Ann Allergy Asthma Immunol 112 (2014) 132e139
Contents lists available at ScienceDirect
Perinatal influences on the development of asthma and atopy in childhood Sabina Illi, PhD *; Juliane Weber, MD *; Anne Zutavern, MD *; Jon Genuneit, MD y; Rudolf Schierl, PhD z; Christine Strunz-Lehner, MPH *; and Erika von Mutius, MD * * University
Children’s Hospital, Munich, Germany Institute of Epidemiology and Medical Biometry, University of Ulm, Ulm, Germany z Institute and Outpatient Clinic for Occupational, Social and Environmental Medicine, University Hospital of Munich, Munich, Germany y
A R T I C L E
I N F O
Article history: Received for publication June 25, 2013. Received in revised form November 19, 2013. Accepted for publication November 21, 2013.
A B S T R A C T
Background: In most children with asthma and atopy, onset of disease occurs early in life, indicating a crucial role of in utero and early childhood environment. However, only a small part of this burden of disease established early in life has been explained. Objective: To examine the effects of early environmental exposures on the development of asthma and atopy within the setting of an affluent urban population. Methods: The authors followed 526 German children from birth to 5 years of age. Parental interviews in pregnancy and then yearly assessed the health of the child and environmental characteristics. Endotoxin and allergens in house dust were measured at 3 months. Atopic sensitization was assessed at 1 and 5 years. Results: In atopic mothers, acute atopic symptoms during pregnancy were associated with increased risk of early atopic dermatitis (adjusted odds ratio [aOR] 1.74, 95% confidence interval [CI] 1.00e3.02) and allergic rhinitis at 5 years (aOR 2.11, 95% CI 1.01e4.41). Further, maternal illnesses during pregnancy (ie, repeated common colds) increased the risk of asthma at 5 years (aOR 2.31, 95% CI 1.12e4.78). Endotoxin in the child’s mattress was inversely associated with atopic sensitization (aOR 0.79, 95% CI 0.64e0.97) and asthma (aOR 0.71, 95% CI 0.55e0.93). A contrasting effect of early endotoxin and mite exposure was observed for mite sensitization: mite exposure increased the risk of mite sensitization at 5 years (aOR 1.30, 95% CI 1.11e1.53), whereas endotoxin exposure was inversely associated with mite sensitization (aOR 0.73, 95% CI 0.57e0.95). Conclusion: Factors affecting the in utero environment, such as maternal atopy and infections, and bacterial exposure in pregnancy or early life may act as immunomodulators enhancing or inhibiting the development of asthma and atopy in childhood. Ó 2014 American College of Allergy, Asthma & Immunology. Published by Elsevier Inc. All rights reserved.
Introduction In most children with asthma and atopy, the onset of disease occurs early in life, indicating a crucial role of the in utero environment and of determinants acting in early childhood.1,2 However, only a small part of this burden of disease established early in life has been explained. Among others, the role of maternal illnesses during pregnancy has not been resolved, and the role of bacterial and allergen exposure in early life coherently has not been clarified. The German PAULA (Perinatal Asthma and Environment Longterm Allergy Study) birth cohort was designed within the framework of the Asthma Multicentre Infants Cohort Study (AMICS), a consortium of 4 separate birth cohorts in England, Spain, and
Germany with common study protocols.3e6 The aim of these studies was to determine the relation between environmental factors relating to the early development and severity of atopy and the early indications of the development of an asthmatic phenotype. The specific focus was on environmental influences and infection as etiologic factors in atopy and asthma in young children. The German PAULA study comprises a comprehensive assessment of pregnancy-related and early life factors and early endotoxin and allergen exposures. The objective was to shed light on the association of early environment and the outcome of asthma and atopy up 5 years of age in an urban and affluent population. Methods
Reprints: Sabina Illi, PhD, Dr von Hauner Children’s Hospital, Ludwig-MaximiliansUniversity Munich, Member of the German Center for Lung Research, Lindwurmstrasse 4, 80337 Munich, Germany; ; E-mail: [email protected]. Disclosures: Dr von Mutius has served as consultant to GlaxoSmithKline, Novartis, ALK Abelló, and Astellas Pharma Europe. Funding: This work was supported by the Bavarian Ministry of Environment and Health (Bayerisches Ministerium für Umwelt und Gesundheit; grant 76b-8736.61999/7).
