Low vitamin B-12–high folate status in adolescents and pregnant women may have deleterious effects on health of the offspring

Dear Editor

We read with interest the paper by Bailey et al. (1) and the associated editorial by Molloy (2). They report a significant interaction between low vitamin B-12 and high folate status in relation to lower cognitive functioning in elderly participants of the NHANES cohort. There is a suggestion that unmetabolized folic acid might be related to such outcomes, raising a concern about dose of folic acid in fortified foods and supplements.

We wish to highlight the possible detrimental effects of maternal imbalance of these 2 vitamins (low vitamin B-12–high folate) on the health of the offspring, both in the short term and long term. Indians are predominantly vegetarian because of multigenerational cultural and socioeconomic influences. Low intake of animal-origin foods in a predominantly vegetarian population contributes to low vitamin B-12–high folate status. The national anemia-control program (now called Intensified-National Iron Plus Initiative) provides iron and folic acid but no vitamin B-12 to children, adolescents, and women of reproductive age. Obstetricians use large-dose folic acid supplements (5 mg) for the prevention of neural tube defects (NTDs) and other purported benefits (although the recommended dose for prevention of a first occurrence of NTDs is only 400 μg).

In the Pune Maternal Nutrition Study (PMNS), a preconceptional birth cohort established in 1993, two-thirds of pregnant mothers had vitamin B-12 deficiency (plasma vitamin B-12 <150 pmol/L) and 90% had elevated methylmalonic acid (>0.26 μM) (3). The mothers were folate replete (<1% had RBC folate <283 nmol/L) even before iron and folic acid were started. Higher frequencies of intake of green-leafy vegetables and fruits (both rich in folate) and maternal erythrocyte folate concentrations were associated with larger birth size of the offspring (4). However higher maternal folate concentrations in pregnancy were associated with higher adiposity in the offspring at 6 y of age (3). Offspring born to mothers with the lowest vitamin B-12 and highest folate concentrations had the highest insulin resistance at 6 y. In the Parthenon cohort from Mysore, maternal vitamin B-12 deficiency was associated with higher BMI, higher prevalence of gestational diabetes (GDM), and higher risk of permanent diabetes 5 y after delivery. Prevalence of GDM in vitamin B-12–deficient women progressively increased with higher folate concentrations (5). A hospital-based cohort study from Bangalore reported the highest risk of small-for-gestational-age infants in women who received high-dose folic acid supplements (> 1000 μg/d) but were in the lowest tertile of vitamin B-12 to folate intake (6).

In the Pune studies we found a positive association between maternal vitamin B-12 status and neurocognitive performance of the child at 2 and 9 y of age (7). Controlled trials of vitamin B-12 supplementation (50 μg/d) during pregnancy from Bangalore demonstrated an improvement in neurocognitive development in the offspring at 9 and 30 mo of age (8). Preliminary findings from a preconceptional controlled trial of low-dose vitamin B-12 supplementation (2 μg/d) to adolescents showed an improvement in neurocognitive performance in the offspring at 2 y of age (9). A trial in children aged 6–30 mo from North India showed that children who received vitamin B-12 (1.8 μg/d) and folic acid (150 μg/d) for 6 mo showed better neurocognitive development than those who received only vitamin B-12 or only folic acid or placebo (10).

Bailey et al. and Molloy rightly comment that, although there is a biological plausibility, the current observational evidence does not support causality. Randomized controlled trials may provide an answer, but careful ethical considerations will be essential, and it will take considerable time and effort. Instead, it will be possible to use genetic markers for vitamin B-12 and folate status for a Mendelian randomization study. Genetic determinants for vitamin B-12 and folate are well known and fairly similar in different populations. This technique has been used to report a possible causal association of maternal vitamin B-12 dietary intakes with offspring intelligence at 8 y of age (11). We used Mendelian randomization analysis to support a causal association between maternal homocysteine concentrations and fetal growth restriction (12). We have also reported an association between maternal holo-transcobalamin concentrations (but not folate) and NTDs in Indians. Causality was supported by the association between maternal transcobalamin 2 genotype (but not methylenetetrahydrofolate reductase) with NTDs (13). Such an approach might help us overcome the current uncertainty, as highlighted by Molloy.

Thus, observations from India and other countries suggest that a low vitamin B-12–high folate pattern in early life may adversely impact neurodevelopmental and metabolic-endocrine processes. These observations expand the scope for improving the health of the population across the life course rather than only in the elderly. Improving vitamin B-12 status in deficient populations, and avoiding inadvertent vitamin B-12–folate imbalance, should be an important consideration for clinicians and public health specialists.