Iron deficiency in infants is one of the most common nutritional deficiency disorders worldwide. In general, the cause of iron deficiency can be ascribed to a negative balance between iron intake and iron loss. Whenever there is rapid growth, as occurs during infancy, iron requirements are increased, and most cases of iron deficiency are directly related to low levels of absorbable iron. There are two independent pathways whereby dietary iron enters the gastrointestinal mucosal cell as non-haem and haem iron. Nonhaem iron is the largest component of the diet, but its absorption is influenced by a large number of promotors and inhibitors. In contrast, haem iron is absorbed into the mucosal cell as the intact porphyrin complex (1,2).
The prevention of iron deficiency by fortification of infant foods is especially appealing because it is easily aimed at the age group at greatest risk and is less costly and more effective than the detection and treatment of anaemia in individual infants. The most common method of iron fortification in industrialized countries is the addition of nonhaem iron to formula and weaning foods, but haem iron deserves more consideration as a compound for use in iron fortification. It is well tolerated by the gastrointestinal tract (3), does not interfere with the absorption of other trace elements such as zinc (4), and is not influenced to any great extent by inhibitors of iron absorption such as phytate (5). Studies have been undertaken to investigate iron bioavailability of haem iron as an alternative fortifier of weaning food by adding it to extruded rice flour (6) and baking it into cookies for a school programme (7).
The purpose of this study was to compare the absorption of iron as haem iron concentrate with the well-absorbed salt ferrous sulphate when added to an infant weaning food and to determine whether the haem iron concentrate has an enhancing effect on iron absorption from the nonhaem iron pool.
SUBJECTS AND METHODS
In this study, 16 healthy infants were recruited from the local birth register and divided into two groups of eight. Parents were visited at home and were provided details about the purpose and procedures of the study, and their written, informed consent was obtained. An initial sample of blood was taken by heel prick to analyse for haematological indices. Subjects with iron-deficiency anaemia (haemoglobin concentration <100 g/l) were not accepted for the study. Anthropometric measurements (head circumference, length, birth weight, and body weight) were recorded, together with the type of milk that was regularly consumed by each infant (breast milk or formula). The study was approved by the ethics committees of the Institute of Food Research and the Norwich Health District.
Three meat and cereal or meat and vegetable commercial foods were used for the study, according to the parents' choice: chicken with rice (0.42 mg Fe/100 g), lamb with vegetables (0.72 mg Fe/100 g), and veal with carrots (0.91 mg Fe/100 g), all supplied by Hero S.A., Alcantarilla, Spain. The foods were provided in 250-g jars and had neither iron nor vitamin C added. Test meals were prepared at the infant's home by adding 2.5 mg of iron as haem iron concentrate (1.32 g of dry powder, Aprocat S.A., Granollers, Spain) or as ferrous sulphate (0.742 ml of a 3.367-mg Fe/ml solution) to approximately 100 g of the infant food and mixing thoroughly just before feeding the infants.
Ferrous sulphate solution was prepared by dissolving 337 mg of iron wire (Aldrich Chemical Co., Gillingham, Dorset, U.K.) in 5 ml of concentrated HNO3 (Aristar grade; BDH, Pode, Dorset, U.K.). The mixture was boiled to dryness on a hot plate and placed in a muffle furnace at 500°C for 30 minutes. The resulting white powder was taken up in a minimum volume of 0.5 M H2SO4 and stirred on a hot plate until it became clear. After cooling, the solution was increased to 100 ml with quartz-distilled water, and then ascorbic acid was added (0.83 mg/mg iron) to reduce all Fe to Fe2+. The solution was dispensed into 3-ml glass screw-capped tubes by sterilized glass pipettes and then autoclaved.
The isotope solution used for the study was prepared from 57Fe-enriched elemental iron (95.1%, AEA Technology, Oxfordshire, UK). A weighed amount (122.72 mg) of 57Fe was dissolved in 5 ml of concentrated HNO3 (Aristar grade) and prepared as the ferrous sulphate solution as described earlier. The isotope solutions (concentration, 1.2 mg Fe/ml) were dispensed in 1-ml aliquots, sealed in glass vials, autoclaved, and tested in the Pharmacy Department of the Norfolk and Norwich Hospital (Norwich, UK).
