Neural tube defects (NTD), manifesting as anencephaly or spina bifida, are to some degree preventable developmental anomalies. The risk of NTD can be reduced nearly 50% with periconceptional maternal exposure to folic acid tablet supplements1 or fortified flour, as we have seen in Canada.2 Despite these remarkable accomplishments, about 6 to 12 in every 10,000 fetuses in Canada have NTD.2,3
A systematic overview of previously published data suggested that a deficiency of vitamin B12 (B12, cobalamin) might also be associated with a higher risk of NTD.4 However, these studies are characterized by relatively few participants, lack of adjustment for maternal folate status, variability in the time of specimen collection during pregnancy, and relatively inaccurate measurement of bioavailable maternal B12.4 As a consequence, there is ongoing debate about whether B12 should be added to folic acid tablet supplements or fortified foods.5 We evaluated serum holotranscobalamin (holoTC)—the fraction of total circulating B12 bound to transcobalamin II. Since holoTC is the only form of B12 taken up by tissues, it should be a sensitive measure of maternal B12 status, as it may be related to NTD risk.6 B12 also appears to be the leading nutritional determinant of hyperhomocysteinemia within folic acid-fortified populations.7 We therefore studied whether this associated risk changed after the fortification of Canadian flour by folic acid.
We performed a population-based case–control study.2,8 Since 1993, under the universal Ontario Health Insurance Plan all pregnant women are offered standardized maternal serum screening at no financial cost. Maternal screening is made available at 15 to 20 weeks’ gestation through a physician or midwife; about 60% of pregnant women are screened. Self-reported maternal date of birth, gravidity, ethnicity, weight at the time of screening, and pregestational diabetes mellitus were recorded on the screening requisition sheet.
Women with a positive screen are referred for counseling at one of 17 genetics centers in Ontario. Each center supplies follow-up data to the Ontario Maternal Serum Screening Database. Open NTD cases are detected antenatally by ultrasonography or fetal autopsy, or postnatally through data linkage of the mother's insurance plan number with that of her infant during the delivery hospitalization, through the Canadian Institute for Health Information Discharge Abstract Database.2,8
Two large metropolitan genetics centers in Southern Ontario—the North York General Hospital (Toronto) and The Credit Valley Hospital (Mississauga)—handle about half of all serum specimens acquired at the time of screening and stored frozen. Blood samples were collected in a red top vacutainer or serum separator tube, and the serum transferred into a plastic transport tube, without a preservative, and then frozen at 70°C.
Cases were all women who had a pregnancy affected by a myelomeningocele or anencephaly, and whose screening database record corresponded to one of the 2 genetics centers. For each case, 5 maternal controls with a healthy pregnancy were randomly selected from these 2 centers within a plus-or-minus 12 months of the case specimen. Serum HoloTC was determined using a radioimmunoassay, according to the manufacturer's protocol (Axis-Shield, Ireland).9 Analytic sensitivity is about 5 pmol/L, with a between-assay imprecision of 8.5% at 32 pmol/L and 7.6% at 94 pmol/L. Serum folate was analyzed using the Bayer Centaur immunoassay analyzer method (Bayer Diagnostics Division, Toronto, ON). The distributions of the holoTC and folate values were positively skewed, so we log-transformed these measures and used inverse transformations to generate geometric means and standard deviations.
