American J Clinical Nutrition 2009 (Dec); 90 (6): 1693–1698 ~ FULL TEXT
Kristin Marmsjö, Helen Rosenlund, Inger Kull, Niclas Håkansson, Magnus Wickman, Göran Pershagen, and Anna Bergström
Institute of Environmental Medicine,
BACKGROUND: Multivitamins are frequently consumed by children, but it is unclear whether this affects the risk of allergic disease.
OBJECTIVE: We sought to study the association between multivitamin supplementation and allergic disease in 8-y-old children.
DESIGN: Data were obtained from a Swedish birth cohort study. Information on lifestyle factors, including use of vitamin supplements, environmental exposures, and symptoms and diagnoses of allergic diseases, was obtained by parental questionnaires. In addition, allergen-specific IgE concentrations of food and airborne allergens were measured in blood samples collected at age 8 y. A total of 2423 children were included in the study. The association between use of vitamin supplements and the selected health outcomes was analyzed with logistic regression.
RESULTS: Overall, no strong and consistent associations were observed between current multivitamin use and asthma, allergic rhinitis, eczema, or atopic sensitization at age 8 y. However, children who reported that they started taking multivitamins before or at age 4 y had a decreased risk of sensitization to food allergens (odds ratio: 0.61; 95% CI: 0.39, 0.97) and tendencies toward inverse associations with allergic rhinitis. In contrast, there was no consistent association among children who started to use multivitamins at or after age 5 y.
CONCLUSION: Our results show no association between current use of multivitamins and risk of allergic disease but suggest that supplementation with multivitamins during the first years of life may reduce the risk of allergic disease at school age.
From the FULL TEXT Article:
The escalating rates of asthma and allergic disease seen in the past decades are most evident in industrialized countries and have recently been proposed to partly be a consequence of changed dietary habits characterized by a fall in consumption of fish, fresh fruit, and vegetables rich in antioxidants and minerals [1, 2]. Antioxidants — for example, vitamin C, vitamin E, vitamin A, and selenium — have been the most widely studied nutrients with regard to allergic diseases, either as individual nutrients or in analyses assessing fruit and vegetable intakes.
A protective role of antioxidants in allergic disease is plausible because of their ability to scavenge free radicals generated by the inflammatory response with can exacerbate the disease process [2, 3]. Moreover, epidemiologic studies support an association between antioxidants and asthma as well as allergic disease [4–7].
The consumption of dietary supplements among children has become quite common [8, 9].
Data from the 1999–2002 National Health and Nutrition Examination Survey (NHANES) show that >30% of children in the United States take dietary supplements regularly, with the most frequent use among 4- to 8-y-old children . Multivitamins and multiminerals were the most commonly used supplements (18%). Likewise, a Swedish national survey performed in 2003 showed that 33% of the children consumed vitamin supplements, most often multivitamins . Information on dietary supplement intake in relation to allergy in children is scarce; however, current evidence indicates that supplementation with antioxidants may have an influence on asthmatic symptoms  and lung function in asthmatic children exposed to high levels of ozone or air pollution . The relation between vitamin supplement use and health status still remains unclear for most dietary supplements but excessive consumption could be of questionable benefit [12–15].
In our study of 2423 Swedish 8-y-olds, there was no strong and consistent association between use of any vitamin supplement during the past 12 mo and allergic disease. Likewise, no association was observed when multivitamin supplementation was analyzed separately. However, for children who reported that they started to use multivitamins at age =4 y, we observed an inverse association for sensitization to food allergens and tendencies toward inverse associations for allergic rhinitis. In contrast, there was no consistent association among children who started to use multivitamins at age =5 y. The use of vitamin supplements was common, ie, 40% reported having taken any vitamin supplements during the past year. Also, children who used supplements displayed more hereditary traits of allergy, had parents with higher socioeconomic status, and more often had symptoms of eczema during the first year of life when compared with nonusers.
The strengths of the present study included the large number of participants, limited loss to follow-up, and detailed assessment of phenotypes as well as of exposure, including vitamin supplements. However, misclassification of exposure might affect our results. We only asked for the frequency of consumption during the past 12 mo, and parents of vitamin supplement users were asked to remember when the child first started taking the supplements. If parents of children with allergic diseases recall previous supplement use differently from other parents, it may give rise to recall bias. The use of dietary supplements among children as reported in a FFQ has not been validated to our knowledge. FFQs have, however, proven to be adequate when ranking and assessing dietary macronutrient, vitamin D, and calcium intakes in children [20, 21]. For adults, on the other hand, a brief questionnaire can accurately and reproducibly capture data on supplement use for frequently consumed products, but it may perform less well for products used less often or more intermittently [22, 23].
