Journal of Nutrition OpenSOurce Diets- www.ResearchDiets.com

Home Help [Feedback] [For Subscribers] [Archive] [Search] [Contents]
QUICK SEARCH:   [advanced]


     

This Article
Right arrow Abstract Freely available
Right arrow Full Text (PDF)
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
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 HighWire
Right arrow Citing Articles via Google Scholar
Google Scholar
Right arrow Articles by Fomon, S. J.
Right arrow Articles by Frantz, J. A.
Right arrow Search for Related Content
PubMed
Right arrow PubMed Citation
Right arrow Articles by Fomon, S. J.
Right arrow Articles by Frantz, J. A.

The Journal of Nutrition Vol. 127 No. 1 January 1997, pp. 83-88
Copyright ©1997 by the American Society for Nutritional Sciences

Erythrocyte Incorporation of Iron Is Similar in Infants Fed Formulas Fortified with 12 mg/L or 8 mg/L of Iron1,2

Samuel J. Fomon3, Ekhard E. Ziegler, Robert E. Serfass*, Steven E. Nelson, and Joan A. Frantz

Department of Pediatrics, University of Iowa, Iowa City, IA 52242 and * Center for Designing Foods to Improve Nutrition, Department of Food Science and Human Nutrition, Iowa State University, Ames, IA 50011

ABSTRACT
INTRODUCTION
MATERIALS AND METHODS
RESULTS
DISCUSSION
FOOTNOTES
LITERATURE CITED

ABSTRACT

Although feeding of formulas with iron concentration of 215 µmol/L (12 mg/L) is a reliable means of preventing iron deficiency, high intakes of iron may adversely affect absorption of copper and zinc. Because data are not available to establish whether fortification at a lower level would result in equivalent iron absorption, we tested the hypothesis that iron absorption is greater by infants fed formulas with an iron concentration of 215 µmol/L (12 mg/L) than by those fed formulas with an iron concentration of 143 µmol/L (8 mg/L). Fifty-two normal infants entered the study at 112 ± 4 d of age, and 46 of these were successfully studied until 196 d of age. Using the stable isotope 58Fe, we determined erythrocyte incorporation of iron by infants fed Formula 8 [iron approximately 143 µmol/L (8 mg/L)] and by infants fed Similac with Iron® [iron approximately 215 µmol/L (12 mg/L)]. On each of three test days beginning at 154 d of age, a major portion of the formula was labeled with 58Fe. Geometric mean erythrocyte incorporation of iron adjusted for plasma ferritin concentration at 168 d of age was 4.82 µmol/d (0.269 mg/d) by infants fed Formula 8 and 5.21 µmol/d (0.291 mg/d) by infants fed Similac with Iron. Corresponding values at 196 d of age were 5.12 and 5.41 µmol/d (0.286 and 0.302 mg/d). The differences in quantity of iron incorporated into erythrocytes by infants fed Formula 8 and Similac with Iron were not statistically significant (P = 0.66 at 168 d of age, P = 0.75 at 196 d of age) and were judged to be nutritionally trivial. Because we were unable to provide support for our hypothesis that iron absorption is greater by infants fed formulas providing 215 µmol (12 mg) of iron per liter than by those fed formulas providing 143 µmol (8 mg) of iron per liter, we conclude that, pending the results of further studies, it is reasonable to decrease the iron concentration of iron-fortified infant formulas.

Key words: human infants, infant formulas, iron absorption, 58Fe.