We compared the year of screening for cases and controls using a χ2 test for trend. The mean serum folate and mean holoTC concentrations among cases and controls were compared using an unpaired t test. An abnormally low holoTC was defined, a priori, as a concentration less than or equal to the bottom quartile value in the control group. For the main analysis, we calculated a crude odds ratio (OR) and 95% confidence interval (CI). We used logistic regression to estimate the adjusted OR, with the highest quartile concentration of holoTC as the referent. We adjusted for maternal age (1-year increments), gravidity, race (white versus nonwhite), weight, low-income status, presence of pregestational diabetes mellitus and serum folate concentration (as a continuous variable) at the time of screening. Low-income status was based on data from the 2001 Canadian Census.10 This method uses the first 3 digits of each woman's 6-digit postal code to designate a geographical region, and then determines the prevalence of households in that same region who spend 55% or more of their income on food, shelter, and clothing.10 A secondary analysis, limited to the period after March 1, 1997, when folic acid food fortification was begun,11 used the same multivariate model as above. Finally, we estimated a population attributable risk percent of NTD in relation to low holoTC, using the adjusted OR.12,13
All variables were included in the model a priori. The research protocol was approved by the Ministry of Health and Longterm Care in Ontario, as well as the Research Ethics Boards of Sunnybrook Health Sciences Centre, St. Michael's Hospital and The North York General Hospital, with participant identifiers removed from the data set prior to analysis.
There were 317 cases of NTD in Ontario during the period of study. Of these, 89 cases originated from the North York General and Credit Valley Hospitals. Of the 434 unaffected pregnant controls selected from these 2 centers, 12 were excluded due to an insufficient volume of serum, leaving 422 controls (Table 1). About half (n = 248) of all participants were enrolled after fortification of flour had begun. Of 89 NTD cases, 67 (75%) were diagnosed antenatally. No difference was observed between cases and controls in the year of specimen collection (χ2 test for trend: P = 0.99) (Table 1). The mean concentration of serum folate did not differ appreciably between cases and controls across the entire study period (13.3 versus 13.9 nmol/L; P = 0.07).
The overall geometric mean serum holoTC level was substantially lower among cases (67.8 pmol/L) than controls (81.2 pmol/L), a mean of difference of 13.4 pmol/L (95% CI = 13.0–13.8) (Table 1). Comparing the lowest with the highest quartile of maternal holoTC concentration, the crude OR for NTD was 2.0 (95% CI = 1.1–3.9), increasing to 2.9 (1.2–6.9) in the adjusted model (Table 2). The corresponding population-attributable risk for NTD in relation to low holoTC was 34%.
In the period before fortification, the mean ± SD serum folate concentrations were 8.8 ± 3.5 nmol/L among the cases and 10.3 ± 3. nmol/L in the controls. After fortification, serum folate concentrations rose to 20.9 ± 2.0 nmol/L for cases and 19.4 ± 2.2 nmol/L for controls. Similarly, the levels of holoTC were 58.5 ± 1.9 pmol/L in cases and 72.7 ± 1.8 pmol/L among the controls and after fortification 78.8 ± 2.0 pmol/L in cases and 91.3 ± 1.8 pmol/L in controls. Restricting the analyses to the time after food fortification with folic acid, the association of low holoTC with risk of NTD remained strong (adjusted OR = 3.2; CI = 0.94–11.0).
Using a sensitive and specific measure of bioavailable cobalamin, we observed nearly a tripling in the risk of NTD in the presence of low maternal B12 status. In a Canadian setting of moderate folic acid tablet supplement use14 and universal folic acid flour fortification,2 these data suggest that about 34% of NTD may be due to low B12.
Our study has some limitations. The collection of maternal specimens about 15 weeks after conception would be expected to dilute any true relationship between B12 insufficiency and NTD risk.15 We do not know how many cases or controls were taking folic acid or B12-containing tablet supplements periconceptionally (ie, before and after conception), nor could we assess a “contamination” effect on measured holoTC concentration from B12 supplement use started after conception. We adjusted for serum folate concentration at the time of screening—which differed little between cases and controls—as well as maternal age and socioeconomic status, each predictors of early access to antenatal care and periconceptional vitamin supplement use.14 There were more nonwhite women among cases than controls, and we adjusted for nonwhite ethnicity. This heterogeneous group of women (mainly Asian and black) may differ from whites in terms of dietary intake and metabolism of B12.16
Surveys suggest that the majority of women in Ontario take a periconceptional tablet supplement containing folic acid,14,17 in keeping with Canadian guidelines.18 The small quantity of B12 contained in each prenatal multivitamin tablet in Canada (2.5 μg) is unlikely to explain the observed difference in holoTC levels (for example, if a greater number of controls than cases had initiated multivitamin tablets soon after conception).
