Secondary Logo

Journal Logo

Original Articles

Meat Intake and Iron Status in Late Infancy: An Intervention Study

Engelmann, Mads D. M.; Sandström, Brittmarie; Michaelsen, Kim F.

Author Information
Journal of Pediatric Gastroenterology & Nutrition: January 1998 - Volume 26 - Issue 1 - p 26-33
  • Free


Iron deficiency is one of the most abundant micronutrient deficiencies worldwide. One of the major risk groups is infants in late infancy. In healthy, term infants, iron endowment at birth will protect the infant from iron deficiency during the first 4 to 6 months. After the fourth month of life, increasing body size begins to deplete iron stores, and the infant becomes increasingly dependent on exogenous sources of iron. Despite efficient absorption and use of iron in breast milk, the absolute amount of iron absorbed from breast milk is less than that required during the second half of the first year, and exclusive breast-feeding beyond 6 months of age is associated with an increased risk of developing iron deficiency(1-3). The consequences of iron deficiency in infancy can be serious, causing impaired immune function(4,5) and delayed cognitive development(6).

Prevention of iron deficiency can be achieved by iron fortification of infant foods or by medicinal iron supplements. Uncritical iron fortification may, however, have adverse effects on the utilization of other minerals, particularly zinc and copper (7-9), and compliance with the administration of medicinal iron is often poor(10). Alternative ways of ensuring adequate iron nutrition in infants have not been explored in detail. Regular consumption of a weaning diet with highly bioavailable iron may preserve iron status during late infancy. Highly bioavailable iron is found in animal tissue protein(meat, fish, and poultry), which contains a proportion of iron in the heme form that is easily absorbed. There are also unidentified factors in meat that promote heme (11,12) and nonheme iron absorption(11,13,14). However, intervention studies examining the effect of meat intake on iron status parameters in infancy are sparse. A higher meat intake would also improve zinc intake, which is desirable because study results suggest zinc deficiency or low intake of zinc in late infancy (15,16). A potentially less beneficial consequence of increased meat consumption could be increased protein intake; several negative effects of high protein intake have been suggested (17-19).

To investigate the effect of increased meat intake on hemoglobin concentration (Hb), serum ferritin (SF), and serum transferrin receptors (TfR) during weaning, we conducted an intervention study on 8-month-old, partially breast-fed infants. Partially breast-fed infants were considered a risk group because results of other studies have shown that prolonged exclusive breast-feeding is associated with an increased risk of iron deficiency(1-3). In addition to iron status parameters, we evaluated the effect of an increased meat intake on serum zinc concentration, growth, and illness during the 2-month intervention period.


The study was a blind, parallel intervention study lasting 2 months beginning in September 1994. Fifty-five healthy, term infants were recruited at 8 months of age from the municipalities of Frederiksberg and Copenhagen. The infants fulfilled the following inclusion criteria: term birth, birth weight of 2,500 g or more, no neonatal disease or malformation, no severe current illness, and partially breast-fed at the time of the recruitment. Infants were classified as partially breast-fed as long as they were breast-fed at least once a day. Stratified randomization with two strata (sex, hemoglobin) was used to ensure that infants in each of the two intervention groups had similar characteristics. Examinations including medical history, anthropometry, and blood sampling were performed on the first and last days of the intervention. Iron deficiency anemia was defined as Hb <105 g/l and SF<10 μg/l. The study was approved by the Ethics Committee for Copenhagen and Frederiksberg.


