• atopic eczema;
  • children;
  • endotoxin;
  • β(1[RIGHTWARDS ARROW]3)-glucans


  1. Top of page
  2. Abstract
  3. Methods
  4. Results
  5. Discussion
  6. Acknowledgments
  7. References

Background:  There is still uncertainty about the determinants of atopic eczema (AE). To explain the heterogeneity of the disease, different phenotypes of AE have been suggested.

Methods:  The cross-sectional PARSIFAL study included 14 893 school-age children of farmers or children attending Steiner schools and their respective reference groups. A detailed questionnaire was completed, and house dust was collected for the measurement of endotoxin and glucans. Atopic sensitization was defined by allergen-specific IgE levels in the serum.

Results:  In multivariate analyses, helping with haying was the only variable related to a farming environment having a consistent inverse association with both current symptoms and a doctor’s diagnosis of AE [aOR = 0.65 (95% CI: 0.46–0.93) and 0.73 (0.51–1.05)], respectively. Severe lower respiratory tract infections (LRTI) in the first 2 years of life and usage of antibiotics ever were found to be positively related only to asthma-associated AE, whereas the effect of LRTI on AE without asthma had an opposite effect. Levels of β(1[RIGHTWARDS ARROW]3)-glucans in mattress dust were inversely related to a doctor’s diagnosis of asthma-associated AE [aOR = 0.75 (0.57–0.98)], and endotoxin levels to current symptoms of asthma-associated AE [aOR = 0.73 (0.57–0.94)].

Conclusions:  The analyses of the PARSIFAL study revealed two different phenotypes of AE, depending on the association with asthma and wheezing ever. With regard to the hygiene hypothesis, help with haying, exposure to β(1[RIGHTWARDS ARROW]3)-glucans and endotoxin were found to be inversely associated with the AE phenotype associated with asthma and wheezing.

Atopic eczema (AE) is a chronic inflammatory disease of the skin, often manifesting in early infancy and with a natural course varying considerably over time (1). Unlike other atopic diseases, little is known about the determinants of AE. Gender, socio-economic status, family size, infant feeding and environmental pollutants have been found to be determinants of the disease in some studies but not in others (2–4). Genetic factors appear to be important in the multifactorial pathogenesis of AE, and in recent years significant associations have been reported (5, 6).

The ‘hygiene hypothesis’ has shifted attention from the adverse health effects to the potential beneficial effects of microbial exposures (7). It is thought that such exposures may protect from developing atopic sensitization and asthma; the association with AE has been less clear. However, clinical studies demonstrating a beneficial effect of probiotics for the prevention of AE have stimulated interest in a possible role of microbial compounds also for AE (8–10).

Previous studies regarding microbial agents and atopic diseases in children have mainly focused on bacterial endotoxin (11, 12). However, it has been suggested that protection against asthma and allergy might also result from exposure to other microbial agents, such as mould β(1[RIGHTWARDS ARROW]3)-glucans, which have known immunomodulatory effects (13).

One of the suggestions for the different phenotypes of AE is the presence of an allergic extrinsic form (EAE) and a nonallergic intrinsic form (IAE) (14). There is, however, uncertainty whether this is a useful distinction as not all studies have shown distinct environmental or genetic determinants for EAE and IAE, respectively (15). In turn, children with AE and asthma may represent a distinct phenotype as their skin lesions are more severe, and they show a specific sensitization pattern (16). Some genetic studies support this concept, as e.g. polymorphisms of SPINK5 and NOD1 have been found to be associated with concomitant expression of asthma and AE (17–19).

The aim of the current analysis was to examine the associations of environmental exposures related to the hygiene hypothesis, e.g. farm-related exposures or microbial components in mattress dust, on AE symptoms and diagnosis. Furthermore, sub-phenotypes of AE with respect to atopic sensitization, hay fever, or asthma, were assessed.


  1. Top of page
  2. Abstract
  3. Methods
  4. Results
  5. Discussion
  6. Acknowledgments
  7. References

Study population

The PARSIFAL study was set up as a cross-sectional study to identify environmental and genetic factors potentially explaining previous observations that children living on farms or with an anthroposophic lifestyle have a lower risk for asthma and allergic diseases than their peers (20). The study included 14 893 children, aged 5–13 years, from farm families, children attending Steiner schools, children from rural nonfarming, and urban environments in Austria, Germany, the Netherlands, Sweden, and Switzerland. The study was approved by the national ethical boards of each center and written informed consent was obtained from parents of each child.


