Iron metabolism in children: Confounding factors

Abstract

Characterization of iron metabolism in infants and children may be confounded by the diverse effects of developmental, genetic, and acquired influences on iron metabolism and laboratory measurements of iron status, especially in areas with intense perennial transmission of Plasmodium falciparum malaria. In the Pemba iron and folic acid supplementation trial, the coadministration of folic acid with iron is a further confounding factor. Because the design of the Pemba iron and folic acid supplementation study did not include a group that received iron supplementation without folic acid, the observed increase in serious adverse events cannot be ascribed unequivocally to iron alone, to folic acid alone, or to the combination of the two. In interpreting the results from the Pemba iron and folic acid supplementation trial, additional analyses of existing data from the trial and from earlier studies in the area could help clarify the roles of iron and folic acid.

Introduction

Characterization of iron metabolism in infants and children may be confounded by a variety of developmental, genetic, and acquired influences, especially in areas with intense perennial transmission of Plasmodium falciparum malaria.

Iron metabolism in infants and children

The concise account of iron metabolism provided by Lönnerdal and Kelleher [1] emphasizes the distinctive features of the handling of iron by infants and children. Iron absorption in young infants initially seems independent of iron intake or body iron stores and then, in older infants, gradually assumes a more adult-like pattern [2]. The ontogeny of iron homeostasis in infancy involves progressive modifications in the regulation, expression, and localization of the proteins involved in iron metabolism. The amount and form of iron presented to the upper small intestine are also important determinants of absorption. In some studies, iron supplements given to iron-replete infants impaired growth [3, 4]. The mechanisms are unknown but may involve compromise of zinc status, with effects on growth hormone and immune function [1]. Iron given as a food fortificant seems to avoid these adverse effects.

Hepcidin, a small antimicrobial peptide principally produced by the liver, has been identified as the central regulator of iron homeostasis, controlling absorption from the intestine, export from macrophages, and release from body stores [5]. Hepcidin acts to decrease iron exit from cells (enterocytes, macrophages, hepatocytes) by binding to ferroportin, the iron export protein, leading to its internalization and degradation. The synthesis of hepcidin is decreased by anemia and hypoxemia but increased by inflammation and by iron overload.

The extent to which developmental changes in the regulation of hepcidin expression govern the pattern of iron absorption in infancy is still uncertain, but progress has been made in identifying the underlying signal transduction mechanisms. The multifunctional BMP/Smad signaling pathway is involved (bone morphogenetic protein/protein Smad, named from a combination derived from two homologues, the Caenorhabditis elegans protein SMA and the Drosophila protein MAD [mothers against decapentaplegic]) [6–9], providing a link to a variety of developmental processes. BMPs are cytokines that are members of the transforming growth factor-β (TGF-β) superfamily and have a variety of roles in control of cell proliferation, differentiation, and apoptosis. A parallel interleukin-6 (IL-6)/signal transducer and activator of transcription-3 (STAT3) pathway is important for transcriptional regulation of hepcidin expression in response to inflammation and infection [10]. Further examination of the interactions of these, and probably other, transduction pathways affecting hepcidin expression is needed for a more detailed understanding of the regulation of iron absorption in infancy and childhood.

In areas with intense perennial transmission of P. falciparum malaria, malarial infection has profound and complex effects on erythropoiesis and iron metabolism [11, 12]. In many of the areas with P. falciparum malaria, gene variants that affect the structure or function of erythrocytes are common, apparently because these have provided a survival advantage. These variants include the thalassemias and hemoglobinopathies as well as polymorphisms of genes for red cell enzymes and membrane proteins [13–15]. Haptoglobin genotypes also seem to be involved in iron homeostasis [16] and to influence the risk of anemia in malarious areas [17]. The Q248H mutation in ferroportin is associated with iron accumulation in adults and might provide some protection against iron deficiency in African children [18].

In populations living in areas with perennial transmission of P. falciparum, our ability to interpret laboratory measurements of iron status is limited by the confounding effects on iron homeostasis of malaria itself, other infections, developmental changes in infancy and childhood, and the influences of polymorphisms for genes affecting red blood cell structure and function [14, 19–21].

Mechanisms of iron toxicity

Hershko [22] emphasizes the importance of non-transferrin-bound plasma iron (NTBI) [23] in the pathogenesis of a variety of toxic effects of iron. As elaborated by Ratledge [24], the availability of NTBI could favor the growth of a variety of microorganisms. Consequently, if given in a highly bioavailable form that produces plasma iron concentrations that exceed the plasma transferrin iron-binding capacity, administration of supplemental iron might favor infection by microbial pathogens. In contrast, as Hershko describes in detail [23], the growth of P. falciparum in infected red blood cells seems to be independent of host iron status. Effects of supplemental iron on the course of malaria are more likely to be the result of other effects of iron, such as altering the balance between pro- and antiinflammatory cytokines during the course of malarial infection.

The natural history of juvenile hemochromatosis provides perspective on the importance of the control of plasma iron in determining susceptibility to infection. Juvenile hemochromatosis is most often caused by a mutation in HJV, the gene for hemojuvelin (on chromosome 1q21[25]). The remaining cases have been identified as the result of mutations in HAMP, the gene encoding hepcidin (on chromosome 19q13 [26, 27]). Urinary hepcidin concentrations are greatly decreased [25] or absent [28]. Having little or no hepcidin-mediated regulation of iron recycling and absorption, plasma iron is increased, with near or full saturation of plasma transferrin, and severe iron overload develops by the second decade of life [29]. Nonetheless, at least in the environments in the developed countries where these patients have been examined, no increased risk of infection has been reported. Patients with thalassemia major and transfusional iron overload do have an increased susceptibility to infection with certain organisms (including Yersinia enterocolitica and Klebsiella spp.), but splenectomy, a variety of immune abnormalities related to blood transfusion, treatment with a siderophore (deferoxamine), and other factors [30] complicate interpretation of the role of increased plasma iron.

