Fetal growth restriction, most often estimated by incidence of a birth weight that is small for gestational age (SGA), is a major public health issue across the globe.1,2 Fetuses with growth restriction are at higher risk of death and serious neonatal morbidities,3 and alarmingly, health risks continue into adulthood.4 SGA is associated with a range of maternal factors including nutritional status, obesity, age, smoking, and infection, although there are few effective interventions for prevention.5
Vitamin D deficiency continues to be a public health issue in the United States, particularly among women of reproductive age, and deficiency rates have been increasing.6 Vitamin D is unique among essential micronutrients because it can be produced by the body subcutaneously after exposure to ultraviolet B radiation. Vitamin D receptors have been identified in tissues throughout the body allowing for a plethora of hormonal roles for the biologically active 1,25-hydroxyvitamin D. Maternal vitamin D deficiency is related to a range of poor pregnancy outcomes, including preterm birth, preeclampsia, and SGA.7 Vitamin D could potentially be related to fetal growth through calcium metabolism and bone growth8 or altering placental function.9,10
Several observational studies have linked maternal 25-hydroxyvitamin D concentrations and risk of SGA in general obstetric populations.11–15 Little is known about this association in high-risk pregnancies in which competing risk factors could augment or dampen the vitamin D–SGA relationship. As well, populations that are geographically and racially diverse are important in vitamin D research to observe a range of solar radiation exposure, cutaneous vitamin D production, and dietary intake. The objective of this study was to examine the association between maternal vitamin D status at 12–26 weeks of gestation and risk of SGA in a multicenter U.S. cohort of women at high risk for preeclampsia who delivered singleton live neonates.
MATERIALS AND METHODS
This was an observational study that used data and blood samples from the High-Risk Aspirin Study, a randomized controlled trial of low-dose aspirin for the prevention of preeclampsia. The trial was conducted in women at high risk for preeclampsia in 12 medical centers across the United States (1991–1995).16 Women who had prepregnancy, insulin-treated diabetes; chronic hypertension; preeclampsia in a previous pregnancy; or multifetal gestation were enrolled at 12–26 weeks of gestation. Further details of enrollment criteria are published.16 Exclusion criteria included planned delivery elsewhere, significant bleeding or bleeding disorders, aspirin allergy, current drug or alcohol abuse, renal failure, active hepatitis, uncontrollable hypertension, fetal anomalies incompatible with life, and fetal hydrops fetalis. After providing informed, written consent, women were randomized to receive 60 mg aspirin or placebo daily until delivery or the development of preeclampsia. The trial found no effect of aspirin on the incidence of preeclampsia, preterm birth, or SGA.16 Data were collected at enrollment on women's medical histories and sociodemographics. Nonfasting blood samples were taken before randomization (ie, before treatment) and neonates were weighed at birth.
Of 1,851 eligible women with singleton pregnancies enrolled, 839 had a serum sample at 26 weeks of gestation or less available for vitamin D assessment. We excluded 29 stillbirths, 10 pregnancies missing birth weight, and eight missing prepregnancy body mass index (BMI, calculated as weight (kg)/[height (m)]2) for a final analytic sample of 792 singleton live births. Our study used deidentified data and was approved by the University of Pittsburgh institutional review board.
Circulating serum 25-hydroxyvitamin D represents vitamin D from oral intake and cutaneous production and is considered the best biomarker of vitamin D status.17 In a DEQAS (Vitamin D External Quality Assessment Scheme)-proficient laboratory, total serum 25-hydroxyvitamin D (25-hydroxyvitamin D2+25-hydroxyvitamin D3) was measured using liquid chromatography–tandem mass spectrometry according to National Institute of Standards and Technology specifications.18 This method has a 2-ng/mL lower limit of detection and no upper limit. The intra-assay coefficients of variation were 8.2% and 5.9% for 25-hydroxyvitamin D2 and 25-hydroxyvitamin D3, respectively. Because there is no universally accepted definition of vitamin D deficiency during pregnancy,19,20 we examined 25-hydroxyvitamin D by multiple cut points from 25 to 80 nmol/L and used splines to investigate nonlinearity. We presented cut points of 30 and 50 nmol/L per the Institute of Medicine's definition of risk of vitamin D deficiency and inadequacy, respectfully19 and 75 nmol/L per the Endocrine Society's definition of insufficiency.20
We defined SGA as less than the 10th percentile of birth weight for gestational age based on a fetal standard.21 We chose the fetal standard because we had a high incidence of preterm birth (26%), and SGA is underestimated in preterm births when birth weight standards are used.22,23 Defining SGA based on a fetal standard has shown to better predict adverse outcomes.24–26 Gestational age was determined by ultrasonography or by date of last menstrual period in the absence of ultrasonography.
