AJCN EB Program 2010 Early Registration
HOME HELP FEEDBACK SUBSCRIPTIONS ARCHIVE SEARCH TABLE OF CONTENTS
QUICK SEARCH:   [advanced]


     

This Article
Right arrow Abstract Freely available
Right arrow Full Text (PDF)
Right arrow Supplemental data
Right arrow Purchase Article
Right arrow View Shopping Cart
Right arrow Alert me when this article is cited
Right arrow Alert me if a correction is posted
Right arrow Citation Map
Services
Right arrow Similar articles in this journal
Right arrow Similar articles in PubMed
Right arrow Alert me to new issues of the journal
Right arrow Download to citation manager
Right arrow reprints & permissions
Citing Articles
Right arrow Citing Articles via Google Scholar
Google Scholar
Right arrow Articles by Hay, G.
Right arrow Articles by Refsum, H.
Right arrow Search for Related Content
PubMed
Right arrow PubMed Citation
Right arrow Articles by Hay, G.
Right arrow Articles by Refsum, H.
Agricola
Right arrow Articles by Hay, G.
Right arrow Articles by Refsum, H.
American Journal of Clinical Nutrition, Vol. 88, No. 1, 105-114, July 2008
© 2008 American Society for Nutrition

ORIGINAL RESEARCH COMMUNICATION

Folate and cobalamin status in relation to breastfeeding and weaning in healthy infants1,2,3

Gry Hay1, Carole Johnston1, Andrew Whitelaw1, Kerstin Trygg1 and Helga Refsum1

1 From the Department of Nutrition, Institute of Basic Medical Sciences, University of Oslo, Oslo, Norway (GH, KT, and HR); the Department of Physiology, Anatomy, and Genetics, University of Oxford, Oxford, United Kingdom (CJ); and the Department of Clinical Science, University of Bristol, Bristol, United Kingdom (AW)

2 Supported by The Throne Holst Foundation for Nutrition Research, The Norwegian Meat Marketing Board, TINE Norwegian Dairies BA, Axellus AS, Mills DA, Freia Chocolate Medical Foundation, Norwegian Women's Public Health Association, Anders Jahre's Fund for the Promotion of Science, The Eckbo Foundation, SCA Hygiene Products AS, and AstraZeneca AS.

3 Reprints not available. Address correspondence to G Hay, Department of Nutrition, University of Oslo, PO Box 1046 Blindern, N-0316 Oslo, Norway. E-mail: gry.hay{at}medisin.uio.no.


   ABSTRACT
TOP
 ABSTRACT
INTRODUCTION
 SUBJECTS AND METHODS
 RESULTS
 DISCUSSION
 REFERENCES
 
Background: Folate and cobalamin status changes markedly during infancy.

Objective: We aimed to examine the influence of breastfeeding on folate and cobalamin status in healthy infants.

Design: In a longitudinal study, we measured serum folate, cobalamin, holotranscobalamin, holohaptocorrin, methylmalonic acid, and homocysteine at birth and at ages 6, 12, and 24 mo (n = 361, 262, 244, and 224, respectively). Breastfeeding status and nutrient intake were assessed by using questionnaires and 7-d weighed-food records (at 12 mo).

Results: All indexes changed significantly from birth to age 24 mo (P < 0.001). Folate was high until age 6 mo and then declined. At age 6 mo, folate was positively correlated with duration of exclusive breastfeeding ({rho} = 0.29; P < 0.001). Cobalamin status declined after birth in breastfed but increased in nonbreastfed infants. Thus, holotranscobalamin (pmol/L) was lower in breastfed than in nonbreastfed children at age 6 mo [geometric x: 37 (95% CI: 33, 40) and 74 (64, 86), respectively], at 12 mo [51 (46, 56) and 76 (70, 82), respectively], and at 24 mo [65 (50, 83) and 90 (85, 97), respectively; P < 0.05 for all]. Complementary feeding did not increase (6 mo) or modestly increased (12 mo) cobalamin status in breastfed children. At 12 mo, cobalamin intake (µg/d), excluding breast milk cobalamin, was lower in breastfed than in nonbreastfed infants [geometric x: 1.4 (1.3, 1.6) and 2.4 (2.1, 2.6), respectively; P < 0.001]. However, after adjustment for total cobalamin intake, cobalamin status (ie, holotranscobalamin) remained significantly lower in breastfed than in nonbreastfed infants [54 (49, 59) and 70 (64, 78), respectively; P < 0.001].

Conclusions: Low cobalamin status is a characteristic finding in breastfed children. Reference limits according to age and breastfeeding status should be considered in early childhood.


   INTRODUCTION
TOP
 ABSTRACT
 INTRODUCTION
SUBJECTS AND METHODS
 RESULTS
 DISCUSSION
 REFERENCES
 
Folate and cobalamin are necessary cofactors in the synthesis of RNA and DNA, and cobalamin is necessary for maintaining the nervous system. These micronutrients are therefore critical to the rapid growth and development of the early years of life (1). Poor folate status has recently been shown to be an independent risk factor for respiratory infections in young children (2), and neonatal cobalamin deficiency causes failure to thrive, hematologic abnormalities, and neurologic manifestations, which may become irreversible (3, 4).

In the developed world, folate and cobalamin deficiencies are rare in infants. Breast milk and breast milk substitutes are usually high in folate and thereby protect the infant against folate deficiency. The amount of cobalamin in breast milk is strongly affected by maternal status or intake (5), and neonatal cobalamin deficiency is usually observed only in exclusively breastfed infants of vegetarian mothers (3, 4). Several studies reported cobalamin and folate status in newborns and during infancy (2, 6-12). Most of these studies showed that infants have lower serum cobalamin and higher methylmalonic acid (MMA) than do older children and adults, and concern has been expressed that cobalamin deficiency is common in otherwise healthy infants of nonvegetarian mothers (3, 13). However, reference limits for indexes of cobalamin and folate status in infants are lacking, and it is therefore difficult, if not impossible, to interpret the results.