The prevalence of diseases and symptoms were assessed by questions on the validated and translated International Study of Asthma and Allergies in Childhood (ISAAC) and the Swedish BAMSE study (21, 22). The questionnaires were distributed and collected from October 2000 to May 2002. Current AE symptoms were considered present, if the child ever had had an itchy rash intermittently for at least 6 months, and if the child had had this rash at any time during the last 12 months. Children with an intermittent itchy rash for at least 6 months and who had ever been diagnosed with AE were considered to have a doctor’s diagnosis of AE. Children with reported physician-diagnosed asthma once or obstructive bronchitis more than once in their lifetime were defined as having asthma ever. Children with wheezing once in their lifetime were defined as having wheezing ever. The parental questionnaire included questions on exposure to farm environment, animals, pets, passive smoking, housing, nutrition, breastfeeding, socio-demographic background, history of infections, and family history of atopic diseases. Previous usage of antibiotics and antipyretics were also asked. In addition, information on the child’s farm activities as well as the mother’s farm exposures during pregnancy was available. The children’s activities on farms occurring at least weekly were defined as ‘regularly’. The four study groups were defined as follows: farmers’ children were defined as children whose parents answered yes to the question ‘does your child live on a farm?’ and whose family ran the farm. Children living in the same villages as the farm children but not living on a farm run by their family were defined as ‘farm reference’ children. Children with an anthroposophic lifestyle were recruited through Rudolf Steiner schools and termed Steiner children. Steiner-reference children lived in the neighborhood of Steiner children, mainly in sub-urban areas, but did not attend a Rudolf Steiner school. Questions on exposures and lifestyle factors related to living on a farm were largely based on questionnaires developed for the Allergy and Endotoxin (ALEX) study (23).

Measurement of allergen-specific IgE

Blood analysis was performed in a sub-sample of 4049 children, which did not differ substantially from the whole sample with respect to atopic diseases (data not shown). All samples were screened with a mix of common inhalant allergens (Phadiatop; Phadia AB, Uppsala, Sweden) and a mix of common food allergens (fx5; ImmunoCAP System; Phadia AB). All analyses were performed at the Department of Clinical Immunology, Karolinska University Hospital, Stockholm, Sweden. Atopic sensitization was defined as at least one allergen-specific IgE result ≥0.35 kU/l against common inhalant and/or food allergens.

Measurement of endotoxin, EPS and β(1[RIGHTWARDS ARROW]3)-glucans in dust samples

The study design included collection of house dust in a sub-sample of farm children, children of Steiner schools and reference groups randomly selected for atopy and wheeze status (24). The sub- population for dust sampling had similar characteristics with the total PARSIFAL study population (24). Dust from mattresses was collected by the parents of the participating children. Endotoxin, β(1[RIGHTWARDS ARROW]3)-glucans and fungal extra-cellular polysaccharides (EPS) were measured in mattress dust samples of 933 children.

In short, endotoxin was analyzed with the kinetic chromogenic limulus amebocyte lysate test (Bio Whittaker, Walkersville, MD, USA), β(1[RIGHTWARDS ARROW]3)-glucans with an inhibition enzyme immunoassay (EIA) (25), and EPS with a specific sandwich EIA for EPS of Aspergillus and Penicillium spp. (26). All samples were analyzed in one laboratory (IRAS, Utrecht, the Netherlands). Values for exposure levels were obtained by division of concentrations by weight of collected mattress dust followed by log-transformation, resulting in normal distribution.

Statistical analysis

Data analysis was performed using SAS statistical software (version 9.1.3; SAS Institute, Cary, NC, USA). In the first step the effects of potential exposures and confounders were analyzed adjusting for center and study group (Table 1); then variables with P-values < 0.2 in these models were included in mutually adjusted models (Tables 2–4). Factors reported in Table 1 that were interpreted as confounders (e.g. gender, parental education) or as representing avoidance measures (e.g. wall-to-wall carpeting, having pets at present) were included in further models for adjustment, but are not presented in the subsequent tables. Variables with missing values or sub-sample analyses caused lower numbers of observations in the multivariate models presented in Tables 2–4.