Infants in the first group were fed two meals daily of approximately 100 g of the test meals using the haem iron concentrate as fortifier, followed by 100 ml of apple juice containing 40 mg vitamin C. In the second group, ferrous sulphate was added as fortifier, followed by the same quantity of apple juice. A 7-day metabolic balance was performed in each baby to determine iron retention from the fortifier. To mark the stools of the balance period, the infants received a 200-mg oral dose or carmine (mixed into the meal) at the beginning and end of each collection period. Duplicate samples of every meal were collected for iron analysis. The milk intake of the breast-fed infants was determined by weighing the child before and after each nursing, using an electronic balance. Samples of breast milk were collected using breast pumps to analyse for iron concentration. Mothers were provided with iron-free diaper liners and disposable diapers (Proctor and Gamble, Newcastle-on-Tyne UK) for the duration of the study. These were collected daily from homes, except during the weekends when cool-boxes were provided, and then stored in a freezer at -18°C until analysis of the faeces for iron.
To determine whether or not the haem iron concentrate had an enhancing effect on nonhaem iron absorption, a single dose of 1.2 mg 57Fe as ferrous sulphate was mixed with the fortifier on days 2 and 3 of the study. This test meal was fed to all subjects except numbers 1, 2, 9, and 10 (who had begun the study protocol before this procedure could be performed). Infants were fed the 57Fe test meal after a minimum fast of 2 hours. They were allowed to drink only apple juice and water for 2 hours after the meal. Any uneaten food was collected and analysed to calculate the exact dose of isotope consumed.
Duplicate diets, faeces and breast milk were analysed for total iron by thermal ionization quadruple mass spectrometry (8). Stools were separated from the diaper (while still frozen) and autoclaved. Food samples and stools were freeze dried and ground to fine powder. Bulked samples were prepared by mixing the whole 7-day collection in a homogeniser for 30 minutes.
Iron in Food and Faeces
Subsamples of dried food (2 g) and stools (1 g) were spiked with an enriched isotope solution of 54Fe (0.123 mg Fe/ml; 99.4% 54Fe) and then ashed in porcelain crucibles at 480°C for 48 hours. The residue was dissolved in 5 ml concentrated HCl (12 mol/1; Aristar), heated in a hot plate until dry, and redissolved in 5 ml 4 M HCl (Aristar). The solution was applied to an acid-washed 100-mm column packed with a resin 200-400 mesh (AG 1-X8; Bio-Rad Laboratories, Richmond, CA. U.S.A.) previously washed with 60 ml 2 M HNO3 in an airtight system using a peristaltic pump. The resin was regenerated to the chloride form using 60 ml 4 M HCl, and then the samples were loaded. The columns were washed again with 30 ml 4 M HCl, and then the samples were eluted with 0.5 M HCl and collected in 1.5-ml polyethylene microtubes. The tubes were dried in a laminar flow cabinet under a 1-kw lamp. The residue was redissolved in 0.050 ml 0.2 M HNO3 and analysed by thermal ionization quadruple mass spectrometry.
Iron in Breast Milk
Approximately 10 ml of breast milk was spiked with the same enriched isotope solution of 54Fe used for food and faecal samples, and 4 ml concentrated HNO3 (Aristar) was added to the mixture and stored overnight in a loosely capped vial. The mixture was transferred to a Teflon (duPont, Wilmington, DE, U.S.A.) digestion vessel, capped, and heated in a microwave oven (MDS 2000; CEM Microwave Ltd., Buckingham, U.K.) in three stages of 20, 40, and 60 psi, during a total period of 40 minutes. The vessels were cooled, 0.5 ml H2O2 (Aristar) was added and heated again at the same conditions of pressure and time. The digest was transferred to polytetrafluoroethylene beakers, the acid evaporated in a laminar flow-fume cupboard, and the residue redissolved in 5 ml 4 M HCl (Aristar). The remaining minerals were removed by anion exchange, as described above, and the residue analysed for total iron by thermal ionization quadruple mass spectrometry.
The iron concentration of duplicate diets and breast milk comprised a mixture of the iron originating from the 54Fe-enriched solution and the Fe from endogenous sources (natural abundance). Faecal samples also contained iron from the 57Fe-enriched meal. The isotopic ratio R (54/56Fe) in faecal iron is arrived at by the following equation: where A is isotopic abundance, W is average molecular weight, x is weight of all isotopes of natural Fe, y is weight of all isotopes of 54Fe and z is weight of all isotopes of 57Fe. Similar equations were derived for R (57/56Fe). The values of R and y were known, so by combining the two simultaneous equations, values for x and z could be calculated.
Apparent iron retention was calculated from the difference between total iron intake and total iron excretion and 57Fe absorption calculated from the difference between isotope dose and faecal excretion.
Iron retention data were analysed by analysis of variance with variables including sex, type of fortifier, and milk type. The correlations between iron retention and hemoglobin concentration, mean cell haemoglobin, birth weight, body weight, and iron intake were examined by regression analysis using a statistical program (GENSTAT (9)). Similar analyses were performed with the 57Fe absorption data as the variable.