There is no gold standard for assessment of folate and cobalamin status. Hematologic variables such as hemoglobin and mean corpuscular volume (MCV) are affected at a late stage of deficiency (8). Conventionally, low folate and cobalamin status is detected by a combination of various indexes, including a finding of low serum concentrations of folate or cobalamin, combined with elevated concentrations of functional markers such as total homocysteine (tHcy) (in both folate and cobalamin deficiency) and MMA (in cobalamin deficiency) (8). Total cobalamin in serum is bound to 2 main transport proteins, haptocorrin and transcobalamin (14). Cobalamin bound to transcobalamin, holotranscobalamin (holoTC), is the biologically active cobalamin in serum and is available to all cells. In contrast, holohaptocorrin (holoHC) has no known function (15). Increasingly, holoTC is used to assess cobalamin status, and it is a more sensitive indicator of change in cobalamin intake and absorption than is serum cobalamin (14).

Most previous studies on folate and cobalamin status in infants have been confined to newborns (8, 10, 11, 16, 17). In studies with data from later infancy, the effect of breastfeeding has often not been reported (7), or the studies have included a limited number of folate and cobalamin indexes (6, 9, 12, 18). With only one exception—a study in newborn infants (8)—none of the studies have reported reference limits for these indexes during infancy. The objectives of the current study were to determine how breastfeeding and weaning affect folate and cobalamin status and to report reference limits for healthy children, according to age and breastfeeding status.


   SUBJECTS AND METHODS
TOP
 ABSTRACT
 INTRODUCTION
 SUBJECTS AND METHODS
RESULTS
 DISCUSSION
 REFERENCES
 
Subjects
This longitudinal study followed Norwegian children from birth until age 2 y (19, 20). The study was designed to examine nutrient status in healthy children. Invitations were sent to 471 pregnant women of Norwegian or other Nordic descent who were scheduled for delivery at Aker University Hospital (Oslo, Norway) between April and June 1997. A total of 364 mothers fulfilled the following criteria: singleton birth, gestational period of 37–43 wk, and birth weight higher than the 2.5th percentile [2600 and 2700 g for girls and boys, respectively, according to Norwegian growth charts (21)]. Information on the newborn child's birth weight, length, head circumference, and sex was collected from the child's medical record. The subject families were invited to participate again when the children were 6, 9, 12, 18, and 24 mo old; blood sampling was done at ages 6, 12, and 24 mo. A total of 197 children (54% of those included at birth) participated in all 4 blood sampling sessions. Dietary assessment was performed at ages 6, 9, 12, 18, and 24 mo. At all infant ages, the parents completed a questionnaire about the child's food intake. At age 6 mo, the questionnaire included information about breastfeeding status at that time; the duration of exclusive breastfeeding; the infant's age when formula, complementary foods, or both were introduced; the type and amount of formula and solids; and the length of time that the various food items were used. At age 12 mo, we used a semiquantitative questionnaire from a nationwide dietary survey among 12-mo-old Norwegian infants (22). At 9 and 18 mo, a simple questionnaire was administered, including questions about breastfeeding. These questionnaires were in a few cases used to give information about breastfeeding that was lacking at age 6 or 12 mo. It was assumed that infants breastfed at ages 9 and 18 mo were also breastfed at ages 6 and 12 mo, respectively. Breastfed infants were defined as infants who received breast milk alone or in combination with complementary foods (breast milk substitutes, solids, or both). Breastfed infants were further categorized into exclusively breastfed (breast milk only) and partially breastfed (complementary foods in addition to breast milk) infants. A proportion of the infants in all 3 groups also received vitamin supplements.

At 12 and 24 mo, a 7-d weighed food record was completed. Only the data from the 12-mo record were used in this study. The average daily intakes of cobalamin and energy were computed by using a food database and software system developed at the Department of Nutrition, University of Oslo, Norway. The food database is mainly based on the official food composition table of Norway (23), and it is continuously supplemented with data on new food items and nutrients. Corrections for losses in cooking are made when nutrient contents were calculated. The use of cod liver oil and vitamin and mineral supplements is included in the nutrient calculation. The content of cobalamin in one brand of infant dinner was obtained from the manufacturer (Nestlé Norway, Sandvika, Norway). The content was calculated by the manufacturer with the use of Nordic food composition tables with an estimated 10% loss of cobalamin due to preparation. On the basis of 12 different variants, the average cobalamin content in an infant dinner with meat was estimated to be 0.13 µg/100 g. The only infant dinner with fish had a cobalamin content of 0.23 µg/100 g.

After each visit, the families were provided with free diapers and other infant products but no food products. After each blood sampling, the mother was informed of her child's iron status.

Written informed consent was obtained from each child's parent or parents. The study was approved by the Regional Committee for Research Ethics and the Norwegian Data Directorate.

Use of formulas, cow milk, and vitamin supplements in infants 6–12 mo old
In Norway, there is no fortification of flour or other food items except baby foods. At the time of the present study, there was no official recommendation on periconceptional folic acid supplementation (instituted in 1998). The 2 commercial formulas used by the children in the present study (NAN; Nestlé; and Collett; Nycomed, Pharma AS, Asker, Norway) contained 50 µg folic acid and 1.0–1.1 µg cyanocobalamin/100 g powder, which corresponded to 6.3 µg folic acid and 0.13–0.15 µg cyanocobalamin/100 mL ready-to-drink formula. The nutrient content was based on information from the manufacturers. Typical content per 100 g cow milk is 0.4 µg cobalamin and 4–5 µg folate (23). Only 3 (6-mo-old) and 4 (12-mo-old) children received other types of breast milk substitutes. There was a general recommendation that all infants were given a vitamin D supplement, preferably in the form of cod liver oil, from the age of 4 wk. The supplements used included cod liver oil (Möller's tran; Axellus AS, Lysaker, Norway), Spedbarnsvitaminer (now called Nycoplus MultiBarn vitaminmikstur; Nycomed Pharma AS), Sanasol (Axellus AS), AD-dråper (bought in Sweden; manufacturer not specified), Floradix formula (Salus Haus, Bruckmuhl/Obb, Germany), and Apovit D (now called Nycoplus D-vitamin droplets; Nycomed Pharma AS). Among these, only Floradix formula, which was used by 2 infants at age 12 mo, contained folic acid (50 µg/5 mL, which is equal to the daily recommended dose). Supplements containing cobalamin were used by 71 and 85 infants at age 6 and 12 mo, respectively, and included Sanasol (2.5 µg/10 mL), Spedbarnsvitaminer (1.0 µg/5 mL), and Floradix formula (0.3 µg/5 mL). Until 2003, cobalamin was not added to instant infant cereals. The content of folate, including added folic acid, was 40–60 µg/100 g powder in cereals that are to be reconstituted with water.

