetus increases the risk of embryopathy, including the following: fetal macrosomia, birth trauma, increased risk of neonatal intensive care unit admission, and perinatal death.12 According to a study performed in Saudi Arabia, GDM was a major contributor to perinatal death, congenital malformation, and neonatal jaundice.13 Fetal macrosomia, or excessive birth weight, is present in 20% of newborns from mothers with GDM, but it occurs in only 12% of pregnancies in which GDM is absent.12 These macrosomic newborns are at increased risk of developing hypoglycemia, jaundice, polycythemia, hypocalcemia, hypomagnesemia, and respiratory distress syndrome.12

In addition, upon reaching adulthood, offspring from mothers with GDM are at increased risk of developing obesity, impaired glucose tolerance, T2D,14 and cardiovascular disease.12 Long-term follow-up of the offspring of Pima Indian mothers, the population with the world's highest prevalence of type 2 diabetes, strikingly supports this relationship between GDM and T2D.5 The offspring born to mothers who were already diabetic upon pregnancy had a 45% likelihood of developing T2D by the age of 20–24 years, whereas the likelihood amongst the offspring of mothers who developed diabetes after pregnancy was only approximately 9%.4

Risk Factors for GDM

The known risk factors for GDM include maternal overweight and obesity, being of a particular race/ethnicity, prior history of GDM, family history of T2D, history of previous fetal death, previous delivery of a macrosomic infant, and increasing maternal age.15 Recent studies have also indicated that vitamin D deficiency might also be a modifiable risk factor for the development of GDM. This suggestion is of interest because vitamin D status is readily amenable to treatment with vitamin D supplements. Some additional points to consider are the following: vitamin D deficiency and insufficiency is quite common among pregnant women around the world; vitamin D deficiency is associated with abnormal glucose metabolism in both nonpregnant and pregnant subjects; and women who develop GDM are more likely to be vitamin D deficient compared to pregnant women who do not develop GDM.8,11,1619 The connection between poor vitamin D status and GDM may be confounded by the known association of obesity and vitamin D deficiency, as well as the increased risk of vitamin D deficiency in darker-skinned ethnic and racial groups.

Role of Vitamin D in Pregnancy

The likely importance of vitamin D during pregnancy is suggested by the presence of nuclear vitamin D receptors and the vitamin D-activating 1-α-hydroxylase enzyme in pregnancy-specific tissues such as the decidua and placenta.20 Serum 25-hydroxyvitamin D concentration is currently considered to be the best biomarker of vitamin D status, and it reflects both dietary vitamin D intake and cutaneous vitamin D production. According to the most recent version of the dietary reference intakes for vitamin D from the Institute of Medicine of the National Academies, the role of vitamin D in pregnancy and fetal development is unclear.21 Moreover, pregnancy does not appear to increase the need for vitamin D. Based on analysis of available information, the current recommended dietary allowance (RDA) for vitamin D in pregnancy is 600 IU/day, the same as for nonpregnant women. This level of vitamin D intake is estimated to be capable of achieving a serum 25-hydroxyvitamin D concentration of approximately 50 nmol/L, which is deemed to be a sufficient concentration to prevent vitamin D deficiency (based on bone health) in 97.5% of the population.21 Serum vitamin D deficiency is generally considered to reflect a serum 25-hydroxyvitamin D concentration of <50 nmol/L (20 ng/mL). Prevalence estimates of vitamin D deficiency in this paper are, therefore, based on this cut-off value.