The characteristics of the infants fed with the two fortifiers, haem iron concentrate and ferrous sulphate, are summarised in Table 1. There were no differences in weight or in any of the haematologic indices between the groups. The 7-day balance data for the haem concentrate and ferrous sulphate groups are shown in Table 2. One of the infants in group 2 had a negative iron balance. The analysis of variance showed that there was no significant difference in iron retention between the groups. There was a large interindividual variation in iron retention in both groups that was not dependent on iron status, sex, or body weight (p > 0.05). Iron intake from breast milk is shown in Table 3. Mean iron intake in infants fed breast milk (8.6 ± 1.2 mg/day) was significantly lower than in infants fed formula (14.7 ± 2.8 mg/day; p < 0.05; Fig. 1). The amount of iron retained was related to iron intake (Fig. 2), with a coefficient of regression of (b = +1.135; p < 0.05; SE 0.350). Iron excretion was also influenced by iron intake (b = +1.007; p < 0.05; SE 0.205).
Haem concentrate did not enhance the absorption of 57Fe administered as ferrous sulphate solution (Table 2). However, the number of subjects was too small to draw any firm conclusions. The chemical composition of the haem iron concentrate is presented in Table 4.
Maintaining positive iron balance in infancy is a major problem. It is a period of rapid growth, and the infant must be provided with a diet containing iron that can be well absorbed once iron stores are depleted (10). Many manufactured infant foods are fortified with the iron salt ferrous sulphate that is cheap and bioavailable, but its high solubility may cause the production of coloured compounds in fortified foods (11). Conversely, some forms of iron that are stable during storage, such as iron orthophosphate and iron pyrophosphate, are so poorly absorbed that they are ineffective (12). In this study, the haem concentrate was evaluated as an alternative iron fortifier. No significant difference was found between dietary iron retention in infants fed this compound (mean, 28.8%) and those fed the well-absorbed salt ferrous sulphate (mean, 24.4%; Table 2). The use of haem concentrate by the infant food industry could provide iron that is highly bioavailable and in addition could contribute to protein intake (13). The protein content of the haem iron concentrate was 60% (Table 4).
An improvement in the iron status of school children participating in a nationwide program for 2 years in which they were fed bovine haemoglobin-fortified cookies was reported by Walter et al. (14) Higher haemoglobin concentrations and fewer cases of low iron stores (expressed as low serum ferritin values) were found in these children compared with that control groups.
The wide variation in retention among individual infants in both groups may be related to the behaviour of the intestinal mucosa and almost certainly reflects individual iron requirements (15). In this study iron retention was linearly related to iron intake (Fig. 1), showing that it is possible to increase the absolute amount of iron absorbed by increasing iron intake. The same linear correlation has been reported by Bothwell et al. (16) for ferrous and ferric iron salts, and by Hallberg et al. (17) for haem iron. These authors reported a decrease in the percentage absorption with increasing dose, but no clear correlation between these parameters can be deduced from our data. The most important conclusion is that iron retention can be improved by increasing the total iron in the diet.
Iron retention, calculated as a total (milligrams) or as a percentage, was unrelated to haemoglobin concentration. That there is no correlation between body iron levels and iron absorption has been reported in previous studies in infants (8,18), suggesting that there is no simple correlation between the efficiency of iron absorption and body iron in infants.
All the infants (except subject 3) received iron in excess of the reference nutrient intake for 6 to 12 months of age (7.8 mg/day) and there was a significantly higher iron intake in the formula-fed infants (Fig. 2) than in the breast-fed infants. Although the iron in breast milk has been reported to be much better absorbed than iron in formulas (19), breast milk contains only 0.3 mg iron/l at 4 to 6 months and after the age of 6 months, the amount of iron in breast milk is insufficient to meet physiological requirements (20). The iron content of breast milk in our study ranged from 0.3 mg/l to 0.6 mg/l (Table 3) and contributed less than 3% of the total iron intake. In the United Kingdom, the iron content of infant formulas, fortified with ferrous sulphate, ranges from 5 to 7 mg/l. The amount of iron provided by both fortifiers under evaluation was relatively small compared with the total iron intake in formula-fed infants, because of its high iron concentration, and this may have masked any difference in total iron retention.