Blood sampling and blood analyses
The procedures for blood sampling and the handling of the samples were reported previously (20). Serum was obtained from cord blood at birth and from venous blood samples at ages 6, 12, and 24 mo. The serum fraction was kept cold and transported on ice; aliquots were stored (without additives) at –70 °C for an average of 8 y before analysis of folate and cobalamin indexes. The original number of blood samples was 364 at birth (cord samples), 287 at age 6 mo, 249 at age 12 mo, and 231 at age 24 mo. After analyses of other nutrients (eg, iron and fat-soluble vitamins), serum samples were available for B vitamin analyses from 361, 262, 244, and 224 infants at ages 0, 6, 12, and 24 mo, respectively. Furthermore, because of limited sample volume, the number of available results varied for each marker, being lowest for holoTC and holoHC. The numbers of measurements are therefore given together with the results in each case.

Serum concentrations of cobalamin and folate were measured by using microbiological assays with a colistin sulfate–resistant strain of Lactobacillus leichmanii (24, 25) and L. casei (26), respectively. Serum folate was measured in samples that previously had not been thawed. Serum holoTC concentrations were measured by using magnetic beads (microspheres) with immobilized monoclonal antibody specific for human transcobalamin to isolate transcobalamin, followed by a microbiological assay for cobalamin (27). HoloHC—ie, cobalamin bound to haptocorrin—was measured by subtracting holoTC from cobalamin. The CV for serum folate and total serum cobalamin in the laboratory is 5%. The CV for holoTC is 5–8% (27). Serum MMA and tHcy were analyzed by using a modified gas chromatography–mass spectrometry method based on ethylchloroformate derivatization (28). The CV for these measurements is < 5%.

Statistical analyses
Because of the skewed distribution of the serum vitamin indexes, the results are presented as geometric means (and 95% CIs). For energy intake and intake of food items and estimates of folate and cobalamin intakes from the various food groups, medians and interquartile ranges are presented. In addition, total energy intake and total intake of cobalamin and folate are presented as geometric mean (and 95% CIs) to allow for adjustment for relevant factors.

Reference limits are presented as 5th–95th percentiles. Student's t test, Mann-Whitney U test, and analysis of variance (ANOVA) were used to compare independent groups. Paired-sample t test or repeated-measures ANOVA was used for comparison of indexes in the different age groups. Adjustment for multiple comparisons was performed by using the Bonferroni correction. Bivariate correlations were examined by using the Spearman rank-order correlation test. Linear regression analyses and ANOVA were used to estimate the relative influence of various factors on folate and cobalamin indexes, with control for potential confounders. To test for interaction between breastfeeding status and age, repeated-measures ANOVA was carried out in the group of children who did not change breastfeeding status between age 6 and 12 mo. P < 0.1 for interactions was considered significant. Statistical analyses were performed by using SPSS for WINDOWS software (version 14.02; SPSS Institute, Chicago, IL). P < 0.05 was considered significant.


   RESULTS
TOP
 ABSTRACT
 INTRODUCTION
 SUBJECTS AND METHODS
 RESULTS
DISCUSSION
 REFERENCES
 
Characteristics
The characteristics of the mothers and newborns were reported previously (20). Mean birth weight was 3673 ± 455 g, and there was no significant difference between boys and girls. The mean age of the 364 mothers was 29.9 ± 4.4 y; 3 were teenagers (18–19 y old), and 5 were 40–42 y old. Seventeen percent of the mothers were daily smokers. Mean first-trimester body mass index (in kg/m2) was 23.5. Seventy-two percent of the mothers had ≥12 y of education, and 49.5% were primipara.

Values at birth compared with values at ages 6, 12, and 24 mo in the total group
Indexes of folate and cobalamin status, from birth until 24 mo of age, are shown in Figure 1Go. For all of the indexes, there were significant changes from birth to age 24 mo (P < 0.001). Serum folate was high at birth and at age 6 mo, but thereafter it declined and reached concentrations of <50% of the birth values at age 24 mo. Total serum cobalamin and holoTC decreased from birth to age 6 mo and then gradually increased; birth concentrations were reached (holoTC) or surpassed (cobalamin) at age 24 mo. The decline in holoTC was much more pronounced than that in total cobalamin, and further investigation showed that cobalamin bound to haptocorrin (holoHC) increased modestly but significantly from birth to age 6 mo and increased further by age 12 and 24 mo. The changes in tHcy and MMA mirrored the changes observed for cobalamin and holoTC—ie, they occurred in the opposite direction. Relative to values at birth, tHcy and in particular MMA were significantly higher at 6 mo, but then they decreased to below the birth values by age 12 and 24 mo.


Figure 1
View larger version (11K):
[in this window]
[in a new window]

  		FIGURE 1.. Geometric mean (95% CI) of folate and cobalamin indexes in serum from birth to age 24 mo in the total group. The number of available results at 0, 6, 12, and 24 mo are the following: Cobalamin, n = 359, 262, 243, 223; holohaptocorrin (holoHC), n = 325, 186, 223, 219; holotranscobalamin (holoTC), n = 328, 186, 223, 219; folate, n = 360, 258, 242, 223; total homocysteine (tHcy), n = 361, 254, 243, 224; methylmalonic acid (MMA), n = 361, 253, 242, 222. Significance was first tested by repeated-measures ANOVA for the period of 0–24 mo (P < 0.001 for all indexes), followed by pairwise comparison with Bonferroni correction between the concentrations at 6, 12, and 24 mo and the concentrations at birth. At age 6 mo, the concentrations of all indexes, except holoHC, were significantly different from the concentrations at birth (P < 0.001). At age 12 mo, all indexes were significantly different from those at birth (P < 0.003). At age 24 mo, all indexes except holoTC were significantly different from those at birth (P < 0.001).

 
Breastfed compared with nonbreastfed children
Both serum concentrations and dietary information (from questionnaires) were available for 262, 243, and 215 children at age 6, 12, and 24 mo, respectively. Of these children, 221 (84%), 115 (47%), and 15 (7%) were breastfed at those respective ages. For 2 children at 6 mo and 2 different children at 12 mo, breastfeeding status was obtained from a questionnaire at a later age, and more detailed dietary information for these 4 children is lacking. Only 4 children were never breastfed. At age 6 mo, 11 infants were exclusively breastfed, and the rest of the infants received both milk and solids in combination. All nonbreastfed infants received milk formula; cow milk was consumed by 8 children (in combination with breast milk or formula). At age 12 mo, all children received solids, 31% received formula, and 68% received cow milk as drinking milk (alone or in combination with other milk). For simplicity, formula and unmodified cow milk are referred to as breast milk substitutes.