INTRODUCTION

In the United States, iron-fortified formulas provide 7.7 µmol of iron per 100 kJ (1.8 mg of iron per 100 kcal) (label claim), equivalent to 215 µmol (12 mg) of iron per liter at standard dilution, whereas in Western Europe the iron concentration of such formulas is generally 107 to 143 µmol/L (6 to 8 mg/L). Feeding of formulas with iron concentration of 215 µmol/L (12 mg/L) or more is a reliable means of preventing iron deficiency (Andelman and Sered 1966, Gorten and Cross 1964, Hertrampf et al. 1986, Marsh et al. 1959, Pizarro et al. 1991, Walter et al. 1993), and such formulas have gained wide acceptance in the United States. However, because high intakes of iron may adversely affect absorption of copper and zinc (Haschke et al. 1986, Lönnerdal 1989, Lönnerdal and Hernell 1994, O'Dell 1989), it is important to determine whether the feeding of formulas with iron concentration of 215 µmol/L (12 mg/L) results in appreciably greater absorption of iron than the feeding of formulas with a somewhat lower iron concentration.

In the present study we determined erythrocyte incorporation of iron from formulas providing approximately 215 µmol (12 mg) of iron per liter and formulas providing approximately 143 µmol (8 mg) of iron per liter. In normal adults, erythrocyte incorporation of iron is a surrogate for iron absorption because an average of about 93% of absorbed iron is promptly (i.e., within 14 d) incorporated into erythrocytes (Larsen and Milman 1975). Although erythrocyte incorporation of iron by infants may not promptly reach the same high percentage as in normal adults, it seemed likely that under identical conditions of study the prompt erythrocyte incorporation of iron would account for approximately the same percentage of newly absorbed iron by all infants in the study. Thus, for the purpose of comparing groups of infants fed formulas at the two levels of fortification, the quantity of iron promptly incorporated into erythrocytes can serve as an indicator of the quantity of iron absorbed.

Test meals of known iron concentration labeled with the stable isotope 58Fe were fed to normal infants under standardized conditions. We hypothesized that erythrocyte incorporation of iron (micromoles per day or milligrams per day) would be greater by infants fed formula with an iron concentration of 215 µmol/L (12 mg/L) than by infants fed formula with an iron concentration of 143 µmol/L (8 mg/L) and that the extent of the difference would be nutritionally significant [defined as 1.8 µmol/d (0.1 mg/d) or more].

MATERIALS AND METHODS

Subjects. The study protocol was reviewed and approved by the University of Iowa Committee on Research Involving Human Subjects. The study procedures were explained to one or both parents, and written consent was obtained. Before entering the present study at 112 d of age, nearly all of the infants participated in other protocols and were fed a variety of experimental and commercially available formulas. Fifty-two infants entered the study, and 46 were successfully studied until 196 d of age. Because the six infants who failed to complete the study dropped out before the first test feeding, we consider the reasons for dropping out to be irrelevant.

Feedings. We desired to limit the study to infants in good iron nutritional status and therefore attempted to ensure adequate intake of iron for all infants. Before enrollment in the study at 112 d of age, infants had been fed formula from birth or had initially been breast-fed for some time before being fed formula. The formulas fed before 112 d of age included commercially prepared milk-based or isolated soy protein-based formulas. With some exceptions, all infants received either an iron-fortified formula [i.e., label claim 215 µmol/L (12 mg/L)] or an iron supplement [134 µmol (7.5 mg) of elemental iron per day in the form of ferrous sulfate] from the first few weeks of life until 112 d of age. The exceptions were three infants enrolled in a previous study (Fomon et al. 1995) who were breast-fed or fed a formula low in iron and did not receive an iron supplement from 59 to 112 d of age, and six infants who were breast-fed but not enrolled in a previous study and who probably did not receive an iron supplement.