Table 1.   Potential determinants of current symptoms and a doctor’s diagnosis of atopic eczema adjusted for center and group (OR are given with 95% CI)
 Current symptoms of atopic eczema n = 1351/14721Doctor’s diagnosis of atopic eczema n = 1500/14699
Age (per year)0.99 (0.96–1.02)1.00 (0.97–1.03)
Female vs male gender1.32 (1.18–1.48)1.08 (0.97–1.20)
Having older siblings (more than 2 vs less)1.04 (0.86–1.27)0.86 (0.70–1.05)
Maternal asthma and/or rhinoconjunctivitis: yes vs no1.59 (1.40–1.80)1.85 (1.64–2.08)
Paternal asthma and/or rhinoconjunctivitis: yes vs no1.39 (1.22–1.59)1.61 (1.42–1.82)
Parental education:
 Elementary vs university0.91 (0.75–1.10)0.73 (0.61–0.88)
 Gymnasium vs university1.02 (0.89–1.17)0.93 (0.81–1.06)
Exclusive breastfeeding for at least 5 months vs less1.05 (0.93–1.19)1.18 (1.04–1.33)
Current visits to animal sheds: at least once a week vs less0.97 (0.83–1.14)0.89 (0.76–1.04)
Current visits to hay lofts: at least once a week vs less0.86 (0.71–1.04)0.73 (0.61–0.89)
Farm milk consumption ever vs never0.80 (0.69–0.94)0.82 (0.70–0.95)
Farm milk consumption in first year of life: yes vs no0.77 (0.66–0.91)0.79 (0.68–0.93)
Mother worked in stable during pregnancy: at least once a week vs less0.69 (0.56–0.86)0.70 (0.56–0.87)
Current help with haying: at least once a week vs less0.62 (0.48–0.81)0.66 (0.51–0.86)
Antibiotic usage ever vs never1.56 (1.34–1.82)1.48 (1.29–1.71)
Antipyretic usage ever vs never1.38 (1.17–1.64)1.33 (1.13–1.55)
Having pets now: yes vs no0.84 (0.75–0.95)0.81 (0.73–0.91)
Lower respiratory tract infections in the first 2 years of life: yes vs no1.57 (1.37–1.80)1.87 (1.65–2.12)
Sharing bedroom now: yes vs no0.93 (0.82–1.05)0.89 (0.79–1.01)
Environmental tobacco smoke exposure: yes vs no0.88 (0.76–1.02)0.79 (0.68–0.91)
Having moved to another flat ever vs never1.17 (1.03–1.32)1.21 (1.07–1.35)
Wall-to-wall carpeting at present: yes vs no0.74 (0.65–0.85)0.75 (0.67–0.85)
Dampness in the bathroom at present: yes vs no1.25 (1.02–1.53)1.08 (0.88–1.32)
Atopic sensitization to food allergens (0.35 kU/l and more vs less)1.84 (1.40–2.42)2.16 (1.65–2.83)
Atopic sensitization to inhalant allergens (0.35 kU/l and more vs less)2.34 (1.89–2.89)2.87 (2.31–3.56)
Sensitization to inhalant or food allergens (0.35 kU/l and more vs less)2.18 (1.77–2.70)2.61 (2.11–3.24)
 n = 62/853n = 85/851
Sub-sample with dust analysis
 β(1[RIGHTWARDS ARROW]3)-glucans0.75 (0.53–1.06)0.69 (0.53–0.90)
 Endotoxin0.73 (0.58–0.91)0.82 (0.68–1.00)
 Extra-cellular polysaccharides0.76 (0.65–0.90)0.91 (0.75–1.11)
Table 2.   The determinants of current symptoms and a doctor’s diagnosis of atopic eczema (OR’s are given with 95% CI)
 Current symptoms of atopic eczema n = 870/9441Doctor’s diagnosis of atopic eczema n = 960/9428
  1. The OR are adjusted for gender, having older siblings, family history of asthma or rhinoconjunctivitis, parental education, breastfeeding, having pets now, sharing bedroom, environmental tobacco smoke exposure, having moved to another flat, wall-to-wall carpeting, dampness in the bathroom, study center, and study group.

  2. LRTI, lower respiratory tract infections.

Current visits to animal sheds1.08 (0.84–1.37)1.05 (0.83–1.34)
Current visits to hay lofts1.06 (0.80–1.41)0.82 (0.61–1.09)
Farm milk consumption ever0.91 (0.75–1.11)0.99 (0.83–1.19)
Mother worked in stable during pregnancy0.88 (0.67–1.16)0.97 (0.73–1.28)
Current help with haying0.65 (0.46–0.93)0.73 (0.51–1.05)
LRTI in the first 2 years of life1.37 (1.15–1.64)1.59 (1.35–1.88)
Antibiotic usage ever1.33 (1.08–1.62)1.15 (0.95–1.38)
Antipyretic usage ever1.17 (0.94–1.47)1.18 (0.96–1.46)
Table 3.   The determinants of current symptoms and a doctor’s diagnosis of extrinsic atopic eczema (EAE) and intrinsic atopic eczema (IAE) in multivariate analysis (OR are given with 95% CI)
 Current symptoms of atopic eczemaDoctor’s diagnosis of atopic eczema
Extrinsic n = 128/2533Intrinsic n = 154/2559Extrinsic n = 136/2556Intrinsic n = 128/2548
  1. The OR are adjusted as in Table 2.

