| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
Pediatric Clinical Chemistry |
1 Department of Pediatrics and2
Department of Pharmacology, and3
LOCUS for Homocysteine and Related Vitamins, University of Bergen, N-5021 Bergen, Norway.
4 Department of Pharmacology, University of Oxford, Oxford, UK.
aAddress correspondence to this author at: Department of Pediatrics, Haukeland University Hospital, N-5021 Bergen, Norway. Fax 47-55-975147; e-mail albm{at}online.no.
 |
Abstract |
|---|
 Top Abstract IntroductionParticipants and MethodsResultsDiscussionReferences |
|---|
Methods: Serum cobalamin, serum and whole-blood folate, and plasma MMA and tHcy were determined in a cross-sectional study of 700 children, ages 4 days to 19 years.
Results: During the first 6 months, serum cobalamin was lower than and plasma MMA, tHcy, and serum folate were higher than the concentrations detected in the other age groups. In infants 6 weeks to 6 months of age, median MMA and tHcy concentrations were >0.78 and >75 µmol/L, respectively. In older children (>6 months), serum cobalamin peaked at 3–7 years and then decreased, median plasma MMA remained low (<0.26 µmol/L), median plasma tHcy was low (<6 µmol/L) and increased from the age of 7 years on, and serum folate gradually decreased. Plasma MMA was inversely associated with cobalamin (r = -0.4) in both age groups, but across the whole range of cobalamin concentrations, MMA was markedly higher in infants (
6 months) than in older children. Plasma tHcy showed a strong negative correlation to cobalamin (r = -0.52) but not to serum folate in infants 6 months. In older children, tHcy showed the expected association with both cobalamin (r = -0.48) and folate (r = -0.51).
Conclusions: In infants 6 weeks to 6 months, concentrations of the metabolic markers MMA and tHcy were higher than in the other age groups and strongly correlated to cobalamin, whereas in older children, both makers showed correlations to cobalamin and folate concentrations documented in adults. Whether this metabolic profile in infants is explained by impaired cobalamin status, which in turn may have long-term effects on psychomotor development, remains to be addressed in intervention studies.
|
Introduction |
|---|
|
TopAbstractIntroductionParticipants and MethodsResultsDiscussionReferences |
|---|
Several case reports, mainly from developing countries (1)(5)(6), demonstrate the importance of maintaining adequate cobalamin concentrations during periods of rapid growth and development. In infancy, cobalamin deficiency may present as failure to thrive, developmental delay and regression, progressive neurologic disorders, or hematologic changes. The symptoms may be evident as early as 3–4 months of age (7), but are often nonspecific and difficult to detect.
Cobalamin is a coenzyme in a methyl transfer reaction that converts homocysteine to methionine and in a separate reaction converts L-methylmalonyl-CoA to succinyl-CoA (8). This explains why increased total homocysteine (tHcy) and/or methylmalonic acid (MMA) in the blood are measures of impaired cobalamin status, which may occur even in the presence of normal serum cobalamin concentrations and the absence of the classic signs myelopathy or megaloblastosis (8).
Recent data indicate that cobalamin and folate status, including the metabolic markers MMA and tHcy, undergoes marked changes during the first year of life (9). In newborns and infants, tHcy shows a strong relationship with serum cobalamin, but not with serum or erythrocyte folate (10)(11). In older children, the associations are similar to those documented in adults: plasma tHcy correlates strongly with serum folate and less with serum cobalamin (12)(13)(14)(15). tHcy and MMA as markers of cobalamin and folate status in infancy and childhood have been the subject of a recent review article (9).
MMA concentrations are frequently high in the blood or urine in infants. Specker et al. (16) noted more than 10 years ago that the reference interval for urinary MMA in infants was wider and higher than that for adults and older children. Reports based on newborn-screening programs indicate the common occurrence of a benign or less severe form of methylmalonic aciduria not related to inborn errors of the MMA-CoA mutase reaction (17)(18). We recently reported the occurrence of low serum cobalamin and increased plasma MMA in a significant portion of infants born to healthy, nonvegetarian mothers (11). Taken together, these data may indicate that impaired cobalamin status in newborns may be more prevalent than formerly recognized.
In the present work we investigated cobalamin concentrations and the association to folate and the metabolic markers in 700 healthy children 4 days to 19 years of age, who consumed a typical Western diet.
|
Participants and Methods |
|---|
|
TopAbstractIntroductionParticipants and MethodsResultsDiscussionReferences |
|---|
For the age group 4 days to 1 year, blood samples were collected from the mothers (n = 308), who also completed a questionnaire on maternal indices, including data on current and former pregnancies. This information was checked against the Medical Birth Register of Norway.