Infants were randomized into two groups: a low-meat group (LMG) receiving a diet with a meat content aimed at the average found in an observational study of infants from the same area (20), and a high-meat group(HMG) receiving a diet aimed at a meat content approximately three times higher than that of the diet in the LMG. Purées and sandwich fillings were prepared at the department and distributed weekly to the parents. The parents could choose from seven purées containing vegetables and muscle-based foods (beef, pork, lamb, turkey, and cod) each prepared in a low-meat and a high-meat version (Table 1). The remaining diet (milk, cereals, bread, fruits), which did not contain meat, was chosen freely by the parents. Energy density (Table 1) was kept constant (≈3.9 kj/g) between the low-meat and the high-meat purées by replacing energy from meat with energy from vegetables, corn oil, and corn starch in the low-meat purées. The sandwich fillings were products typically consumed by Danish infants, and each group had a choice of five fillings: cheese spread, fruit spread, cod roe, and fish pâté in the LMG, and cold cuts (ham and turkey), chicken sausage, liver pâté, and fish pâté in the HMG.

On the first day of the intervention, all parents received an electronic scale with a precision of 1 g (Soehnle attaché, Montlingen, Switzerland) and were instructed in keeping food records by a trained dietitian. The same dietitian kept weekly contact with the parents to ensure proper food registration. The parents recorded total dietary intake once a week by 24-hour weighed food records. Nutrient intake was calculated using the Dankost computer program (version 1.3), developed by the Danish National Food Agency. The number of breast-feedings per 24 hours was recorded at the start and the end of the intervention.

Blood Sampling and Laboratory Analyses

The blood sampling procedure was planned as 2 ml venous blood from a cubital vein, but because of difficulties in venipuncture, 37 infants had finger-prick and 13 had venipuncture; in the remaining 5 infants, blood sampling was not successful. The number of infants who had venipuncture in the LMG (n = 7) was not significantly different (p = 0.9) from the number in the HMG (n = 6). Because mean Hb, SF, and TfR levels in finger-prick blood were not significantly different from mean Hb, SF, and TfR levels in venous blood, results were pooled in the statistical analysis.

Hemoglobin concentration and mean corpuscular volume were measured on the day of sampling. Serum was stored at -80°C for the other analyses, which were made within the following 6 months. All analyses were performed in duplicate, and mean values were used in the statistical analysis. Mean corpuscular volume and Hb (Drabkin's method) were determined on a Cobas Minos-ST (Hoffmann-LaRoche; Nutley, NJ, U.S.A.). Serum concentrations of ferritin and TfR were determined by commercial enzyme immunoassay kits: NovaPath ferritin determination kit (Bio-Rad; Segrate, Italy) calibrated against WHO-1 st international standard (IS) 80/602, and TfR determination kit(Ramco Laboratories; Houston, TX, U.S.A.). Serum zinc was determined after dilution 1:4 with deionized distilled water by atomic absorption spectrophotometry (Model 360, Perkin Elmer, Norwalk, CT, U.S.A.). Levels in reference serum (Seronorm, Nyegaard; Oslo, Norway) were measured simultaneously and were within certified values. Standards were prepared from standard solution (Titrisol, Merck; Darmstadt, Germany). In 32 of the included infants, blood smears were made and differential counts done manually. In the remaining 9 infants, blood smears were not performed because of insufficient amounts of blood in their samples.

Portions of freeze-dried purées were analyzed in duplicate for energy, nitrogen, zinc, and iron. Energy content was analyzed by combustion calorimetry (calorimeter system C 4000, IKA, Heitersheim, Germany). Nitrogen content was determined by N2 combustion (Dumas method, NA 1500 Carlo Erba, Fisons Instruments, Italy). Protein content was calculated from total nitrogen content, using the conversion factor 6.25 g protein: 1 g nitrogen. Zinc and iron were determined by atomic absorption spectrophotometry(21).


Body weight was measured to the nearest 1 g with an electronic scale(Sartorius IP 65, Göttingen, Germany). Crownheel length was measured to the nearest 0.1 cm with an electronic measuring board (Force Institutes, Glostrup, Denmark). Mid-upper-arm circumference was measured with a nonstretchable fiberglass tape to the nearest 0.1 cm and triceps skinfold with Harpenden skinfold calipers (CMS Weighing Equipment LTD, London, UK) to the nearest 0.1 mm.