Women were interviewed at baseline and self-reported prepregnancy weight, parity, marital status, smoking habits, age, and total years of schooling. Women self-reported predominant race, which we classified as black (non-Hispanic black and Hispanic black) or white (non-Hispanic white, Hispanic white, and Native American or Alaskan). Height was measured and date of blood draw was recorded. Data were also available on the season of blood draw (winter [December to February], spring [March to May], summer [June to August], or fall [September to November]), latitude of study site, and neonatal sex.
The distribution of 25-hydroxyvitamin D was approximately normal by visual inspection with a Kernel density plot, and we used the t test to compare mean levels of 25-hydroxyvitamin D for SGA and non-SGA. We used log-binomial models to estimate risk ratios, risk differences, and their respective 95% confidence intervals (CIs) for the association between 25-hydroxyvitamin D and SGA. We assessed nonlinearity with restricted cubic splines and tested the spline terms with the Wald test. Biologic interaction was tested on the additive scale for prepregnancy BMI, race, gestational age at blood sample, season of blood sample, parity, and neonatal sex in full models with all potential confounders using the synergy index.27,28 Statistical interaction terms were included in the regression model when biological interaction was present. Potential confounders (race, prepregnancy BMI, height, gestational age at blood sampling, season at blood sampling, marital status, smoking, education, age, latitude, and neonatal sex) were identified using a theory-based causal graph.29 Confounding was considered present if variable removal from the full model changed the 25-hydroxyvitamin D and SGA association by greater than 10%. None of the potential confounders met this criterion. We retained BMI and race out of convention and included treatment group, risk group (prepregnancy diabetes; chronic hypertension; previous preeclampsia), and study latitude to account for the design of the high-risk aspirin trial.
In sensitivity analysis, we tested associations in conditional logistic regression models conditioned on study site, because the sample size did not provide adequate power to include 11 indicator variables for study site in final models.
There were no differences in maternal 25-hydroxyvitamin D levels, birth weight, or SGA incidence in women excluded (n=47) compared with included (n=792; data not shown). Most women in the study were non-Hispanic black, 20–29 years old, parous, not married, nonsmokers, and had a high school education (Table 1). In this high-risk population, only one third of women were of normal weight (BMI less than 25), whereas one fourth were severely obese (BMI 35 or greater). Thirteen percent of neonates were SGA at birth. A higher proportion of SGA neonates were born to smokers and women with less education (Table 1). Twenty-six percent of deliveries were less than 37 weeks of gestation. The mean (standard deviation) of maternal 25-hydroxyvitamin D at study entry (ie, before randomization) was 63.9 (29.5) nmol/L. Overall, 11.2%, 36.6%, and 67.9% of women had serum 25-hydroxyvitamin D concentrations of less than 30, less than 50, and less than 75 nmol/L, respectively.
In unadjusted analyses, mean (standard deviation) 25-hydroxyvitamin D concentrations were lower in women who delivered SGA neonates compared with women delivering non-SGA neonates (57.9 [29.9] compared with 64.8 [29.3] nmol/L, P=.028). We observed a nonlinear relationship between 25-hydroxyvitamin D and risk of SGA (P<.04; Fig. 1). After adjustment for race, pregravid BMI, latitude, treatment group, and risk group in restricted cubic spline models, the risk of SGA declined as 25-hydroxyvitamin D concentrations increased up to approximately 50 nmol/L and leveled off thereafter. In categorical analysis, 25-hydroxyvitamin D concentrations of 50–74 nmol/L and 75 nmol/L or greater compared with less than 30 nmol/L were associated with 43% and 54% reductions in risk of SGA and 8.4 and 10.7 fewer cases of SGA per 100 births, respectively (Table 2). Additional adjustment for gestational age at blood sampling, season at blood sampling, height, marital status, smoking, education, age, and neonatal sex did not meaningfully change the findings nor did use of conditional logistic models conditioned on site (data not shown).