Newborns born to mothers on regular drug treatment, with diabetes, or who experienced Rh incompatibility during pregnancy were excluded. In addition, children with malformations or serious, chronic diseases were excluded.
Ethics approval of the protocol was granted by the local Committee on Medical Research Ethics, and one parent of each child gave written, informed consent.
data collection
Data on each child’s diet and intake of vitamin supplements and drugs were obtained through a questionnaire filled out by a parent for the younger age groups or by self-reporting in the older age groups (>13 years).
In the age group 4 days to 1 year, additional data on breastfeeding, formula feeding, use of solid foods, and maternal diet, vitamin supplement intake, and smoking habits were collected. For this age group, information on current and former pregnancies was obtained from the Medical Birth Registry of Norway.
Intake of vitamin supplements containing cobalamin and/or folate was categorized as daily and never/sporadic; the latter category was used for those taking no supplement or supplements containing neither folate nor cobalamin. Nutrition during the first year of life was classified as nonexclusive (additional formula and/or solid food) and exclusive breastfeeding (breast milk with no additional formula or food). Smoking habits were recorded as either nonsmoker or daily smoker.
blood sampling and storage
Fasting blood samples were obtained by antecubital venipuncture. For infants up to 1 year, the prandial state was not registered, but all samples were collected in the morning. The samples used for tHcy determinations were placed in ice water, and plasma was separated from the blood cells within 2 h. Plasma and serum samples were stored at -20 °C until analysis.
biochemical analyses
Serum cobalamin was determined by a Lactobacillus leichmannii microbiological assay (19) and serum and whole-blood folate by a L. casei microbiological assay (20). Both cobalamin and folate assays were adapted to a microtiter plate format (21) and carried out by a robotic workstation (Microlab AT plus 2; Hamilton Bonaduz AG). The microbiological assays correlated with immunoassays (Immuno 1; Bayer) for estimation of serum (n = 373; r = 0.91) and whole-blood folate (n = 253; r = 0.61) and serum cobalamin (n = 380; r = 0.96).
Plasma MMA concentrations were assayed using a gas chromatography–mass spectrometry method based on ethylchloroformate derivatization (22), and plasma tHcy was assayed by a liquid chromatography–tandem mass spectrometry method (23).
statistical analyses
Results are presented as medians and interquartile ranges. Medians were compared by the Mann–Whitney test. Differences in vitamin and metabolite concentrations by age groups were assessed by the Kruskal–Wallis test. Correlation was assessed by the Spearman correlation.
Serum folate and serum cobalamin concentrations as determinants of high tHcy (upper quartile for the total group, >7.85 µmol/L) at the ages below and above 6 months were assessed by ROC analysis. Two-sided P values <0.05 were considered statistically significant.
The SPSS statistical package (Ver. 10) was used for all statistical analyses.
|
Results |
|---|
|
TopAbstractIntroductionParticipants and MethodsResultsDiscussionReferences |
|---|
. A total of 700 children 4 days to 19 years of age were studied; 56% were boys. All children >1 year consumed an omnivorous diet. Only 17% of the total population were daily users of vitamin supplements that contained cobalamin and/or folate. Fifty-six percent of the supplements used contained both vitamins, and this percentage was equally distributed among the age groups.
|
Children 4 days to 1 year of age (n = 315) had a median gestational age at birth of 40 weeks (range, 39–41 weeks), and the median birth weight was 3690 g (range, 3340–4000 g). Daily use of multivitamin supplements increased during the first year and was 11% at 6 weeks and 33% at 12 months. Exclusive breastfeeding decreased from 73% at 6 weeks to 35% at 6 months, and at 12 months, all infants received additional solid foods.
Fifty-nine percent of the infants were born to multiparous women who had a mean of 1.5 other children (range, 1–10 children). Eighteen percent of the mothers took multivitamin supplements daily, and 17% were daily smokers.
vitamins and metabolic markers
The median (interquartile range) concentrations of cobalamin, MMA, tHcy, and folate for each age group are listed in Table 2
and presented visually in Fig. 1
. Apart from a significantly higher median plasma MMA (0.32 vs 0.28 µmol/L; P = 0.012) in girls compared with boys at day 4 after birth, there were no sex-related differences, and the data for both genders were analyzed together.