Surprisingly, pregnancy has no obvious effect on vitamin D status, and important pregnancy-associated changes in calcium economy appear to be independent of vitamin D status.21 The fetus requires a relatively large amount of calcium, about 30 g, during development, with the majority required in the third trimester,22 and total serum calcium concentrations are known to decline as pregnancy progresses.15 However, the change in serum calcium may reflect a dilution effect due to pregnancy-associated changes in plasma volume. In addition, maternal serum 1,25-dihydroxyvitamin D, the active hormonal form of vitamin D, rises from early in the first trimester and increases progressively during gestation, rising to twice as high in late pregnancy than in postpartum or in nonpregnant controls.22,23 The reason for this increase of 1,25-dihydroxyvitamin D remains unclear; some think it is linked to elevated 1α-hydroxylase activity in the maternal kidney,24 with possibly some additional input from the fetal kidney, placenta, and decidua by extrarenal 1-α hydroxylase enzyme activity.22,23 The changes in serum 1,25-dihydroxyvitamin D concentration during pregnancy could reflect a rise in blood vitamin D-binding protein concentration. Free 1,25-dihydroxyvitamin D concentrations do not rise until the third trimester. On the other hand, these changes in 1,25-dihydroxyvitamin D do not appear to be linked to changes in intestinal calcium absorption. Animal studies have shown that maternal hormones such as prolactin and placental lactogen can stimulate intestinal calcium absorption independently of 1,25-dihydroxyvitamin D, indicating that these hormones could be responsible for the observation that intestinal calcium absorption doubles in humans and rodents early in pregnancy, well before free 1,25-dihydroxyvitamin D concentrations increase late in pregnancy.24 Clearly, the role of vitamin D in pregnancy on maternal-fetal calcium transfer is open to further study. However, according to a recent evaluation of the evidence concerning dietary requirements for calcium and vitamin D made by the Institute of Medicine, the “bulk of the evidence suggests that calcium is moved to the fetus without requiring calcitriol.”21

There is no clear indication that pregnancy influences vitamin D status or vitamin D requirements above those of nonpregnant women. However, some studies do report declines in vitamin D status throughout pregnancy and in the postpartum period. For example, Haliloglu et al.25 recently collected blood from 30 healthy Turkish pregnant women in each trimester and during the postpartum period (6 weeks after delivery); they reported that serum 25-hydroxyvitamin D concentration decreased significantly as follows: first trimester, 48 ± 18 nmol/L; second trimester, 39 ± 18 nmol/L; third trimester, 28 ± 22 nmol/L; and further decreases in the postpartum period (17 ± 11 nmol/L). However, the authors did not address the contribution of any potential confounders, including possible seasonal effects, on vitamin D status.

Low Vitamin D Status is Common in Pregnant Women

In the absence of vitamin D supplementation, vitamin D status during pregnancy is likely to be a reflection of a woman's usual vitamin D status. Moreover, to a very large extent, fetal vitamin D status is a reflection of maternal vitamin D status. Vitamin D deficiency is common among pregnant women.19 Vitamin D deficiency during pregnancy is geographically widespread and can occur in up to 95% of pregnant women, depending on country of residence and other factors.26 Residence in more northern or southern latitudes is associated with lower UVB radiation exposure and reduced cutaneous synthesis of vitamin D, leading to an increased risk of developing vitamin D deficiency. In Belfast, Northern Ireland (latitude 55°N), Holmes et al.26 found that the prevalence of vitamin D deficiency was approximately 95% among both pregnant and nonpregnant women. Yu et al.27 compared serum 25-hydroxyvitamin D at gestational week 27 among 180 pregnant women of four ethnic groups living in London (latitude 51°N). They found that Asian (47%), Middle Eastern (64%), and black women (58%) each had a higher prevalence of very poor vitamin D status, with serum 25-hydroxyvitamin D concentrations of <25 nmol/L, compared with Caucasian women (13%). The prevalence of secondary hyperparathyroidism in these women followed a similar ethnic/racial pattern. In addition, using multiple regression analysis, they found that ethnic group, age, parity, and daily sun exposure were significant predictors of vitamin D levels. Similarly, Johnson et al.28 recently investigated the prevalence of vitamin D deficiency during early pregnancy (<14 weeks) in 494 women from a southern city of the United States (Charleston, South Carolina; latitude 32°N) and found that, despite plenty of sunshine, 41% of the women were vitamin D deficient. As expected, race/ethnicity was an important determinant of vitamin D deficiency, which was found in approximately 75% of African-Americans, 30% of Hispanics, and 10% of Caucasian women. In Shepparton, Victoria, in Australia (latitude 32°S), Teale and Cunningham29 assessed vitamin D status in 330 pregnant women at approximately week 28 of gestation and found that vitamin D deficiency was present among 26% of the pregnant women; this was despite abundant sunshine and a latitude that provided adequate ultraviolet light for vitamin D synthesis throughout the year. Studies from other parts of Australia, The Netherlands, and New Zealand also identified a high prevalence of vitamin D deficiency in pregnant women; in particular, those with darker skin or limited skin sun exposure were found to be severely vitamin D deficient.29 Such data led to the call for pregnant women, especially those who are dark-skinned or veiled, to be screened and treated for vitamin D deficiency. However, in many countries, this is not a routine practice. In a 1984 study from Saudi Arabia, a region with plenty of sunlight (latitude 25°N) but where cutaneous sun exposure is often limited, especially in women, it was revealed that 25% of mothers and 68% of their neonates had undetectable to low serum 25-hydroxyvitamin D levels (0–25 nmol/L) at delivery, which predisposes the mothers and infants to bone diseases such as osteomalacia and rickets, respectively.30 Similarly, a recent study reported in 2011 continued to show a high prevalence (72%) of vitamin D deficiency among Saudi women of childbearing age.31