Cohort Recruitment and Follow-up The aim of the longitudinal PAULA study was to examine the effects of pre- and postnatal environmental and lifestyle-based exposures on the development of asthma and atopy in childhood within the fairly affluent and urban setting of Greater Munich, Germany. Recruitment took place from October 1999 to December
1081-1206/13/$36.00 - see front matter Ó 2014 American College of Allergy, Asthma & Immunology. Published by Elsevier Inc. All rights reserved. http://dx.doi.org/10.1016/j.anai.2013.11.019
S. Illi et al. / Ann Allergy Asthma Immunol 112 (2014) 132e139
2001. All pregnant women presenting for antenatal care in the obstetric department of 1 inner city hospital were invited to participate in the study. Exclusion criteria were hospital admission during pregnancy, no permanent residence within Greater Munich, not having a telephone, and no proficient German-language capability. Five hundred thirteen eligible women agreed to participate, resulting in 526 live births. The cohort was followed until the fifth birthday of each child. The study was approved by the ethics committee of the Bavarian Medical Association. Written informed consent was obtained from all study participants. The Munich birth cohort is part of the AMICS that previously recruited 3 separate birth cohorts in Ashford (United Kingdom), Barcelona (Spain), and Menorca Island (Spain).3e6 Questionnaires During pregnancy and when the child was 3 and 12 months old and then yearly up to 5 years of age, parents were interviewed to assess the health status of the child and the environmental and socioeconomic characteristics of the household. Questions concerned asthmatic and atopic symptoms and diagnoses and factors concerning hygiene, environmental factors, and socioeconomic factors. Among others, parental history of atopic diseases, maternal illnesses during pregnancy, level of parental education, parental smoking, pet ownership, household size, living space, and number of siblings were assessed. Definition of Atopic and Wheezing Outcomes Study outcomes were defined in analogy to other AMICS populations.3e5 Symptoms of atopic dermatitis in the first year of life were defined as an itchy rash that came and went or that lasted at least 3 months. In the second year of life, symptoms of atopic dermatitis were defined as an itchy rash with scratching and rubbing on the inner elbows, back of the knees, ankles, neck, or around the eyes. Early atopic dermatitis was defined as present if at least 1 of the following applied in the first 2 years of life: (1) reported doctor’s diagnosis, (2) parental reporting of symptoms of atopic dermatitis, and (3) visible atopic dermatitis at the time of the 1year follow-up based on a positive Scoring Atopic Dermatitis index. Early wheeze was defined as any wheezing in the first 3 years of life. Current wheeze at 4 to 5 years of age was defined as any wheezing in the previous 12 months at the 4- or 5-year follow-up, and current allergic rhinitis at 4 to 5 years of age was defined as sneezing or a runny, blocked, or itchy nose without a cold in the previous 12 months at the 4- or 5-year follow-up. A doctor’s diagnosis of asthma ever was defined as a doctor’s diagnosis of asthma at least once or a doctor’s diagnosis of recurrent wheezy bronchitis ever with repeated (2) episodes of wheeze up to 5 years of age. Early food sensitization, early atopic dermatitis, current allergic rhinitis, and current inhalant sensitization were grouped as “atopic nonwheezing outcomes” and early and late wheeze and a doctor’s diagnosis of asthma were grouped as “wheezing outcomes.” Atopic Sensitization of the Child At the 1- and 5-year follow-ups, the children were invited to the clinic for the assessment of atopic sensitization. At 1 year blood samples were taken and serum total IgE levels and specific IgE antibodies to relevant food allergens (cow’s milk and egg white) were determined by CAP-RAST FEIA (Pharmacia & Upjohn, Freiburg, Germany). Early food sensitization was defined as specific IgE antibodies of at least 0.35 kU/L against either allergen. At 5 years the children underwent skin prick testing to common inhalant allergens: house dust mites (Dermatophagoides pteronyssinus and Dermatophagoides farinae), cat, Alternaria species, grass mix, birch, and hazelnut. A child was defined as sensitized to the respective
133