The Pemba iron and folic acid supplementation trial

In the Pemba iron and folic acid supplementation trial [31], the coadministration of folic acid with iron is a further confounding factor, especially because of the use in the study of an antifolate as antimalarial therapy. Because the design of the Pemba iron and folic acid study did not include a group that received iron supplementation without folic acid, the observed increase in serious adverse events cannot be ascribed unequivocally to iron alone, to folic acid alone, or to the combination of the two. As described by Metz [32], growth of some strains of P. falciparum may be enhanced by host folate. Folic acid supplementation may increase the risk of both malarial infection and of treatment failure associated with administration of the antifolate antimalarial sulfadoxine-pyrimethamine [33]. In interpreting the results from the Pemba iron and folic acid trial supplementation trial, further analyses of existing data from the trial and from earlier studies in the Pemba area could help elucidate the roles of iron, folic acid, and antifolate antimalarial treatment.

Reexamination of antimalarial therapy given in the Pemba iron and folic acid supplementation substudy

To ensure that the antimalarial sulfadoxine–pyrimethamine therapy administered to the Pemba substudy participants did not contribute to the observed differences in rates of hospitalization and death, reexamination of the study data would be helpful to determine whether the number and timing of antimalarial treatments were evenly allocated among the three study groups. At least 487 courses of sulfadoxine–pyrimethamine were administered by the study group to the 2,413 substudy children for “confirmed malaria” during the trial [31]. Because of the long elimination half-life of the sulfadoxine–pyrimethamine combination [34], a single treatment has a prolonged effect. Consequently, differences in the number or timing of the antifolate antimalarial treatments could have had a profound influence on the risk of adverse events related to malaria.

Moreover, effects of iron and folic acid supplementation on the behavior of children conceivably could influence the likelihood of receiving antimalarial treatment. In Pemba, an earlier study found that over 80% of children 4 to 71 months of age were infected with P. falciparum throughout the year. In children less than 30 months old, about half had a parasite density > 5,000/mm³ [35] during the 12 months of the study. In other words, the majority of children in Pemba had persistent parasitemia, and approximately half would have had a high (> 5,000/mm³) parasite count during the course of a year. In the Pemba substudy, any child with a parasite count above 5,000/mm³ and an axillary temperature above 37.5°C was give a dose of sulfadoxine–pyrimethamine [31]. It seems possible that behavioral effects of treatment of iron deficiency could influence the likelihood of receiving antimalarial therapy. For example, with placebo treatment, anemic and iron-deficient children could have been more likely to be withdrawn and apathetic. As a consequence, a febrile illness would have been less likely to be noticed by their caregivers, and these children would have been less likely to receive medical attention. By contrast, if treated with iron and folic acid, anemic and iron-deficient children might then have been more alert and active, fussing and complaining with a febrile illness, and more likely to have been taken for treatment. Given the high prevalence of parasitemia, antimalarial therapy might well have been given preferentially to the anemic and iron-deficient children treated with iron and folic acid both for malaria and for other febrile illnesses that coincided with a high parasite count.

In the Pemba trial, the investigators state that “we did not find any association of recovery from sulfadoxine–pyrimethamine-treated malaria episodes or subsequent recrudescence with iron and folic acid versus placebo (unpublished data)” [31]. Given the evidence elsewhere in Tanzania of the rapid emergence of resistance to sulfadoxine–pyrimethamine [36] and of studies finding that folic acid supplementation may increase the risk of treatment failure with this antifolate antimalarial [37–40], a full presentation of these data would be valuable.

Reanalysis of a previous study in Pemba of iron supplementation without folic acid

Between 1996 and 1997, an earlier study in Pemba [35, 41, 42] examined the effects of 12 months of low-dose (10 mg/day) iron supplementation without folic acid in a community-based sample of 614 children 4 to 71 months old, using a double-blind, placebo-controlled, randomized design. Although this smaller study did not examine clinical malarial outcomes, iron supplementation had no overall effect on either the prevalence of malaria infection or parasite densities. At baseline and at the end of the 12-month trial, both hemoglobin and zinc protoporphyrin were measured in samples of venous blood. Accordingly, the effects of iron supplementation without folic acid on the prevalence of malaria infection and on parasite densities could be examined with respect to iron status and anemia. As an example, if a lower prevalence of malarial infection, lower parasite densities, or both were found only in the “iron-deficient and anemic” group that received iron, then this reanalysis would provide additional evidence supporting a strategy of targeting iron supplementation to children with evidence of iron deficiency.

In addition, in the 2002–2003 Pemba iron and folic acid supplementation substudy, after 12 months of supplementation, venous blood samples were available for 308 children in the iron and folic acid group without zinc and for 327 children in the placebo group. Comparisons with the similarly sized iron (without folic acid) and placebo groups in the 1996–1997 Pemba study would permit an assessment of the contribution of folic acid to the hematological response to supplementation. If no significant beneficial effect were found, this result would provide further evidence for excluding folic acid from universal supplementation of infants and children, as suggested by Metz [32], i.e., in areas with a high prevalence of anemia, intense perennial malarial transmission, and limited medical facilities using antifolate antimalarials as first-line therapy.

These additional examinations of existing data would be invaluable in clarifying the contributions of iron, folic acid and antifolate antimalarial treatment to the pathogenesis of the adverse events reported in the Pemba iron and folic acid supplementation trial.