Serum folate did not differ significantly between breastfed and nonbreastfed children at age 6 mo [56 (95% CI: 53, 60) and 52 (45, 59) nmol/L, respectively], at age 12 mo [34 (32, 37) and 31 (28, 34) nmol/L, respectively], or at age 24 mo [24 (19, 29) and 19 (18, 20) nmol/L, respectively]. Infants who were breastfed at age 6 mo had significantly lower cobalamin, holoHC, and holoTC and higher MMA and tHcy concentrations than did nonbreastfed infants at age 6 mo (P < 0.001 for all). The pattern was the same at age 12 mo (Figure 2Go). At age 24 mo, only 15 children were still breastfed, but they had significantly lower holoTC concentrations than did children not breastfed at that age [65 (50, 83) and 90 (85, 97) pmol/L, respectively; P = 0.010].


Figure 2
View larger version (19K):
[in this window]
[in a new window]

  		FIGURE 2.. Geometric mean (95% CI) of folate and cobalamin indexes in breastfed (BF) and nonbreastfed (NBF) infants at age 6 and 12 mo. The number of available results at 6 mo in BF and in NBF infants and at 12 mo in BF and in NBF are: Cobalamin, 221, 41, 114, 128; holohaptocorrin (holoHC) and holotranscobalamin (holoTC), 158, 28, 104, 118; folate, 218, 40, 113, 128; total homocysteine (tHcy), 213, 41, 115, 127; methylmalonic acid (MMA), 212, 41, 115, 126. Student's t test was used to compare groups. At age 6 mo, all cobalamin indexes differed significantly between BF and NBF infants (P < 0.001 for cobalamin, holoTC, MMA, and tHcy; P = 0.002 for holoHC). At age 12 mo, all cobalamin indexes differed significantly between BF and NBF infants (P < 0.001 for cobalamin and holoTC; P = 0.006 for holoHC; P = 0.019 for MMA; and P = 0.002 for tHcy). The dotted line represents the concentration at birth (in cord serum). To test for a breastfeeding status x age interaction, repeated-measures ANOVA was carried out in the group of children that did not change breastfeeding status between age 6 and 12 mo (n = 94 BF and 34 NBF). In these analyses, the effect of breastfeeding status remained highly significantly for all indexes (P < 0.001). The P value for interaction was < 0.1 for all indexes and thus significant: Total cobalamin, P < 0.001; holoTC, P = 0.001; folate, P = 0.034; holoHC, P = 0.044; MMA, P = 0.045; tHcy, P = 0.059.

 
In nonbreastfed children, serum cobalamin and holoHC increased from birth to 12 mo of age (Figure 2Go) and then increased further, until the highest concentration was reached at 24 mo [cobalamin: 420 (396, 445) pmol/L; holoHC: 314 (293, 337) pmol/L]. In contrast, holoTC remained relatively unchanged for the entire period, with nonsignificant changes. In the breastfed children, serum cobalamin declined significantly from birth to age 6 mo; thereafter, it increased markedly, reaching the highest concentration at age 24 mo [378 (308, 464) pmol/L]. For holoTC, there was reduction between birth age age 6 mo to less than one-half of the birth concentration and then a gradual increase until age 24 mo [65 (50, 83) pmol/L], but it still remained lower than the birth values. HoloHC did not change by age 6 mo, but, thereafter, it increased gradually until age 24 mo [306 (243, 386) pmol/L]. Thus, the decline in cobalamin after birth in breastfed children was explained by the decline in holoTC.

Changes in cobalamin and folate status after the introduction of breast milk substitutes and solids
The Spearman correlations between folate and cobalamin indexes and various variables reflecting breastfeeding and weaning are shown in Table 1Go. Serum folate at age 6 mo was positively correlated with duration of exclusive breastfeeding and negatively correlated with the time (in d) since the introduction of other food than breast milk. Serum folate at age 12 mo did not correlate with diet. For cobalamin, holoTC, and holoHC, the correlations at both ages 6 and 12 mo were consistent in the direction of lower values with breastfeeding and higher values with the number of days since cessation of breastfeeding and introduction of breast milk substitutes or solids. The associations were stronger for holoTC than for cobalamin and holoHC. For tHcy and MMA, the associations were similar to those for the cobalamins but in the opposite direction. All of the cobalamin indexes were more strongly correlated with the introduction of breast milk substitutes than with the introduction of solids. Indeed, only holoTC was significantly associated with the introduction of solids. In a linear regression model including days since introduction of breast milk substitutes and days since introduction of solids and with log holoTC as dependent variable, only the introduction of breast milk substitutes remained significantly associated (P < 0.001).


View this table:
[in this window]
[in a new window]

  		TABLE 1. Influence of weaning on folate and cobalamin status1

 
Exclusively breastfed compared with partially breastfed and nonbreastfed infants
Only 11 infants were exclusively breastfed at age 6 mo. None of the measured indexes differed between the exclusively breastfed and the partially breastfed infants. Because many infants were exclusively breastfed until shortly before age 6 mo, we expanded this group by including infants who had been exclusively breastfed until 2 wk before the blood collection at age 6 mo (n = 17), and we compared this group of infants (exclusively breastfed for 5.5–6 mo; n = 28) with other breastfed (n = 193) and nonbreastfed (n = 41) infants. None of the cobalamin indexes, including tHcy, differed between the infants exclusively breastfed for 5.5–6.0 mo and the partially breastfed group. However, these exclusively breastfed infants had significantly higher serum folate [72 (62, 82) nmol/L] than did the partially breastfed [54 (50, 58) nmol/L] or nonbreastfed [52 (45, 59) nmol/L; P = 0.005] infants.

Furthermore, we compared folate and cobalamin status at age 6 and 12 mo in children according to various feeding patterns (breast milk, breast milk substitutes, or solid foods) (Table 2Go). The data further support the findings in the other analyses—ie, that folate status is highest in the exclusively breastfed group, whereas the cobalamin variables differed primarily between breastfed and nonbreastfed infants. In breastfed infants at age 6 mo, the introduction of solid foods, breast milk substitutes, or both did not result in any significant change in the cobalamin indexes, whereas, at age 12 mo, the introduction to breastfed infants of breast milk substitutes in addition to solid foods was associated with modestly higher holoTC and holoHC. It was, however, only in the nonbreastfed infants that serum cobalamin and holoTC were high and metabolites were low.