Determination of the extent of illness was based on parental observation and was classified by number of episodes with otitis media, pneumonia, catarrhal inflammations, asthmatic bronchitis, urinary tract infections, gastroenteritis, sore throat, and number of fever episodes. The number of episodes from birth to the 8-month termination, during the intervention, and during the week previous to the 8- and 10-month exterminations were registered. In addition, an objective evaluation of current illness was made on the day of examination.

Statistical Analysis

The Wilk-Shapiro test and the normal probability plot were used to assess whether data were compatible with a normal distribution. The paired Student'st-test and Wilcoxon two-sample test were used, where appropriate, to test for differences in mean or median values between groups. A pairedt-test was used to test for changes in hematologic parameters in infants within groups. Fisher's exact test was used to test for differences in proportions. Because of skewed distribution, SF was logarithmically transformed. The association between change in hematologic values and initial concentrations were analyzed using Pearson correlation coefficients. The effects on change in hematologic values of group relationship and episodes of illness were evaluated by analysis of covariance. All analyses were performed using SAS Statistical Software (22).


Of the 55 infants recruited, 1 infant was withdrawn because the parents did not want to participate in the second examination, 5 were excluded because blood sampling was not successful, and 8 infants were excluded because Hb was below 105 g/l at the start of the study. The 8 infants with initial Hb below 105 g/l were treated with medicinal iron. In the remaining 41 infants, there were no significant differences between the groups in sex distribution, anthropometric measurements, and hematologic values (Table 2). Infants in the LMG had significantly longer birth length (p = 0.04) and higher birth weight (p = 0.04) than infants in the HMG(Table 2). However, there were no correlations between birth weight or birth length and hematologic parameters.

Food Intake

Food consumption was divided into eight dietary categories(Table 3). Mean meat intake was significantly higher in the HMG (26.7 g/day) than in the LMG (10.1 g/day; Table 3). In Denmark, formula and gruels are the only iron-fortified products. There was a tendency (p = 0.06) toward a higher formula intake in the HMG (52.3 g/day) compared with that in the LMG (22.3 g/day). However, this difference was counteracted by a significantly (p = 0.03) higher intake of iron-fortified gruel in the LMG (60.8 g/day) than in the HMG (17.2 g/day). For the remaining dietary categories no significant differences were observed (Table 3). Initial median frequencies of breast-feeding were 4 times/day (range, 0-7 times) in the LMG and 3 times/day(range, 1-15 times) in the HMG. In two infants from the LMG, breast-feeding was stopped in the period between recruitment and start of the intervention. At the end of the study frequencies were 2 times/day in both groups (range, 0-7 times in the LMG and 0-10 times in the HMG). There were no significant differences in frequency of breast-feeding between the two groups at the beginning or at the end of the study. Infants with high breast-feeding frequencies had low intakes of weaning food. In the HMG one infant, who was breast-fed 15 times a day at the start of the intervention, had a mean total weaning food intake of only 49.6 g/day (Table 3).

Nutrient Intake

The contribution of iron and protein from the meat category was, as expected, significantly higher in the HMG compared with that in the LMG(Table 3). However, the LMG consumed significantly more iron from gruel and significantly more protein from both the gruel and the bread and cereals category (Table 3). Consequently, there were no significant differences in total iron and protein intake between the groups. Furthermore, there were no significant differences between the HMG and the LMG in total food intake and in intake of energy, fat, carbohydrates, vitamin C, calcium, zinc, and vitamin B12(Tables 3, 4).


At the beginning of the intervention, no significant differences were found in Hb, SF, TfR and serum zinc concentrations between the HMG and the LMG(Table 2). At the beginning and at the end of the intervention, mean corpuscular volume was within normal limits (70-86 fl) in all infants except 1, who had microcytosis at the beginning of the study. All of the 32 infants who had blood smears at the beginning and the end of the study had normal differential counts, and no hypersegmented nuclei were observed in the polymorphonuclear leukocytes.


