The role of the farm environment and animal contact for the development of asthma and allergies.

Abstract

There is increasing interest in the role of the farm environment and more specifically of animal contact in influencing the development of asthma and allergies since the first observations of a reduced risk of hayfever, asthma and atopy in children from farming families, as compared to their peers from non-farming families, were published in this journal [1-3]. The reduction in risk associated with growing up on a farm was substantial, such that hayfever prevalence was reduced by between 37 and 72% in farmers' children. The findings demonstrate that there are still groups of individuals within developed western societies who are at a much lower risk of allergic diseases, the risk reduction being comparable to the ones observed between post-socialist eastern European and western industrialized countries [4, 5]. Hence, the farming environment provides a valuable source for studying environmental determinants for the development of asthma and allergies. The farms in Switzerland, Austria and southern Bavaria are typically small and run by the members of the family and only occasionally by farm workers. More than 95% of these farmers have cattle and other farm animals like pigs, sheep and poultry. Due to the climatic conditions with strong winters the cattle are usually kept in stables from late autumn to spring. It is therefore not clear whether this protective effect of the farming environment can be generalized or whether it is confined to the environmental conditions encountered on dairy farms. A recent study from Australia provided an initial answer to this question [6]. The risk of atopy was assessed in children from two rural towns, one in a mixed farming region with livestock farming (Wagga) and one in a crop farming region (Moree). In Wagga, the prevalence of atopy was significantly lower in farm children (adjusted OR 0.47, 95% CI 0.32–0.72) compared with non-farm children, no such effect was observed in the crop farming region. Thus, the presence of livestock seems to be an essential part for the observed protective ‘farm effect’. It has been suggested that the ‘farm effect’ may result from elevated exposures to bacterial compounds in the microbial environment of farm stables. Previous reports have indicated that several exposures are particularly high in stables such as exposures to moulds, ammonia, faeces, animal proteins, constituents of feed and endotoxin [7]. Endotoxin is an intrinsic part of the outer membrane of gram-negative bacteria, thereby activating Th1-type immune responses and thus interfering with the development of Th2-type immune responses resulting in atopy. This interpretation was originally based on the observation of a strong inverse, dose–dependent association between exposure to livestock and atopic disorders in the Bavarian study [3] and the finding that Austrian children in regular contact with farm animals, but not growing up on a farm, had a significantly lower prevalence of atopic sensitization than control children (13.5% vs. 34.8%) [2]. In a recent pilot study conducted in rural areas of southern Germany and Switzerland environmental endotoxin exposures were measured in homes of farmers' children, children with regular contact to livestock and control children with no contact to farms [8]. Samples of settled and airborne dust were collected in stables, and of settled dust indoors from kitchen floors and the children's mattresses. Endotoxin concentrations were determined by a kinetic Limulus assay. Endotoxin concentrations were highest in stables of farming families, but were also significantly higher indoors in dust from kitchen floors (143 EU/mg vs. 39 EU/mg, P < 0.001) and children's mattresses (49479 EU/m2 vs. 9383 EU/m2, P < 0.001) as compared to control children from non-farming families. In addition, endotoxin levels were also marginally higher in mattresses of children from non-farming families who had regular contact to farm animals (23340 EU/m2 vs. 9383 EU/m2, P = 0.12) as compared to control subjects. Endotoxin and other compounds of microbial origin have been demonstrated to regulate a variety of processes in the immune system such as production of IL-12 and IFN-γ, thereby counteracting allergic sensitization via the creation of a milieu that selects against Th2 cells [9-12]. The predominant type of response (Th1-like or Th2-like) to a given antigen is determined at the time of the primary encounter with the antigen, a life period during which the Th2-polarization characteristic of the fetal immune system is being progressively replaced by the Th1-dominance [13, 14]. Whether the timing of exposure to a farming environment might be important to confer protection from the development of asthma, hayfever and allergic sensitization has recently been investigated by the ALEX study group (ALEX being an acronym for ‘Allergy and Endotoxin’), an international cooperation of the research teams of Switzerland, Austria and southern Germany [15]. In a cross-sectional study