Study formulas. For the formula providing iron at 215 µmol/L (12 mg/L) we used batches of Similac with Iron® (Ross Products Division, Abbott Laboratories, Columbus, OH) that had the lowest iron concentration of available batches. Formula 8 was a specially made formula identical to Similac with Iron except for the lower iron concentration. The formulas were supplied to us by the manufacturer in ready-to-feed form in 0.24-L feeding units and provided the following nutrients per 100 kcal (label claim): 2.14 g protein, 5.40 g fat (from soybean and coconut oils); 10.7 g carbohydrate (as lactose), 73 mg calcium, 56 mg phosphorus, 9 mg ascorbic acid (per 100 kJ: 0.511 g protein, 1.29 g fat; 2.56 g carbohydrate, 0.44 mmol calcium, 0.43 mmol phosphorus, 2.15 mg ascorbic acid), and other nutrients within the limits specified by the U.S. Food and Drug Administration (1985). The iron concentration of Formula 8 was 149 µmol/L (8.3 mg/L) (first batch, fed to Cohort 1) and 129 µmol/L (7.2 mg/L) (second batch, fed to Cohort 4). The iron concentration of Similac with Iron was 220 µmol/L (12.3 mg/L) (first batch, fed to Cohort 2 and to all but two infants in Cohort 3) and 235 µmol/L (13.1 mg/L) (second batch, fed to the last two infants in Cohort 3).

From 112 until 196 d of age, all infants were fed Similac with Iron or Formula 8. Formula served as the sole source of energy from the time of enrollment until 140 d of age. Beginning at 140 d of age, strained pears were permitted, and after 161 d of age other beikost items (foods other than milk or formula) were permitted.

Study design. Normal infants of either sex with gestational age 37 wk or more and birth weight of 2500 g or more entered the present study at 112 ± 4 d of age and completed the study 84 d later at 196 d of age. For the entire study period we supplied the families with formulas of known iron concentration. Test meals labeled with the stable isotope 58Fe were fed for three consecutive days, starting within 4 d of age 154 d. Blood samples were obtained within 4 d of ages 140 d (baseline), 168 d (14 d after the first test meal) and 196 d of age (42 d after the first test meal). Erythrocyte incorporation of iron (µmol/d) adjusted for plasma ferritin concentration was compared between the two feeding groups 14 and 42 d after consumption of the test meals.

We determined that to detect a difference in iron incorporation of 1.79 µmol/d (0.1 mg/d) at alpha  = 0.05 and power of 0.9, 30 subjects per group were needed. To complete the study, several batches of each formula were required. Because Similac with Iron was commercially available and merely required testing to identify batches with the desired iron concentration, batches of this formula were more readily available than the specially made Formula 8. Because of this and other practical considerations, the following cohorts of infants were studied either concurrently or successively: Cohort 1, 10 infants (4 males and 6 females) fed Formula 8; Cohort 2, 10 infants (7 males and 3 females) fed Similac with Iron; Cohort 3, 10 infants (2 males and 8 females) fed Similac with Iron; Cohort 4, 16 infants (11 males and 5 females) fed Formula 8. 

An interim analysis of the data performed after completion of study of Cohort 4 led to the conclusion that studying additional infants fed Similac with Iron would not materially alter the study outcome, and the decision was made to terminate the study.

Test feedings. For three consecutive days beginning within 4 d of age 154 d, each infant was admitted for 6 to 8 h to the Lora N. Thomas Pediatric Metabolic Ward. On these days, strained pears were omitted from the infant's diet. Metallic 58Fe-enriched iron (93.243 weight % 58Fe, obtained from U.S. Services Inc., Summit, NJ) was converted to ferrous sulfate as described previously (Janghorbani et al. 1986) and the solution stored in a nitrogen-purged rubber-stoppered vial. On each of the three test days, a precisely weighed amount (about 300 mg) of the solution [providing approximately 5.73 µmol (0.32 mg) of iron, including 5.18 µmol (0.30 mg) of 58Fe] was added to 0.24 L of formula. This labeled formula was fed quantitatively in two equal portions by a research nurse to the infant. Additional (unlabeled) formula was fed at home, and the amount consumed during the day was determined by weighing the containers. Thus, a total of 15.5 µmol (0.9 mg) of 58Fe was administered during the three test days. From the quantity of formula consumed and the determined iron concentration of the formula, iron intake during the 3 d was calculated.

Procedures. Capillary blood was obtained by heel-stick using disposable blades (Tenderfoot, International Technidyne Corporation, Edison, NJ) and collected in heparinized Microvette tubes (CB 1000 S, Sarstedt, Newton, NC). Plasma was separated from cells within 30 min of collection.

