Published ahead of print on July 25, 2002, doi:10.1164/rccm.200203-256OC
© 2002 American Thoracic Society
House Dust Endotoxin and Allergic Sensitization in ChildrenGSF National Research Center for Environment and Health, Institute of Epidemiology, Neuherberg; Ludwig-Maximilians-University of Munich, Institute of Medical Data Management, Biometrics, and Epidemiology, Chair of Epidemiology, Munich; Friedrich-Schiller-University of Jena, Institute of Occupational, Social, and Environmental Medicine, Department of Indoor Climatology (ark), Erfurt; and Friedrich-Schiller-University of Jena, Institute of Clinical Immunology, Jena, Germany Correspondence and requests for reprints should be addressed to Ulrike Gehring, GSF-National Research Centre for Environment and Health, Institute of Epidemiology, Ingolstaedter Landstr. 1, D-85764 Neuherberg, Germany. E-mail: gehring{at}gsf.de
A higher exposure to endotoxin was hypothesized to contribute to lower prevalence of allergic sensitization and hay fever in children growing up on a farm. We studied the association between house dust endotoxin and allergic sensitization. We randomly selected 740 children, aged between 5 and 10 years, from a group of children who participated in two cross-sectional surveys performed in Saxony-Anhalt, Germany, from 1992 to 1993 and from 1995 to 1996, such that 50% of the children were atopic or had a diagnosis of asthma. From 1996 to 1998, we collected living-room floor dust in the homes of 454 of these children (61%). The content of endotoxin in house dust was quantified using a chromogenic kinetic limulus amoebocyte lysate test and was related with health outcomes measured in the preceding cross-sectional surveys. Multiple logistic regression analyses adjusted for place of residence, sex, age, parental education, parental atopy, and pet ownership showed a negative association between exposure to endotoxin and sensitization to one or more allergens (aOR [95% CI] 0.95 [0.83; 1.10]) and two or more allergens (aOR [95% CI] 0.80 [0.67; 0.97]) using 0.35 kU/L as the cutoff value for sensitization. The protective effect was strengthened with increasing degree of sensitization. In conclusion, exposure to higher levels of house dust endotoxin is associated with lower prevalence of allergic sensitization in children.
Key Words: endotoxin • house dust • atopy • school children
Several different hypotheses have been proposed to explain the rise in asthma and allergic diseases. One is the so-called hygiene hypothesis (1, 2) that is still under debate (3, 4). It states that increased cleanliness in modern societies may be partly responsible for the increase in atopic diseases. Endotoxin has been considered as a potential marker for the hygienic level (5). Endotoxins are cell wall components of the outer membrane of gram-negative bacteria. They are known to have strong immune-stimulatory and proinflammatory properties (6). Endotoxins are believed to promote asthma by inducing airway inflammation on the one hand (7, 8) and to have protective effects on atopy development by activating type-1 T-helper cell responses on the other (9). Endotoxins are more or less ubiquitous in the environment and are present in normal indoor environments as constituents of house dust (7, 10–12). Increased concentrations of house dust endotoxin were found in homes of farming families, in households where children had regular contact with farm animals (13), and where animals were kept indoors (5, 11, 12, 14). Several cross-sectional studies have shown a consistently lower risk of allergic sensitization (15–20), allergic rhinitis, and hay fever (15–17, 21, 22) in children who grew up on farms. Other studies have shown a protective effect of exposure to pets on cat allergy, in particular, for children with atopic heredity (23–25). Nafstad and coworkers (26) reported protective effects of exposure to pets early in life on the development of allergic diseases such as atopic eczema and allergic rhinitis (26). Furthermore, one cross-sectional study has demonstrated a negative association between presence of pets in the home during childhood and adult atopy (27). Therefore, it was hypothesized that the level of environmental exposure to endotoxin and other bacterial wall components early in life might be protective in the development of allergic diseases in childhood (2, 13). At present, few studies can be found relating objectively measured exposures to endotoxin in house dust with allergies in children. One small study reported a negative association between exposure to house dust endotoxin and allergic sensitization (9). The authors found exposure to high levels of house dust endotoxin to be protective with respect to atopy by the activation of type-1 T-helper cell responses (9). Other studies have shown that high levels of house dust endotoxin are associated with an increased risk of wheezing (12), increased peak flow variability (11), and exacerbation of asthma (10). Therefore, the objective of the present work was to analyze associations between house dust endotoxin levels and allergic sensitization in children aged between 5 and 10 years, using data from a study designed to assess influences of indoor factors and genetics in asthma.