allergen if it provoked a mean wheal diameter at least 2 mm larger than that from a negative (saline) control. Current inhalant sensitization was defined as any positive prick test result at 5 years of age. Parental Atopy and Illnesses At recruitment in pregnancy, health information was collected from the parents, including a history of allergic diseases and current maternal symptoms, illnesses, and medication during pregnancy. Parental atopy was defined as a reported doctor’s diagnosis of asthma, hay fever, or atopic dermatitis. Current maternal atopic symptoms in pregnancy were defined as symptoms of asthma, hay fever, or atopic dermatitis during pregnancy. The number of maternal colds during pregnancy was assessed by the question: “Since beginning of the pregnancy, how often did you have a common cold, eg, runny or stuffed nose, cough, hoarseness, or tonsillitis?” Furthermore, blood samples were taken from both parents to assess specific IgE to 20 common food and inhalant allergens as determined by CAP-RAST FEIA (Pharmacia & Upjohn). Parental atopic sensitization was defined as a specific IgE level of at least 0.35 kU/L to any of the tested allergens. Allergens and Endotoxin in Dust At 3 months of age, each family was visited at home by a trained field worker and dust samples were collected. These samples were assayed for concentrations of endotoxin, house dust mite (Der p 1 and Der f 1), and cat (Fel d 1) allergens using standard techniques as described previously.6,7 In brief, separate dust samples were collected by vacuuming the child’s and mother’s mattresses and the living room floor. The main focus of dust analysis was on the assessment of endotoxin. When limited dust was available for a specific site, endotoxin was analyzed first and then the remaining dust was analyzed for the allergens. All allergens were measured by monoclonal antibody enzyme-linked immunosorbent assays and resulting allergen levels were expressed as micrograms per gram of dust. The concentrations of Der p 1 and Der f 1 were added and are presented as the “mite allergen level.” Endotoxin content of the dust samples was determined by a kinetic Limulus assay (kineticQCL, Lonza Ltd, Basel, Switzerland) with potency against Escherichia coli O55:B5 of 13 endotoxin units per nanogram. Resulting endotoxin levels were expressed as endotoxin units per milligram of dust. To take different sleeping situations into account, the authors asked the mother at the child’s age of 3 months where the child mostly slept: in its own or in the mother’s bed. Based on this information, the respective endotoxin and allergen levels from the child’s and mother’s mattresses were combined and a new exposure variable, “where the child sleeps,” was defined by taking the respective value from the mother’s mattress if the child mostly slept there and from the child’s mattress otherwise. Low and high exposures of endotoxin and mite allergen levels were defined as levels below or above the respective median. Based on these dichotomized exposures, 4 distinct exposure groups were defined to assess the combined effect on relevant outcomes: (1) high mite allergen/low endotoxin, (2) high mite allergen/high endotoxin, (3) low mite allergen/low endotoxin, and (4) low mite allergen/high endotoxin exposure. This combined variable was based on endotoxin from the child’s mattress and defined separately for mite allergen from the mother’s and from the child’s mattress. Statistical Analyses All children with at least 1 nonmissing outcome were included in the statistical analyses. Owing to missing values for questionnaire-based items and for laboratory measurements, the total sample size for various variables may vary. The association of early life exposures with outcomes was assessed by logistic
134
S. Illi et al. / Ann Allergy Asthma Immunol 112 (2014) 132e139
Results
Table 1 Characteristics and atopic outcomes of study population
Characteristics and Outcomes
Characteristics
n/N
%
Parental atopy (diagnosis of asthma, hay fever, or atopic dermatitis) Maternal atopy Paternal atopy Parental atopic sensitization Maternal atopic sensitization Paternal atopic sensitization Maternal atopic symptoms in pregnancy Maternal colds in pregnancy, n 0 1 2 3 High maternal education (college, university) High paternal education (college, university) Maternal smoking ever at birth Cat at home During pregnancy At 1 y of age Dog at home During pregnancy At 1 y of age Older siblings, n 0 1 2 Atopic and wheezing outcomes Early in life Early sensitization to egg or milk (1 y) Early atopic dermatitis (2 y) Early wheeze (3 y) At end of follow-up Doctor’s diagnosis of asthma ever Current wheeze (4e5 y) Current allergic rhinitis (4e5 y) Any current inhalant sensitization (5 y) Current sensitization to indoor allergens (5 y) Current sensitization to pollen (5 y)
311/508
61.2%
194/513 194/502 333/431 216/473 203/386 125/513
37.8% 38.6% 77.3% 45.7% 52.6% 24.4%
123/509 213/509 129/509 44/509 225/513 300/504 203/513
24.2% 41.8% 25.3% 8.6% 43.9% 59.5% 39.6%
49/503 30/458
9.7% 6.5%
27/496 22/456
5.4% 4.8%
279/513 191/513 43/513
54.4% 37.2% 8.4%
32/370 167/476 108/505
8.6% 35.1% 21.4%
42/501 48/504 68/494 82/361 50/361 60/361
8.4% 9.5% 13.8% 22.7% 13.8% 16.6%