Laboratory methods. The iron concentration of the formulas was determined after dry-ashing by flame atomic absorption spectrophotometry (model 560, Perkin Elmer, Norwalk, CT). Heparinized blood was analyzed for hemoglobin and hematocrit by Coulter Counter (model M430, Coulter Electronics, Hialeah, FL). Plasma was analyzed for ferritin using the RIANEN Assay System ferritin [125I] radioimmunoassay kit (catalogue no. NEA-078, Du Pont, Billerica, MA) for the first 27 subjects. Subsequent determinations were performed with the Quantimune kit (catalogue no. 190-2001, Bio-Rad Laboratories, Hercules, CA). Because the RIANEN assay gave systematically lower values in parallel determinations, values obtained with this assay were multiplied by a factor of 1.15. Quality control data for plasma ferritin determinations were obtained by monitoring control serum furnished with the RIANEN kits. With control serum in the range of 89 to 109 µg/L, accuracy was 1.26 ± 8.74 µg/L and precision was 8.92% relative standard deviation (RSD). Corresponding values with the Quantimune kits were 2.64 ± 6.03 µg/L and 6.58% RSD.

For Cohorts 1 and 2, the 58Fe to 57Fe ratio in blood samples was determined by inductively coupled plasma mass spectrometry (ICP/MS), using the Elan 250 ICP/MS system as described by Janghorbani et al. (1986). These results were calculated, as in our early studies (Fomon et al. 1988, 1989 and 1993), as mass isotope ratios (MIR58/57). The 58Fe:57Fe ratio in blood samples from the remaining infants was also determined by ICP/MS but with the use of different instrumentation (Finnigan Sola, Finnigan MAT, Ltd., Hemel Hempstead, U.K.) and with some refinements in sample processing as described by Fomon et al. (1995). The results of these determinations were calculated, as in our most recently reported study (Fomon et al. 1995), as atom isotope ratios (IR58/57) rather than mass isotope ratios.

Calculations. The quantity of 58Fe label in the blood at ages 140, 168 and 196 d of age was calculated as previously described by Fomon et al. (1989) for the studies in which MIR58/57 was determined, and as described by Fomon et al. (1995) for the studies in which IR58/57 was determined. The quantity of total circulating iron was calculated on the basis of an assumed blood volume of 0.065 L/kg body wt, the determined hemoglobin concentration (g/L) and the iron concentration of hemoglobin [62.1 µmol/g (3.47 mg/g)]. From the isotopic enrichment and the amount of circulating iron, the amount of 58Fe label in the circulation was calculated. The quantity of iron (µmol/d) incorporated into erythrocytes was determined by multiplying the percentage of incorporation of the 58Fe label for each infant by the quantity of iron consumed (daily average for the 3 d on which the test feedings were given).

Statistical analyses. Descriptive statistics, regressions, and two-way repeated measures analyses of variance and covariance were performed using SAS (version 6.08, SAS Institute, Cary, NC). Plasma ferritin concentration and erythrocyte incorporation of iron (percentage of label and µmol/d) were log transformed before statistical analysis to compensate for non-normal distributions (Cook et al. 1969). Covariate analysis adjusting for mean plasma ferritin concentration of each infant at 140 and 168 d of age was used in comparing iron incorporation (µmol/d) between treatment groups. P values are presented for (two-tailed) F tests.

RESULTS

Table 1 presents for each infant the total iron intake (µmol/d average) during the three test days, and (for ages 140, 168 and 196 d) the body weight, hemoglobin concentration, plasma ferritin concentration and MIR58/57 (Cohorts 1 and 2) or IR58/57 (Cohorts 3 and 4), and erythrocyte incorporation of 58Fe (percentage of dose) and of total iron (µmol/d) at each age of testing. All infants were in satisfactory hematologic and iron nutritional status. Although a few hemoglobin values less than 110 g/L were observed, each of these values was preceded and/or followed by one or more values above 110 g/L.