Subjects We conducted two cross-sectional surveys from 1992 to 1993 and from 1995 to 1996 to study the long-term effects of ambient air pollution on the health of German school children aged between 5 and 14 years and living in three areas of Saxony-Anhalt (participation rates: n = 2,470 [89.1%] and 2,814 [74.7%] children, respectively) (28). We selected a random sample of 740 children, aged between 5 and 10 years, from both survey populations for the present study on influences of indoor factors and genetics in asthma. Children were selected such that 50% of the children were either atopic, determined by skin prick testing or measuring serum-specific IgE (CAP-RAST-FEIA method: IgE > 0.35 kU/L; Pharmacia Diagnostics, Freiburg, Germany), or had a physician diagnosis of asthma at any time before the survey. We used a stratified random sampling approach in two age groups (5–6 and 8–10 years) from three residential areas. Parents of 454 of the 740 children (response rate was 61%) consented to dust collection in their homes. Blood samples originating from the preceding cross-sectional surveys were available for 444 of these children. The present study includes these 444 children for whom blood and dust samples were available for analysis. The study protocol was approved by the Ethics Committees of the University of Rostock and the University of Munich. Written informed consent was obtained from the parents of all participating children.
IgE Measurement
Questionnaire Data
Dust Sampling, Extraction, and Analysis Endotoxin was assayed with a quantitative kinetic chromogenic Limulus Amebocyte Lysate method (14), using Escherichia coli endotoxin (lot no. 0L4940; Bio Whittaker, Walkersville, UK) as the standard endotoxin. Additional detail on the analysis of endotoxin is provided in the online data supplement. Sample allergen content of house dust mite (D. pteronyssinus [Der p1] and Dermatophagoides farinae [Der f1]) and cat (Fel d1) was measured using a two-site monoclonal enzyme-linked immunosorbent assay, following the manufacturer's instructions (Indoor Biotechnologies, Clwyd, UK) (29).
Statistical Methods Generalized additive models (30) were used to investigate the functional relationship between allergic sensitization measured in the cross-sectional surveys and endotoxin measurements conducted in the study on influences of indoor factors and genetics in asthma. The additive effect of the endotoxin was estimated nonparametrically with loess smoothers in S-Plus version 6.0 (Insightful Corporation, Seattle, WA). Pointwise ± 2 SE bands for the smooth curves were computed. Due to the linear relationship between the log-odds and natural log(ln)-transformed endotoxin concentrations, a parametric approach using linear logistic regression analyses (31) with ln-transformed endotoxin levels was applied subsequently. We controlled for the influence of potentially confounding factors and other risk factors such as place of residence, sex, age, parental education, parental atopy, and pet ownership (dog and/or cat) by including them in the regression models. Statistical significance was set at 5% level. Results are presented as crude and adjusted odds ratios (OR) with 95% confidence intervals (CI) associated with an increase in exposure of two geometric standard deviation values. Computations were performed using SAS for Windows version 8.2 (SAS Institute, Cary, NC).
Description of the Study Population An overview of the study population is presented in Table 1 . The number of boys was slightly more than that of girls in the study population. Sixty-two percent were first graders aged between 5 and 6 years. Twenty percent of the parents had a higher education (at least 12th grade). Twenty-three percent of the parents were atopic. Twenty-two percent of the families had a cat and/or a dog indoors. One-third of the children were living in the same home since birth. In addition, Table 1 presents the prevalence of allergic symptoms and diseases.