performed in rural areas of Austria, Germany and Switzerland data for 812 children aged 6–12 including a parental questionnaire, a detailed home interview on the child's activities related to the farm environment during different life spans, and blood samples for determination of specific IgE were available for analyses. The risk for asthma ever, current asthma symptoms and atopic sensitization was reduced to about one-third if the child had first been exposed to stables during the first year of life as compared to first exposure to stables during school age or no exposure (Adj. OR for asthma ever 0.27 (95% CI 0.11–0.68), current wheeze OR 0.36 (95% CI 0.18–0.70), and atopic sensitization OR 0.33 (95% CI 0.19–58). First exposure to stables between the 2nd to 5th year of life as compared to later or no exposure was not associated with a significant risk reduction (Adj. OR 0.93 (95% CI 0.49–1.81), 0.73 (95% CI 0.40–1.31), and 0.87 (95% CI 0.53–1.42), for asthma ever, current wheeze and atopic sensitization, respectively). In addition, consumption of farm milk in the first year of life was independently associated with a risk reduction for asthma ever (Adj. OR 0.33 (95% CI 0.17–0.64), current wheeze (Adj. OR 0.50 (95% CI 0.30–0.85), and atopic sensitization (Adj. OR 0.25 (95% CI 0.15–0.39). These results suggest that the time window for the exposure to farming environmental factors seems crucial and, in addition, that different routes of exposure might be important, namely inhalation and/or swallowing. Studies in farming populations so far have appeared to show more consistent protection from hayfever and allergic sensitization than from asthma [16]. In this most recent study, however, exposure to stables in the first year of life was protective against asthma as well as allergic sensitization. The importance of the timing of exposure for the development of asthma and allergies is supported by results from experimental immunological studies. Experiments with rats exposed to aerosolized lipopolysaccharide (LPS) 1 day before, and 1, 4, 6, 8 and 10 days after intraperitoneal sensitization with ovalbumin showed that a single exposure to LPS up to 4 days after sensitization protected against production of specific IgE, bronchial hyper-responsiveness and inflammation in bronchial lavage fluid. However, after day four of sensitization, LPS inhalation had detrimental effects and led to increases in the allergic responses [12]. More anecdotecal evidence for the importance of early exposure to the farming environment can also be found in the recent Canadian study reporting reduced odds ratios for current wheeze and atopy in children associated with being raised on a farm [17]. The authors describe that children raised on family farms in Quebec are exposed in early life to environments such as dairy barns, pig sties, and aviaries and that most farm women participate actively in farm chores so that small children are frequently allowed to play in these buildings. In this issue of the journal, Barnes et al.[18] contribute interesting new findings to this debate which are based on a study on childhood allergy performed in an urban area (Heraklion) and in rural areas of Crete. Atopy was twice as common among urban children as compared to rural ones (19.6% vs. 9.6%). Among urban children the authors found significant gradients in the prevalence of atopy across several categories of animal contact and consumption of unpasteurized farm milk products before the age of 5 years. In addition, a significant interaction between contact to farm animals and consumption of unpasteurized milk early in life could be demonstrated. However, no such gradient across exposure to farm animals or milk consumption could be shown among rural children who in general had a low prevalence of atopic sensitization. Important differences exist between farms in Crete and those in central Europe. Farms in Crete are generally small, often some distance from the family home and in most cases concerned with the cultivation of olives and the tending of goats whereas on typical farms in central Europe livestock is kept in stables built in close proximity to the farm houses. A recent case control study on indoor exposures and childhood asthma in Nepal [19] demonstrated that keeping cattle inside of the family home was associated with a risk reduction for asthma whereas keeping cattle outside of the house had no protective effect. Thus, it is conceivable that exposure to microbial products is much higher inside stables and that therefore the question relating to time spent in stables in the central European studies capture the relevant differences in exposure among rural children. The question regarding contact with farm animals in rural Crete obviously does not differentiate between more and less exposed children. It is not known from the Barnes paper whether contact with farm animals in rural Crete takes place in stables or outdoors. Given the mild climate of Crete it seems possible that the goats spend most of their time outdoors. If this were true, exposure to higher