The relationship of maternal vitamin D status and risk of SGA was modified by race and obesity status (Table 2). Prevalence of 25-hydroxyvitamin D less than 50 nmol/L differed by race (16% in white women compared with 49% in black women, P<.001) and obesity (30% in nonobese compared with 45% in obese, P<.001). We observed a different curvilinear relationship between 25-hydroxyvitamin D and risk of SGA in white compared with black mothers and nonobese compared with obese mothers. White women and nonobese women had higher risks of SGA with lower maternal 25-hydroxyvitamin D status (P<.01; Fig. 2). Furthermore, in adjusted models, white women and nonobese women with 25-hydroxyvitamin D 50 nmol/L or greater compared with less than 50 nmol/L had reduced risks of delivering a neonate who was SGA (Table 2). There was no association between 25-hydroxyvitamin D and risk of SGA in black or obese mothers. There was only moderate overlap of white mothers and nonobese mothers. Sixty-two percent (n=184) of white women and 53% (n=260) of black women were not obese. There were an insufficient number of SGA cases to stratify results into four race or obesity groups. Associations were not modified by gestational age at study entry, parity, or neonatal sex.
In this multicenter study of women at high risk for preeclampsia, we found that maternal second-trimester vitamin D status was inversely associated with risk of SGA in singleton pregnancies. The association seemed to be driven by white mothers and nonobese mothers. In both of these groups, 25-hydroxyvitamin D less than 50 nmol/L was associated with an increased risk of SGA after adjustment for confounders.
There are few randomized trials of maternal vitamin D supplementation, and we identified two that studied pathologic fetal growth.30,31 In 126 Asian women living in Britain, daily 1,000 international units in the third trimester compared with placebo reduced risk of SGA by 13%, although this difference was not statistically significant (0.05<P<.10).30 In another British population (n=179), no effect of daily 800 international units or a large single dose of 200,000 international units at 27 weeks of gestation compared with no treatment was observed.31 This trial was not placebo-controlled. A recent Cochrane review reported that there is limited evidence to assess the effect of vitamin D supplementation on SGA.32
In observational studies, maternal vitamin D deficiency has been associated with risk of SGA in several general obstetric populations across the United States and Europe.11–15 The overall association is similar to the magnitude of effect we observed in this high-risk population with 25-hydroxyvitamin D concentrations of 25–37.5 nmol/L appearing to be an important cutoff for increased risk. Two smaller studies, one in a general obstetric population33 and one in women at high risk for preeclampsia,34 found no association between SGA and vitamin D status but were likely underpowered with only 46 and 13 cases of SGA, respectively.
Our results of an association between low 25-hydroxyvitamin D and SGA in white mothers but no association among black mothers is consistent with one previous study.12 Two other studies examined an interaction by maternal race and reported no black–white differences.11,13 Although more research is needed to examine whether 25-hydroxyvitamin D has a different association with SGA by maternal race, it is possible that differences in genetic variability play a role.12,35 We did not have a large enough sample size to study the interaction between vitamin D and polymorphisms in genes related to its metabolism, but such studies would be fascinating. Some have speculated that the lower range of 25-hydroxyvitamin D values in black women may limit the ability to observe associations. The distribution of 25-hydroxyvitamin D was left-shifted in black compared with white women in this study, but the range was still wide, with approximately 50% of black women having values above 50 nmol/L. Thus, it does not seem that the range or distribution was a limitation in this case. Others have posed that blacks may not be affected by low vitamin D status or may have different adaptive responses compared with whites.36 For example, black women with vitamin D deficiency have a decreased osteoporosis risk and increase in bone mass compared with white women with vitamin D deficiency, indicating there may be a lower threshold for parathyroid hormone induced bone turnover in black women.36
We are unaware of any previous vitamin D fetal growth study to have assessed differences by prepregnancy BMI. However, our findings are consistent with a growing body of observational data suggesting that nutritional exposures may be protective against adverse pregnancy and birth outcomes such as SGA birth only among lean women.37 Obese mothers have altered metabolism and are at risk of multiple micronutrient deficiencies, including vitamin D deficiency.38 It is possible that higher doses of nutrients are needed in obese women to have effects similar to lean women.37
The biologic mechanisms that may connect maternal vitamin D status to fetal growth remain elusive. A plausible mechanism for the effect of maternal vitamin D on fetal growth is placental vascularization, which has received considerable attention in its association with fetal growth.39–41 Several observational studies have connected poor vitamin D status with higher risk of preeclampsia,7,42 which, like fetal growth restriction, has placental origins related to angiogenesis and uterine blood flow.39,43 Mice raised on vitamin D-deficient diets have placentas with narrower fetal vessels in the placental labyrinth compared with mice fed vitamin D-sufficient diets, indicating dysregulated vascularization.44 We have demonstrated an inverse relationship between maternal 25-hydroxyvitamin D and risk of placental vascular lesions in pregnancies with male fetuses10 and others have documented associations between vitamin D and biomarkers of angiogenesis.45–47 More basic science research is needed in this area as well as studies with multiple measurements of fetal growth and placental vascularization.