  2. LRTI, lower respiratory tract infections.

Current visits to animal sheds1.12 (0.62–2.03)1.35 (0.83–2.21)1.16 (0.63–2.12)1.16 (0.67–1.99)
Current visits to hay lofts1.13 (0.56–2.28)1.53 (0.89–2.62)0.61 (0.29–1.30)1.03 (0.56–1.90)
Farm milk consumption ever0.96 (0.58–1.57)0.75 (0.48–1.16)1.11 (0.69–1.80)1.14 (0.72–1.83)
Mother worked in stable during pregnancy0.67 (0.35–1.28)0.75 (0.45–1.26)0.90 (0.46–1.76)0.84 (0.46–1.52)
Current help with haying0.79 (0.37–1.70)0.69 (0.39–1.22)1.11 (0.50–2.43)0.70 (0.36–1.37)
LRTI in the first 2 years of life1.50 (0.97–2.33)1.93 (1.28–2.90)1.61 (1.05–2.46)2.17 (1.42–3.31)
Antibiotic usage ever1.31 (0.75–2.30)0.70 (0.44–1.11)0.94 (0.57–1.55)0.84 (0.50–1.40)
Antipyretic usage ever1.08 (0.62–1.91)1.75 (1.00–3.07)0.89 (0.53–1.51)1.41 (0.80–2.51)
Table 4.   The determinants of current symptoms and a doctor’s diagnosis of atopic eczema with and without a doctor’s diagnosis of asthma and a history of wheeze in multivariate analysis (OR are given with 95% CI)
 Current symptoms of atopic eczemaDoctor’s diagnosis of atopic eczema
With asthma or wheeze n = 276/8847Without asthma and wheeze n = 572/9143With asthma or wheeze n = 363/8831Without asthma and wheeze n = 570/9038
  1. The OR are adjusted as in Table 2.

  2. LRTI, lower respiratory tract infections.

Current visits to animal sheds1.14 (0.73–1.78)1.06 (0.79–1.42)0.87 (0.57–1.33)1.18 (0.88–1.57)
Current visits to hay lofts0.65 (0.37–1.14)1.22 (0.88–1.70)0.65 (0.38–1.12)0.87 (0.61–1.23)
Farm milk consumption ever1.14 (0.82–1.60)0.83 (0.65–1.04)1.24 (0.93–1.67)0.88 (0.69–1.11)
Mother worked in stable during pregnancy0.64 (0.38–1.09)0.97 (0.70–1.34)0.73 (0.45–1.18)1.11 (0.79–1.57)
Current help with haying0.55 (0.28–1.08)0.68 (0.45–1.02)0.58 (0.30–1.14)0.78 (0.50–1.20)
LRTI in the first 2 years of life3.46 (2.68–4.47)0.63 (0.48–0.82)3.83 (3.05–4.81)0.65 (0.49–0.85)
Antibiotic usage ever2.46 (1.59–3.81)1.11 (0.88–1.40)1.44 (1.03–2.00)1.08 (0.86–1.35)
Antipyretic usage ever1.19 (0.80–1.77)1.16 (0.89–1.51)1.03 (0.73–1.44)1.34 (1.03–1.75)

Symptoms and diagnosis of AE were also explored separately for children with and without a reported physician’s diagnosis of asthma or wheezing ever; with (EAE) or without (IAE) atopic sensitization [IgE antibodies to common food (fx5) and/or common inhalant (Phadiatop) allergens’; with or without atopic sensitization to either common food or inhalant allergens; or with and without symptoms or a reported physician’s diagnosis of rhinoconjunctivitis.

Levels of microbial compounds were only available for a sub-sample selected for wheezing and atopy status (24), therefore stratified weighted logistic regression was performed (27). For microbial compounds, multivariate models were established by backward elimination (level of stay P = 0.05). Analysis of microbial compounds was not possible stratified for a reported physician’s diagnosis of asthma or wheezing symptoms ever, as children with AE but without wheezing were excluded from dust sampling (24). Statistical significance was defined as a P-value < 0.05.


  1. Top of page
  2. Abstract
  3. Methods
  4. Results
  5. Discussion
  6. Acknowledgments
  7. References

Of the 21 905 children invited to take part in the study, questionnaires were completed for 15 137 children, implying an overall response rate of 69%. A total of 237 children were excluded because they were outside the age limits (5- to 13-year old) and seven children because of missing information regarding sex or which group they belonged to, leaving 14 893 children for the analyses. In total, 4049 of all invited children from a random sub-sample (n = 4854) (83.4%) provided a blood sample.

