The novel proposition of a causal link between the quantity of iron intake in the first 4 months of life and future development of T1DM is plausible considering the known strong association between ferritin and Type 2 diabetes in adults [38
], a disorder sometimes diagnostically indistinct from T1DM with similar surrogate autoimmune markers and medical histories. Because in genetically susceptible subjects β-cell auto antibodies may materialize in infancy [40
], this observation buttresses the argument that the environmental trigger for diabetes autoimmunity occurs early in life.
The physiological anemia of infancy seen in the first 2-3 months of life is not due to iron deficiency and does not respond to iron supplementation [35
]. A normal term infant has sufficient iron stores for the first 4-6 months of life. Hence, the introduction of iron fortified formula in the critical period of 4-6 months could be potentially baneful. There has been a significant reduction in the prevalence of iron deficiency anemia in 6 to 60 month-old children after the introduction of iron-fortified baby formulas. However, there is limited clinical evidence which can be adduced to support the early introduction of iron fortified formulas in the first 4 months of life.
Although the mechanism remains unresolved, numerous studies have confirmed that prolonged breast feeding protects against T1DM [23
], and breast feeding activists advocate the postponement of formula feeding. As regards to iron intake with breast feeding, even though 50% of the iron is absorbed, which is incontestably a high proportion when compared with cow's milk or iron fortified formula, breast milk per unit volume contains roughly 1/10th
of the iron content of the "low" iron formulas [27
]. The net result is that breast fed babies absorb quantitatively far less iron than formula-fed babies, and this fact alone may explain the putative effect of breast milk.
Another confounding variable is that breast fed babies may delay feeding formula, cow's milk, cereals or other solid foods. This practice may avert exposure of the infant to foreign food antigens, such as gluten and bovine insulin, which may induce a cross-reacting immune response, particularly if gut permeability is altered by microbial infection [41
]. Contrary to other studies [35
], our study did not show that the number of weeks of breast milk ingestion conferred protection, most likely due to the small sample size in our study. Moreover, only a minority of children were on low-iron formula and the majority of children who had low total iron intake in the first 4 months had at some point also received breast milk. The absence of statistically significant association between the number of weeks of breast milk intake and having diabetes in our study is not enough to abandon the tenet that breast feeding is a protective factor against T1DM.
Type 1 diabetes involves the inexorable destruction of β cells secondary to autoimmunity. In iron overload conditions such as hemochromatosis, there is excessive iron storage in solid organs and selective destruction of β cells. Iron overload form hereditary hemochromatosis is associated with diabetes and impaired glucose tolerance and decreased insulin secretory capacity, which is reportedly reversed with phlebotomy therapy [42
]. Excess pancreatic β-cells iron reportedly impairs mitochondrial function and glucose-stimulated insulin secretion [43
The etiology of T1DM is likely multifactorial with a complex interplay between genetic, autoimmune, environmental, infectious and dietary factors. The role of iron in the infant formula in manifestation of T1DM was explored in this study. We could not find any studies examining relationship of iron status before onset of T1DM in genetically susceptible individuals. However, as in other autoimmune disorders, there are reports suggesting iron status may play a role in manifestation of systemic lupus erythematosis in murine models [45
]. Iron is involved in the proliferation of T cells and other cells of immune system [46
]. Similarly, ferritin reportedly has immune regulatory roles as well [47
]. The pathoetiology by which fortified iron formulas could predispose to development of autoimmune diabetes in genetically susceptible individuals is uncertain. Whether the development of T1DM appertains to the actual quantity of iron ingested or is somehow attributable to the type of the fortified iron (ferrous sulfate vs. ferrous fumarate), or duration and timing of iron intake is likewise unclear. Excess cellular uptake of ferrous iron could in theory initiate a oxidation cascade which overwhelms the numerous intracellular protective anti-oxidation biochemical strategies [48
]. It is thereby conceivable that the oxidative stress from immoderate iron exposure in the first 4 months in genetically susceptible infants may promote autoimmunity due to overproduction of free radical oxidative species. Parenthetically, ferrous fumarate is thought to produce minimal or no oxidative stress in vitro compared with ferrous sulfate [34
]. And, interestingly, lactoferrin in breast milk reportedly attenuates iron-induced oxidative damage in vitro[34
]. Finally, in the gut an overabundance of iron may modify the gastrointestinal immune response or its micronutrient milieu [49
]. In conclusion, large-scale prospective studies are warranted to determine whether low iron or lactoferrin enriched formulas forefend against T1DM in high risk individuals.
Limitations of the study
The major methodological pitfall of this retrospective study is that the data are self-reported. Hence, there is a potential chance of error insofar as this information relies on maternal dietary recall over a long duration. A memory bias may exist between that of the child with diabetes and non- diabetic sibling and that could have caused differential misclassification. Even though the letter accompanied the survey did not specify our hypothesis on iron content of formula, it mentioned we are investigating a link between T1DM and any of the minerals, vitamins, infant formulas or maternal nutrition. As the surveys were anonymous we do not have any data (clinical, demographic or medical) about the non-responders. We have excluded all ambiguous surveys. The choice of infant formula after birth depends on parental preferences, cost, co-morbid conditions like colic, acid reflux disease, lactose intolerance, milk protein allergy, type of formula the mother is first exposed too, etc. Consequently, the choice of low iron formula could be a proxy for socioeconomic characteristics or other factors which were not assessed in the surveys. Another confounding factor is that early introduction of bovine milk formula, besides increasing iron intake, may introduce previously proposed diabetogenic protein antigens.