Protection Against Atopic Diseases by Measles—A Rash Conclusion?

A major question facing modern medicine is why atopic diseases such as asthma, allergic rhinitis, and atopic dermatitis, which have in common the overproduction of allergen-specific or total IgE, are increasing in prevalence. Atopic diseases are influenced by genetic predisposition and environmental exposures. However, since the spontaneous genetic mutation rate is quite slow, it is clear that changes in the environment are responsible for the increased prevalence of atopic diseases. But which environmental factor(s) is the culprit?

Many theories, based on epidemiologic or experimental data, have been advanced, including improved hygiene^{1 - 2} ; changes in diet^{3 - 4} ; changes in intestinal flora due to increased use of antibiotics⁵ and altered patterns of infant feeding⁶ ; greater exposure to allergens⁷ ; obesity and reduced physical activity⁸ ; and changes in the prenatal environment.⁹ One interesting theory is that the reduction of early childhood infectious diseases, due to widespread vaccination of children or increased use of antibiotics, has led to an increased prevalence of allergic diseases. This theory, known as the "hygiene hypothesis," was advanced by Strachan,¹ who noted an association between large family size and reduced rates of allergy and asthma; this finding has been verified in several epidemiologic surveys.^{10 - 12}

To explain these findings requires an appreciation of the developing immune system. T helper cell (T_H2)–like cytokines (interleukin [IL] 4, IL-13, and IL-5) produced in the uterine environment induce similar T_H2-like responses in the immature immune system of the newborn, which increases the likelihood that postnatal (and possibly even prenatal) exposure to allergens will lead to production of allergen-specific IgE and eosinophilic inflammation.¹³ According to the hygiene hypothesis, infections with viruses and perhaps other intracellular organisms also influence the developing immune system: the T-cell responses to these infections generate T_H1-like cytokines such as IL-12 and interferon γ that down-regulate T_H2 responses. Repeated infections, through the activities of T_H1-like cytokines, would help T-cell immune responses to mature into a balanced phenotype that would be less likely to favor allergen sensitization in childhood. Following this line of reasoning, if the frequency of infections in infancy were reduced by multiple vaccinations or by limiting the exposure to infectious diseases, maturation of the immune system would be delayed and the risk of developing atopic diseases would be increased. The ironic implications of this theory are that the advances in public health that have been so successful in eliminating or drastically reducing the frequency of severe childhood infections, notably vaccinations, antibiotics, and indoor plumbing, might also be responsible for the significant increase in allergic diseases over the past 30 years.

This theory is attractive for several reasons. First, the increase in allergic diseases and the decrease in childhood infectious diseases have occurred during roughly the same period. Second, the worldwide prevalence of atopic diseases is unevenly distributed: the United States, Western Europe, Australia, and New Zealand have high rates of atopic diseases, whereas atopic diseases are less common in developing regions such as Eastern Europe, China, and India.¹⁴ In contrast to the low prevalence of atopic diseases, these developing countries have relatively high rates of serious infections in infants and children. Moreover, several epidemiologic studies have demonstrated inverse relationships between certain childhood infections, such as measles, mycobacteria, and hepatitis A, and the risk of atopic diseases.^{15 - 17} These studies add a new wrinkle to the hygiene hypothesis by raising the possibility that specific illnesses might reduce the subsequent risk of allergic diseases.

A study of the relationship between measles infection and subsequent atopy in a village of Guinea-Bissau stimulated considerable debate.¹⁶ The study subjects were children aged 6 years or younger who experienced an epidemic of measles in 1978-1980. During the epidemic, the World Health Organization established a campaign to vaccinate everyone who was not infected; consequently, excellent records were available regarding the history of measles and measles vaccination. Fifteen years later when these same individuals were reevaluated, those with a history of measles were found to be one third as likely to have positive allergen-specific skin test results. One criticism of this study was that 25% of the children younger than 3 years who contracted measles died of the disease, which raised the possibility that the apparent protective effect of the measles infection was due to greater mortality in allergy-prone individuals. However, the theory that measles infection could reduce the risk for atopic diseases also was supported by a cross-sectional study of Scottish schoolchildren,¹⁵ in which a history of measles after the age of 3 years was associated with reduced risk for allergic rhinitis. Together, these 2 studies were widely interpreted as evidence of a causal relationship between measles infection and a reduced risk of allergy.

In this issue of THE JOURNAL, a third and contrasting study of the long-term effect of measles infections is reported by Paunio and colleagues.¹⁸ The authors evaluated computerized medical records of a nationwide Finnish campaign to vaccinate children with an improved measles-mumps-rubella vaccine in 1982-1986. Although children of all ages were enrolled in the study, 2 age groups (14-18 months and 6 years) were targeted for vaccination in the public health campaign and constituted a large proportion of the database. At the time of vaccination, parents were asked whether the children had ever had measles and whether a health care practitioner had ever diagnosed allergic rhinitis, atopic dermatitis, or asthma in the child. In the cross-sectional analysis of 6-year-old children, which most cleanly excluded the confounding effects of age on the analysis, naturally acquired measles infection was associated with increased prevalence of asthma, and trends toward increases in allergic rhinitis and atopic dermatitis. This study has important limitations in that the rates of both measles and atopic diseases were estimated by parental recall, the order of the events was not determined, and there were no objective measures of atopy. Nevertheless, the results of this study, which are remarkable for the scope and size of the database, seem to contradict the other 2 studies of measles and atopy.

Can the findings in the 3 studies of measles and atopy be reconciled? In considering the apparent discordant findings, it may be helpful to revisit the original observation that larger family size is associated with reduced atopy. If the protective factor associated with larger families is rate of infections, the key factor might be the overall frequency of childhood illnesses, rather than any single infectious disease. When viewed in this light, specific infections such as measles or hepatitis A may represent markers of overall exposure to infectious diseases, rather than events that can singularly reduce the risk of atopic diseases. It is possible that measles infection is not a good indicator of overall exposure to infectious disease in a Western environment such as Finland.

Second, it should not be assumed that all infectious diseases have the same effects on the risk for allergies and asthma. For instance, an increasing number of studies indicate that lower respiratory tract infections, including respiratory syncytial virus bronchiolitis, pneumonia, perhaps pertussis, and measles in childhood may increase the subsequent risk of childhood asthma without modifying the likelihood of sensitization to allergens.^{19 - 21} One possible explanation for this effect is that severe lower respiratory tract infections either damage lung tissues or impair the development of the lung in such a way as to promote asthma. The timing of the infection or vaccination also may be relevant and have specific effects on immune system ontogeny, lung development, or both. In addition, the effects of bacterial and fungal infections, or even exposure, need to be considered in light of recent evidence that children who grow up in an agricultural lifestyle are at lower risk for developing respiratory allergies.²²

Epidemiologic studies comparing family size and infectious diseases with the prevalence of atopic disorders have initiated new areas of research regarding the relationships between lifestyle, the immune system, and the development of atopic disorders. What is needed are prospective studies to evaluate effects of childhood infections on development of the immune system, and prospective data on the effects of viral and bacterial infections on the synthesis of IgE and allergic inflammation. Further studies may lead to greater understanding of the pathogenesis of allergic disease and may provide new targets to stem the rising tide of allergic diseases.