View this table:
[in this window]
[in a new window]

  		TABLE 2. Indexes of folate and cobalamin status at 6 and 12 mo according to feeding patterns1

 
Finally, we examined the effect of the various milk types and the use of cobalamin-containing supplements on the folate and cobalamin indexes. (See Tables S1 and S2 under "Supplemental data" in the current online issue.) Some of the milk intake groups become very small, but the pattern remains that breastfed infants have lower cobalamin status. There was no significant difference in cobalamin status between formula and cow-milk groups. However, at age 12 mo, children drinking cow milk had significantly lower folate and higher tHcy (P < 0.001 and P = 0.002, respectively) than did children receiving formula. With respect to intake of cobalamin-containing supplements, there was a modest but significant effect on most of the cobalamin indexes in the expected direction. However, even breastfed children receiving cobalamin supplements had lower holoTC than did nonbreastfed children not receiving supplements (P < 0.01).

Cobalamin status at age 12 mo according to cobalamin intake from complementary foods
To further evaluate whether the low serum cobalamin in the breastfed infants could be explained by low dietary intake, we used data from the 7-d weighed food record at age 12 mo. Total intakes of energy, folate, and cobalamin and the dietary sources of the vitamins in the breastfed and nonbreastfed groups are presented elsewhere. (See Table S3 under "Supplemental data" in the current online issue.)

The geometric mean (95% CI) cobalamin intake was 1.4 (1.3, 1.6) in the breastfed group and 2.4 (2.1, 2.6) µg/d in the nonbreastfed group (P < 0.001). The main sources of cobalamin were meat and dairy products; as expected, the breastfed children received significantly less cobalamin from dairy products (milk, yogurt, and cheese) than did the nonbreastfed children. Nine of 104 breastfed children (9%) had a cobalamin intake from complementary foods that was below adequate intake for age 6–12 mo—ie, 0.5 µg/d (29)—whereas that was not the case for any of the nonbreastfed children (0/115).

Geometric mean serum cobalamin was 328 (306, 352) pmol/L (n = 114) and 396 (371, 423) pmol/L (n = 127), and holoTC was 51 (46, 56) pmol/L (n = 104) and 76 (70, 82) pmol/L (n = 117) in the breastfed group and in the nonbreastfed group, respectively (P < 0.001 for all). Adjustment for total energy intake from complementary foods and for body weight at age 12 mo did not change the results. When total cobalamin intake from complementary foods was added to the model, geometric mean serum cobalamin and holoTC remained significantly lower in the breastfed than in the nonbreastfed groups [cobalamin: 329 (304, 355) and 392 (362, 424) pmol/L, respectively (P = 0.004); holoTC: 54 (49, 59) and 70 (64, 78) pmol/L, respectively (P = 0.001)].

Finally, we used linear regression to evaluate the association by cobalamin intake from complementary foods as a determinant of serum cobalamin and holoTC. In unadjusted analyses, cobalamin intake was significantly associated with both serum cobalamin (r = 0.30, P < 0.001) and holoTC (r = 0.34, P < 0.001). In a linear regression model with serum cobalamin as the dependent variable and with total cobalamin intake and the number of breast feedings (in categories) as independent variables, partial r was –0.25 (P < 0.001) for the number of breast feedings and partial r was 0.15 (P = 0.030) for total cobalamin intake. When the model was repeated for holoTC, it showed partial r = –0.33 (P < 0.001) for the number of breast feedings and partial r = 0.25 (P = 0.001) for total cobalamin intake. Thus, the lower serum cobalamin status in breastfed infants was independent of their lower cobalamin intake from complementary foods.

Hematologic variables
At age 6 mo, there was no significant difference between breastfed (n = 237) and nonbreastfed (n = 46) infants in hemoglobin (11.5 in both groups; P = 0.98) or MCV (77.3 and 78.2, respectively; P = 0.09). Also at age 12 mo, the differences between breastfed (n = 115) and nonbreastfed (n = 132) infants were not significant; hemoglobin was 11.0 and 11.2, respectively (P = 0.06), and MCV was 76.7 and 76.5, respectively (P = 0.53). Likewise, neither hemoglobin nor MCV differed according to breastfeeding cessation (data not shown).

Reference intervals
Because breastfed children have a markedly different pattern for the cobalamin indexes, we have presented reference intervals for the different age groups in the total group combined and also according to breastfeeding status (Table 3Go). These reference intervals are based on a population without general folic acid fortification and with limited use of cobalamin and folic acid supplements; for their total intake of these vitamins at 12 mo of age, see Table S3 in "Supplemental data" in the current online issue.


View this table:
[in this window]
[in a new window]

  		TABLE 3. Reference intervals for folate and cobalamin status according to age and breastfeeding status1

 

   DISCUSSION
TOP
 ABSTRACT
 INTRODUCTION
 SUBJECTS AND METHODS
 RESULTS
 DISCUSSION
REFERENCES
 
We have investigated folate and cobalamin status in a group of healthy infants in an affluent Western society. We confirm that there are marked changes in serum concentrations of cobalamin, folate, tHcy, and MMA during infancy (7), and our data show that these changes are, to a great extent, attributable to feeding mode, in particular breastfeeding.

Folate
As others have reported from most studies of healthy Western infants (7-9, 12, 18, 30, 31), we found that serum folate was higher than in adults and older children. The concentrations increased modestly from birth to age 6 mo, and there was a subsequent decline until age 24 mo. Serum folate was similar in the breastfed and nonbreastfed groups, but exclusively breastfed infants had the highest concentrations, and serum folate was positively associated with exclusive breastfeeding.