regression analysis and are presented as odds ratio (OR) or as OR adjusted for the potential confounders maternal atopy and older siblings (aOR) and 95% confidence interval (95% CI). Endotoxin and allergen levels are presented as median and interquartile range and as geometric mean and 95% CI. For inclusion in regression models, endotoxin and allergen levels were logarithmically transformed. Statistical analysis was performed with SAS 9.2 (SAS Institute, Cary, North Carolina); a P value of .05 was considered significant. Owing to the exploratory character of the analysis, corrections for multiple testing were not performed.8
Of the 526 children in the PAULA cohort, 513 (97.5%) had at least 1 nonmissing outcome and thus were included in the analyses. Mean gestational age at the time of study inclusion and pregnancyrelated questionnaire was 36.8 1.8 weeks. Mean completed weeks of gestation at birth was 39.7 1.5, and 20 infants (3.9%) were preterm. Of the 513 study children, 136 (26.5%) were born by cesarean section. Mean birthweight was 3,359.7 522.4 g. Table 1 lists the characteristics of the study population. A large proportion of families reported an atopic background: in 61.2% of families, either parent had a doctor’s diagnosis of asthma, hay fever, or atopic dermatitis. This large percentage of parental atopy also was observed when validating the data by specific IgE: in 77.3% of 431 families with nonmissing parental IgE data, at least 1 parent showed elevated specific IgE to at least 1 of the tested allergens. The relative affluence of the study population is depicted by the high level of education of both parents. Furthermore, more than half the study children were firstborns. As presented in Table 1, the high prevalence of parental atopy was in contrast to the fairly low prevalence in the children (early sensitization to egg or milk 8.6%, early atopic dermatitis 35.1%, early wheeze 21.4%, doctor’s diagnosis of asthma ever 8.4%, current wheeze 9.5%, current allergic rhinitis 13.8%, any current inhalant sensitization 22.7%). To assess potential confounding, the association of parental smoking (in pregnancy and currently), mode of delivery, gestational age at birth, birthweight, parental education, pet keeping, and older siblings with all outcomes was analyzed. Older siblings were associated with a lower risk of atopic outcomes, significantly so for early atopic dermatitis (Table 2). None of the other factors showed a significant association with any of the outcomes (data not shown). Maternal History of Atopy Parental atopy was associated with an increased risk for nonwheezing atopic outcomes, especially with those assessed at 5 years of age (ie, allergic rhinitis and atopic sensitization; Table 2). In contrast, no association was observed between parental atopy and wheezing outcomes in the child (Table 3). For the nonwheezing atopic outcomes, the effect of maternal atopy was strongest not only if the mother reported a doctor’s diagnosis of atopy but also if current symptoms during pregnancy were prevalent. These associations remained significant even in the subgroup of mothers with an underlying history of atopy (Table 4). This effect was mediated mainly by the association of current maternal symptoms of hay fever and atopic dermatitis in pregnancy and the respective outcome in the child (aOR 2.42, 95% CI
Table 2 Association of characteristics of study population with atopic outcomes
Parental atopy Maternal atopy Paternal atopy Parental atopic sensitization Maternal atopic sensitization Paternal atopic sensitization Maternal atopic symptoms in pregnancy Maternal colds in pregnancy, n 0 1 2 3 Older siblings
Early food sensitization
Early atopic dermatitis
Current allergic rhinitis
Current inhalant sensitization
OR
95% CI
P value
OR
95% CI
P value
OR
95% CI
P value
OR
95% CI
P value
1.73 2.43 2.04 2.03 2.11 1.62 2.78
0.78e3.86 1.17e5.06 0.97e4.29 0.68e6.02 0.99e4.49 0.72e3.66 1.32e5.84
.178 .018 .059 .204 .053 .246 .007
1.36 1.30 1.43 1.67 1.25 1.51 1.60
0.92e2.02 0.88e1.90 0.97e2.10 0.99e2.82 0.84e1.86 0.97e2.34 1.05e2.45
.126 .185 .075 .053 .266 .067 .029
2.36 1.73 2.18 2.99 1.48 2.28 2.19
1.30e4.26 1.04e2.90 1.30e3.65 1.24e7.18 0.86e2.55 1.22e4.27 1.27e3.75
.005 .036 .003 .014 .155 .010 .005
2.81 2.28 2.04 3.34 3.99 1.66 2.78
1.58e5.00 1.38e3.76 1.23e3.39 1.52e7.32 2.32e6.85 0.94e2.91 1.65e4.69
Contents lists available at ScienceDirect
Perinatal influences on the development of asthma and atopy in childhood Sabina Illi, PhD *; Juliane Weber, MD *; Anne Zutavern, MD *; Jon Genuneit, MD y; Rudolf Schierl, PhD z; Christine Strunz-Lehner, MPH *; and Erika von Mutius, MD * * University
Children’s Hospital, Munich, Germany Institute of Epidemiology and Medical Biometry, University of Ulm, Ulm, Germany z Institute and Outpatient Clinic for Occupational, Social and Environmental Medicine, University Hospital of Munich, Munich, Germany y
A R T I C L E
I N F O
Article history: Received for publication June 25, 2013. Received in revised form November 19, 2013. Accepted for publication November 21, 2013.