Table 1. Body weight, iron intake, serum concentrations of hemoglobin and plasma ferritin concentrations, ratio of 58Fe to 57Fe, and erythrocyte incorporation of iron from test feedings of infants fed formula containing 8 or 12 mg iron/L1,2

[View Table]

Baseline values for MIR58/57 were elevated (>0.150) in three infants in Cohort 1 (subjects 5039, 5090 and 5099) and in one infant in Cohort 2 (subject 5046) (Table 1). Although the elevated baseline values were unexplained, we used them in calculating erythrocyte incorporation of iron for these infants. Because of the uncertainty of the results of these calculations, however, mean values presented in the tables were obtained by excluding data from these four subjects. The exclusion of data from these subjects did not alter the conclusions (see comments regarding Table 2).

Table 2. Erythrocyte incorporation of iron from test feedings of infants fed formula containing 8 or 12 mg iron/L

[View Table]

As may also be noted from Table 1, baseline values for IR58/57 in Cohorts 3 and 4 ranged from 0.1330 to 0.1350, with the exception of three infants (subject 5654 in Cohort 3 and subjects 5645 and 5659 in Cohort 4) whose elevated baselines were explained by their participation in an earlier study (Fomon et al. 1995) in which they received 58Fe. Values for IR58/57 were not obtained for subject 5408 at 140 and 168 d of age, and the mean baseline value (0.1338) for the other subjects in Cohort 3 was used in calculating erythrocyte incorporation of iron by this subject at 196 d of age.

Table 2 summarizes data on plasma ferritin concentration, erythrocyte incorporation of 58Fe (percentage of administered label), and erythrocyte incorporation of total dietary iron (µmol/d) unadjusted and adjusted for plasma ferritin concentration. Plasma ferritin concentration was significantly lower in infants fed Formula 8 than in infants fed Similac with Iron (P = 0.046 at age 140 d, P = 0.004 at age 168 d, P = 0.041 at age 196 d). Because the difference in plasma ferritin concentration was present at 140 d of age, although there had been little prior difference in iron intake by the two feeding groups, and because the change in plasma ferritin concentration from 140 to 196 d of age was similar in the two feeding groups, it seems unlikely that the differences in plasma ferritin concentrations between the two feeding groups at 168 and 196 d of age were related to a difference in iron absorption from the study formulas.

In our previous studies of infants (Fomon et al. 1988, 1989, 1993 and 1995), erythrocyte incorporation of the 58Fe label (percentage of intake) was inversely related to plasma ferritin concentration, and this relation was similar although not statistically significant in the present study. The plasma ferritin concentrations considered relevant were those obtained in closest temporal proximity to the age at which the test meals were fed. Thus, the average of the plasma ferritin concentrations at 140 and 168 d of age (or a single value when only one was available) was related to incorporation at both 168 and 196 d of age. For the pooled data (Formula 8 and Similac with Iron), the correlation of erythrocyte incorporation of iron with plasma ferritin concentration (partial correlation coefficient adjusted for formula) was not significant at 168 d of age (r = -0.28, P = 0.094) or 196 d of age (r = -0.16, P = 0.32).

For infants at 168 d of age, the unadjusted geometric mean erythrocyte incorporation of iron was 5.10 µmol/d (0.285 mg/d) by infants fed Formula 8 and 4.80 µmol/d (0.268 mg/d) by infants fed Similac with Iron. Corresponding values at 196 d of age were 5.34 and 5.05 µmol/d (0.298 and 0.282 mg/d). After adjustment for plasma ferritin concentration, erythrocyte incorporation of iron at 168 d of age was 4.82 µmol/d (0.269 mg/d) by infants fed Formula 8 and 5.21 µmol/d (0.291 mg/d) by infants fed Similac with Iron; corresponding values at 196 d of age were 5.12 and 5.41 µmol/d (0.286 and 0.302 mg/d).