We did not have information on vitamin D dietary intake, supplement use, or sun exposure to inform the determinants of 25-hydroxyvitamin D in this population. Our study was limited by one measurement of 25-hydroxyvitamin D between 12 and 26 weeks of gestation. Because others have shown that exposure to vitamin D is differentially associated with SGA depending on trimester of sampling,11 measurements of 25-hydroxyvitamin D across gestation would have been ideal to allow us to study other critical windows outside this range. Although the choice of a weight standard to classify SGA is controversial, we chose a fetal weight standard to better capture SGA in preterm births because, in our study, more than one fourth of neonates were born at less than 37 weeks of gestation.
It is unclear how our findings in high-risk mothers translate to a general obstetric population. However, the vitamin D–SGA association we observed was similar to that reported in other general obstetric populations, which may demonstrate broad importance of maternal vitamin D status even in pregnancies with many competing risk factors for poor fetal growth. We were not able to study Hispanic women as a separate race or ethnicity group as a result of small sample size (n=54), notwithstanding our study benefits from a having a well-characterized and geographically and racially diverse population and use of gold standard methodology for assessing 25-hydroxyvitamin D.
Our cohort study adds to the body of evidence suggesting that maternal vitamin D status in pregnancy is related to fetal growth. Much of the supporting literature is based on observational studies, and only randomized controlled trials can determine whether this association is causal. Until data from trials are available, obstetricians should follow the American College of Obstetricians and Gynecologists guidelines for vitamin D screening and supplementation during pregnancy.38
1. de Onis M, Blössner M, Villar J. Levels and patterns of intrauterine growth retardation in developing countries. Eur J Clin Nutr 1998;52(suppl 1):S5–15.
2. Morisaki N, Esplin MS, Varner MW, Henry E, Oken E. Declines in birth weight and fetal growth independent of gestational length. Obstet Gynecol 2013;121:51–8.
3. McIntire DD, Bloom SL, Casey BM, Leveno KJ. Birth weight in relation to morbidity and mortality among newborn infants. N Engl J Med 1999;340:1234–8.
4. Gluckman PD, Hanson MA, Cooper C, Thornburg KL. Effect of in utero and early-life conditions on adult health and disease. N Engl J Med 2008;359:61–73.
5. Kramer MS. Intrauterine growth and gestational duration determinants. Pediatrics 1987;80:502–11.
6. Looker AC, Johnson CL, Lacher DA, Pfeiffer CM, Schleicher RL, Sempos CT. Vitamin D status: United States, 2001–2006. NCHS Data Brief 2011:1–8.
7. Aghajafari F, Nagulesapillai T, Ronksley PE, Tough SC, O'Beirne M, Rabi DM. Association between maternal serum 25-hydroxyvitamin D level and pregnancy and neonatal outcomes: systematic review and meta-analysis of observational studies. BMJ 2013;346:f1169.
8. Specker BL. Does vitamin D during pregnancy impact offspring growth and bone? Proc Nutr Soc 2012;71:38–45.
9. Thorne-Lyman A, Fawzi WW. Vitamin D during pregnancy and maternal, neonatal and infant health outcomes: a systematic review and meta-analysis. Paediatr Perinat Epidemiol 2012;(26 suppl 1):75–90.