|
|
The most notable findings were low median cobalamin (<350 pmol/L) and high median plasma MMA (>0.26 µmol/L) during the first 6 months (Fig. 1
). Median serum cobalamin decreased markedly after birth (median, 314 pmol/L) to reach a nadir (median, 217 pmol/L) between 6 weeks and 6 months (P <0.001). Thereafter, serum cobalamin increased to a maximum concentration at 
3–7 years, after which the median concentration gradually decreased toward the concentrations observed in adults (24). The period of low serum cobalamin was associated with a marked increase in plasma MMA (from a median of 0.29 to 0.78 µmol/L; P <0.001) and a slight increase in plasma tHcy (from a median of 6.22 to 7.47 µmol/L; P <0.001) and serum folate (from a median of 27.0 to 31.6 nmol/L; P = 0.02; Table 2
). At the age of 1 year, all three indices had decreased to the median concentrations observed in early childhood. Thereafter, plasma MMA and tHcy were stable, and after age 7 years, plasma tHcy increased toward adult concentrations, whereas serum folate gradually decreased up to the age of 15 years (Table 2
and Fig. 1
). Whole-blood folate was high (median, 532 nmol/L) in newborns, remained moderately high (median, 299 nmol/L) during the first 6 months, and reached stable median concentrations of 220 nmol/L from 3 years on (Table 2
).
simple correlations
Spearman rank correlation coefficients demonstrated that the relationships between vitamins and the metabolites MMA and tHcy were dependent on age (Table 3
). Notably, during the first 6 months of life, tHcy was inversely and strongly related to serum cobalamin, but not related to serum folate. Another unique trait in this age group was serum folate, which was positively related to MMA.
|
In children older than 6 months, tHcy showed an equally strong relationship with both cobalamin and folate (Table 3
). We observed a strong correlation between serum cobalamin and plasma MMA (r = -0.4) and a significant correlation between tHcy and MMA in both age groups (Table 3
).
We plotted the plasma tHcy concentrations according to serum folate quintiles (for the whole study group) in children younger and older than 6 months of age (Fig. 2
, top left panel). The inverse dose–response relationship between tHcy and folate was similar in the age groups, but serum folate in the younger children was distributed to higher concentrations and included few individuals with low folate (Fig. 2
, bottom left panel). tHcy concentrations were significantly higher in infants 6 months than in older children across each quintile in folate quintiles 2–4 (P <0.01), but not in quintile 1.
|
A similar graphic presentation of plasma MMA concentrations according to serum cobalamin quintiles in the two age groups is shown in the top right panel of Fig. 2
. Cobalamin and MMA were inversely associated up to a serum cobalamin concentration of 572 pmol/L in both age groups. Notably, MMA concentrations in the infants 6 months were markedly higher than in the older children across each quintile in cobalamin quintiles 1–4 (P <0.001). The difference was largest (0.53 vs 0.19 µmol/L) in individuals with low cobalamin (quintile 1; Fig. 2
, top right panel).
increased plasma THCY according to vitamin concentration
The age-related association between high plasma tHcy (>75 percentile, 7.85 µmol/L) and the vitamins was confirmed and visualized by ROC analyses (Fig. 3
). Below the age of 6 months, the curve for folate was close to the 45-degree diagonal, showing that serum folate was not related to tHcy, whereas the curve for cobalamin was located in the upper triangle throughout the whole range of sensitivity-specificity pairs, demonstrating a significant association (P <0.001) between tHcy and cobalamin. In children older than 6 months, folate and cobalamin both showed a significant (P <0.001) and equal discriminatory power, and both were related to increased tHcy.
|
|
Discussion |
|---|
|
TopAbstractIntroductionParticipants and MethodsResultsDiscussionReferences |
|---|
study design and limitations
We investigated a large number of children (n = 700) recruited to obtain an even age distribution from 4 days to 19 years. The study was mainly cross-sectional, and only 46 children were followed from birth to 6 weeks (15). The study population included presumably healthy infants and children, and the subgroup recruited from the outpatient clinic for minor ear, nose, and throat surgery had normal hematologic indices.
In older children, the blood samples were taken in the fasting state. This may decrease the intraindividual variability in tHcy, because tHcy may vary according to prandial status in children, as has been demonstrated previously in adults (25). For the infant group, we could not obtain fasting samples. Whether the time since last meal influences tHcy in infants is unknown.
vitamin and metabolite concentrations in children
We found relatively high median serum folate concentrations (>25 nmol/L) during the first 6 months of life. Similar observations have been made previously by others (26)(27). The concurrent low median serum cobalamin (<350 pmol/L) described by us has been demonstrated in some (10)(28), but not all (27) published studies. The cobalamin and folate concentrations in children older than 6 months reported here (Fig. 1
and Table 2
) are in accordance with data published by others (27)(29)(30)(31).