Statistical comparison of erythrocyte incorporation of iron by infants fed the two formulas was performed by analysis of covariance with plasma ferritin concentration as the covariate. The difference in quantity of iron incorporated into erythrocytes by infants fed Formula 8 and Similac with Iron was not significant at 168 d of age (P = 0.66) or 196 d of age (P = 0.75). The difference in the adjusted values for erythrocyte incorporation of iron was 0.39 µmol/d (0.022 mg/d) at 168 d of age and even less [0.29 µmol/d (0.016 mg/d)] at 196 d of age. As already mentioned, we had hypothesized that erythrocyte incorporation of iron would be at least 1.8 µmol/d (0.1 mg/d) greater by infants fed Similac with Iron than by infants fed Formula 8. With an estimated requirement for absorbed iron of 9.8 to 13.4 µmol/d (0.55 to 0.75 mg/d) (Fomon 1993), the observed difference in erythrocyte incorporation of iron [0.39 or 0.29 µmol/d (0.022 or 0.016 mg/d)] is only 3 to 4% of the estimated requirement. Therefore, we consider the observed differences to be not only nonsignificant but nutritionally trivial.

The results were little affected by the exclusion of data for the four subjects in Cohorts 1 and 2 with unexplained elevated baseline values for MIR58/57. With the inclusion of data for these four infants, geometric mean erythrocyte incorporation of iron adjusted for plasma ferritin concentration at 168 d of age was 4.60 µmol/d (0.257 mg/d) by infants fed Formula 8 and 4.71 µmol/d (0.263 mg/d) by infants fed Similac with Iron. Corresponding values at 196 d were 4.76 µmol/d (0.266 mg/d) by infants fed Formula 8 and 4.89 µmol/d (0.273 mg/d) by infants fed Similac with Iron. The differences were not significant (P = 0.91 and P = 0.89, respectively).

DISCUSSION

Using the stable isotope 58Fe, we determined erythrocyte incorporation of dietary iron by infants fed formulas with different concentrations of iron. Formula 8 provided 149 or 129 µmol (8.3 or 7.2 mg) of iron per liter, and Similac with Iron provided 220 or 235 µmol (12.3 or 13.1 mg) of iron per liter. The quantity of iron incorporated into erythrocytes at 168 d of age (i.e., 14 d after the administration of the 58Fe-labeled formula) was 4.82 µmol/d (0.269 mg/d) by infants fed Formula 8 and 5.21 µmol/d (0.291 mg/d) by infants fed Similac with Iron, a difference (0.39 µmol/d [0.022 mg/d]) that was not only nonsignificant but one that we consider nutritionally trivial.

There are few previous reports of the effect of the level of iron fortification on iron absorption or erythrocyte incorporation of iron. Saarinen and Siimes (1976) determined iron absorption using whole-body counting 14 d after feeding 59Fe-labeled iron-fortified infant formulas to approximately 1-y-old infants. From the 0.05-L test feeding, retention of iron averaged 0.77 µmol (0.043 mg) by 10 infants fed a formula providing 229 µmol (12.8 mg) of iron per liter and 0.57 µmol (0.032 mg) by nine infants fed a formula providing 122 µmol (6.8 mg) of iron per liter. In view of the limited number of subjects per group, the wide differences in absorption between infants in the same feeding group and the lack of adjustment for plasma ferritin concentration, failure to detect a significant difference seems inconclusive. Using 59Fe as a label, Stekel et al. (1986) over a period of several years studied erythrocyte incorporation of iron by 5- to 18-mo-old infants fed a large number of iron-fortified experimental and commercially available infant formulas. Although the studies were performed to determine the influence of various formula characteristics rather than to explore the effect of different levels of iron fortification, one milk product was studied at two levels of iron fortification [179 and 269 µmol (10 and 15 mg) Fe/L]. The difference between the two formulas in 59Fe incorporation into erythrocytes was not significant, but no adjustment for the widely varying iron nutritional status of the subjects was made.