10. Gernand AD, Bodnar LM, Klebanoff MA, Parks WT, Simhan HN. Maternal serum 25-hydroxyvitamin D and placental vascular pathology in a multicenter US cohort. Am J Clin Nutr 2013;98:383–8.
11. Gernand AD, Simhan HN, Klebanoff MA, Bodnar LM. Maternal serum 25-hydroxyvitamin D and measures of newborn and placental weight in a U.S. multicenter cohort study. J Clin Endocrinol Metab 2013;98:398–404.
12. Bodnar LM, Catov JM, Zmuda JM, Cooper ME, Parrott MS, Roberts JM, et al.. Maternal serum 25-hydroxyvitamin D concentrations are associated with small-for-gestational age births in white women. J Nutr 2010;140:999–1006.
13. Burris HH, Rifas-Shiman SL, Camargo CA Jr, Litonjua AA, Huh SY, Rich-Edwards JW, et al.. Plasma 25-hydroxyvitamin D during pregnancy and small-for-gestational age in black and white infants. Ann Epidemiol 2012;22:581–6.
14. Ertl R, Yu CK, Samaha R, Akolekar R, Nicolaides KH. Maternal serum vitamin D at 11–13 weeks in pregnancies delivering small for gestational age neonates. Fetal Diagn Ther 2012;31:103–8.
15. Leffelaar ER, Vrijkotte TG, van Eijsden M. Maternal early pregnancy vitamin D status in relation to fetal and neonatal growth: results of the multi-ethnic Amsterdam Born Children and their Development cohort. Br J Nutr 2010;104:108–17.
16. Caritis S, Sibai B, Hauth J, Lindheimer MD, Klebanoff M, Thom E, et al.. Low-dose aspirin to prevent preeclampsia in women at high risk. National Institute of Child Health and Human Development Network of Maternal-Fetal Medicine Units. N Engl J Med 1998;338:701–5.
17. Seamans KM, Cashman KD. Existing and potentially novel functional markers of vitamin D status: a systematic review. Am J Clin Nutr 2009;89:1997S–2008S.
18. Holick MF, Siris ES, Binkley N, Beard MK, Khan A, Katzer JT, et al.. Prevalence of Vitamin D inadequacy among postmenopausal North American women receiving osteoporosis therapy. J Clin Endocrinol Metab 2005;90:3215–24.
19. Institute of Medicine. Dietary reference intakes for calcium and vitamin D. Washington (DC): The National Academies Press; 2011.
20. Holick MF, Binkley NC, Bischoff-Ferrari HA, Gordon CM, Hanley DA, Heaney RP, et al.. Evaluation, treatment, and prevention of vitamin D deficiency: an Endocrine Society clinical practice guideline. J Clin Endocrinol Metab 2011;96:1911–30.
21. Hadlock FP, Harrist RB, Martinez-Poyer J. In utero analysis of fetal growth: a sonographic weight standard. Radiology 1991;181:129–33.
22. Ott WJ. Intrauterine growth retardation and preterm delivery. Am J Obstet Gynecol 1993;168:1710–5.
23. Hutcheon JA, Platt RW. The missing data problem in birth weight percentiles and thresholds for ‘small-for-gestational-age.’ Am J Epidemiol 2008;167:786–92.
24. Zaw W, Gagnon R, da Silva O. The risks of adverse neonatal outcome among preterm small for gestational age infants according to neonatal versus fetal growth standards. Pediatrics 2003;111:1273–7.
25. Lackman F, Capewell V, Richardson B, daSilva O, Gagnon R. The risks of spontaneous preterm delivery and perinatal mortality in relation to size at birth according to fetal versus neonatal growth standards. Am J Obstet Gynecol 2001;184:946–53.
26. Cooke RW. Conventional birth weight standards obscure fetal growth restriction in preterm infants. Arch Dis Child Fetal Neonatal Ed 2007;92:F189–92.
27. Skrondal A. Interaction as departure from additivity in case-control studies: a cautionary note. Am J Epidemiol 2003;158:251–8.
28. Rothman KJ. The estimation of synergy or antagonism. Am J Epidemiol 1976;103:506–11.
29. Greenland S, Pearl J, Robins JM. Causal diagrams for epidemiologic research. Epidemiology 1999;10:37–48.
30. Brooke OG, Brown IR, Bone CD, Carter ND, Cleeve HJ, Maxwell JD, et al.. Vitamin D supplements in pregnant Asian women: effects on calcium status and fetal growth. Br Med J 1980;280:751–4.
