ARTICLE |
a Obstructive Lung Disease in Northern Sweden (OLIN) Studies, Department of Medicine, Sunderby Central Hospital of Norrbotten, Luleå, Sweden
b Department of Respiratory Medicine and Allergy, University of Umeå, Umeå, Sweden
c Department of Environmental Health Sciences, Columbia University, New York, New York
d Asthma and Allergic Diseases Center, University of Virginia, Charlottesville, Virginia
e Lung and Allergy Research, National Institute of Environmental Medicine, Karolinska Institutet, Stockholm, Sweden
| ABSTRACT |
|---|
|
|
|---|
METHODS. A population-based cohort of 3430 (97% of invited) 7- to 8-year-old school children participated in an expanded International Study of Asthma and Allergy in Childhood survey, and two thirds were skin-prick tested. Heredity was defined as a family history of (1) asthma and (2) atopy (allergic rhinitis or eczema). Multivariate analyses corrected for known risk factors for asthma.
RESULTS. At ages 7 to 8, prevalence of asthma was 5.3% among the children and 9.0% among the parents. In children without parental asthma or parental atopy, the prevalence of asthma was 2.8%. Corrected for parental asthma, parental atopy was a weak but significant risk factor. There were minor differences in the impact of parental disease between sensitized and nonsensitized children and between boys and girls.
CONCLUSIONS. As risk factors for childhood asthma, there were major differences between parental asthma and parental atopy. Sibling asthma was only a marker of parental disease. Interactions between parental disease and the child's allergic sensitization or gender were not statistically significant. Asthma in both parents conferred a multiplicative risk, whereas the effect of parental atopy was additive, however limited. Asthma and atopy, despite their causal relationship, are separate entities and could be inherited differently. This large, population-based, and well-characterized cohort study does not confirm parent-of-origin effects found in previous studies.
Key Words: allergy asthma family history prevalence risk
Abbreviations: ISAAC—International Study of Asthma and Allergy in Childhood SPT—skin-prick test OR—odds ratio CI—confidence interval
The global increase in asthma and allergic diseases is well documented1 and has also occurred in Swedish populations.2 Recent data suggest that this development may have plateaued in some communities.3 Incidence and remission of asthma peak during childhood.4,5 Childhood asthma is heterogeneous, and several distinct phenotypes have been suggested.6 Although an important risk factor, a family history of asthma is neither necessary nor sufficient for predisposing an individual to developing asthma, implicating a significant environmental exposure component.7 A parental history of asthma does not identify a specific asthma phenotype in children but is associated with persistent and seemingly more severe disease.8 It is likely that a family history of asthma also includes environmental and social factors.9
A better understanding of the mechanisms through which heredity increases the risk for asthma could possibly assist in early diagnosis and proper management or even guide preventive measures. Individuals with a family history of atopic disease are considered at higher risk for developing asthma, and this is commonly used as an inclusion criterion in prevention studies. Although most studies show a positive association between childhood asthma and a family history of asthma, several studies have failed to show a significant association for parental atopic disease other than asthma.10,11 Patterns of inheritance of atopic diseases are seemingly complex. Differences in inheritance patterns have been observed between asthma and other allergic diseases,12,13 as well as differences in risk factors.14–16
A literature review on asthma heredity by Burke et al17 concluded that having 1 parent with asthma increased the child's risk for having asthma by 2 to 4 times but that a family history of asthma alone could not identify children who are at risk for developing asthma. As stated by the authors, the next step will be to study gene–environment interactions. Although progress has been made in the experimental field identifying candidate genes for asthma and pathogenetic mechanisms,18 epidemiologic knowledge of gene–environment interaction is still limited. Maternal smoking and levels of indoor mite and cat allergen seem to exert different effects on the development of atopic disease in the child depending on parental atopic status.5,19
In a population-based cohort of 3525 school children in northern Sweden who were followed yearly by expanded International Study of Asthma and Allergies in Childhood (ISAAC) questionnaires and by skin-prick tests (SPTs),14,20 asthma inheritance was studied. Independent relationships for hereditary factors were calculated using previously identified risk factors as confounders. The aims were to investigate in detail the influence of a family history of asthma and/or atopy on asthma development and to study interactions with other known risk factors.
| METHODS |
|---|
|
|
|---|
Questionnaire
An expansion of the ISAAC protocol,21 which has been described and discussed previously,14,20 was used. The questionnaire was distributed through the children's teachers, and in 1996, 3430 (97%) children participated. A clinical validation study of the asthma cases identified by questionnaire in this cohort has previously been described.14
Skin-Prick Tests
The children in 2 of the municipalities (Kiruna and Luleå, n = 2454) were invited to SPTs in 1996, with 88% participation rate (n = 2149). Skin-prick testing followed European Academy of Allergology and Clinical Immunology recommendations22 and was conducted by 2 nurses specifically trained for this study.