The average requirement for absorbed iron during the first year of life has been estimated to be 9.8 to 13.4 µmol/d (0.55 to 0.75 mg/d) (Fomon 1993). Assuming for the age interval 154 to 196 d a daily weight gain of 0.022 kg, blood volume of 0.065 L/kg, hemoglobin concentration of 120 g/L and iron concentration of hemoglobin 62.1 µmol/g (3.47 mg/g), the increase in circulating hemoglobin iron is estimated to be 10.6 µmol/d (0.59 mg/d). Thus, the quantity of dietary iron incorporated into erythrocytes in the present study [approximately 5.4 µmol/d (0.3 mg/d)] is only about half of the calculated increase in circulating iron. The remaining half of the increase in circulating iron is presumably obtained from body iron storage sites.

If infants promptly incorporated into erythrocytes more than 90% of absorbed iron, as is the case for normal adults (Larsen and Milman 1975), the quantity of absorbed iron would have been approximately 5.4 µmol/d (0.3 mg/d), and one would anticipate that a number of infants fed formulas providing 215 µmol (12 mg) of iron per liter would become iron-deficient by 1 y of age. The fact that nearly all infants fed iron-fortified formulas maintained good iron nutritional status (Andelman and Sered 1966, Gorten and Cross 1964, Hertrampf et al. 1986, Marsh et al. 1959, Pizarro et al. 1991, Walter et al. 1993) suggests that the percentage of absorbed iron promptly incorporated into erythrocytes is substantially less in infants than in adults. Erythrocyte incorporation of 59Fe administered intravenously to three infants 16 to 25 d of age ranged (judging from graphic data) from 25 to 55% of the dose 10 d after administration to 35 to 85% of the dose 90 d after administration (Garby et al. 1964). Similar studies of older term infants have not been reported. In low-birth-weight infants, mean erythrocyte incorporation has been reported to range from 17 to 52% of absorbed iron (Ehrenkranz et al. 1992, Gorten et al. 1963, Zlotkin et al. 1995).

The daily turnover of hemoglobin iron in adults is approximately 6 mg/L of whole blood (Bothwell et al. 1979), and the turnover rate in infants may be of similar magnitude. Thus, the daily turnover of hemoglobin iron of a 7-kg infant with blood volume of 0.45 L is likely to be more than 36 µmol/d (2 mg/d), a value much greater than the requirement for absorbed iron. Although the adult needs only to replace ongoing iron losses, the infant must also expand total body iron, especially in circulating erythrocytes and in myoglobin. It is possible that iron recovered from senescent erythrocytes in the reticuloendothelial system of the infant has some priority for hemoglobin formation over newly absorbed iron, whereas newly absorbed iron has priority for incorporation into myoglobin and enzymes.

Although there is little evidence that infant formula with iron concentration of 215 µmol/L (12 mg/L) will adversely affect absorption of other minerals, it is important to note that all manufacturers provide somewhat more iron (spoken of as "overage") than indicated by the label claim, and the upper limit for iron in infant formulas currently permitted by the United States Food and Drug Administration is 12.8 µmol/100 kJ (3 mg/100 kcal) (Food and Drug Administration 1985), equivalent to approximately 358 µmol (20 mg) of iron per liter at standard dilution. Unfortunately, the safety of formulas providing 269 to 358 µmol (15 to 20 mg) of iron per liter is untested. If label claims for iron concentration of infant formulas were 125 or 143 µmol/L (7 or 8 mg/L) rather than 215 µmol/L (12 mg/L), the Food and Drug Administration would presumably set the upper limit for iron in infant formulas at a much more modest value.