31. Yu CK, Sykes L, Sethi M, Teoh TG, Robinson S. Vitamin D deficiency and supplementation during pregnancy. Clin Endocrinol (Oxf) 2009;70:685–90.
32. De-Regil LM, Palacios C, Ansary A, Kulier R, Pena-Rosas JP. Vitamin D supplementation for women during pregnancy. The Cochrane Database of Systematic Reviews 2012, Issue 2. Art. No.: CD008873. DOI: 10.1002/14651858.CD008873.pub2.
33. Fernández-Alonso AM, Dionis-Sánchez EC, Chedraui P, González-Salmerón MD, Pérez-López FR; Spanish Vitamin D and Women’s Health Research Group. First-trimester maternal serum 25-hydroxyvitamin D(3) status and pregnancy outcome. Int J Gynaecol Obstet 2012;116:6–9.
34. Shand AW, Nassar N, Von Dadelszen P, Innis SM, Green TJ. Maternal vitamin D status in pregnancy and adverse pregnancy outcomes in a group at high risk for pre-eclampsia. BJOG 2010;117:1593–8.
35. Swamy GK, Garrett ME, Miranda ML, Ashley-Koch AE. Maternal vitamin D receptor genetic variation contributes to infant birthweight among black mothers. Am J Med Genet A 2011;155A:1264–71.
36. Cosman F, Nieves J, Dempster D, Lindsay R. Vitamin D economy in blacks. J Bone Miner Res 2007;22(suppl 2):V34–8.
37. Bodnar LM, Parrott MS. Intervention strategies to improve outcome in obese pregnancies: micronutrients and dietary supplements. In: Gillman MW, Poston L, editors. Maternal obesity. Cambridge (UK): Cambridge University Press; 2012.
38. Vitamin D: screening and supplementation during pregnancy. Committee Opinion No. 495. American College of Obstetricians and Gynecologists.Obstet Gynecol 2011;118:197–8.
39. Staff AC, Benton SJ, von Dadelszen P, Roberts JM, Taylor RN, Powers RW, et al.. Redefining preeclampsia using placenta-derived biomarkers. Hypertension 2013;61:932–42.
40. Zhang J, Merialdi M, Platt LD, Kramer MS. Defining normal and abnormal fetal growth: promises and challenges. Am J Obstet Gynecol 2010;202:522–8.
41. Turan S, Miller J, Baschat AA. Integrated testing and management in fetal growth restriction. Semin Perinatol 2008;32:194–200.
42. Christesen HT, Falkenberg T, Lamont RF, Jørgensen JS. The impact of vitamin D on pregnancy: a systematic review. Acta Obstet Gynecol Scand 2012;91:1357–67.
43. Bhattacharyya SK, Kundu S, Kabiraj SP. Prediction of preeclampsia by midtrimester uterine artery Doppler velocimetry in high-risk and low-risk women. J Obstet Gynaecol India 2012;62:297–300.
44. Liu NQ, Ouyang Y, Bulut Y, Lagishetty V, Chan SY, Hollis BW, et al.. Dietary vitamin D restriction in pregnant female mice is associated with maternal hypertension and altered placental and fetal development. Endocrinology 2013;154:2270–80.
45. Wei SQ, Audibert F, Luo ZC, Nuyt AM, Masse B, Julien P, et al.. Maternal plasma 25-hydroxyvitamin D levels, angiogenic factors, and preeclampsia. Am J Obstet Gynecol 2013;208:390.e1–6.
46. Grundmann M, Haidar M, Placzko S, Niendorf R, Darashchonak N, Hubel CA, et al.. Vitamin D improves the angiogenic properties of endothelial progenitor cells. Am J Physiol Cell Physiol 2012;303:C954–62.
© 2014 by The American College of Obstetricians and Gynecologists. Published by Wolters Kluwer Health, Inc. All rights reserved.
47. Levine MJ, Teegarden D. 1alpha,25-dihydroxycholecalciferol increases the expression of vascular endothelial growth factor in C3H10T1/2 mouse embryo fibroblasts. J Nutr 2004;134:2244–50.