The cross-sectional design, which used both exposure and outcome information obtained at 8 y of age, is a drawback of this study. In epidemiologic reports on allergic diseases in childhood, disease-related modification of exposure may be a major bias, which is not always easy to control for and may lead to misinterpretation of the results. This has been discussed in several articles investigating behavioral factors, such as breastfeeding , vaccination , pet ownership , and consumption of specific food items  in relation to allergic diseases. Therefore, we cannot exclude a disease-related modification of exposure, ie, that the observed risk estimate was affected by the fact that children started to take dietary supplements (or avoided to do so) because of their allergic symptoms, rather than because of a true effect of dietary supplements on allergic diseases. We tried to investigate this by grouping individuals according to age at the start of multivitamin consumption and by adjusting for early symptoms of wheeze and eczema in the multivariate analyzes. These adjustments had no major influence on the observed ORs among children who started to use multivitamins at age 4 y or earlier, whereas the nonsignificant increments in risk observed for allergic rhinitis, eczema, and any allergic disease among children who started to consume multivitamins at age =5 y were somewhat attenuated. Furthermore, it should be noted that most of the data on other risk factors used in our analyses, such as breastfeeding and parental smoking habits, where obtained before onset of disease.
Multivitamins are the most commonly purchased and regularly used supplement by adults and children [8, 28, 29]. Despite this, there is limited evidence on multivitamin use in relation to allergic disease in children. Milner et al  reported an increased risk of asthma among infants who consumed multivitamin supplements during the first 6 mo of life, although restricted to black infants. However, they found no association between multivitamin use at 3 y of age and asthma. Likewise, in a Norwegian study, intake of vitamin supplements during the first year of life was associated with a borderline significant increased risk of sensitization at school age . In our study there was no increased risk of allergic diseases among children who started to use multivitamins at an early age. Instead, we observed an inverse association with sensitization to food allergens in this group. There may be several explanations for the differences in results. First, the earlier studies investigated the effect of vitamin supplementation during the first year of life, whereas most of the children in the present study started to consume multivitamins after age 1 y. It should be noted, though, that 98% of the children in our study received supplementation of vitamins A and D from 2 wk of age and through the first years of life . Second, the composition, dose, and duration of the consumed supplements may differ between the studies. Third, differences in nutritional status between the populations might cause different responses to vitamin supplementation. Fourth, Milner et al did not consider disease-related modification of exposure in their analyses, whereas the Norwegian study excluded children with asthma during the first year of life; thus, their results are less likely to be affected by this potential bias.
An association between use of multivitamins and allergic disease is biologically plausible because of the antioxidant and immunomodulatory properties of certain vitamins. Many of the pathophysiological changes associated with allergic diseases are produced by activated inflammatory cells, which generate an excessive amount of oxygen free radicals. A postulated hypothesis is that antioxidants, because of their ability to quench free radicals, prevent chain reactions that could result in lipid peroxidation and damage to cell membranes, ie, of immune cells  or DNA, both of which have been suggested to be involved in the allergic disease process [30, 31]. On the other hand, it has been proposed that several antioxidant nutrients can act as pro-oxidants at higher doses through a reduction of transition metal ions . Vitamins have also been postulated to exert nonantioxidant immunomodulatory effects [33, 34]. Previous research has shown that a variety of vitamins commonly found in multivitamins can cause naive T cells to differentiate toward the extremes of the T helper type 1 and T helper type 2 phenotypes [3, 33], each characterized by different cytokine profiles, which leads to physiologic states that may increase the odds of an allergic response when encountering certain antigens [13, 14, 34, 35].
Overall, the scientific evidence regarding the beneficial or harmful effects of vitamin supplements in relation to allergic diseases appears contradictory. In previous studies in which the effect of single nutrient supplementation has been investigated and found not to be effective, it has been suggested that perhaps a combination of antioxidants would have given different results , a finding not supported by the present study. Rather, dose and duration of supplementation seem to be important conditioning factors [37, 38]. Furthermore, Feary and Britton  propose that perhaps vitamin supplements only work in nutritionally deplete populations and that no additional beneficial effect will occur in well-fed and consequently oversupplemented individuals. The variability in individual susceptibility to oxidative stress may also explain the conflicting results between studies on antioxidant supplementation . Hence, the effectiveness of vitamin supplementation is still an unresolved matter that warrants rigorous scientific evaluation [38, 41, 42], and further exploration of the association between vitamin supplementation and allergic diseases is very important from a public health point of view.
In conclusion, our study of 2423 Swedish boys and girls showed no association between current multivitamin use and risk of allergic disease, but suggests that supplementation with multivitamins in early life may reduce the risk of allergic disease at school age. The cross-sectional nature of the data implies that the findings should be interpreted with caution.
We express our gratitude to the BAMSE cohort participants, nurses, and research team, especially Stina Gustavsson, Eva Hallner, and André Lauber (Institute of Environmental Medicine, Karolinska Institutet, Stockholm, Sweden).
The authors’ responsibilities were as follows—MW (principal investigator of the BAMSE project) and GP: initiated the BAMSE project; KM, HR, IK, and AB: planned the study; IK: supervised the data collection; KM, HR, IK, NH, and AB: prepared the data for analyses; KM and AB: performed the analyses; KM, HR, IK, and AB: interpreted the data; KM: drafted the manuscript; and HR, IK, NH, MW, GP, and AB: provided critical review of the manuscript. None of the authors had a conflict of interest related to the data in the manuscript.