The prevalence of current symptoms of AE in the total study population was 9.1% (n = 1351), and 10.2% (n = 1500) of children had a previous doctor’s diagnosis of AE. About half of them had EAE. The non-asthma associated phenotype was more frequent (6% for both diagnosis and symptoms) than the asthma associated phenotype (4% for diagnosis and 3% for symptoms). Prevalence of asthma or wheeze ever was 21%, of atopic sensitization was 31%, and of hay fever symptoms or diagnosis was 9%.

Potential determinants of current symptoms and a doctor’s diagnosis of AE adjusted for center and study group are shown in Table 1. When establishing multivariate models (Table 2), help with haying was the only variable related to a farming environment having a significant inverse association with current symptoms of AE, and a trend towards an inverse association with a doctor’s diagnosis of AE. A history of lower respiratory tract infections (LRTI) was positively associated with both outcomes, and usage of antibiotics ever with current symptoms of AE and nonsignificantly with a doctor’s diagnosis of AE (Table 2).

In the sub-sample with dust analyses a significant inverse association in multivariate analysis was seen for endotoxin [OR = 0.72 (0.56–0.93)] with current symptoms of AE, whereas the levels of EPS and β(1[RIGHTWARDS ARROW]3)-glucans were not significantly associated. In turn, the levels of β(1[RIGHTWARDS ARROW]3)-glucans were found to be inversely related to a doctor’s diagnosis of AE [OR = 0.75 (0.57–0.98)], whereas the levels of endotoxin and EPS had no significant effect on the doctor’s diagnosis of AE. The models included further covariables for current symptoms (LRTI, paternal asthma and/or rhinoconjunctivitis, help with haying) and doctor’s diagnosis (center, LRTI, maternal asthma and/or rhinoconjunctivitis, having pets at home, help with haying, antibiotic usage). Because of the study design, sub-phenotypes of AE could not be studied in the sub-sample with dust analyses.

The determinants of EAE and IAE did not differ (Table 3). Furthermore, a differential analysis for sensitization to food or inhalant allergens, or for hay fever diagnosis or symptoms did not reveal any differential effects of the investigated exposures (data not shown). In contrast, stratification by wheezing and asthma revealed differential effects (Table 4). In the AE phenotype associated with asthma, OR of all protective and risk factors became stronger. Probably because of a lower sample size, the effect of help with haying did not reach the significance level. This was as well the case for an inverse relation of visits to hay lofts, which was only seen in this phenotype. When using a combined variable for any contact to hay (i.e. visits to hay lofts or help with haying) in the same models as in Table 4, significant effects on both, symptoms and diagnosis of AE associated with asthma or wheeze were seen [OR = 0.51 (0.29–0.89) and OR = 0.57 (0.34–0.97), respectively]. In contrast to the asthma-related phenotype, the effect of severe LRTI in the first 2 years of life was found to be inversely related to AE without asthma.


  1. Top of page
  2. Abstract
  3. Methods
  4. Results
  5. Discussion
  6. Acknowledgments
  7. References

The PARSIFAL study was set up to identify potential protective factors for the prevention of atopic diseases in a population of farm and Steiner children and their respective references. With respect to AE, only help with haying was found to be inversely related. In a sub-sample analysis, β(1[RIGHTWARDS ARROW]3)-glucans were inversely related to a doctor’s diagnosis of AE, and endotoxin levels to current symptoms of AE. When assessing different phenotypes of AE, the phenotypes defined by concomitant presence or absence of asthma or wheezing ever explained more heterogeneity of AE than the dichotomization into EAE and IAE. Severe LRTI in the first 2 years of life and usage of antibiotics ever were found to be positively related only to AE with asthma, whereas the effect of LRTI on AE without asthma had an opposite effect.

One of the major limitations of this study was the challenge of defining AE by questionnaire data. However, a previous study had shown considerable concordance between the same set of parent-administered questions as used in our study and skin examinations performed by dermatologists (15). Yet, repeated examinations may give better agreement between questionnaire and clinical examination. Because of the cross-sectional design, microbial exposures were measured simultaneously with the assessment of the clinical outcomes; only cohort studies may allow for comparing childhood outcomes with objective measurements of exposure early in life.

There is much debate about distinct phenotypes of AE. The discussion whether AE should be differentiated into an allergic (EAE) and non-allergic intrinsic form (IAE) is still ongoing (14, 15, 28, 29). In a recent study, it has been shown that the risk factor profile was similar for IAE and EAE suggesting that these phenotypes may not completely explain the heterogeneity of AE (15). In fact, in the PARSIFAL study, risk factor profiles for EAE and IAE were also comparable.