Adverse Health Effects of Low Vitamin D Status on Mother and Infant

There is considerable evidence that low maternal serum 25-hydroxyvitamin D is associated with adverse outcomes for both mother and fetus in pregnancy, as well as for the neonate and child after pregnancy.22 These adverse effects can include alterations in calcium homeostasis, which causes hypocalcemia and abnormal softening or thinning of the skull (craniotabes) and high bone turnover in the neonate as well as osteomalacia and hypovitaminosis D myopathy in the mother.27 Aside from the adverse skeletal effects, vitamin D deficiency during pregnancy has been linked with a number of maternal problems, including infertility, preeclampsia, gestational diabetes, and an increased rate of caesarean section.22,23,32 Stores of vitamin D in newborns are dependent on maternal vitamin D status.26 Maternal serum and cord blood 25-hydroxyvitamin D are highly correlated,33 and both maternal serum and cord blood 25-hydroxyvitamin D increase after maternal vitamin D supplementation.20 It has been suggested that low serum 25-hydroxyvitamin D in the mother might cause reduced transfer of 25-hydroxyvitamin D to the fetus, leading to impaired growth, low infant birth weight, delayed bone ossification and congenital rickets, abnormal tooth enamel formation, and lower bone mineral content.25 Case reports show that neonatal complications from extreme maternal vitamin D deficiency can be life threatening, e.g., severe hypocalcemic fits with high risks of resultant brain damage and neonatal heart failure.20 Long-term complications of poor vitamin D status for the offspring are not well understood, but may relate to changes in fetal programming that could lead to increased risk of osteopenia,16 type 1 diabetes,22 asthma, and schizophrenia in later life.23