The folate content in breast milk is relatively independent of maternal folate status (32-34), and breastfed infants are usually well protected against folate deficiency (18, 31, 35). The bioavailability of folate from breast milk is high, possibly beause of the folate-binding protein, which may facilitate absorption from the gastrointestinal tract (36-40). Serum folate in formula-fed infants is often higher than that in breastfed infants (6, 12, 31), depending on the folate content of the formula (18, 31, 41). In the present study, serum folate was not higher in the formula-fed infants, probably because of the relatively low content in Norwegian formulas. Introduction of foods other than breast milk changes the gastrointestinal pH and bacterial flora, which may lead to changes in folate bioavailability (42), and a drop in folate status was shown to parallel the use of solid foods (18, 31). Pasteurization may alter the folate-binding protein and thereby reduce bioavailability (42, 43), which could explain the lower serum folate reported in infants fed cow milk (12). With the reservation related to small group size, our results support the view that exclusive breastfeeding for 6 mo will maintain adequate folate nutrition (44).

Cobalamin
Cord serum cobalamin indexes in our infants were similar to results in other Western newborns (7-11), and our data confirm that concentrations change markedly during the first year of life (7). As has also been reported by others (2, 6, 9, 12), breastfed infants in the present study had cobalamin indexes consistent with a cobalamin status lower than that of nonbreastfed infants. This finding may be related to the lower cobalamin content in breast milk than in substitute foods (12). Breast milk from women with adequate cobalamin status contains only {approx}0.4 µg cobalamin/L (45), whereas the common formula used in the present study (NAN; Nestlé) contains 1.5 µg cobalamin/L and cow milk contains 4 µg cobalamin/L (23).

At age 12 mo, breastfed infants had lower energy and cobalamin intakes from complementary foods than did nonbreastfed infants. If we assume that the energy difference was due to breast milk, this amount of energy corresponds to {approx}335 g breast milk, yielding 0.13 µg cobalamin. Thus, even when breast milk was taken into account, cobalamin intake was lower in the breastfed group than in the nonbreastfed group.

Our data, however, suggest that cobalamin status during infancy is also determined by breastfeeding itself. For instance, in breastfed infants, cobalamin indexes did not differ by complementary feeding. The use of cobalamin-containing supplements only modestly increased cobalamin status. Furthermore, the effect of breastfeeding on cobalamin status remained after adjustment for cobalamin intake from other foods at age 12 mo. We also observed that low serum cobalamin during breastfeeding was explained by a low holoTC concentration, whereas the holoHC concentration did not change. Breast milk contains numerous bioactive components, such as hormones (42), and it is possible that one or more of these components stimulated the utilization of holoTC. If so, the pattern of low holoTC with high holoHC during breastfeeding may reflect an efficient use of cobalamin rather than a deficiency. Under such conditions, holoHC with its long half-life may be a stable and, hence, better marker of cobalamin supply than may the rapidly utilized holoTC. In adults, holoHC and holoTC have half-lives of 10 d and <2 h, respectively (46, 47).

In Western countries, the prevalence of cobalamin deficiency in infancy is unknown (4), but it probably is low, and deficiency usually occurs in exclusively breastfed infants of vegetarian mothers (3, 4). However, on the basis of low cobalamin and high MMA concentrations and of the belief that this pattern reflects poor cobalamin function, it has been suggested that cobalamin deficiency is also widespread in breastfed infants of nonvegetarian mothers (3). In the present study, hematologic status was not better in the nonbreastfed infants, despite their substantially higher cobalamin status. We did not investigate psychomotor function, but a large meta-analysis from Western countries showed that breastfed infants have significantly higher cognitive scores than do formula-fed infants (48). Breastfeeding also has other important effects in relation to infant and long-term health (42, 49). Whatever the mechanism for the low holoTC and high MMA concentrations in healthy breastfed infants, this serum profile could prove to be appropriate and beneficial. Thus, before concluding that breastfeeding by nonvegetarian mothers causes deficiency, one needs further evidence derived from controlled studies with proper endpoints related to health and morbidity. In the meantime, one should target high-risk groups, such as children of vegetarian mothers (4).

Reference limits
Assessment of cobalamin status in infancy is usually relevant only in children with symptoms of deficiency. Our data suggest that age and breastfeeding status should be taken into account in evaluations of the results. After the cessation of breastfeeding, there is likely to be a period of some weeks before the child obtains a stable nonbreastfed status. The holoTC and MMA results are particularly difficult to interpret because they are extremely variable during the first year of life. In breastfed infants, the observed MMA concentrations, which are higher than those at any time later in life, may be misdiagnosed as cobalamin deficiency or even methylmalonic aciduria if proper reference limits are not applied. The lower reference limits for folate are high at birth and age 6 mo, and then they decline markedly, particularly in the nonbreastfed infants. A recent study suggested that the protective effect of breastfeeding against infections is partly mediated by the higher folate concentrations in breastfed infants (2). This possibility suggests that the reference values for folates in breastfed infants may be most appropriate in relation to infant health.

Strengths and limitations of the study
The strengths of the present study include the size of the population, the longitudinal study design, and the measurements of several indexes of folate and cobalamin status. Furthermore, extensive data collection allowed us to investigate the feeding mode, with emphasis on breastfeeding status. Our study was confined to healthy, term infants of predominantly nonvegetarian mothers, which excluded infants at particular risk of cobalamin deficiency (3). Thus, our findings cannot be generalized to these high-risk infants. We do not have data on maternal diet, but, in Norway, <1.5% of the population is vegetarian (50), and the intakes of meat, fish, milk, cheese, and eggs are high (51). In this study, we have mainly examined the effect of age and breastfeeding. In future studies, we will present further data, including the effect of maternal factors on cobalamin and folate status at birth.

Conclusions
Healthy infants, particularly if they are breastfed, have different serum folate and cobalamin concentrations than do older children and adults. Our data suggest that relatively low serum cobalamin and holoTC and high serum MMA concentrations are normal findings in breastfed infants; thus, the reference limits will differ according to both the age of the child and the breastfeeding status. Further studies are required to determine whether this pattern is solely attributed to the cobalamin content in breast milk or whether metabolic effects of breast milk or breastfeeding cause a change in cobalamin homeostasis. To ensure sufficient folate and cobalamin status in infancy, the strategy should be to encourage pregnant and lactating women to have a diversified diet, to advise vegetarian mothers to take a cobalamin supplement during pregnancy and lactation, and to include cobalamin-containing foods in the weaning diet.