In turn, children with AE and asthma were suggested to have a distinct phenotype characterized early in life by severe AE and wheezing or a specific pattern of atopic sensitization (16). Therefore, we stratified the patients with AE according to the presence and absence of a diagnosis of asthma and wheeze ever. In fact, in the PARSIFAL study population, risk factor profiles for the two strata differed, suggesting that the presence of asthma may play a role in the expression of the heterogeneity of AE.

In this context, the opposite effect of LRTI on AE in the two different strata requires attention. Whereas the asthma-associated phenotype was found to be positively associated with LRTI, the phenotype without asthma was negatively related to the same exposure. This paradox may be resolved by any of the following three explanations: (i) in the tenor of the hygiene hypothesis, LRTI may be a proxy for a higher frequency of unhygienic contacts, which may account for the inverse relation of LRTI to AE without wheezing; (ii) LRTI might also result from a minor immunodeficiency (30). This hypothesis may explain the positive relation of LRTI to asthma-associated AE and is supported by the higher frequency of antibiotic usage in children with this phenotype; and (iii) asthma may just operate as a confounder in the alleged association of LRTI and AE. Adjustment for asthma, however, does not fully abolish the effect of LRTI on AE with asthma [OR = 1.77 (1.33–2.36) for current symptoms and OR = 1.88 (1.45–2.43) for diagnosis], ultimately not excluding the afore mentioned options. Moreover, the effects of antibiotic usage were confounded only in part by asthma (data not shown). However, frequency of physician contacts may account for residual confounding, as frequent physician contacts would increase both the probability of a prescription of medication and the probability of a doctor’s diagnosis of AE.

In contrast to other health outcomes such as asthma and atopic sensitization, where many significant associations with farm-related exposures were found in the same population (31), few associations of AE with these exposures were detected. Given the large sample size of the PARSIFAL study, this seems an important finding. The only clearly related exposure in the farming context for diagnosis and current symptoms of AE, was help with haying on a regular basis. Of note, the effect was neither confounded by other farm-related exposures such as visits to animal sheds or barns, either by asthma or by wheeze (data not shown). A likely explanation for the effect of help with haying may be exposure to contaminating microorganisms or their components during this activity. An alternative explanation relates to potential reverse causation as children with AE were possibly not able or willing to help their parents with farm activities for manifest or anticipated allergic problems. However, there was no change in the effect of haying on AE after adjusting for atopic sensitization to grass pollen or avoidance of help with farm activities (data not shown), ultimately rendering reverse causation rather improbable. Furthermore, there is some evidence for an immunological effect of haying, as gene expression levels of Toll-like receptors in children helping with haying are elevated independently of other farm-related exposures (31). These receptors are part of innate immunity and recognize microbial substances.

Although Table 1 may suggest associations of predominant farm milk consumption ever with AE, these assumed associations have not been confirmed when models were adjusted for other farm-related exposures such as help with haying (Tables 2–4). When focusing on predominant farm milk consumption in the first year of life, effects did not change (data not shown). A thorough analysis on the effects of consumption of milk and other dairy products in the PARSIFAL population is provided in a separate publication (32).

We also studied exposure to microorganisms or their components by measuring the levels of biocontaminants in mattress dust. Elevated levels of endotoxin, an intrinsic part of the outer membrane of gram-negative bacteria have been suggested to be an important protective factor not only in the farming environment (11). However, no association was observed between endotoxin exposure and a doctor’s diagnosis of AE in the present study, which is in accordance with previous results (12). Yet, exposure to endotoxin may protect from developing current symptoms of AE. This obvious discrepancy may be a consequence of different diagnostic criteria (doctor’s diagnosis vs parents’ observations) or different time frames: the doctor’s diagnosis of AE reflects the life time prevalence of AE, whereas current symptoms refer to a period prevalence of 12 months.

It has been demonstrated that β(1[RIGHTWARDS ARROW]3)-glucans are nonallergenic structural cell wall components of most fungi (25). Previous population studies suggested some association between exposure to β(1[RIGHTWARDS ARROW]3)-glucans, airway inflammation and symptoms; however, results were controversial and potential underlying inflammatory mechanisms associated with exposure could not be identified (33). In this study, exposure to higher concentrations of β(1[RIGHTWARDS ARROW]3)-glucans was found to be protective for a doctor’s diagnosis of AE. Interestingly, exposure to endotoxin or β(1[RIGHTWARDS ARROW]3)-glucans, respectively, as well as help with haying stayed in the final models for current symptoms or a doctor’s diagnosis of AE, respectively, indicating independent effects of the two microbial components and the exposure to haying on the prevalence of AE. Still, our results are not applicable to AE without asthma and wheeze because of the study design.