Vitamin D and Glucose Homeostasis

A connection between vitamin D status and insulin and glucose metabolism in animals has been appreciated for approximately 30 years.34,35 Early animal studies suggested that vitamin D deficiency causes impaired insulin release from the rat pancreas. With increasing concern about the obesity epidemic and concomitant comorbidities, such as type 2 diabetes, as well as growing widespread concern about possible nonskeletal adverse effects of low vitamin D status, there has been renewed research interest in the possible role of vitamin D status in glucose metabolism. However, the evidence for a possible role of vitamin D in glucose homeostasis in humans is still controversial. The relationship was recently reviewed by Pittas and Dawson-Hughes,36 who summarized several mechanisms based on findings in both animal and human studies. In brief, as pointed out by these authors, the effect of vitamin D on regulation of pancreatic β-cell function and insulin secretion could be by a direct or indirect effect of the binding of the circulating active hormonal vitamin D form, 1,25-dihydroxyvitamin D, to the β-cell vitamin D receptor, which could regulate the balance between the extracellular and intracellular β-cell calcium pools.36 Vitamin D action may also be needed to ensure a normal rate of calcium influx across cell membranes and maintenance of an adequate intracellular cytosolic calcium pool, which is important for insulin-mediated intracellular processes in insulin-responsive tissues.19 In addition, vitamin D action could enhance insulin sensitivity by stimulating insulin receptor expression, thereby enhancing insulin-mediated glucose transport.36 Clinical evidence from a few small, and non-randomized, studies suggested that vitamin D supplementation can increase insulin secretion.16 During pregnancy, observational studies have found that maternal vitamin D deficiency is associated with an increased risk of several adverse pregnancy events, including gestational diabetes.22 To date, however, only a limited number of studies have investigated the association of vitamin D with glucose homeostasis during pregnancy and in the development of GDM. These studies are discussed below.

Do women with gestational diabetes have poorer vitamin D status?

Several cross-sectional studies have reported an association between low maternal vitamin D status and GDM (summarized in Table 1).

View this table:
Table 1

Observational studies investigating an association between vitamin D status and glucose metabolism in pregnant women

Soheilykhah et al.18 investigated serum 25-hydroxyvitamin D at weeks 24–28 of gestation among 204 consecutively enrolled pregnant Iranian women. Exclusion criteria included women with pregestational diabetes, multiple pregnancies, fetal abnormality, chronic disease, hypertension, and history of consumption of anticonvulsant drugs. GDM was defined as at least two of four abnormally elevated blood glucose concentrations in a 100 g oral glucose tolerance test (OGTT). A person with only one abnormal value on the OGTT was considered as having impaired glucose tolerance. Based on the oral glucose tolerance test, 26% of this group had GDM and 19% had IGT. Vitamin D deficiency was found in 78% of the cohort. The study participants were then divided into three groupings: 54 women with GDM, 39 women with impaired glucose tolerance (IGT), and 111 women with a normal glucose tolerance test. Women with GDM and IGT had a significantly lower median serum 25-hydroxyvitamin D concentration than normal control subjects (24 and 17 nmol/L versus 32 nmol/L, respectively). Women with GDM had a twofold higher (non-significant) risk of having vitamin D deficiency, and a significant 2.7-fold (95% CI, 1.26–5.6) increased risk of having more severe vitamin D deficiency (serum 25-hydroxyvitamin D < 37.7 nmol/L) compared with the control group. Serum 25-hydroxyvitamin D had no significant correlation with age, parity, BMI, or fasting blood glucose level in these subjects.

Maghbooli et al.11 investigated in 741 pregnant Iranian women the relationship at weeks 24–28 of gestation between serum 25-hydroxyvitamin D concentrations and insulin resistance, using the homeostasis model assessment index (HOMA) equation to obtain estimates. They found that in this pregnancy cohort there was a 71% prevalence of vitamin D deficiency (defined in this study as a serum 25-hydroxyvitamin D level <25 nmol/L), and the prevalence of severe vitamin D deficiency (<12.5 nmol/L) was 29% in the entire cohort. GDM developed in 7% of all pregnancies. Participants were classified into three groups according to their glucose homeostasis: GDM, IGT, and normal. There was a significant difference in serum 25-hydroxyvitamin D concentration between the GDM and normal groups (16 ± 10 versus 23 ± 18 nmol/L) and between the IGT and normal groups (19 ± 12 nmol/L versus 23 ± 18 nmol/L). Also, severe vitamin D deficiency was more common in the GDM group than in the IGT and normal groups (44%, 33%, and 23%, respectively.) These observations would be consistent with the notion that poorer vitamin D status might increase the risk of GDM. Likewise, in regard to the question of vitamin D deficiency and insulin resistance in this study, the prevalence of a HOMA index ≥3 (indicative of insulin resistance) in those with vitamin D deficiency was higher than in those with normal vitamin D status (43% versus 31%). Serum 25-hydroxyvitamin D was significantly correlated with the HOMA index, but no correlation with serum 25-hydroxyvitamin D was observed with age, parity, and BMI. In all subjects combined, there was a significant positive correlation found between the HOMA index and BMI and parity, but there was no correlation with age. However, the association between insulin resistance and serum 25-hydroxyvitamin D remained the same, even after adjustment for age and BMI.