A B S T R A C T
Background: In most children with asthma and atopy, onset of disease occurs early in life, indicating a crucial role of in utero and early childhood environment. However, only a small part of this burden of disease established early in life has been explained. Objective: To examine the effects of early environmental exposures on the development of asthma and atopy within the setting of an affluent urban population. Methods: The authors followed 526 German children from birth to 5 years of age. Parental interviews in pregnancy and then yearly assessed the health of the child and environmental characteristics. Endotoxin and allergens in house dust were measured at 3 months. Atopic sensitization was assessed at 1 and 5 years. Results: In atopic mothers, acute atopic symptoms during pregnancy were associated with increased risk of early atopic dermatitis (adjusted odds ratio [aOR] 1.74, 95% confidence interval [CI] 1.00e3.02) and allergic rhinitis at 5 years (aOR 2.11, 95% CI 1.01e4.41). Further, maternal illnesses during pregnancy (ie, repeated common colds) increased the risk of asthma at 5 years (aOR 2.31, 95% CI 1.12e4.78). Endotoxin in the child’s mattress was inversely associated with atopic sensitization (aOR 0.79, 95% CI 0.64e0.97) and asthma (aOR 0.71, 95% CI 0.55e0.93). A contrasting effect of early endotoxin and mite exposure was observed for mite sensitization: mite exposure increased the risk of mite sensitization at 5 years (aOR 1.30, 95% CI 1.11e1.53), whereas endotoxin exposure was inversely associated with mite sensitization (aOR 0.73, 95% CI 0.57e0.95). Conclusion: Factors affecting the in utero environment, such as maternal atopy and infections, and bacterial exposure in pregnancy or early life may act as immunomodulators enhancing or inhibiting the development of asthma and atopy in childhood. Ó 2014 American College of Allergy, Asthma & Immunology. Published by Elsevier Inc. All rights reserved.
Introduction In most children with asthma and atopy, the onset of disease occurs early in life, indicating a crucial role of the in utero environment and of determinants acting in early childhood.1,2 However, only a small part of this burden of disease established early in life has been explained. Among others, the role of maternal illnesses during pregnancy has not been resolved, and the role of bacterial and allergen exposure in early life coherently has not been clarified. The German PAULA (Perinatal Asthma and Environment Longterm Allergy Study) birth cohort was designed within the framework of the Asthma Multicentre Infants Cohort Study (AMICS), a consortium of 4 separate birth cohorts in England, Spain, and
Germany with common study protocols.3e6 The aim of these studies was to determine the relation between environmental factors relating to the early development and severity of atopy and the early indications of the development of an asthmatic phenotype. The specific focus was on environmental influences and infection as etiologic factors in atopy and asthma in young children. The German PAULA study comprises a comprehensive assessment of pregnancy-related and early life factors and early endotoxin and allergen exposures. The objective was to shed light on the association of early environment and the outcome of asthma and atopy up 5 years of age in an urban and affluent population. Methods
Reprints: Sabina Illi, PhD, Dr von Hauner Children’s Hospital, Ludwig-MaximiliansUniversity Munich, Member of the German Center for Lung Research, Lindwurmstrasse 4, 80337 Munich, Germany; ; E-mail: [email protected]. Disclosures: Dr von Mutius has served as consultant to GlaxoSmithKline, Novartis, ALK Abelló, and Astellas Pharma Europe. Funding: This work was supported by the Bavarian Ministry of Environment and Health (Bayerisches Ministerium für Umwelt und Gesundheit; grant 76b-8736.61999/7).