In children older than 6 months, we measured plasma tHcy concentrations (Fig. 1
and Table 2
) that agreed with published data, which consistently demonstrated a plasma tHcy of 4–8 µmol/L at age <12 years and an increase in tHcy as a function of age (9). We observed no sex differences. Some (32)(33)(34), but not all (14)(35)(36)(37) published studies have demonstrated a slightly higher plasma tHcy concentration in boys than in girls, and this gender effect is enhanced during and after puberty.
Published data on plasma MMA in infants are scarce (4)(11)(38). In infants 6 weeks to 6 months of age, we observed median plasma MMA concentrations 3 times higher (Fig. 1
and Table 2
) than the upper reference limits established in adults (39). Similarly, Specker and coworkers reported markedly higher urinary MMA excretion in infants than in adults, in particular in infants fed human milk (16) and breast-fed infants of vegetarian mothers (40). In children older than 6 months, we observed plasma MMA concentrations (Table 2
) comparable to those established in adults.
homocysteine and vitamin status by age
In infants <6 months of age, plasma tHcy was strongly correlated with serum cobalamin, but not with serum folate, whereas in older children, serum concentrations of both vitamins were associated with tHcy (Table 3
and Figs. 2
and 3
). A similar observation of serum cobalamin as the main tHcy determinant in neonates has been reported by Minet et al. (10), and there are consistent reports showing that serum folate is strongly associated with tHcy in older children (9)(15)(32).
The lack of correlation between tHcy and serum folate coincided with a period (first 6 months of life) of high serum folate (Fig. 1
and Table 2
). During this period, whole-blood folate was essentially normal (Table 2
) and showed a significant inverse association with tHcy (Table 3
).
possible mechanisms and implications of increased metabolites in infants
The increased metabolite concentrations in infants may be a harmless phenomenon related to developmental, physiologic, or nutritional factors or may reflect the common occurrence of impaired cobalamin function in infants.
High concentrations of plasma tHcy during the first 6 months of life correlated to serum cobalamin but not serum folate. The lack of a relationship between tHcy and folate may reflect that the high serum folate in infants (Fig. 2
and Table 2
) exceeds the subnormal or deficient folate concentrations associated with increased tHcy in adults (8). Alternatively, low cobalamin and high MMA may reflect impaired cobalamin status (8), which would lead to inhibition of methionine synthase and thus to methylfolate trapping and increased 5-methyltetrahydrofolate in serum/plasma (41). The observations that serum folate in infants (but not in older children) tended to be inversely associated with serum cobalamin and positively related to plasma MMA (Table 3
) support this contention. However, long-term methylfolate trapping is expected to cause folate depletion (42), whereas these infants had high-normal whole-blood folate (Table 2
).
The possible effects of constituents in human milk on plasma tHcy should be considered. The methionine content in human milk is actually lower than in milk-based formulas, and breastfeeding has been associated with low plasma methionine (43). In adults, ingestion of large amounts of methionine during methionine loading increases tHcy (44), whereas dietary methionine does not seem to affect plasma tHcy (45). Thus, dietary methionine may not explain the variations in plasma tHcy in infants.
In infants 6 months, MMA was inversely related to cobalamin, but the MMA concentrations were higher than in older children at each cobalamin quintile and therefore through the range of cobalamin concentrations (Fig. 2
). The higher plasma MMA concentrations in infants than in older children (Fig. 1
and Table 2
) and adults (39) may be related to processes independent of cobalamin function, such as propionate production by intestinal bacteria (46) and liver (47) and kidney immaturity (48). The tubular system responsible for secretion of organic acids is not fully developed at birth (48). Impaired secretion of MMA or its precursor(s) in infants may explain high plasma MMA concentrations, but certainly not the increased urinary MMA excretion, which is reported to be inversely related to maternal cobalamin status (16)(40).
Because urinary MMA excretion in infants is related to breastfeeding (16), dietary causes of increased MMA should be considered. Cholesterol (49) and odd-chain fatty acids (50) are constituents of human milk and, together with some amino acids (methionine, isoleucine, threonine, and valine), are precursors of propionic acid and MMA (8). Total blood cholesterol is higher in breast-fed than formula-fed infants (51)(52), but there are indications that the high cholesterol may be related to enhanced synthesis rather than intake (53). Fecal concentrations of propionic acid have been determined to be lower rather than higher in breast-fed infants compared with infants given milk formula (54). Finally, protein intake on a weight basis may be high in infants, but the plasma concentrations of methionine, isoleucine, and threonine are lower in breast-fed than in formula-fed infants (43). Thus, the data cited above do not provide any firm evidence that high plasma MMA in infants is related to intake of compounds degraded to propionic acid and MMA.