Clifton-Bligh et al.16 studied the association during the second or third trimester of gestation between serum 25-hydroxyvitamin D and glucose metabolism, and the effect of ethnicity on this relationship, in a convenience cohort of 307 pregnant women attending a prenatal clinic in metropolitan Sydney, Australia. Ethnic groups were Europeans, Southeast Asians, Asians, and Middle Easterners. In this cohort, 48% of the women were vitamin D deficient, while GDM was evident in 32% of the cohort. In the entire group, serum 25-hydroxyvitamin D was negatively correlated with fasting plasma glucose, fasting insulin, and insulin resistance (calculated by HOMA-IR). The association between serum 25-hydroxyvitamin D and fasting glucose remained significant after accounting for ethnicity, age, and BMI in multivariate analyses. However, the associations between 25-hydroxyvitamin D and fasting insulin and HOMA-IR were no longer significant after accounting for the previous confounders. A possible problem in interpretation of these findings is that the role of the confounding factors, such as ethnicity and obesity, in this relationship is unclear. Multivariate analysis, with its assumption of a linear relationship between the confounders, may also not be sufficient to answer this question; moreover, it should be noted that certain ethnic groups may be predisposed to both vitamin D deficiency and fasting hyperglycemia through independent mechanisms. The high proportion of GDM in this group (32%) was likely due to a bias toward women with a greater risk of having GDM because it included all women who had a blood test at this obstetrical clinic, which also included women specifically referred for a follow-up OGTT. Although mean serum 25-hydroxyvitamin D was lower in those with GDM, the odds ratio (OR) for GDM risk in those with vitamin D deficiency (OR: 1.92) did not reach statistical significance. Nevertheless, the overall findings in this study are consistent with the hypothesis that poor vitamin D status is a risk factor for poor glucose control in pregnant women.

Lau et al.8 investigated during the third trimester (mean gestational age of 35 ± 2 weeks) the association between serum 25-hydroxyvitamin D and glycated hemoglobin (HbA1c), an integrated measure of blood glucose control, in 147 pregnant Australian women with GDM. Interestingly, despite the fact that all women were advised to take daily prenatal multivitamins that contained either 400 IU or 500 IU vitamin D (cholecalciferol), 41% of the participants had vitamin D deficiency. Ethnicity, occupational status, and season significantly influenced serum 25-hydroxyvitamin D, but BMI did not. Indian subcontinent and Middle Eastern groups had similar levels of serum 25-hydroxyvitamin D (median levels: 49 and 38 nmol/L, respectively), which were significantly lower than those for the East or Southeast Asian and Caucasian groups (median levels: 63 and 62 nmol/L, respectively). Other confounders that may affect vitamin D status, such as sunlight exposure, physical activity, dietary vitamin D, and calcium intake, were not examined. Serum 25-hydroxyvitamin D was inversely associated with fasting and 2-h blood glucose concentration during an oral glucose tolerance test, as well as with log[HbA1c]. The latter relationship also remained significant after removal of HbA1c outliers that were =7%, as well as after removal of HbA1c values that were =6%, even though this excluded 38% (n = 56) of the study sample. Moreover, excluding both low (<5%) and high (=7%) HbA1c outliers also did not negate the observed inverse relationship between serum 25-hydroxyvitamin D and HbA1c. Multivariable analysis identified serum 25-hydroxyvitamin D and blood glucose concentration during the oral glucose tolerance test as independent predictors of fasting HbA1c levels. These findings suggest that, even within a patient group with established GDM, poorer vitamin D status is associated with poorer blood glucose control. However, the multiple ethnic groups in this cohort make a clear interpretation of these findings uncertain.

