A study of iron nutrition and immunity in infancy

Master Thesis


Permanent link to this Item
Journal Title
Link to Journal
Journal ISSN
Volume Title

University of Cape Town

Motivation and study design: Iron deficiency is a common condition in infancy, particularly in lower socio-economic groups. In Cape Town it remains a problem in spite of public health measures taken against it: a recent survey found a prevalence of iron deficiency anaemia of 34% in healthy 1-year old term infants who had ready access to a municipal health clinic where iron fortified milk formula is sold at subsidized prices. The consequences of iron deficiency extend beyond anaemia- to involve all organ systems including the immune system. Since Helen Mackay's report in 1928 of a striking decrease in incidents of infection in infants treated with iron, clinicians have assumed that iron deficiency predisposes to infection. Despite a sound theoretical basis for this belief, the clinical evidence for the assumption is poor as studies to date have displayed methodological deficiencies. On the other hand, iron is also essential for the growth of micro-organisms. As such, supplemental iron may predispose to infection. Indeed, there is much laboratory and clinical evidence to show that excess iron can result in the recrudescence of quiescent infections and increase the virulence of newly acquired infections. Thus, the competition between host and parasite may sometimes hinge on the relative availability of iron and it has been speculated that excess iron in infant milk formula may increase susceptibility to infectious diarrhoeal disease. The problem addressed by this thesis was to determine the utility of increasing the level of iron fortification of infant milk formula. Three questions were posed: Does increasing the level of iron fortification of conventional infant milk formula improve the iron nutrition of normal infants fed on the formula? Does increased iron fortification of infant milk formula alter immunity as reflected by incidence of infection and laboratory tests of immune function? Are there any handful effects of increasing the quantity of iron in conventional infant milk formula? A double blind randomized trial was carried out in 1983 and 1984 to answer these questions. A group of 149 healthy, well-nourished infants from a lower socio-economic community of so called Cape Coloureds were followed from the age of 3 months to 1 year. Half of the infants, the Control group, were given a commercially available infant milk formula (Lactogen Full Protein) which has 8.3 mg Fe/ 100 g formula and 37 mg ascorbic acid/ 100 g. The other half of subjects, the Test group, were given the same milk formula but fortified with iron to a concentration of 40 mg Fe/ 100 g. The children were examined every 3 or 4 weeks and any infection or history of infection was noted. Laboratory tests were done at the start of the trial and again on completion. During the trial, laboratory tests were performed only if clinically indicated. The tests included full blood count and differential analysis, red cell zinc protoporphyrin, plasma ferritin, plasma and hair zinc and lymphocyte subtyping with monoclonal antibodies. Within each group, half of the infants were randomly selected for assay of neutrophil bactericidal activity. The other half were assayed for lymphocyte blastogenic response to stimulation with phytohaemagglutinin. Tests of delayed cutaneous hypersensitivity to Candida antigen and PPD were done and all children and their mothers had antibodies to tetanus and polio determined. Results: 74 infants in the Control group started the trial and 62 completed it. In the Test group, 75 infants began and 70 completed the study. Intake of milk and solid foods was not quantified, but the ages of weaning and of introduction of new foods were determined. The Control and Test groups did not differ significantly on any test item. The mean age of completion of weaning was 3.60 months for the Control group and 4.04 months for the Test group. The Control group was first given meat or fish at a mean age of 5.19 months; the Test. group had meat or fish introduced to their diets at a mean age of 4.36 months. These differences were not statistically significant. The children in the Control group were lighter and shorter than the Test group at the end of the year. Mean standard deviation scores for weight were 0.23 and 0.48 respectively (P = 20%), while for length the SD scores were -0.13 and 0.06 (P = 20%).