Allergic Sensitization Prevalence of allergic sensitization using different cutoff values is presented in Table 2 . Forty percent of the children had at least one elevated serum-specific IgE against five allergens tested; 20% had multiple sensitization using 0.35 kU/L as the cutoff value. Sensitization to grass pollen was most frequent, followed by house dust mite, birch pollen, and cat dander. Increasing the cutoff value for sensitization to 3.5 kU/L resulted in sensitization rates of 19% for at least one allergen and 7% for multiple sensitization. Prevalence of allergen-specific sensitization, except sensitization to grass pollen, decreased to 5% or less.
Exposure to Endotoxin and Allergens Endotoxin levels in living-room floor dust ranged from 160 to 2,670,001 EU/m2, with geometric mean (geometric standard deviation) of 24,221 (4.1) EU/m2. Mite (sum of Der p1 and Der f1) and cat allergen levels ranged from below the limit of detection (24 and 30% of the homes, respectively) to 182,144 and 5,882,000 ng/m2, with geometric mean (geometric standard deviation) of 456 (16.6) and 251 (26.3) ng/m2, respectively.
Associations between Endotoxin and Allergic Sensitization
Crude and adjusted logistic regression results describing the association between allergic sensitization and ln-transformed living-room floor dust endotoxin are presented in Table 3 . All odds ratios (except the crude OR for C. herbarum using a cutoff value of 3.5 kU/L) were below one, indicating a protective effect of higher levels of house dust endotoxin. Adjusted odds ratios differ in most cases only slightly from the crude odds ratios, indicating that the observed effects were not confounded by other covariates included in the model.
There was a statistically significant decrease in the risk of multiple sensitization (aOR = 0.80; 95% CI = 0.67–0.97) and sensitization to C. herbarum (aOR = 0.70; 95% CI = 0.51–0.97) using a cutoff value of 0.35 kU/L for sensitization. On increasing the cutoff value to 0.70 and 3.50 kU/L, respectively, the protective effects of endotoxin on sensitization to at least one allergen and sensitization to multiple allergens were strengthened. In particular, regarding allergen-specific sensitization, increasing the cutoff value strengthened the effects on sensitization to grass and cat allergens. Using 3.50 kU/L as the cutoff value, the effects of endotoxin on sensitization to at least one allergen (aOR = 0.82; 95% CI = 0.68–0.98) and sensitization to grass pollen became statistically significant (aOR = 0.76; 95% CI = 0.62–0.94). We conducted sensitivity analyses and compared children who had lived in the same home since birth with children who had moved at least once during their life. The protective effects of endotoxin on sensitization to the indoor allergens of mite and cat were found to be stronger in the former group compared with the latter one. The results of these analyses using 0.70 kU/L as the cutoff value for sensitization are shown in Figure 2 . On using 0.35 kU/L as the cutoff value for sensitization, the results are very similar (data not shown), with differences between the two groups being only slightly less pronounced. Using the cutoff value of 3.50 kU/L, the effects on sensitization to mite, cat, and C. herbarum were not estimable for the group of children living in the same home since birth because of sample size limitations.
In addition, we looked for effect modification by parental atopy and sex. Stratified analyses showed no systematic differences in endotoxin effects between boys and girls and children with and without parental atopy, respectively (data not shown).
Associations between Endotoxin and Allergic Symptoms and Diseases
Our study demonstrates for the first time that exposure to higher levels of house dust endotoxin is associated with a decreased prevalence of allergic sensitization in school children. However, associations between physician-diagnosed atopic symptoms and diseases such as asthma, hay fever, and eczema were not found.