Cohort Recruitment and Follow-up The aim of the longitudinal PAULA study was to examine the effects of pre- and postnatal environmental and lifestyle-based exposures on the development of asthma and atopy in childhood within the fairly affluent and urban setting of Greater Munich, Germany. Recruitment took place from October 1999 to December
1081-1206/13/$36.00 - see front matter Ó 2014 American College of Allergy, Asthma & Immunology. Published by Elsevier Inc. All rights reserved. http://dx.doi.org/10.1016/j.anai.2013.11.019
S. Illi et al. / Ann Allergy Asthma Immunol 112 (2014) 132e139
2001. All pregnant women presenting for antenatal care in the obstetric department of 1 inner city hospital were invited to participate in the study. Exclusion criteria were hospital admission during pregnancy, no permanent residence within Greater Munich, not having a telephone, and no proficient German-language capability. Five hundred thirteen eligible women agreed to participate, resulting in 526 live births. The cohort was followed until the fifth birthday of each child. The study was approved by the ethics committee of the Bavarian Medical Association. Written informed consent was obtained from all study participants. The Munich birth cohort is part of the AMICS that previously recruited 3 separate birth cohorts in Ashford (United Kingdom), Barcelona (Spain), and Menorca Island (Spain).3e6 Questionnaires During pregnancy and when the child was 3 and 12 months old and then yearly up to 5 years of age, parents were interviewed to assess the health status of the child and the environmental and socioeconomic characteristics of the household. Questions concerned asthmatic and atopic symptoms and diagnoses and factors concerning hygiene, environmental factors, and socioeconomic factors. Among others, parental history of atopic diseases, maternal illnesses during pregnancy, level of parental education, parental smoking, pet ownership, household size, living space, and number of siblings were assessed. Definition of Atopic and Wheezing Outcomes Study outcomes were defined in analogy to other AMICS populations.3e5 Symptoms of atopic dermatitis in the first year of life were defined as an itchy rash that came and went or that lasted at least 3 months. In the second year of life, symptoms of atopic dermatitis were defined as an itchy rash with scratching and rubbing on the inner elbows, back of the knees, ankles, neck, or around the eyes. Early atopic dermatitis was defined as present if at least 1 of the following applied in the first 2 years of life: (1) reported doctor’s diagnosis, (2) parental reporting of symptoms of atopic dermatitis, and (3) visible atopic dermatitis at the time of the 1year follow-up based on a positive Scoring Atopic Dermatitis index. Early wheeze was defined as any wheezing in the first 3 years of life. Current wheeze at 4 to 5 years of age was defined as any wheezing in the previous 12 months at the 4- or 5-year follow-up, and current allergic rhinitis at 4 to 5 years of age was defined as sneezing or a runny, blocked, or itchy nose without a cold in the previous 12 months at the 4- or 5-year follow-up. A doctor’s diagnosis of asthma ever was defined as a doctor’s diagnosis of asthma at least once or a doctor’s diagnosis of recurrent wheezy bronchitis ever with repeated (2) episodes of wheeze up to 5 years of age. Early food sensitization, early atopic dermatitis, current allergic rhinitis, and current inhalant sensitization were grouped as “atopic nonwheezing outcomes” and early and late wheeze and a doctor’s diagnosis of asthma were grouped as “wheezing outcomes.” Atopic Sensitization of the Child At the 1- and 5-year follow-ups, the children were invited to the clinic for the assessment of atopic sensitization. At 1 year blood samples were taken and serum total IgE levels and specific IgE antibodies to relevant food allergens (cow’s milk and egg white) were determined by CAP-RAST FEIA (Pharmacia & Upjohn, Freiburg, Germany). Early food sensitization was defined as specific IgE antibodies of at least 0.35 kU/L against either allergen. At 5 years the children underwent skin prick testing to common inhalant allergens: house dust mites (Dermatophagoides pteronyssinus and Dermatophagoides farinae), cat, Alternaria species, grass mix, birch, and hazelnut. A child was defined as sensitized to the respective
133