Analysis of the differences in Hb concentration between the two examinations (ΔHb) showed that the LMG had a highly significant decrease in Hb (ΔHb ± SD = -4.9 ± 5.4 g/l; p = 0.0008). In the HMG the reduction in Hb concentration was insignificant (-0.6 ± 4.9 g/l; p = 0.58). Furthermore, the decrease in Hb was significantly greater in the LMG compared with that in the HMG(Table 5).

Serum Ferritin, Serum Transferrin Receptors, and Transferrin Receptor-Ferritin Ratio

There was no significant difference between the two groups in the proportion of infants with depleted iron stores-that is, with SF less than 10μg/l (Table 6)-and no infant had iron deficiency anemia at the beginning or the end of the study. The final SF and TfR values and the transferrin receptor-ferritin ratio (TfR:SF) were not significantly different between the groups. Neither group relationship nor hemoglobin concentration was significantly associated with change in SF (ΔSF), change in serum TfR (ΔTfR) or change in transferrin receptor-ferritin ratio(ΔTfR:SF). Furthermore, ΔHb was not significantly correlated withΔSF, ΔTfR, or ΔTfR:SF. To exclude abnormal elevations of SF caused by infection, the analyses were repeated after exclusion of infants who showed any sign of illness 7 days before blood samples were collected. This exclusion had only minor effects on p values.


Changes in the following anthropometric measures: weight, length, arm circumference, and triceps skinfold thickness were not significantly associated with group relationship (Table 5). One infant from the LMG lost 1 kg during the intervention but weighed 14.7 kg at 8 months.


The number of episodes of illness during the intervention (otitis media, pneumonia, catarrhal inflammations, asthmatic bronchitis, urinary-tract infections, gastroenteritis, sore throat, and number of fever episodes) were pooled into one variable in the statistical analysis. This variable was not significantly different (Fisher's exact test; p = 0.2) in the LMG(mean = 4) and the HMG (mean = 3), and had no impact on change in hematologic values when analyses of covariance were performed with illness and group relationship as predictor variables.

Serum Zinc

The final serum zinc concentrations were not significantly different between the groups, and group relationship was not significantly associated with change in serum zinc (ΔS-zinc) (Table 5). No significant correlations were found between iron status parameters andΔS-zinc.


In the current study, infants who were still breast-fed at 8 months maintained their Hb concentration during the 2-month intervention period if they consumed a diet containing 27 g meat per day, whereas infants consuming a diet with 10 g meat per day had a significant decrease in Hb concentration. Ten grams of meat per day is close to the average meat intake found in an earlier study of weaning diet in Denmark (20).

In contrast to the significant changes found in Hb level, there was no association between meat intake and change in SF concentrations. It has been hypothesized that if changes in SF are to be used as a measure for iron absorption, it is a prerequisite that the infants have their individual optimal Hb concentration (23). If this criterion is not fulfilled, the additional absorbed iron will be used preferentially for hemoglobin synthesis rather than for iron stores. The partially breast-fed, 8-month-old infants included in this study were considered a risk group for suboptimal Hb concentrations, because prolonged breast-feeding is associated with an increased risk of developing iron deficiency(1-3). Thus, in the partially breast-fed 8-month-old infants, one could expect a high prevalence of infants with suboptimal Hb concentrations and a restoring of Hb in preference to an increase in SF concentrations.

The interpretation of SF as an index of iron status or iron ingestion has not been sufficiently evaluated in infancy, and SF may not be a reliable parameter for the transition from optimal to suboptimal iron status in infancy. Thus, infants with suboptimal Hb concentrations may have SF values within the normal reference range. In addition, it has been proposed that the use of SF as a measure of iron stores may require a steady state(23), a condition that is far from fulfilled in late infancy.