In conclusion, we identified two phenotypes of AE according to the presence and absence of a diagnosis of asthma and wheeze ever with different risk factors. This differentiation may be more relevant than EAE and IAE to determine the heterogeneity of AE. With regard to the hygiene hypothesis, help with haying and exposure to β(1[RIGHTWARDS ARROW]3)-glucans and endotoxin were found to be protective for the respective AE phenotypes.


  1. Top of page
  2. Abstract
  3. Methods
  4. Results
  5. Discussion
  6. Acknowledgments
  7. References

This study was supported by a research grant from the European Union QLRT 1999-01391 and by funding from the Swedish Foundation for Health Care Science and Allergy Research, the Swiss National Foundation, grant Nr. 32-100324, the Kühne-Foundation, and by the EU Framework program for research (contract n° FOOD-CT-2004-506378, Global Allergy and Asthma European Network (GA2LEN)). Bülent Karadag is the recipient of a European Respiratory Society fellowship (no. 338).


  1. Top of page
  2. Abstract
  3. Methods
  4. Results
  5. Discussion
  6. Acknowledgments
  7. References
  • 1
    Akdis CA, Akdis M, Bieber T, Bindslev-Jensen C, Boguniewicz M, Eigenmann P et al. Diagnosis and treatment of atopic dermatitis in children and adults: European Academy of Allergology and Clinical Immunology/American Academy of Allergy, Asthma and Immunology/PRACTALL Consensus Report. Allergy 2006;61:969987.
  • 2
    Kerkhof M, Koopman LP, Van Strien RT, Wijga A, Smit HA, Aalberse RC et al. Risk factors for atopic dermatitis in infants at high risk of allergy: the PIAMA study. Clin Exp Allergy 2003;33:13361341.
  • 3
    Strachan DP. Family size, infection and atopy: the first decade of the “hygiene hypothesis”. Thorax 2000;55(Suppl 1):S2S10.
  • 4
    Von Mutius E. Risk factors for atopic dermatitis. In: BieberT, LeungD, editors. Atopic dermatitis. New York: Marcel Dekker Inc., 2002:111122.
  • 5
    Weidinger S, Illig T, Baurecht H, Irvine AD, Rodriguez E, Diaz-Lacava A et al. Loss-of-function variations within the filaggrin gene predispose for atopic dermatitis with allergic sensitizations. J Allergy Clin Immunol 2006;118:214219.
  • 6
    Palmer CN, Irvine AD, Terron-Kwiatkowski A, Zhao Y, Liao H, Lee SP et al. Common loss-of-function variants of the epidermal barrier protein filaggrin are a major predisposing factor for atopic dermatitis. Nat Genet 2006;38:441446.
  • 7
    Martinez FD, Holt PG. Role of microbial burden in aetiology of allergy and asthma. Lancet 1999;354(Suppl. 2):SII12SII15.
  • 8
    Weston S, Halbert A, Richmond P, Prescott SL. Effects of probiotics on atopic dermatitis: a randomised controlled trial. Arch Dis Child 2005;90:892897.
  • 9
    Kalliomaki M, Salminen S, Arvilommi H, Kero P, Koskinen P, Isolauri E. Probiotics in primary prevention of atopic disease: a randomised placebo-controlled trial. Lancet 2001;357:10761079.
  • 10
    Kalliomaki M, Salminen S, Poussa T, Isolauri E. Probiotics during the first 7 years of life: a cumulative risk reduction of eczema in a randomized, placebo-controlled trial. J Allergy Clin Immunol 2007;119:10191021.
  • 11
    Braun-Fahrlander C, Riedler J, Herz U, Eder W, Waser M, Grize L et al. Environmental exposure to endotoxin and its relation to asthma in school-age children. N Engl J Med 2002;347:869877.
  • 12
    Gehring U, Bischof W, Fahlbusch B, Wichmann HE, Heinrich J. House dust endotoxin and allergic sensitization in children. Am J Respir Crit Care Med 2002;166:939944.
  • 13
    Osborne M, Reponen T, Adhikari A, Cho SH, Grinshpun SA, Levin L et al. Specific fungal exposures, allergic sensitization, and rhinitis in infants. Pediatr Allergy Immunol 2006;17:450457.
  • 14
    Schmid-Grendelmeier P, Simon D, Simon HU, Akdis CA, Wuthrich B. Epidemiology, clinical features, and immunology of the “intrinsic” (non-IgE-mediated) type of atopic dermatitis (constitutional dermatitis). Allergy 2001;56:841849.