In a cross-sectional study of 559 nondiabetic pregnant women living in South India, Farrant et al.37 investigated the associations between maternal vitamin D status and risk of having gestational diabetes at 30 weeks gestation; as well as associations with newborn size at delivery, and cord blood insulin concentration at delivery. Vitamin D deficiency was present in 66% of the women in this study, and 31% were found to have more severe vitamin D deficiency (serum 25-hydroxyvitamin D <28 nmol/L). GDM was found in 7% of the cohort. Median serum 25-hydroxyvitamin D concentrations were similar in women with and without GDM (∼38 nmol/L). In this study, there was no association found between maternal vitamin D status and risk of gestational diabetes. Observations from this study do not support the idea that GDM is associated with a lower vitamin D status, or that vitamin D deficiency increases the risk of having GDM. The only significant association found was that, within the group of mothers with vitamin D deficiency, there was an inverse association between serum 25-hydroxyvitamin D and 30-min glucose concentrations after an oral glucose load and a positive association between serum 25-hydroxyvitamin D and fasting proinsulin concentrations, including after adjustment for maternal age, fat mass, and GDM status. This observation would be consistent with the idea that low vitamin D status may affect blood glucose metabolism by influencing the early phase of glucose-induced insulin secretion following an oral glucose load.

Overall, these five cross-sectional studies are generally supportive of the hypothesis that women with gestational diabetes have poorer vitamin D status than pregnant women with normal glucose metabolism. However, caution should be exercised in concluding that this intriguing association is causal, rather than merely associative. As cross-sectional studies, it is not possible to predict causality and the associations observed may be influenced by ambiguous or incomplete control of important confounding variables such as ethnicity, seasonality, physical activity, parity, prepregnancy BMI, or other risk factors for GDM and/or vitamin D deficiency. Factors leading to a greater risk of developing vitamin D deficiency in some women may also be independently associated with an increased risk of glucose abnormalities during pregnancy. Also, in each of the studies examined above, measurement of serum 25-hydroxyvitamin D was in late pregnancy when GDM had already developed. It would be important to know the extent to which vitamin D status early in pregnancy might influence the risk of developing GDM. To our knowledge, this question has been addressed in only one prospective cohort study by Zhang et al.19

Does vitamin D deficiency increase the risk of developing gestational diabetes?