Previous intervention studies in infants have focused on the effect of meat in combination with conventional iron fortification(24,25), which made the interpretation of the effect of meat per se on iron status difficult. In a case-control study in twelve 36-month-old children, a low heme iron intake was associated with a risk of low SF concentrations (26). In adults, results of cross-sectional studies show a positive association between meat consumption and SF (27-29) level. However, a recent controlled feeding study produced results indicating a negative association between SF values and meat intake, and the investigators suggested that SF may not be a sensitive indicator of iron ingestion (30).

The finding of suboptimal Hb concentrations in infants without definite iron deficiency (SF less than 10 μg/l) agrees with results of several other studies of iron supplementation of infants. Findings in these studies have shown similar increases in Hb concentrations in subjects with no iron deficiency anemia (31-34).

Low Hb values can be caused by a number of conditions besides iron deficiency. In healthy Danish term infants, iron deficiency is likely to be the most important nutritional factor leading to anemia. However, folate, vitamin B12, and copper deficiencies could also result in anemia. Folate and vitamin B12 deficiencies can be seen in premature infants: The anemia is megaloblastic, and hypersegmentation of the nuclei of polymorphonuclear leukocytes is nearly a rule (35). In the current study, there were no premature infants, none had megaloblastic anemia, and no hypersegmentation was found in the 32 infants on whom blood smears were performed. Anemia related to copper deficiency is very rare, and there is generally a predisposing cause (antecedent malnutrition, malabsorption, prematurity, or dietary insufficiency) often caused by copper-deficient total parenteral nutrition (36), none of which were seen in this study. The anemia of copper deficiency is a sideroblastic, microcytic, hypochromic anemia, which cannot be distinguished from iron deficiency anemia per se; but in copper deficiency, neutropenia is common (36). Neutropenia was not observed in any of the 32 infants on whom blood smears were performed; thus, it is unlikely that the maintenance in Hb concentration seen in the HMG can be ascribed to nutritional factors other than iron. Furthermore, meat is not a major source of folate, vitamin B12, or copper.

Growth and illness also affect Hb concentration. Total body iron-the sum of circulating iron, iron in myoglobin, and the quantity of storage iron-is closely related to the body weight of the infant; consequently, the need for iron in infancy is closely related to growth. However, in this study, changes in anthropometric measures were not significantly different in the two groups. Low Hb concentrations can also be caused by chronic infections and inflammations (37). Great care was taken in including only healthy infants and in evaluating illness during the intervention period. There was no significant difference between the two groups in the total number of episodes of illness during intervention, and this parameter had no impact on change in Hb when group relationship was taken into consideration. Thus, neither growth nor differences in episodes of illness can explain the significant differences in change in Hb concentration between the LMG and the HMG.

Bearing these considerations in mind and the fact that meat never has been shown to stimulate erythropoiesis per se, iron is the most likely nutritional factor responsible for the maintenance of the Hb level in the HMG. The decrease in Hb level in the LMG is in agreement with results of other longitudinal studies of healthy infants receiving no medicinal iron from any source (2,38). In these studies, there is a primary decline in hemoglobin concentration during the first 2 months of life and a secondary decline between 8 and 10 months of life. The primary decrease may be related to suppression of the bone marrow as a result of the increased oxygenation of the blood after birth. The secondary decrease is probably related to iron deficiency (39). The fact that there was no significant difference in total iron intake between the LMG and the HMG implies that absorption of iron was better in the HMG than it was in the LMG. Difference in iron absorption between the LMG and the HMG could be explained by different intake of iron absorption inhibitors and iron enhancers. Major inhibitors of iron absorption include calcium, polyphenols, and phytic acid(40). There was no significant difference between the LMG and the HMG in calcium intake. The most important sources of polyphenols in infant diet are plant foods-that is, fruits and vegetables. Intake of plant foods was not significantly different between the two groups during the intervention. Phytic acid is of widespread occurrence in cereal grains and legume seeds (bread, cereal, and gruel). However, the gruel consumed in the current study was based on rice and corn, which have a low content of phytic acid, and there was no significant difference in intake of bread and cereals. The major enhancers of iron absorption are vitamin C and animal tissue protein(10). In the current study, there were no differences in intake of vitamin C or total iron between the two groups. Thus, the iron-absorption-enhancing effect of animal muscle protein is the most likely explanation for the maintained Hb values, in the HMG.