The concentration of holoTC defining the bottom quartile among our pregnant controls (55 pmol/L) was comparable to the lower-quartile values described in other studies of pregnant women19 and healthy adults.9,20 A lower reference limit of 50 pmol/L is recommended for a diagnosis of B12 deficiency in young and middle-aged adults.21 Together, this reinforces the robustness of the holoTC threshold that we used to define “abnormal,” and the comparability of our controls to other study populations of young adults.
Though not a central focus of our study, we were surprised that the concentration of serum folate was only slightly lower among the cases than controls. This may have been due to the dilutional effect of pregnancy on folate levels, a similar prevalence of use of folic acid tablet supplements among cases and controls around the time of maternal screening, or perhaps the fact that nearly half of all study participants were screened after universal folic acid flour fortification. Other Canadian studies have found that serum folate levels have increased by more than 60% after the introduction of flour fortification, with 99% of women now replete.11
Previous studies have directly or indirectly assessed an association between low maternal B12 and NTD risk (Table 3). One small case–control study has observed an association of low holoTC with risk of NTD (crude OR = 2.9; 95% CI = 0.9–9.2).19 Another study,28 comprising 33 cases and 132 controls, found an adjusted OR for NTD of 13 (95% CI = 2.7–66) using another sensitive functional indicator of B12 deficiency (maternal serum methylmalonic acid concentration above the 90th centile).29 The combination of a low B12 and a genetic polymorphism MTRR, which makes of methionine synthase reductase, (an enzyme that activates cobalamin-dependent methionine synthase), was associated with an OR for NTD of 4.8 (95% CI = 1.5–16).27 In general, while there appears to be a consistent association between maternal B12 insufficiency and NTD risk, few studies have examined maternal B12 status using methods of sufficient sensitivity, or controlled for influential maternal risk factors such as folate status (Table 3).
Our findings suggest that as much as 34% of all NTDs in Canada may be attributable to low maternal B12 status, a conclusion which could have profound public health implications, if generalizable. However, only about 60% of pregnant women in Ontario have maternal screening, which limits the generalizability of our data. Women who do not access prenatal screening may differ from those who do,30 with lower rates of pregnancy planning and periconceptional folic acid tablet supplement use.14
B12 and folic acid are the main nutritional determinants of homocysteine metabolism.7 While folic acid supplementation may lower both plasma homocysteine and the risk of NTD, the impact of higher B12 intake on NTD risk is not known. Given our results and those of other studies, a randomized clinical trial of periconceptional B12 may be indicated. Adding B12 to folic acid in food fortification may help prevent NTD, while at the same time reducing concern about masking B12-related neurologic disease, which can occur when fortifying with folic acid alone.31,32
1. Lumley J, Watson L, Watson M, et al. Periconceptional supplementation with folate and/or multivitamins for preventing neural tube defects. Cochrane Database Syst Rev
2. Ray JG, Meier C, Vermeulen MJ, et al. Association of neural tube defects and folic acid food fortification in Canada. Lancet
3. De Wals P, Rusen ID, Lee NS, et al. Trend in prevalence of neural tube defects in Quebec. Birth Defects Res A Clin Mol Teratol
4. Ray JG, Blom HJ. Vitamin B12
insufficiency and the risk of fetal neural tube defects. QJM
5. Finglas PM, de Meer K, Molloy A, et al. Research goals for folate and related B vitamin in Europe. Eur J Clin Nutr
6. Hvas AM, Nexo E. Holotranscobalamin—a first choice assay for diagnosing early vitamin B deficiency? J Intern Med
7. Green R, Miller JW. Vitamin B12
deficiency is the dominant nutritional cause of hyperhomocysteinemia in a folic acid-fortified population. Clin Chem Lab Med
8. Ray JG, Wyatt PR, Vermeulen MJ, et al. Greater maternal weight and the ongoing risk of neural tube defects after folic acid flour fortification. Obstet Gynecol
9. Refsum H, Johnston C, Guttormsen AB, et al. Holotranscobalamin and total transcobalamin in human plasma: determination, determinants, and reference values in healthy adults. Clin Chem
10. Statistics Canada. Income in Canada 2000
. Ottawa: Statistics Canada; Catalogue No. 75–202 XIE, November 2002.