Zhang et al.19 conducted a prospective cohort study among 953 mostly non-Hispanic white pregnant women in Tacoma, Washington in the United States, examining the association between maternal plasma 25-hydroxyvitamin D in early pregnancy (∼16 week) and the risk of developing GDM. These researchers found that among the 953 pregnant women enrolled in their study, 57 women (∼6%) developed GDM. Using a nested case-control design, women with GDM were matched (by season of conception) with 114 control women from the cohort who did not develop GDM. Mean maternal plasma 25-hydroxyvitamin D, measured at approximately 16 weeks of gestation, was significantly lower, by 20%, in the group that developed GDM than in the control group (60.5 versus 75.3 nmol/L), and this difference between groups remained after controlling for established risk factors of GDM, including maternal age, family history of T2D, race/ethnicity, and prepregnancy BMI. Vitamin D deficiency was evident in 33% of the women who developed GDM compared to 14% in controls, which is consistent with most of the other cross-sectional studies discussed above11,16,18 in which GDM patients were more likely to be vitamin D deficient or to have lower serum 25-hydroxyvitamin D. In the Tacoma study, the risk of developing GDM, assessed at week 24–28 of gestation, was 2.66-fold (95% CI: 1.01–7.02) higher in women who were vitamin D deficient at week 16 of gestation compared to the nondeficient women, when controlled for established risk factors of GDM. Moreover, after restricting the analysis to non-Hispanic whites only, who comprised the majority of study participants, vitamin D deficiency at week 16 of gestation was associated with a 3.77-fold increased risk of developing GDM after covariance adjustments. Moreover, these researchers found that each 12.5 nmol/L decrease in serum 25-hydroxyvitamin D was related to a 1.29-fold (95% CI: 1.05–1.60) increase in GDM risk. Additional adjustment for season and physical activity did not substantially change these findings. Overweight, vitamin D-deficient women were at fivefold greater risk of developing GDM compared to lean women with adequate vitamin D status. Importantly, the findings in this study demonstrate that maternal vitamin D deficiency in early pregnancy (∼16 weeks) is significantly associated with an elevated risk of having GDM assessed at week 24–28 of gestation. The researchers cautioned, however, about possible overinterpretation of their findings because women in the study had only a single measurement of serum 25-hydroxyvitamin D, which may not reflect their vitamin D status throughout their pregnancy. In addition, since GDM was not assessed until 2–3 months later, there is the possibility that some of the GDM cases identified at week 24–28 of gestation may have already had undiagnosed glucose intolerance at 16 weeks when blood was drawn for vitamin D measurement. On the other hand, the fact that the study population was mainly non-Hispanic whites suggests that ethnicity was an unlikely potential confounder in their study, as could be claimed in some of the other studies in this research area. Moreover, although there seems to be increasing evidence of an association between poor vitamin D status and GDM risk, it still is not clear how vitamin D is involved with the pathogenesis of GDM and whether increasing vitamin D status in early pregnancy can reduce the risk of developing GDM. These important questions will require more basic research and the findings of vitamin D supplementation trials in pregnant women.

Intervention studies linking vitamin D status to GDM risk

1- α-hydroxyvitamin D treatment.

Rudnicki and Pedersen17 were the first to conduct a clinical trial to study the effect of administration of the active vitamin D hormone, 1,25-dihydroxyvitamin D3, on glucose metabolism in pregnant women with GDM. In that study, 12 GDM patients at approximately 27 weeks of gestation were given the vitamin D analogue 1α-hydroxyvitamin D, which is activated in vivo by hepatic C-25 hydroxylation of the compound to the 1,25-dihydroxyvitamin D hormone. The participants received a single intravenous dose of 1α-hydroxyvitamin D3 (Etalpha®, 2 μg/m2, LEO Pharma, Inc.) and were instructed to take 0.25 μg 1α-hydroxyvitamin D3 orally for the following two weeks and also follow a calcium-restricted diet. An OGTT was conducted three times in each patient: at baseline, 2 h after the intravenous 1α-hydroxyvitamin D3 dose, and then at 14 days after the oral 1α-hydroxyvitamin D3 dosing period.

Interestingly, the fasting serum glucose level immediately decreased significantly from 5.6 to 4.8 mmol/L (100.9 to 86.5 mg/dL) after intravenous treatment with 1α-hydroxyvitamin D, but was 5.3 mmol/L (95.5 mg/dL) after 2 weeks of oral 1α-hydroxyvitamin D. This study found that the serum glucose concentration was not different, but the serum insulin levels tended to be lower after both intravenous and oral 1α-hydroxyvitamin D supplementation, suggesting a possible increase in insulin sensitivity. In a multiple regression model analysis, including serum 1,25-dihydroxyvitamin D, serum insulin, body weight and height as predictors of serum glucose, only serum 1,25-dihydroxyvitamin D and serum insulin remained in the final regression model. Although the study shows that intravenous 1α-hydroxyvitamin D can significantly decrease fasting glucose in GDM, this effect may have been short-lived because the daily oral supplementation of 1α-hydroxyvitamin D did not sustain the initially observed effect on fasting glucose. This might be due to a number of reasons, such as the following: the low prescribed oral 1α-hydroxyvitamin D dose regimen; or increased 1,25-dihydroxyvitamin D breakdown induced by continued daily dosing of 1α-hydroxyvitamin D. It should also be mentioned that the three glucose tolerance tests were done in sequential order with no independent control group, and the vitamin D status of the patients is unknown.