allergen if it provoked a mean wheal diameter at least 2 mm larger than that from a negative (saline) control. Current inhalant sensitization was defined as any positive prick test result at 5 years of age. Parental Atopy and Illnesses At recruitment in pregnancy, health information was collected from the parents, including a history of allergic diseases and current maternal symptoms, illnesses, and medication during pregnancy. Parental atopy was defined as a reported doctor’s diagnosis of asthma, hay fever, or atopic dermatitis. Current maternal atopic symptoms in pregnancy were defined as symptoms of asthma, hay fever, or atopic dermatitis during pregnancy. The number of maternal colds during pregnancy was assessed by the question: “Since beginning of the pregnancy, how often did you have a common cold, eg, runny or stuffed nose, cough, hoarseness, or tonsillitis?” Furthermore, blood samples were taken from both parents to assess specific IgE to 20 common food and inhalant allergens as determined by CAP-RAST FEIA (Pharmacia & Upjohn). Parental atopic sensitization was defined as a specific IgE level of at least 0.35 kU/L to any of the tested allergens. Allergens and Endotoxin in Dust At 3 months of age, each family was visited at home by a trained field worker and dust samples were collected. These samples were assayed for concentrations of endotoxin, house dust mite (Der p 1 and Der f 1), and cat (Fel d 1) allergens using standard techniques as described previously.6,7 In brief, separate dust samples were collected by vacuuming the child’s and mother’s mattresses and the living room floor. The main focus of dust analysis was on the assessment of endotoxin. When limited dust was available for a specific site, endotoxin was analyzed first and then the remaining dust was analyzed for the allergens. All allergens were measured by monoclonal antibody enzyme-linked immunosorbent assays and resulting allergen levels were expressed as micrograms per gram of dust. The concentrations of Der p 1 and Der f 1 were added and are presented as the “mite allergen level.” Endotoxin content of the dust samples was determined by a kinetic Limulus assay (kineticQCL, Lonza Ltd, Basel, Switzerland) with potency against Escherichia coli O55:B5 of 13 endotoxin units per nanogram. Resulting endotoxin levels were expressed as endotoxin units per milligram of dust. To take different sleeping situations into account, the authors asked the mother at the child’s age of 3 months where the child mostly slept: in its own or in the mother’s bed. Based on this information, the respective endotoxin and allergen levels from the child’s and mother’s mattresses were combined and a new exposure variable, “where the child sleeps,” was defined by taking the respective value from the mother’s mattress if the child mostly slept there and from the child’s mattress otherwise. Low and high exposures of endotoxin and mite allergen levels were defined as levels below or above the respective median. Based on these dichotomized exposures, 4 distinct exposure groups were defined to assess the combined effect on relevant outcomes: (1) high mite allergen/low endotoxin, (2) high mite allergen/high endotoxin, (3) low mite allergen/low endotoxin, and (4) low mite allergen/high endotoxin exposure. This combined variable was based on endotoxin from the child’s mattress and defined separately for mite allergen from the mother’s and from the child’s mattress. Statistical Analyses All children with at least 1 nonmissing outcome were included in the statistical analyses. Owing to missing values for questionnaire-based items and for laboratory measurements, the total sample size for various variables may vary. The association of early life exposures with outcomes was assessed by logistic
134
S. Illi et al. / Ann Allergy Asthma Immunol 112 (2014) 132e139
Results
Table 1 Characteristics and atopic outcomes of study population
Characteristics and Outcomes
Characteristics
n/N
%
Parental atopy (diagnosis of asthma, hay fever, or atopic dermatitis) Maternal atopy Paternal atopy Parental atopic sensitization Maternal atopic sensitization Paternal atopic sensitization Maternal atopic symptoms in pregnancy Maternal colds in pregnancy, n 0 1 2 3 High maternal education (college, university) High paternal education (college, university) Maternal smoking ever at birth Cat at home During pregnancy At 1 y of age Dog at home During pregnancy At 1 y of age Older siblings, n 0 1 2 Atopic and wheezing outcomes Early in life Early sensitization to egg or milk (1 y) Early atopic dermatitis (2 y) Early wheeze (3 y) At end of follow-up Doctor’s diagnosis of asthma ever Current wheeze (4e5 y) Current allergic rhinitis (4e5 y) Any current inhalant sensitization (5 y) Current sensitization to indoor allergens (5 y) Current sensitization to pollen (5 y)
311/508
61.2%
194/513 194/502 333/431 216/473 203/386 125/513
37.8% 38.6% 77.3% 45.7% 52.6% 24.4%