One small study assessing the relationship between exposure to house dust endotoxin and allergic sensitization has been published so far (9). The results of the present study, including children with and without parental atopy, are in accordance with the findings of this study published by Gereda and coworkers (9), who found a negative association between exposure to house dust endotoxin and allergic sensitization in infants prone to asthma. In addition, Gereda and coworkers (9) found a correlation between house dust endotoxin concentration and interferon- We compared the effects of house dust endotoxin for different degrees of allergic sensitization. The effects were strengthened using higher cutoff values for sensitization, indicating that exposure to higher levels of house dust endotoxin in particular decreases the risk of more severe sensitization. Although different cutoff values for allergic sensitization (0.35, 0.70, and 3.5 kU/L) were used in the studies on house dust endotoxin, farming environment, and pet ownership cited previously, no direct comparison of the results for different cutoff values has been published so far. In the majority of cases, the protective effects of owning pets on atopy were limited to cat-ownership (23–25). This might be explained by the fact that cat-ownership is associated with increased levels of house dust endotoxin (5, 14) or by the development of allergen-specific IgG4 antibodies as a part of a modified Th2 response, as described by Platts-Mills and coworkers (32). Therefore, we adjusted for cat-ownership and cat allergens in house dust in addition to the other covariates. However, the endotoxin effects remained unchanged after adjustment (data not shown). Both the level of house dust endotoxin and the timing of exposure are important. Tulic and coworkers (33) showed, in an animal model, that exposure to bacterial lipopolysaccharide immediately before or shortly after primary allergen sensitization inhibits the increase of allergen-specific IgE levels, in contrast to the exacerbating effects of a lipopolysaccharide exposure, several days after allergen challenge on the allergic response. Riedler and coworkers (15) demonstrated that exposure to stables in or before the first year of life is crucial for the protective effect. One major limitation of the present study is that exposure to house dust endotoxin was measured when the children were between 5 and 10 years of age. Thus, no information on early exposure to house dust endotoxin is available. We approached this problem by performing separate analyses for children living in the same home since birth and children who moved at any time during their life. The underlying assumption is that endotoxin measurements performed in the homes of children aged between 5 and 10 years yield a better estimate of exposure in early life for children who never moved compared with children who have moved at any time during their life. This assumption is supported by the findings of Park and coworkers (34) and J. Heinrich (personal communication) who showed that between-home variability of endotoxin measurements in floor dust is larger than that within home variability. We found that endotoxin effects on sensitization to any allergen (data not shown) and, in particular, on sensitization to the indoor allergens (mite and cat) were stronger in the former group compared with the latter group. Our interpretation of these findings is that moving introduces some noise in the exposure assessment, which makes endotoxin effects more difficult to detect. Therefore, the odds ratios for the children who have moved tended toward one. Because changes associated with moving from one home to another are more relevant to the indoor environment rather than to outdoor environments, differences in endotoxin effects were more pronounced for indoor allergens compared with outdoor allergens.
Parental atopy is an important risk factor for allergic sensitization and allergic disease in childhood (35). We found an association between allergic sensitization (any specific IgE Several studies have consistently shown a protective effect of the farm environment and of owning pets on allergic rhinitis and hay fever (15, 26). In the present study, no association was found between exposure to endotoxin and parental reported hay fever and symptoms of reddened eyes and/or sneeze attacks and/or runny/stuffy nose, respectively. But the number of children for whom hay fever was reported is very small. The associations between endotoxin and wheezing and asthma are less clear. Some studies have shown an association between exposure to endotoxin and exacerbation of asthma (10, 11) and increased risk of wheezing (12), whereas others have shown protective effects of the farming environment on asthma and wheezing (15, 16, 18, 19, 21, 22). In the present study, we could not find an association between house dust endotoxin and asthma and wheezing, neither for all children with asthma and wheezing nor for sensitized children with asthma and wheezing. But the statistical power for (extrinsic) asthma and (extrinsic) wheezing is very low (less than 10%). Thus, we cannot conclude from the lack of significant association that there is no association at all. More detailed studies are needed to study the association between house dust endotoxin and asthma. The lack of an association between house dust endotoxin and eczema is similar to the results of Tariq and coworkers (35) and Riedler and coworkers (15) for exposure to pets and a farm environment, respectively. Eczema, in the