Ideally, our failure to demonstrate greater erythrocyte incorporation of iron from formulas providing 215 or 233 µmol (12 or 13 mg) of iron per liter than from formulas providing 125 or 143 µmol (7 or 8 mg) of iron per liter should be followed up by a large-scale field study of infants. Iron-fortified formulas at the two levels of iron fortification might be fed from the early weeks of life until 1 y of age, using indices of iron nutritional status as the critical endpoints. In addition, it is desirable to study iron absorption as well as erythrocyte incorporation of iron in infants fed formulas fortified with iron at 215 µmol/L (12 mg/L), 143 µmol/L (8 mg/L) and at some lower level, perhaps 90 or 107 µmol/L (5 or 6 mg/L). Because such studies are difficult and expensive and unlikely to be performed in the near future, we recommend that until such studies have been undertaken and reported, infant formulas be fortified with iron at a level of 125 or 143 µmol/L (7 or 8 mg/L). We recommend that the upper limit of iron in infant formulas be set at 7.7 µmol/100 kJ (1.8 mg/100 kcal) [215 µmol/L (12 mg/L) at standard dilution].

FOOTNOTES

1   Supported in part by Maternal and Child Health Bureau, PHS/DHHS, grant MCJ-190606.
2   The costs of publication of this article were defrayed in part by the payment of page charges. This article must therefore be hereby marked "advertisement" in accordance with 18 USC section 1734 solely to indicate this fact.
3   To whom correspondence should be addressed.

Manuscript received 28 May 1996. Initial reviews completed 1 August 1996. Revision accepted 26 September 1996.

LITERATURE CITED

0022-3166/97 $3.00 ©1997 American Society for Nutritional Sciences



This article has been cited by other articles:


Home page
 J. Nutr.Home page
S. J. Fomon, S. E. Nelson, R. E. Serfass, and E. E. Ziegler
Absorption and Loss of Iron in Toddlers Are Highly Correlated
J. Nutr., April 1, 2005; 135(4): 771 - 777.
[Abstract] [Full Text] [PDF]

Home page
 Am. J. Clin. Nutr.Home page
O. Hernell and B. Lonnerdal
Iron status of infants fed low-iron formula: no effect of added bovine lactoferrin or nucleotides
Am. J. Clinical Nutrition, October 1, 2002; 76(4): 858 - 864.
[Abstract] [Full Text] [PDF]

Home page
 Am. J. Clin. Nutr.Home page
B. Lonnerdal, S. L Kelleher, and E. L Lien
Extent of thermal processing of infant formula affects copper status in infant rhesus monkeys
Am. J. Clinical Nutrition, May 1, 2001; 73(5): 914 - 919.
[Abstract] [Full Text] [PDF]

Home page
 J. Nutr.Home page
S. J. Fomon, R. E. Serfass, S. E. Nelson, R. R. Rogers, and J. A. Frantz
Time Course of and Effect of Dietary Iron Level on Iron Incorporation into Erythrocytes by Infants
J. Nutr., March 1, 2000; 130(3): 541 - 545.
[Abstract] [Full Text]

Home page
 J. Nutr.Home page
S. J. Fomon, E. E. Ziegler, R. E. Serfass, S. E. Nelson, R. R. Rogers, and J. A. Frantz
Less Than 80% of Absorbed Iron Is Promptly Incorporated into Erythrocytes of Infants
J. Nutr., January 1, 2000; 130(1): 45 - 52.
[Abstract] [Full Text]

Home page
 PediatricsHome page
Committee on Nutrition
Iron Fortification of Infant Formulas
Pediatrics, July 1, 1999; 104(1): 119 - 123.
[Abstract] [Full Text]

This Article
Right arrow Abstract Freely available
Right arrow Full Text (PDF)
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
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 HighWire
Right arrow Citing Articles via Google Scholar
Google Scholar
Right arrow Articles by Fomon, S. J.
Right arrow Articles by Frantz, J. A.
Right arrow Search for Related Content
PubMed
Right arrow PubMed Citation
Right arrow Articles by Fomon, S. J.
Right arrow Articles by Frantz, J. A.

Home Help [Feedback] [For Subscribers] [Archive] [Search] [Contents]