The optimal blood-sampling procedure would have been venipuncture in all infants. Some study results have shown significant differences between hemoglobin concentration (41,42) and serum ferritin levels (43) in capillary and venous blood; others have found no differences (44,45). In the current study, mean Hb, SF, and TfR values from finger-prick blood were not statistically different from mean Hb, SF, and TfR values in venous blood. Furthermore, the number of infants with venipuncture was not significantly different in the two groups. Thus, in the statistical analysis, results from capillary and venous blood samples were pooled.

Meat has a high protein content, and an increased meat intake will result in a higher protein intake if the diet is otherwise unchanged. To our surprise there was no difference in total protein intake between the two groups. The HMG had a significantly higher protein intake from meat, but this was compensated by a significantly lower protein intake from gruel and the bread and cereals category (Table 3). The protein intake in this group of partially breast-fed infants was not high and is not likely to have any adverse effects.

Serum TfR are assumed to play an important role in the regulation of cellular iron need. An increase in circulating TfR has been regarded as an early sign of cellular iron deficiency (46). Several studies in adults have evaluated this novel index of iron status(47,48). Studies in infants are sparse(49), and the potential usefulness in infancy is far from established. In the current study, there was no association between meat intake and change in TfR receptor levels.

The serum TfR:SF ratio has been used to portray the entire spectrum of iron status, ranging from large iron stores to severe iron deficiency(46): Low values correspond to large iron stores and vice versa. As for TfR, the potential usefulness of the TfR:SF ratio in infancy remains to be established. In the current study, there was no significant difference between this ratio and group relationship.

Advice on intake of meat, fish, and poultry does not usually play a conspicuous role in the recommendations aimed at prevention of iron deficiency during infancy. However, results of the current study suggest that even a minor increase in meat intake can prevent a decline in Hb concentration in late infancy. Further research of the impact of diet composition, including meat intake on iron status parameters in late infancy, is highly warranted.

Acknowledgment: The study was supported by grants from the Danish Research and Development Programme for Food Technology and The Federation of Danish Pig Producers and Slaughterhouses.

Although this research was partly funded by The Federation of Danish Pig Producers and Slaughterhouses, the contract specified that investigators had total control over data collected and that the Federation would not have any input into publications arising from this project.