11. Ray JG, Vermeulen MJ, Langman LJ, et al. Persistence of vitamin B12
insufficiency among elderly women after folic acid food fortification. Clin Biochem
12. Rockhill B, Newman B, Weinberg C. Use and misuse of population attributable fractions. Am J Public Health
13. Heller RF, Buchan I, Edwards R, et al. Communicating risks at the population level: application of population impact numbers. BMJ
14. Ray JG, Singh G, Burrows RF. Evidence for suboptimal use of periconceptional folic acid supplements globally. BJOG
15. Metz J, McGrath K, Bennett M, et al. Biochemical indices of vitamin B12
nutrition in pregnant patients with subnormal serum vitamin B12
levels. Am J Hematol
16. Saxena S, Carmel R. Racial differences in vitamin B12
levels in the United States. Am J Clin Pathol
17. Tam LE, McDonald SD, Wen SW, et al. A survey of preconceptional folic acid use in a group of Canadian women. J Obstet Gynaecol Can
18. Van Allen MI, Fraser FC, Dallaire L, et al. Recommendations on the use of folic acid supplementation to prevent the recurrence of neural tube defects. Clinical Teratology Committee, Canadian College of Medical Geneticists. CMAJ
19. Afman LA, Van Der Put NM, Thomas CM, et al. Reduced vitamin B12
binding by transcobalamin II increases the risk of neural tube defects. QJM
20. Loikas S, Lopponen M, Suominen P, et al. RIA for serum holo-transcobalamin: method evaluation in the clinical laboratory and reference interval. Clin Chem
21. Lloyd-Wright Z, Hvas AM, Moller J, et al. Holotranscobalamin as an indicator of dietary vitamin B12
deficiency. Clin Chem
22. Kirke PN, Molloy AM, Daly LE, et al. Maternal plasma folate and vitamin B12
are independent risk factors for neural tube defects. QJM
23. Molloy AM, Kirke P, Hillary I, et al. Maternal serum folate and vitamin B12
concentrations in pregnancies associated with neural tube defects. Arch Dis Child
24. van der Put NM, Thomas CM, Eskes TK, et al. Altered folate and vitamin B12
metabolism in families with spina bifida offspring. QJM
25. Groenen PM, van Rooij IA, Peer PG, et al. Marginal maternal vitamin B12
status increases the risk of offspring with spina bifida. Am J Obstet Gynecol
26. Suarez L, Hendricks K, Felkner M, et al. Maternal serum B12
levels and risk for neural tube defects in a Texas-Mexico border population. Ann Epidemiol
27. Wilson A, Platt R, Wu Q, et al. A common variant in methionine synthase reductase combined with low cobalamin (vitamin B12
) increases risk for spina bifida. Mol Genet Metab
28. Adams MJ Jr, Khoury MJ, Scanlon KS, et al. Elevated midtrimester serum methylmalonic acid levels as a risk factor for neural tube defects. Teratology
29. McMullin MF, Young PB, Bailie KE, et al. Homocysteine and methylmalonic acid as indicators of folate and vitamin B12
deficiency in pregnancy. Clin Lab Haematol
30. Kuppermann M, Learman LA, Gates E, et al. Beyond race or ethnicity and socioeconomic status: predictors of prenatal testing for Down syndrome. Obstet Gynecol
31. Czernichow S, Noisette N, Blacher J, et al. Case for folic acid and vitamin B12
fortification in Europe. Semin Vasc Med
32. Eichholzer M, Tonz O, Zimmermann R. Folic acid: a public-health challenge. Lancet