Vitamin D treatment

Recently, a preliminary meeting report (Wagner et al., Pediatric Academic Societies Annual Meeting 2010, Abstract 1665.6, unpublished data) described the findings of a vitamin D treatment trial in 350 pregnant women at week 12–16 weeks gestation. Subjects were first stratified by race (African-American, Hispanic, and Caucasian) and then randomized into one of three vitamin D treatment groups: 400, 2,000, or 4,000 IU vitamin D3/day as dietary supplementation until delivery. According to the report, vitamin D sufficiency was strongly associated with decreased risk of pregnancy comorbidities, including preeclampsia, GDM, infection, preterm labor [PTL]/preterm birth [PTB] <37 weeks gestation). The group with the greatest effect was the 4,000 IU vitamin D/day regimen, which proved to have no adverse effect and raised serum 25-hydroxyvitamin D level to 100 nmol/L. The complete published findings from this study are keenly awaited.

In summary, a critical review of the literature concerning vitamin D and GDM indicates consistent observations in several countries that women with GDM have lower vitamin D status (Table 1). However, there is a need for intervention trials to test the possible beneficial effect of vitamin D supplementation in pregnant women with low vitamin D status to reduce the risk of developing GDM.


GDM imparts significant and long-lasting health risks on mother and baby. Fetal programming in utero is believed to increase the risk of obesity and chronic diseases in offspring of mothers with GDM. Therefore, it is very important that GDM is promptly recognized and appropriately managed to reduce perinatal deaths and to improve the quality of life for both mother and child. Vitamin D deficiency remains a common problem among certain pregnant women of various ethnicities and in many countries, and may have long-term negative consequences on both mother and child. Even some countries with abundant sunshine, such as Australia, have taken steps to screen and treat for vitamin D deficiency among pregnant women, with an emphasis on those who are dark-skinned or veiled.24 This action was taken after a high prevalence of vitamin D deficiency was detected among pregnant women living in Australia; however, not all countries have taken similar action or have agreed on the appropriate course of action. For example, a very recent official statement issued by the American College of Obstetricians and Gynecologists in July 2011 stated that most pregnant women do not need to be screened for vitamin D deficiency, nor do they need to be given additional supplements.38 The authors also noted, “When vitamin D deficiency is identified during pregnancy, most experts agree that 1,000–2,000 IU/day of vitamin D is safe. Recommendations concerning routine vitamin D supplementation during pregnancy beyond that contained in a prenatal vitamin should await the completion of ongoing randomized clinical trials.”38

In light of the evidence available to date, there is an intriguing suggestion that vitamin D deficiency in pregnant women increases the risk for GDM. However, this determination is based largely on only six published observational studies and one short-term intervention study with an active analog form of 1,25(OH)2vitamin D; these seven studies investigated the association between vitamin D status and gestational diabetes or measures of glucose metabolism. The effect of treating pre-existing vitamin D deficiency on the subsequent development of GDM in pregnant women is unknown. This gap in knowledge is an important concern given the widespread occurrence of low vitamin D status in certain population groups and countries and the rising incidence of GDM and T2D in the world. Well-designed, randomized vitamin D supplementation trials are needed in pregnant women to determine optimal vitamin D status and dosing and to evaluate the potential effectiveness of vitamin D supplementation on the risk of developing GDM.


Declaration of interest

The authors have no relevant interests to declare.


View Abstract