The 57Fe absorption results indicated that the haem iron concentrate did not enhance the absorption of nonhaem iron. No significant differences were found between isotope absorption in the two groups. The enhancing effect of meat on nonhaem food iron absorption was first reported by Layrisse et al., (21) but the mechanism of this observation is still unclear. It has been suggested that meat may stimulate the production of gastric acid, thereby promoting iron solubilization within the stomach. The secretion of gastric acid is dependent on the amount of food protein or amino acids and on the protein source. Alternatively, meat may enhance nonhaem iron absorption by forming peptide complexes, the iron from which is efficiently taken up into the mucosal cell. The haem iron concentrate used in our study was prepared from haemolysed erythrocytes from pigs and has a high protein content (60%). The present results suggest that the meat factor is not related to blood proteins, because these do not enhance nonhaem iron absorption. Thus, the enhancing effect must be attributed to other meat proteins. The positive correlation between percentage 57Fe absorption and iron retention indicates that the efficiency of absorption from the whole diet can be predicted from 57Fe absorption from a test meal.
The results obtained predict a promising future for the haem concentrate as a fortifier in infant foods. Further studies are needed to clarify the technical feasibility of using this product.
Acknowledgment: The authors thank the research nurse, Sheila Sills, for assisting with the blood collection and Angela Twaite for her help with the test meals; Hero Baby S.A., Alcantarilla, Murcia, Spain, for supplying the weaning foods and Aprocat S.A. Granollers, Spain, for the haem iron concentrate; and all the mothers and infants who took part in the study.
This work was sponsored in part by Biotechnology and Biological Sciences Research Council, United Kingdom. C.M. was funded by the European Community, Agriculture and Agroindustry programme.
1. Grasbeck R, Majuri R, Kouvonen I, Tenhunen R. Spectral and other studies on the intestinal haem receptor of the pig. Biochim Biophys Acta
2. Lynch SR, Dassenko SA, Morck RA, Beard JL, Cook JD. Soy protein products and heme iron absorption in humans. Am J Clin Nutr
3. Heinrich HC, Gabbe EE. Hemoglobin iron for prophylaxis and treatment of iron deficiency. Klinische Wochenschrift
4. Solomons NW, Jacob RA. Studies on the bioavailability of zinc in humans: Effects of heme and nonheme iron on the absorption of zinc. Am J Clin Nutr
5. Hallberg L, Sölvell L. Absorption of hemoglobin iron in man. Acta Med Scand
6. Calvo E, Hertrampf E, de Pablo S, Amar M, Stekel A. Haemoglobin-fortified cereal: An alternative weaning food with high iron bioavailability. Eur J Clin Nutr
7. Stekel A. Fortification from the laboratory to a national program. Two Chilean experiences.
In: Proceedings of the Annual Meeting of the International Nutritional Anaemia Consultative Group. Section 7. Nutrition Foundation: New York, 1981.
8. Fairweather-Tait S, Fox T, Wharf G, Eagles J. The bioavailability of iron in different weaning foods and the enhancing effect of a fruit drink containing ascorbic acid. Pediatr Res
9. Payne RW, Lane PW, Ainsley AE, et al. Genstat 5 Reference Manual. Clarendon: Clarendon Press, 1987.
10. Dallman PR, Siimes MA, Stekel A. Iron deficiency in infancy and childhood. Am J Clin Nutr
11. MacPhail AP, Bothwell TH. Fortification of the diet as a strategy for preventing iron deficiency. Acta Paediatr Scand Suppl
12. Dallman PR. Iron deficiency in the weanling: A nutritional problem on the way to resolution. Acta Pediatr Scand Suppl
13. Reizenstein P. Cattle haemoglobin-a possible dietary iron supplement. Br J Haematol
14. Walter T, Hertrampf E, Pizarro F, et al. Effect of bovinehemoglobin-fortified cookies on iron status of schoolchildren: A nationwide program in Chile. Am J Clin Nutr
15. Kuhn IN, Monsen ER, Cook JD, Finch CA. Iron absorption in man. J Lab Clin Med
16. Bothwell TH, Charlton RW, Cook JD, Finch CA. Iron metabolism in man
. Oxford, UK: Blackwell Scientific, 1979.
17. Hallberg L, Bjorn-Rasmussen E, Howard L, Rossander L. Dietary heme iron absorption. A discussion of possible mechanisms for the absorption-promotion effect of meat and for the regulation of iron absorption. Scand J Gastroenterol
18. Kastenmayer P, Davidsson L, Galan P, Cherouvrier F, Hercberg S, Hurrell RF. A double stable isotope technique for measuring iron absorption in infants. Br J Nutr
19. Schulz-Lell G, Dorner K, Oldigs HD, Sievers E, Schaub J. Iron availability from an infant formula supplemented with bovine lactoferrin. Acta Paediatr Scand
20. Department of Health. Weaning and the weaning diet
. Report on Health and Social Subjects No 45. London: Her Majesty's Stationery Office, 1995.
21. Layrisse M, Martinez Torres C, Roche M. Effect of interaction of various foods on iron absorption. Am J Clin Nutr