1. Siimes MA, Salmenperä L, Perheentupa J. Exclusive breast-feeding for 9 months: Risk of iron deficiency. J Pediatr 1984;104:196-9.
2. Saarinen UM. Need for iron supplementation in infants on prolonged breast feeding. J Pediatr 1978;93:177-80.
3. Haschke F, Vanura H, Male C, et al. Iron nutrition and growth of breast- and formula-fed infants during the first 9 months of life.J Pediatr Gastroenterol Nutr 1993;16:151-6.
4. Dhur A, Galan P, Hercberg S. Iron status, immune capacity and resistance to infections. Comp Biochem Physiol 1989;94:11-9.
5. Hershko C. Iron and infection. In: Fomon SJ, Zlotkin SH, eds. Nutritional anemias. Nestlé nutrition workshop series. Vol. 30. New York: Vevey/Raven Press, 1992:53-64.
6. Walter T. Early and long-term effect of iron deficiency anemia on child development. In: Fomon SJ, Zlotkin SH, eds. Nutritional anemias. Nestlé nutrition workshop series. Vol 30. New York: Vevey/Raven Press, 1992:81-90.
7. Yip R, Reeves JD, Lönnerdal B, Keen CL, Dallman PR. Does iron supplementation compromise zinc nutrition in healthy infants?Am J Clin Nutr 1985;42:683-7.
8. Haschke F, Ziegler EE, Edwards BB, Fomon SJ. Effect of iron fortification of infant formula on trace mineral absorption. J Pediatr Gastroenterol Nutr 1986;5:768-73.
9. Barclay SM, Aggett PJ, Lloyd DJ, Duffty P. Reduced erythrocyte superoxide dismutase activity in low birth weight infants given iron supplements. Pediatr Res 1991;29:297-301.
10. Fomon SJ. Iron. In: Fomon SJ, ed. Nutrition of normal infants. St. Louis, MO: Mosby-Year Book, 1993:239-60.
11. Martinez-Torres C, Layrisse M. Iron absorption from veal muscle. Am J Clin Nutr 1971;24:531-40.
12. Hallberg L, Bjørn-Rasmussen E, Howard L, Rossander L. Dietary heme iron absorption. A discussion of possible mechanisms for the absorption-promoting effect of meat and for the regulation of iron absorption. Scand J Gastroenterol 1979;14:769-79.
13. Monsen ER, Hallberg L, Layrisse M, et al. Estimation of available dietary iron. Am J Clin Nutr 1978;31:134-41.
14. Cook JD, Monsen ER. Food iron absorption in human subjects. III. Comparison of the effect of animal proteins on nonheme iron absorption. Am J Clin Nutr 1976;29:859-67.
15. Walravens PA, Chakar A, Mokni R, Denise J, Lemonnier D. Zinc supplements in breastfed infants. Lancet 1992;340:683-5.
16. Michaelsen KF, Samuelson G, Graham TW, Lönnerdal B. Zinc intake, zinc status and growth in a longitudinal study of healthy Danish infants. Acta Paediatr 1994;83:1115-21.
17. Axelsson I, Borulf S, Abildskov K, Heird W, Räihä N. Protein and energy intake during weaning. III, effects on plasma amino acids. Acta Paediatr Scand 1988;77:42-8.
18. Axelsson IEM, Ivarsson SA, Räihä NCR. Protein intake in early infancy: Effects on plasma amino acid concentrations, insulin metabolism, and growth. Pediatr Res 1989;26:614-7.
19. Rolland-Cachera M-F, Deheeger M, Bellisle F, Sempé M, Guilloud-Bataille M, Patois E. Adiposity rebound in children: A simple indicator for predicting obesity. Am J Clin Nutr 1984;39:129-35.
20. Michaelsen KF, Milman N, Samuelson G. A longitudinal study of iron status in healthy Danish infants: Effects of early iron status, growth velocity and dietary factors. Acta Paediatr 1995;84:1035-44.
21. Sandström B, Almgren A, Kivistö B, CederbladÅ. Effect of protein level and protein source on zinc absorption in humans. J Nutr 1989;119:48-53.
22. The SAS Statistical Software 1989 SAS/STAT Users Guide. Version 6. 4th ed. Vols. 1 and 2 1989.
23. Hallberg L. Absorption of iron from diet and supplements[in Swedish]. Scand J Nutr 1995;39:20-6.
24. Bergmann R, Bergler H, Moshoudis E, Bergmann E, Lachmann E, Bergmann KE. Preventing iron deficiency in breastfed infants by suitable supplementary nutrition (beikost). A prospective, controlled study [in German]. Monatschr Kinderheilkd 1989;137:775-9.