  • 15
    Zutavern A, Hirsch T, Leupold W, Weiland S, Keil U, Von Mutius E. Atopic dermatitis, extrinsic atopic dermatitis and the hygiene hypothesis: results from a cross-sectional study. Clin Exp Allergy 2005;35:13011308.
  • 16
    Illi S, Von Mutius E, Lau S, Nickel R, Gruber C, Niggemann B et al. The natural course of atopic dermatitis from birth to age 7 years and the association with asthma. J Allergy Clin Immunol 2004;113:925931.
  • 17
    Kabesch M, Carr D, Weiland SK, Von Mutius E. Association between polymorphisms in serine protease inhibitor, kazal type 5 and asthma phenotypes in a large German population sample. Clin Exp Allergy 2004;34:340345.
  • 18
    Moffatt MF. SPINK5: a gene for atopic dermatitis and asthma. Clin Exp Allergy 2004;34:325327.
  • 19
    Weidinger S, Klopp N, Rummler L, Wagenpfeil S, Novak N, Baurecht HJ et al. Association of NOD1 polymorphisms with atopic eczema and related phenotypes. J Allergy Clin Immunol 2005;116:177184.
  • 20
    Alfven T, Braun-Fahrlander C, Brunekreef B, Von Mutius E, Riedler J, Scheynius A et al. Allergic diseases and atopic sensitization in children related to farming and anthroposophic lifestyle – the PARSIFAL study. Allergy 2006;61:414421.
  • 21
    Asher MI, Keil U, Anderson HR, Beasley R, Crane J, Martinez F et al. International Study of Asthma and Allergies in Childhood (ISAAC): rationale and methods. Eur Respir J 1995;8:483491.
  • 22
    Wickman M, Kull I, Pershagen G, Nordvall SL. The BAMSE project: presentation of a prospective longitudinal birth cohort study. Pediatr Allergy Immunol 2002;13(Suppl. 15):1113.
  • 23
    Riedler J, Braun-Fahrlander C, Eder W, Schreuer M, Waser M, Maisch S et al. Exposure to farming in early life and development of asthma and allergy: a cross-sectional survey. Lancet 2001;358:11291133.
  • 24
    Schram D, Doekes G, Boeve M, Douwes J, Riedler J, Ublagger E et al. Bacterial and fungal components in house dust of farm children, Rudolf Steiner school children and reference children – the PARSIFAL Study. Allergy 2005;60:611618.
  • 25
    Douwes J, Doekes G, Montijn R, Heederik D, Brunekreef B. Measurement of beta(1[RIGHTWARDS ARROW]3)-glucans in occupational and home environments with an inhibition enzyme immunoassay. Appl Environ Microbiol 1996;62:31763182.
  • 26
    Douwes J, Van Der Sluis B, Doekes G, Van Leusden F, Wijnands L, Van Strien R et al. Fungal extracellular polysaccharides in house dust as a marker for exposure to fungi: relations with culturable fungi, reported home dampness, and respiratory symptoms. J Allergy Clin Immunol 1999;103:494500.
  • 27
    Chambless L, Boyle K. Maximum likelihood methods for complex sample data: logistic regression and discrete proportional hazerds models. Commun Stat – Theory and Methods 1985;14:13771392.
  • 28
    Novak N, Bieber T. Allergic and nonallergic forms of atopic diseases. J Allergy Clin Immunol 2003;112:252262.
  • 29
    Folster-Holst R, Pape M, Buss YL, Christophers E, Weichenthal M. Low prevalence of the intrinsic form of atopic dermatitis among adult patients. Allergy 2006;61:629632.
  • 30
    Contoli M, Message SD, Laza-Stanca V, Edwards MR, Wark PA, Bartlett NW et al. Role of deficient type III interferon-lambda production in asthma exacerbations. Nat Med 2006;12:10231026.
  • 31
    Ege MJ, Frei R, Bieli C, Schram-Bijkerk D, Waser M, Benz MR et al. Not all farming environments protect from the development of asthma and wheeze in children. J Allerg Clin Immunol 2007;119:11401147.
  • 32
    Waser M, Michels KB, Bieli C, Flöistrup H, Pershagen G, Mutius Ev et al. Inverse association of farm milk consumption with asthma and allergy in rural and suburban populations across Europe. Clin Exp Allergy 2007;37:661670.
  • 33
    Douwes J. (1[RIGHTWARDS ARROW]3)-Beta-d-glucans and respiratory health: a review of the scientific evidence. Indoor Air 2005;15:160169.