loads of microbial products might be more ubiquitously distributed in rural Crete resulting in a general risk reduction for atopic diseases, but no gradient for increasing contact to animals. However, the findings might also be interpreted as showing a protective effect of ‘rurality’ which is not related to exposure to microbial products. As indicated by the authors the evaluation of such a rural factor in their study is hampered by the fact that the study was designed as a two-point comparison, which is not very useful for aetiological research. Although exposures occurring early in life seem to be of great importance for the development of asthma and allergies the signal of a farming childhood can also be detected in young adults. Among Finnish first-year university students aged 18–24 years being raised on a farm during the first five years of life was associated with a reduced risk for physician-diagnosed allergic rhinitis, for diagnosed asthma and/or episodic wheeze [20]. Urban childhood environment, however, did not show an increased risk when compared with rural non-farm residency. This is in contrast to the findings of Filipiak et al. reported in this issue of the journal [21]. Their analyses are based on 25–74-year-old-residents of the city of Augsburg and two surrounding counties in southern Germany. Rural, urban and suburban residents were defined by community size, farming status was based on information about the participants occupation. Only 118 individuals out of the 4856 study participants reported full-time farming as their present occupation. Farmers had lower risks of allergic rhinitis and atopic sensitization than rural non-farming residents, although the differences were statistically not significant. The urban population had an increased risk of allergic rhinitis and atopic sensitization when compared to the rural non-farming population, no difference was observed for asthma. The suburban residents did not differ from urban residents. The protective effect of ‘rurality’ of this study might be in part explained by the fact that possibly a substantial number of rural adults had spent their early years on a farm. As in many parts of Europe the number of active farmers has tremendously decreased over the past decades because of economic constraints. Thus, the small number of today's full-time farmers only partially reflects the number of people exposed to a farming environment during childhood. The Finnish study [20] that did not find an effect of ‘rurality’ had inquired about childhood farm exposure. Alternatively, it is conceivable that farming is an indicator of a more traditional lifestyle which might be more prevalent in rural communities even outside of the farming sector. Specifically the role of dietary factors has not been assessed in most of the farming studies or in urban-rural comparison studies. It has been postulated that changes in the consumption of omega-3 fatty acids and fresh foods containing antioxidants and magnesium may be responsible for the higher prevalence of asthma observed in affluent countries [22, 23]. The Bavaran study [3] included items on dietary habits into the questionnaire. The authors reported that the consumption of whole but not skimmed milk and self-produced foods was inversely related to hayfever and asthma. Adjustment for these factors did not substantially change the association with farming as parental occupation but they might contribute to a ‘rurality’ effect outside of the farming environment. Endotoxin can be detected in the presence of gram-negative bacteria regardless of whether the bacteria are alive, and because these bacteria are ubiquitous, everyone is exposed to at least low levels of environmental endotoxin. It is well known that endotoxin is present in house dust. In occupational health studies endotoxin exposure has been recognized as an important factor in the aetiology of both acute and chronic occupational lung diseases including non-allergic asthma [7]. In addition, several studies have shown that endotoxin in house dust is associated with exacerbation of pre-existing asthma in children and adults [24, 25]. It is only recently that the potential protective effect of endotoxin exposures for the development of asthma and allergies has attracted scientific interest. Two recent birth cohort studies investigated the effect of endotoxin exposure commonly encountered in metropolitan homes on the development of allergen sensitization and type 1 immunity in asthma-prone children [26] and on the occurrence of wheeze in children with a familial predisposition to asthma and allergy [27]. Gereda et al.