Alternatively, high plasma MMA reflects impaired cobalamin function. Several lines of evidence support the latter contention. The relationships between serum folate, serum cobalamin, and metabolites suggest a methylfolate trap, as outlined above. Furthermore, the increase in MMA concurs with increased tHcy and low plasma cobalamin (Figs. 1
and 2
) and is predicted by maternal cobalamin status and factors expected to put a strain on maternal cobalamin status, such as multiparity (11). There are consistent reports demonstrating higher urinary MMA (16) or tHcy (10)(55) in breast-fed infants than in infants given formula, which is usually enriched with cobalamin (56). Likewise, initially increased tHcy in breast-fed neonates has been corrected by cobalamin supplementation (10). Thus, the increased MMA and its relationship with other blood indices suggest the common occurrence of impaired cobalamin status in infants (57).
In conclusion, the combinations of lower serum cobalamin and increased MMA, tHcy, and folate in the first 6 months of life compared with other age groups may reflect normal, transient physiologic processes. However, this metabolic profile is related to factors expected to affect maternal cobalamin balance and seems to be normalized by cobalamin supplementation. These observations point to the possibility that impaired cobalamin status may be common in infants born to nonvegetarian mothers on a typical Western diet, as suggested previously by others (1)(57). Because impaired cobalamin function may have long-term effects related to psychomotor development (58), our findings may motivate intervention studies with cobalamin supplementation in pregnancy and infancy.
|
Acknowledgments |
|---|
|
References |
|---|
|
TopAbstractIntroductionParticipants and MethodsResultsDiscussionReferences |
|---|
The following articles in journals at HighWire Press have cited this article:
|
|
 |
|
  G. Hay, C. Johnston, A. Whitelaw, K. Trygg, and H. Refsum Folate and cobalamin status in relation to breastfeeding and weaning in healthy infants Am. J. Clinical Nutrition, July 1, 2008; 88(1): 105 - 114. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 A.-L. Bjorke-Monsen, I. Torsvik, H. Saetran, T. Markestad, and P. M. Ueland Common Metabolic Profile in Infants Indicating Impaired Cobalamin Status Responds to Cobalamin Supplementation Pediatrics, July 1, 2008; 122(1): 83 - 91. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
T. A Strand, S. Taneja, N. Bhandari, H. Refsum, P. M Ueland, H. K Gjessing, R. Bahl, J. Schneede, M. K Bhan, and H. Sommerfelt Folate, but not vitamin B-12 status, predicts respiratory morbidity in north Indian children Am. J. Clinical Nutrition, July 1, 2007; 86(1): 139 - 144. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 E. D. McLean, L. H. Allen, C. G. Neumann, J. M. Peerson, J. H. Siekmann, S. P. Murphy, N. O. Bwibo, and M. W. Demment Low Plasma Vitamin B-12 in Kenyan School Children Is Highly Prevalent and Improved by Supplemental Animal Source Foods J. Nutr., March 1, 2007; 137(3): 676 - 682. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
R. Obeid, W. Munz, M. Jager, W. Schmidt, and W. Herrmann Biochemical indexes of the B vitamins in cord serum are predicted by maternal B vitamin status Am. J. Clinical Nutrition, July 1, 2005; 82(1): 133 - 139. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
I. M van Beynum, M. den Heijer, C. M. Thomas, L. Afman, D. Oppenraay-van Emmerzaal, and H. J Blom Total homocysteine and its predictors in Dutch children Am. J. Clinical Nutrition, May 1, 2005; 81(5): 1110 - 1116. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 H. Refsum, A. W. Grindflek, P. M. Ueland, A. Fredriksen, K. Meyer, A. Ulvik, A. B. Guttormsen, O. E. Iversen, J. Schneede, and B. F. Kase Screening for Serum Total Homocysteine in Newborn Children Clin. Chem., October 1, 2004; 50(10): 1769 - 1784. [Abstract] [Full Text] [PDF]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
). Categorization represents quintiles for the whole study group (n = 700). The data are given as median values with the 25th and 75th percentiles (shaded areas). (Bottom panels), frequency distributions according to quintiles of folate or cobalamin for the two age groups.