123/509 213/509 129/509 44/509 225/513 300/504 203/513
24.2% 41.8% 25.3% 8.6% 43.9% 59.5% 39.6%
49/503 30/458
9.7% 6.5%
27/496 22/456
5.4% 4.8%
279/513 191/513 43/513
54.4% 37.2% 8.4%
32/370 167/476 108/505
8.6% 35.1% 21.4%
42/501 48/504 68/494 82/361 50/361 60/361
8.4% 9.5% 13.8% 22.7% 13.8% 16.6%
regression analysis and are presented as odds ratio (OR) or as OR adjusted for the potential confounders maternal atopy and older siblings (aOR) and 95% confidence interval (95% CI). Endotoxin and allergen levels are presented as median and interquartile range and as geometric mean and 95% CI. For inclusion in regression models, endotoxin and allergen levels were logarithmically transformed. Statistical analysis was performed with SAS 9.2 (SAS Institute, Cary, North Carolina); a P value of .05 was considered significant. Owing to the exploratory character of the analysis, corrections for multiple testing were not performed.8
Of the 526 children in the PAULA cohort, 513 (97.5%) had at least 1 nonmissing outcome and thus were included in the analyses. Mean gestational age at the time of study inclusion and pregnancyrelated questionnaire was 36.8 1.8 weeks. Mean completed weeks of gestation at birth was 39.7 1.5, and 20 infants (3.9%) were preterm. Of the 513 study children, 136 (26.5%) were born by cesarean section. Mean birthweight was 3,359.7 522.4 g. Table 1 lists the characteristics of the study population. A large proportion of families reported an atopic background: in 61.2% of families, either parent had a doctor’s diagnosis of asthma, hay fever, or atopic dermatitis. This large percentage of parental atopy also was observed when validating the data by specific IgE: in 77.3% of 431 families with nonmissing parental IgE data, at least 1 parent showed elevated specific IgE to at least 1 of the tested allergens. The relative affluence of the study population is depicted by the high level of education of both parents. Furthermore, more than half the study children were firstborns. As presented in Table 1, the high prevalence of parental atopy was in contrast to the fairly low prevalence in the children (early sensitization to egg or milk 8.6%, early atopic dermatitis 35.1%, early wheeze 21.4%, doctor’s diagnosis of asthma ever 8.4%, current wheeze 9.5%, current allergic rhinitis 13.8%, any current inhalant sensitization 22.7%). To assess potential confounding, the association of parental smoking (in pregnancy and currently), mode of delivery, gestational age at birth, birthweight, parental education, pet keeping, and older siblings with all outcomes was analyzed. Older siblings were associated with a lower risk of atopic outcomes, significantly so for early atopic dermatitis (Table 2). None of the other factors showed a significant association with any of the outcomes (data not shown). Maternal History of Atopy Parental atopy was associated with an increased risk for nonwheezing atopic outcomes, especially with those assessed at 5 years of age (ie, allergic rhinitis and atopic sensitization; Table 2). In contrast, no association was observed between parental atopy and wheezing outcomes in the child (Table 3). For the nonwheezing atopic outcomes, the effect of maternal atopy was strongest not only if the mother reported a doctor’s diagnosis of atopy but also if current symptoms during pregnancy were prevalent. These associations remained significant even in the subgroup of mothers with an underlying history of atopy (Table 4). This effect was mediated mainly by the association of current maternal symptoms of hay fever and atopic dermatitis in pregnancy and the respective outcome in the child (aOR 2.42, 95% CI
Table 2 Association of characteristics of study population with atopic outcomes
Parental atopy Maternal atopy Paternal atopy Parental atopic sensitization Maternal atopic sensitization Paternal atopic sensitization Maternal atopic symptoms in pregnancy Maternal colds in pregnancy, n 0 1 2 3 Older siblings
Early food sensitization
Early atopic dermatitis
Current allergic rhinitis
Current inhalant sensitization
OR
95% CI
P value
OR
95% CI
P value
OR
95% CI
P value
OR
95% CI
P value
1.73 2.43 2.04 2.03 2.11 1.62 2.78
0.78e3.86 1.17e5.06 0.97e4.29 0.68e6.02 0.99e4.49 0.72e3.66 1.32e5.84
.178 .018 .059 .204 .053 .246 .007
1.36 1.30 1.43 1.67 1.25 1.51 1.60
0.92e2.02 0.88e1.90 0.97e2.10 0.99e2.82 0.84e1.86 0.97e2.34 1.05e2.45
.126 .185 .075 .053 .266 .067 .029
2.36 1.73 2.18 2.99 1.48 2.28 2.19
1.30e4.26 1.04e2.90 1.30e3.65 1.24e7.18 0.86e2.55 1.22e4.27 1.27e3.75
.005 .036 .003 .014 .155 .010 .005
2.81 2.28 2.04 3.34 3.99 1.66 2.78
1.58e5.00 1.38e3.76 1.23e3.39 1.52e7.32 2.32e6.85 0.94e2.91 1.65e4.69