   ACKNOWLEDGMENTS
 
We are indebted to the families who participated in the study, the laboratory technicians at Aker University Hospital under the leadership of Britt Eieland, and several project workers and Master's degree students who contributed considerably to the study. We especially thank Marianne Hope, Gunn Helene Arsky, Janne Liabø, Elisabeth Lind Melbye, Nina Cecilie Øverby, Torild Lilleaas Grønnerud, Ellen Margrethe Hovland, Kirsti Kverndokk, and Kathrine C Haavardsholm for their invaluable contributions to the data collection and Otto Andreas Bårholm and Cynthia Prendergast for help with the analyses of the cobalamin and folate indexes. We also greatly appreciate Elfrid Blomdal's help with the literature search.

The authors' contributions are as follows—GH: design of the experiment, collection and analysis of data, and writing of the manuscript; CJ: analysis of data; AW: design of the experiment, analysis and interpretation of data, and critical revision of the manuscript; KT: design of the experiment and analysis and interpretation of data; and HR: analysis and interpretation of data and writing of the manuscript. All authors have read and approved the manuscript. The funding sources had no direct influence on the design, collection, and analyses of the data or on the decision to submit this report for publication. None of the authors had a personal or financial conflict of interest.


   REFERENCES
TOP
 ABSTRACT
 INTRODUCTION
 SUBJECTS AND METHODS
 RESULTS
 DISCUSSION
 REFERENCES
 

  1. Obeid R, Herrmann W. Homocysteine, folic acid and vitamin B12 in relation to pre- and postnatal health aspects. Clin Chem Lab Med 2005;43:1052–7.[Medline]
  2. Strand TA, Taneja S, Bhandari N, et al. Folate, but not vitamin B-12 status, predicts respiratory morbidity in north Indian children. Am J Clin Nutr 2007;86:139–44.[Abstract/Free Full Text]
  3. Bjorke Monsen AL, Ueland PM. Homocysteine and methylmalonic acid in diagnosis and risk assessment from infancy to adolescence. Am J Clin Nutr 2003;78:7–21.[Abstract/Free Full Text]
  4. From the Centers for Disease Control and Prevention. Neurologic impairment in children associated with maternal dietary deficiency of cobalamin—Georgia, 2001. JAMA 2003;289:979–80.[Free Full Text]
  5. Allen LH. Multiple micronutrients in pregnancy and lactation: an overview. Am J Clin Nutr 2005;81(suppl):1206S–12S.[Abstract/Free Full Text]
  6. Karademir F, Suleymanoglu S, Ersen A, et al. Vitamin B12, folate, homocysteine and urinary methylmalonic acid levels in infants. J Int Med Res 2007;35:384–8.[Medline]
  7. Monsen AL, Refsum H, Markestad T, Ueland PM. Cobalamin status and its biochemical markers methylmalonic acid and homocysteine in different age groups from 4 days to 19 years. Clin Chem 2003;49:2067–75.[Abstract/Free Full Text]
  8. Refsum H, Grindflek AW, Ueland PM, et al. Screening for serum total homocysteine in newborn children. Clin Chem 2004;50:1769–84.[Abstract/Free Full Text]
  9. Minet JC, Bisse E, Aebischer CP, Beil A, Wieland H, Lutschg J. Assessment of vitamin B-12, folate, and vitamin B-6 status and relation to sulfur amino acid metabolism in neonates. Am J Clin Nutr 2000;72:751–7.[Abstract/Free Full Text]
  10. Obeid R, Morkbak AL, Munz W, Nexo E, Herrmann W. The cobalamin-binding proteins transcobalamin and haptocorrin in maternal and cord blood sera at birth. Clin Chem 2006;52:263–9.[Abstract/Free Full Text]
  11. Murphy MM, Molloy AM, Ueland PM, et al. Longitudinal study of the effect of pregnancy on maternal and fetal cobalamin status in healthy women and their offspring. J Nutr 2007;137:1863–7.[Abstract/Free Full Text]
  12. Davis RE, Icke GC, Hilton JM, Orr E. Serum thiamin, pyridoxal, cobalamin and folate concentrations in young infants. Acta Paediatr Scand 1986;75:402–7.[Medline]
  13. Ueland PM, Monsen AL. Hyperhomocysteinemia and B-vitamin deficiencies in infants and children. Clin Chem Lab Med 2003;41:1418–26.[Medline]
  14. Hvas AM, Nexo E. Diagnosis and treatment of vitamin B12 deficiency. An update. Haematologica 2006;91:1506–12.[Abstract/Free Full Text]
  15. Morkbak AL, Poulsen SS, Nexo E. Haptocorrin in humans. Clin Chem Lab Med 2007;45:1751–9.[Medline]
  16. Frery N, Huel G, Leroy M, et al. Vitamin B12 among parturients and their newborns and its relationship with birthweight. Eur J Obstet Gynecol Reprod Biol 1992;45:155–63.[Medline]
  17. Giugliani ER, Jorge SM, Goncalves AL. Serum vitamin B12 levels in parturients, in the intervillous space of the placenta and in full-term newborns and their interrelationships with folate levels. Am J Clin Nutr 1985;41:330–5.[Abstract/Free Full Text]
  18. Salmenpera L, Perheentupa J, Siimes MA. Folate nutrition is optimal in exclusively breast-fed infants but inadequate in some of their mothers and in formula-fed infants. J Pediatr Gastroenterol Nutr 1986;5:283–9.[Medline]
  19. Hay G, Sandstad B, Whitelaw A, Borch-Iohnsen B. Iron status in a group of Norwegian children aged 6–24 months. Acta Paediatr 2004;93:592–8.[Medline]
  20. Hay G, Refsum H, Whitelaw A, Melbye EL, Haug E, Borch-Iohnsen B. Predictors of serum ferritin and serum soluble transferrin receptor in newborns and their associations with iron status during the first 2 y of life. Am J Clin Nutr 2007;86:64–73.[Abstract/Free Full Text]
  21. Knudtzon J, Waaler PE, Skjaerven R, Solberg LK, Steen J. New Norwegian percentage charts for height, weight and head circumference for age groups 0–17 years. Tidsskr Nor Laegeforen 1988;108:2125–35.[Medline]
  22. Andersen LF, Lande B, Arsky GH, Trygg K. Validation of a semi-quantitative food-frequency questionnaire used among 12-month-old Norwegian infants. Eur J Clin Nutr 2003;57:881–8.[Medline]