present study was not restricted to atopic eczema. Thus, the eczema variable includes both atopic eczema and other types of eczema, which is probably not specific enough. It is common to express endotoxin concentrations either as endotoxin units per square meter (EU/m2) or as endotoxin units per gram of dust (EU/g) dust. Currently, our knowledge on health effects associated with house dust endotoxin is limited, and the role of these two measures regarding health effects is not clear yet. Therefore, we analyzed associations between allergic sensitization and endotoxin levels expressed as EU/m2 as well as EU/g dust (data not shown). The two measures of endotoxin concentration were highly correlated (r = 0.88); therefore, their associations with allergic sensitization were rather similar. However, the associations with endotoxin levels expressed per square meter were somewhat stronger. For this reason we presented results for EU/m2. This is in line with the findings of Douwes and coworkers (11), who reported an association between peak flow variability and house dust endotoxin for concentrations expressed per square meter but not for concentrations expressed per gram of dust. Confounding variables were taken into account, and all results presented were adjusted for these variables. After adjustment for place of residence, sex, age group, parental education, and parental atopy, the associations of allergic sensitization with exposure to house dust endotoxin remained. In addition, we looked for confounding by older siblings, and mite and cat allergens in living-room floor dust, but the estimated effects of house dust endotoxin remained unchanged (data not shown). A potential limitation of the present study is that house dust endotoxin measurements and health outcome measurements were not performed at the same time. The time period between health outcome measurements and exposure assessment ranged from 1.0 to 5.4 years, with a median of 3.4 years. To date, no data has been published on the validity of a single endotoxin measurement for a time period of several years. However, there was no statistically significant difference in median time period between health outcome and exposure measurements between sensitized and nonsensitized children (3.6 and 3.4 years, respectively). If there was a misclassification of exposure due to the time lag between health outcome measurements and exposure assessment, it was nondifferential misclassification. Nondifferential misclassification of exposure biases the association between house dust endotoxin and allergic sensitization toward the null. Hence, in the absence of an association between endotoxin and sensitization, we could not conclude that there was no association at all. Because there is an association, we can conclude that this association is not just a consequence of misclassification of exposure. Another limitation in the design of the present study is that we cannot disentangle the chronologic order of exposure to endotoxin and manifestation of allergic sensitization. This can only be done in the framework of a cohort study. Thus, prospective cohort studies are needed to confirm the results of the present study. In conclusion, exposure to higher levels of house dust endotoxin is associated with lower prevalence of allergic sensitization. The protective effect of endotoxin is more pronounced in children who have lived in the same home since birth. INGA Study Group—GSF-National Research Center for Environment and Health, Neuherberg, Germany: Institute of Epidemiology (H.E. Wichmann, J. Heinrich, P. Schneider, J. Cyrys, I. Groß, A. Houzer, G. Wölke, G. Silbernagl, U. Gehring, B. Jacob, C. Frye), Institute of Ecological Chemistry (I. Gebefügi, G. Lörinci); Friedrich-Schiller-University, Jena, Germany: Institute of Occupational, Social, and Environmental Medicine (W. Bischof, A. Koch, J. Witthauer, K.J. Heilemann), Institute of Clinical Immunology (L. Jäger, B. Fahlbusch, G. Schlenvoigt); Grosshansdorf Hospital, Hamburg, Germany: Center for Pneumology and Thoracic Surgery (H. Magnussen, K. Richter, R. Jörres); Utrecht University, Utrecht, The Netherlands: Environmental and Occupational Health Group (B. Brunekreef, J. Douwes, G. Doekes).
The authors wish to thank Gaby Wölke and Brigitte Hollstein for the coordination of the home visits and all families for their participation, and Dr Gabriele Bolte for her critical comments on the draft manuscript.
The INGA study was supported by the Federal Ministry for Education, Science, Research, and Technology, grant EE 93016, and by the GSF-National Research Center for Environment and Health. This article has an online data supplement, which is accessible from this issue's table of contents online at www.atsjournals.org Received in original form March 27, 2002; accepted in final form July 2, 2002
This article has been cited by other articles:
|
||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
TOP
ABSTRACT
INTRODUCTION
3) as the cutoff level for sensitization.
–producing T-lymphocytes. The authors speculated that indoor exposure to endotoxin early in life might promote Th1 cell differentiation, which might minimize the risk of allergen sensitization. Furthermore, the negative associations between allergic sensitization and exposure to endotoxin are in line with the findings of a number of studies that have found a consistently lower risk of allergic sensitization in children who grew up on farms (
3)-ß-D-glucan and endotoxin in house dust and peak flow variability in children. Am J Respir Crit Care Med 2000;162:1348–1354.