[26] could demonstrate that the homes of allergen-sensitized infants contained significantly lower concentrations of house dust endotoxin than those of non-sensitized infants. In addition, increased house dust endotoxin concentrations correlated with increasing proportions of IFN-γ-producing DC4 T cells. Such concentrations did not correlate with proportions of cells that produced IL-4, -5, or -13. Thus, the results of this study suggest that environmental endotoxin exposures might mitigate the development of allergen sensitization by enhancement of type 1 immunity. Park and coworkers [27] reported a significantly increased risk for repeated wheeze during the first year of life associated with elevated levels of endotoxin in family room dust in a cohort of 499 infants with a family history of asthma and allergies. Wheezing in the first year of life is a frequent symptom for children who are born with small airways and/or when children have upper or lower airway infections and is not necessarily associated with increased risk of asthma later in life [28]. Exposure to low levels of endotoxin in the first year of life may cause airway inflammation, thus triggering wheeze but the question remains open whether these exposures will be associated with an increased or decreased risk of allergic asthma later in life. High endotoxin levels are not only known to occur in animal stables but a few recent reports also suggested that higher house dust endotoxin levels can be found in homes with pets [27, 29, 30]. Heinrich et al. investigated in a study published in this issue of the journal whether the presence of pets or vermin in the home of 454 homes in eastern Germany was associated with higher exposures to bacterial endotoxin [31]. The authors demonstrate that the presence of dogs, cats and cockroaches in the home increased the levels of endotoxin in living rooms, even when adjustment was made for potential confounders like city of residence, season of dust sampling and age of the building. Dogs and cats may be a direct source of bacteria and endotoxin through faecal contamination and they may carry endotoxin from outdoors into the homes. The observation of pets increasing the endotoxin levels indoors is of particular interest as some recent studies have suggested that exposure to cats and dogs in the home early in life decreases the risk of sensitization and asthma [32, 33]. One might therefore speculate that exposure to microbial products is responsible for this protective effect. However, Platts-Mills et al. suggested another mechanism for the observed protective effect of cat exposure [34]. They showed that many children exposed to greater than 20 µg of Fel d 1/g of dust at home made an IgG and IgG4 antibody response to Fel d 1 without IgE antibody and this modified Th2 response was not associated with symptoms. The production of IgG4 antibodies are part of Th2 response contradicting the view that the protective effect of cat exposure on sensitization may be explained by increased exposure to microbial animal products biasing the immune response towards a Th1 response. Striking differences in the prevalence of asthma and allergies have repeatedly been observed between post-socialist eastern European and western industrialized countries. Differences in indoor climate, family size, early exposure to childhood infections due to day-care attendance at young ages have been suggested as explanations for this East-West gradient although none of these factors fully explained the observed differences in prevalence rates. As evidenced by the Tucson cohort study different types of childhood wheezing phenotypes can be distinguished [28, 35]: ‘Transient early wheezing’ limited to the first three years of life and unrelated to increased airway lability, ‘non-atopic wheezing’ of the toddler and early school years associated with positive peak flow variability but not with methacholine hyper-responsivness and ‘IgE associated wheeze/asthma’ associated with persistent wheeze at any age and with methacholine hyper-responsiveness, peak flow variability, and markers of atopy. In the present issue of the journal Annus et al. investigated the framework of the ISAAC phase II study whether the prevalence of wheeze in schoolchildren and the type of wheeze differed between Sweden and Estonia [36]. The prevalence of current wheeze was similar in the Estonian and in the two Swedish centres, while diagnosed asthma and atopic symptoms were more common in Sweden. Wheezing children in Sweden were more likely to be sensitized to common allergens, to have bronchial hyper-responsiveness and to receive anti-asthmatic treatment whereas in Estonia only a small proportion of wheezing children showed markers of atopy. Thus, the type, and not the prevalence of wheeze, differs between the two countries suggesting that different environmental factors are responsible for the occurrence of these symptoms. However, the study could not elucidate which environmental factors might be responsible for the observed differences. Although the farming studies and the comparisons between eastern and western European population illustrate that environmental factors play a major role in the development of asthma and atopy their exact contribution remains to be determined. Future research not only has to specify more precisely the relevant exposures and the underlying effect mechanisms but also has to take into account the timing and the duration of these exposures with the ultimate goal to identify opportunities for prevention.