  23. Rimestad AH, Borgejordet Å, Vesterhus KN, et al. Den store matvaretabellen. (The Norwegian food composition table.) 2nd ed. Oslo, Norway: National Nutrition Council, 2001 (in Norwegian).
  24. Kelleher BP, Walshe KG, Scott JM, O'Broin SD. Microbiological assay for vitamin B12 with use of a colistin-sulfate-resistant organism. Clin Chem 1987;33:52–4.[Abstract/Free Full Text]
  25. Kelleher BP, Broin SDO. Microbiological assay for vitamin B12 performed in 96-well microtitre plates. J Clin Pathol 1991;44:592–5.[Abstract/Free Full Text]
  26. Molloy AM, Scott JM. Microbiological assay for serum, plasma, and red cell folate using cryopreserved, microtiter plate method. Methods Enzymol 1997;281:43–53.[Medline]
  27. Refsum H, Johnston C, Guttormsen AB, Nexo E. Holotranscobalamin and total transcobalamin in human plasma: determination, determinants, and reference values in healthy adults. Clin Chem 2006;52:129–37.[Abstract/Free Full Text]
  28. Husek P. Simultaneous profile analysis of plasma amino and organic acids by capillary gas chromatography. J Chromatogr B Biomed Appl 1995;669:352–7.[Medline]
  29. Food and Nutrition Board, Institute of Medicine. Dietary reference intakes for thiamin, riboflavin, niacin, vitamin B6, folate, vitamin B12, pantothenic acid, biotin, and choline. Washington, DC: National Academies Press, 1998.
  30. Shojania AM. Folic acid and vitamin B12 deficiency in pregnancy and in the neonatal period. Clin Perinatol 1984;11:433–59.[Medline]
  31. Smith AM, Picciano MF, Deering RH. Folate intake and blood concentrations of term infants. Am J Clin Nutr 1985;41:590–8.[Abstract/Free Full Text]
  32. Tamura T, Yoshimura Y, Arakawa T. Human milk folate and folate status in lactating mothers and their infants. Am J Clin Nutr 1980;33:193–7.[Abstract/Free Full Text]
  33. Allen LH. Maternal micronutrient malnutrition: effects on breast milk and infant nutrition, and priorities for intervention. SCN News 1994;11:21–4.
  34. Mackey AD, Picciano MF. Maternal folate status during extended lactation and the effect of supplemental folic acid. Am J Clin Nutr 1999;69:285–92.[Abstract/Free Full Text]
  35. Ek J, Magnus EM. Plasma and red blood cell folate in breastfed infants. Acta Paediatr Scand 1979;68:239–43.[Medline]
  36. Lonnerdal B. Human milk proteins: key components for the biological activity of human milk. Adv Exp Med Biol 2004;554:11–25.[Medline]
  37. Rosenberg IH, Selhub J. Folate and vitamin B12 transport systems in the developing infant. J Pediatr 2006;149:S62–3.
  38. Colman N, Hettiarachchy N, Herbert V. Detection of a milk factor that facilitates folate uptake by intestinal cells. Science 1981;211:1427–9.[Abstract/Free Full Text]
  39. Lonnerdal B, Lien EL. Nutritional and physiologic significance of alpha-lactalbumin in infants. Nutr Rev 2003;61:295–305.[Medline]
  40. Birn H, Zhai X, Holm J, et al. Megalin binds and mediates cellular internalization of folate binding protein. FEBS J 2005;272:4423–30.[Medline]
  41. Ek J, Magnus E. Plasma and red cell folate values and folate requirements in formula-fed term infants. J Pediatr 1982;100:738–44.[Medline]
  42. Lonnerdal B. Breast milk: a truly functional food. Nutrition 2000;16:509–11.[Medline]
  43. Gregory JF 3rd. Denaturation of the folacin-binding protein in pasteurized milk products. J Nutr 1982;112:1329–38.[Abstract/Free Full Text]
  44. Dewey KG, Cohen RJ, Brown KH. Exclusive breast-feeding for 6 months, with iron supplementation, maintains adequate micronutrient status among term, low-birthweight, breast-fed infants in Honduras. J Nutr 2004;134:1091–8.[Abstract/Free Full Text]
  45. WHO/FAO. Vitamin B12. Vitamin and mineral requirements in human nutrition. 2nd ed. Geneva, Switzerland: World Health Organization, 2004:279–88.
  46. Hom BL, Olesen HA. Plasma clearance of 57cobalt-labelled vitamin B12 bound in vitro and in vivo to transcobalamin I and II. Scand J Clin Lab Invest 1969;23:201–11.[Medline]
  47. Chanarin I. The megaloblastic anaemias. 3rd ed. Oxford, United Kingdom: Blackwell Scientific Publications, 1990:23–33.
  48. Anderson JW, Johnstone BM, Remley DT. Breast-feeding and cognitive development: a meta-analysis. Am J Clin Nutr 1999;70:525–35.[Abstract/Free Full Text]
  49. Raisler J, Alexander C, O'Campo P. Breast-feeding and infant illness: a dose-response relationship? Am J Public Health 1999;89:25–30.[Abstract/Free Full Text]
  50. Larsson CL, Klock KS, Astrom AN, Haugejorden O, Johansson G. Food habits of young Swedish and Norwegian vegetarians and omnivores. Public Health Nutr 2001;4:1005–14.[Medline]
  51. Johansson L, Solvoll K, Bjørneboe G-EAa, Drevon CA. Dietary habits among Norwegian men and women. Scand J Nutr 1997;41:63–70.
Received for publication September 5, 2007. Accepted for publication April 1, 2008.





This Article
Right arrow Abstract Freely available
Right arrow Full Text (PDF)
Right arrow Supplemental data
Right arrow Purchase Article
Right arrow View Shopping Cart
Right arrow Alert me when this article is cited
Right arrow Alert me if a correction is posted
Right arrow Citation Map
Services
Right arrow Similar articles in this journal
Right arrow Similar articles in PubMed
Right arrow Alert me to new issues of the journal
Right arrow Download to citation manager
Right arrow reprints & permissions
Citing Articles
Right arrow Citing Articles via Google Scholar
Google Scholar
Right arrow Articles by Hay, G.
Right arrow Articles by Refsum, H.
Right arrow Search for Related Content
PubMed
Right arrow PubMed Citation
Right arrow Articles by Hay, G.
Right arrow Articles by Refsum, H.
Agricola
Right arrow Articles by Hay, G.
Right arrow Articles by Refsum, H.

HOME HELP FEEDBACK SUBSCRIPTIONS ARCHIVE SEARCH TABLE OF CONTENTS