No epidemiological evidence for infant vaccinations to cause allergic disease

Abstract

Context: The prevalence of allergic diseases has increased considerably over the last decades. The hygiene hypothesis has emerged, linking reduced microbial exposure and infections early in life with the development of allergic diseases. Especially some of currently available non-replicating infant vaccines are unlikely to mimic a natural infection-mediated immune response that protects against the development of allergic diseases. Moreover, several studies suggested infant vaccinations to increase the risk of allergic diseases. Objective: To determine whether infant vaccinations increase the risk of developing allergic disease. Data Sources: We searched MEDLINE from 1966 to March 2003 and bibliography lists from retrieved articles, and consulted experts in the field to identify all articles relating vaccination to allergy. Study Selection and Data Extraction: We selected epidemiological studies with original data on the correlation between vaccination with diphtheria, pertussis, tetanus (DPT), measles, mumps, rubella (MMR) and Bacillus Calmette-Guérin (BCG) vaccine in infancy and the development of allergic diseases, and assessed their quality and validity. Data Synthesis: Methodological design and quality varied considerably between the studies we reviewed. Many studies did not address possible confounders, such as the presence of lifestyle factors, leaving them prone to bias. The studies that offer the stronger evidence, including the only randomized controlled trial at issue published to date, indicate that the infant vaccinations we investigated do not increase the risk of developing allergic disease. Furthermore, BCG does not seem to reduce the risk of allergies. Conclusions: The reviewed epidemiological evidence indicates that, although possibly not contributing to optimal stimulation of the immune system in infancy, current infant vaccines do not cause allergic diseases.

Introduction

The prevalence of allergic diseases such as atopic asthma, hay fever and atopic eczema has increased considerably over the past decades [1], [2], [3], [4], [5], [6]. The largest increase was reported from the United States, where self-reported prevalence rates for asthma increased by 75% from 1980 to 1994 [7]. While critical appraisals have shown the difficulty of defining objective measures of asthma and allergic disease in childhood, the increase nevertheless appears to be real [8], [9], [10]. Furthermore, allergic diseases are most common in Western, industrialized countries [11]. In 1989, Strachan proposed an explanation for the rising prevalence of allergic disease [12]. He postulated that microbial exposure and infections in early childhood prevent the development of allergic diseases, a proposition known as the hygiene hypothesis [12]. Since then, many epidemiological findings have supported this theory. For instance, living in a large family, attending day care early in life, and growing up on a farm or in a family with an anthroposophic lifestyle have shown to reduce the risk of asthma and allergies [13], [14], [15], [16], [17]. Recently, being exposed to food-borne and oro-fecal infections at early age was found to be associated with a decreased prevalence of hay fever and asthma [18]. These findings emphasize that living conditions as mentioned above probably predispose to infection and exposure to microbial structures, rather than being direct determinants of allergic disease [14]. In the previous century, improved sanitation and clean drinking water reduced the burden of many infectious diseases. However, a modern hygienic Western lifestyle may contribute to the development of atopic disease by depriving the developing immune system of essential microbial stimuli [18], [19]. The identification of T-helper 1 and 2 lymphocyte subsets (Th1, Th2) with their typical cytokine patterns provided an immunological mechanism underlying the hygiene hypothesis. A predominance of Th2-type mediated responses is implicated in the development of allergic disease [20]. In newborns, the immune response is skewed away from a Th1 profile [21], reflecting the Th2 conditions of pregnancy [22]. Rapid suppression of Th2 responses and deviation towards a Th1 lineage can be seen in non-atopic infants [23]. Infants with genetic predisposition for the development of atopy can be programmed to consolidate Th2 responses to allergens within the first 6 months of life [21], [23], [24]. However, early exposure to certain viral respiratory infections, and to commensals and pathogens from gut flora appears to drive maturation of Th1-type responses [25], [26], [27].

Vaccination with current infant vaccines may contribute to limited induction of Th1 responses [28]. After completing a primary immunization series of acellular Bordetella pertussis vaccine (aP), mixed Th1–Th2 responses were found [29], whereas recipients of whole-cell vaccine (wP) showed only a Th1-type response [30]. Responses to tetanus toxin (TT) after a primary series of diphtheria, tetanus and acellular pertussis (DTaP) vaccination showed a steady Th2-type response that decreased after priming, but resurged after the age of 1 year [31]. After measles vaccination, infants showed reduced Th1 cell responses [32], [33], which may reflect immunosuppression similar to that seen after natural measles infection in children [34]. Infants vaccinated at birth with Bacillus Calmette-Guérin (BCG) vaccine develop Th1-type immune responses to purified protein derivative (PPD) [35], whereas a Th2-type responses predominate in unvaccinated infants [36]. Early BCG vaccination, however, does not drive immune responses to other, unrelated vaccines towards a Th1-type response [37]. Vaccination against hepatitis B at birth, and at 2 and 4 months of age, with concomitant BCG vaccination at birth increased not only Th1 activity, but also a Th2-type response to HBsAg, compared to non-BCG vaccinated controls. BCG vaccination at birth followed by DTP vaccination at the ages of 2, 3, and 4 months also slightly but significantly, increased a Th2-type response to TT when compared to controls [37]. Therefore, it seems that the influence of BCG vaccination on the immune response is more complex than just a mere polarization towards Th1 responses. In general, T cell responses following vaccination of human infants are characterized by limited Th1 responses although, contrary to murine responses, preferential Th2 polarization is uncommon [28]. While the causes of the increasing prevalence of allergic diseases remain to be further uncovered [14], infant vaccinations were suggested to increase the risk of allergic disease [38], [39]. Consequently, the question is whether healthy children vaccinated at early age with current vaccines have increased risk of development of allergic disease. The aim of this review is to answer this question based on the best available epidemiological evidence.