25. Haschke F, Pietschnig B, Vanura H, et al. Iron intake and iron nutritional status of infants fed iron-fortified beikost with meat.Am J Clin Nutr 1988;47:108-12.
26. Mira M, Alperstein G, Karr M, et al. Haem iron intake in 12 - 36 month old children depleted in iron: Case-control study. BMJ 1996;312:881-3.
27. Worthington-Roberts BS, Breskin MW, Monsen ER. Iron status of premenopausal women in a university community and its relationship to habitual dietary sources of protein. Am J Clin Nutr 1988;47:275-9.
28. Leggett BA, Brown NN, Bryant SJ, Duplock L, Powell LW, Halliday JW. Factors affecting the concentrations of ferritin in serum in a healthy Australian population. Clin Chem 1990;36:1350-5.
29. Davis CD, Malecki EA, Greger JL. Interactions among dietary manganese, heme iron, and nonheme iron in women. Am J Clin Nutr 1992;56:926-32.
30. Hunt JR, Gallagher SK, Johnson LK, Lykken GI. High- versus low-meat diets: Effects on zinc absorption, iron status, and calcium, copper, iron, magnesium, manganese, nitrogen, phosphorous, and zinc balance in postmenopausal women. Am J Clin Nutr 1995;65:621-32.
31. Lundström U, Siimes MA, Dallman PR. At what age does iron supplementation become necessary in low-birth-weight infants?J Pediatr 1977;91:878-3.
32. Saarinen UM, Siimes MA. Serum ferritin in assessment of iron nutrition in healthy infants. Acta Paediatr Scand 1978;67:745-51.
33. Driggers DA, Reeves JD, Lo EYT, Dallman PR. Iron deficiency in one-year-old infants: Comparison of results of a therapeutic trial in infants with anemia or low-normal hemoglobin values. J Pediatr 1981;98:753-8.
34. Parks YA, Aukett MA, Murray JA, Scott PH, Wharton BA. Mildly anaemic toddlers respond to iron. Arch Dis Child 1989;64:400-1.
35. Lindenbaum J. Megaloblastic anaemia and neutrophil hypersegmentation. Br J Haematol 1980;44:511-3.
36. Shaw JCL. Copper deficiency in term and preterm infants. In: Fomon SJ, Zlotkin SH, eds. Nutritional anemias. Nestlé nutrition workshop series. Vol. 30. Vevey/Raven Press, 1992:105-17.
37. Gibson RS. Assessment of iron status. In: Gibson RS. ed.Principles of nutritional assessment. Oxford, UK: Oxford University Press, 1990:349-76.
38. Burman D. Haemoglobin levels in normal infants aged 3 to 24 months, and the effect of iron. Arch Dis Child 1972;47:261-71.
39. Burman D. Iron requirements in infancy. Br J Haematol 1971;20:243-47.
40. Hurrell RF. The relevance of food iron bioavailability to iron nutrition. Lebensmittel-Technologie 1996;29:250-8.
41. Mollison PL. Some normal blood findings in newborn infants: Indication for simple transfusion. In: Mollison PL, ed. Blood transfusion in clinical medicine. 3rd ed. Oxford, UK: Blackwell Scientific Publications, 1961:578-95.
42. Moe PJ. Hemoglobin, hematocrit and red blood cell count in “capillary” (skin-prick) blood compared to venous blood in children. Acta Paediatr Scand 1970;49-51.
43. Pootrakul P, Skikne BS, Cook JD. The use of capillary blood for measurements of circulating ferritin. Am J Clin Nutr 1983;37:307-10.
44. Burman D. Anaemia in infancy-is iron important?Proc R Soc Med 1971;64:579-80.
45. Segall ML, Heese Hd, Dempster WS. Estimation of serum ferritin in blood obtained by heelstick. J Pediatr 1979;95:65-7.
46. Skikne BS, Flowers CH, Cook JD. Serum transferrin receptor: A quantitative measure of tissue iron deficiency. Blood 1990;75:1870-6.
47. Kohgo Y, Niitsu Y, Kondo H, et al. Serum transferrin receptor as a new index of erythropoiesis. Blood 1987;70:1955-8.
48. Flowers CH, Skikne BS, Covell AM, Cook JD. The clinical measurement of serum transferrin receptor. J Lab Clin Med 1989;144:368-77.
49. Lönnerdal B, Hernell O. Iron, zinc, copper and selenium status of breast-fed infants and infants fed trace element fortified milk-based infant formula. Acta Paediatr 1994;83:367-73.

Hemoglobin; Protein intake; Serum ferritin; Serum transferrin receptors; Serum zinc

© Lippincott-Raven Publishers