Testing was performed by lancet on the forearm for each of 10 allergens: birch, timothy, mugwort, cat, dog, horse, Dermatophagoides pteronyssinus, Dermatophagoides farinae, Cladosporium herbarum, and Alternaria alternata. Histamine (10 mg/mL) was positive control, and glycerol was negative control (Soluprick; ALK, Hørsholm, Denmark). The potency of the allergens was 10 histamine-equivalent prick test except the 2 molds, which were 1:20 weight/volume. Mean histamine wheal size was 4.0 mm. Allergic sensitization was defined as a mean wheal diameter of
3 mm, which correlated well with specific immunoglobulin E >0.35 IU/mL.23
Definitions
The definitions used for this cohort were described previously,14,20 Definitions of special importance to this study are as follows:
Statistical Analyses
The prevalence of asthma and related conditions was measured in 1996 for all children who participated in the questionnaire. Risk factors that we previously identified as being significantly associated with asthma in this cohort were included as covariates in multivariate analyses: a family history of asthma, living in a damp house, low birth weight, male gender, and respiratory infections.7 Although it was a statistically significant risk factor, allergic sensitization was not included in the multivariate model because it would limit the study cohort to the two thirds skin-prick tested.
Univariate and multivariate relationships are presented as odds ratios (ORs) using the 95% (P < .05) confidence interval (CI) for statistical significance. In univariate analyses, the
2 test was used. Complementary to the analyses stratified by gender and sensitization status, multivariate interaction analysis was used to test interaction between parental history and child's gender and sensitization status. Interaction term maternal x paternal asthma was also tested. All analyses were made by using SPSS 11.5.0 (SPSS Inc, Chicago, IL).
| RESULTS |
|---|
|
|
|---|
|
Children with parental atopy had a higher prevalence of current asthma than those without (7.2% vs 3.2%; P < .01). In children with atopy in both parents or asthma in at least 1 parent, the prevalence of asthma was 14.0% and 12.7%, respectively. Among those with parental asthma and a positive SPT, the prevalence of asthma was 29.4%, compared with 8.3% among those with a negative SPT. The prevalence when asthma was present in both parents was 35.7% (Fig 1).
|
|
10; Table 2). In evaluation of risk factors for wheeze in the last 12 months, ORs were similar but lower in magnitude. Patterns were similar in boys and girls.
|
|
|
Family History as a Risk Factor for Asthma, Stratified Analyses
Multivariate risk factor analysis was also performed after stratification by gender and allergic sensitization (data not shown). Having a family history of asthma had a similar impact on boys as on girls (OR: 3.3 [95% CI: 2.2–5.0] and 3.9 [95% CI: 2.4–6.4], respectively). Maternal asthma tended to have a somewhat higher impact on girls (OR: 4.5; 95% CI: 2.3–8.6) than on boys (OR: 2.5; 95% CI: 1.4–4.4). This difference was not seen for paternal asthma. Parental asthma was a significant risk factor for asthma among both sensitized and nonsensitized children (OR: 3.4 [95% CI: 2.0–6.0] and OR: 3.0 [95% CI: 1.7–5.1]), and differences between the 2 groups were small. As a risk factor for current asthma, no significant interaction was observed between gender of the child and parental asthma (P = .58) or parental atopy (P = .64). There was no interaction between the child's SPT result and parental asthma (P = .68) or parental atopy (P = .51).
Parental atopy was also studied as a risk factor for current asthma after stratification by gender and allergic sensitization of the child. The associations with asthma in the child were generally weak; therefore, these data are not shown. There were no differences between boys and girls in the effect of parental atopy and only minor differences between sensitized and nonsensitized children.
| DISCUSSION |
|---|
|
|
|---|
In this study, childhood asthma was defined by questionnaire, which was clinically validated previously.14 Family history of asthma and atopy has not been validated in the same manner, but asthma in young adults is generally better recognized and less diverse than in children, and questionnaire data have been found valid in studies of adults.
Does a Positive Family History Identify Children Who Are at Risk for Asthma?
The development of asthma is known to be mediated through interaction between genes and the environment. Allergic sensitization is the strongest known risk factor for asthma and increases the risk by 3 to 5 times.7 We found no interaction but did find additive effects between allergic sensitization and parental disease in this study. This finding suggests that the hereditary component of asthma is independent from the development of allergic sensitization.
Data in Figs 1 and 2 illustrate that the increase in prevalence of asthma associated with having 1 parent with atopy is modest unless the other parent has atopy and asthma and that having 2 parents with atopy confers approximately the same risk as having 1 parent with asthma. One third of the children who had 2 parents with asthma have asthma at this age. Expressed as an OR, the effect of asthma in both parents is multiplicative, whereas that of atopy in both parents is additive. Still, when tested by interaction analysis, no interaction between maternal and paternal asthma was found. The different combinations of parental asthma and allergic disease in Table 2 give an overview that parental asthma is much more strongly related to asthma in the child than is parental atopy and that gender of the affected parent is of no major importance in this respect.
Sibling asthma was not significantly associated with asthma in the child when neither parent had asthma. Furthermore, when parental history of asthma was positive, there was no additional effect of having a sibling with asthma. Despite the possibility that sibling asthma could indicate genes with limited penetrance, our findings conclude that sibling asthma is only a marker of parental disease.
Several studies have indicated that maternal asthma confers greater risk,10,24 as well as the opposite,14,25 and the review by Burke et al17 showed that findings of parent-of-origin effects were not conclusive. Several of the studies that reported risk differences by gender of the affected parent are from small and/or selected cohorts. To our knowledge, none has addressed differences by stratification for both child's gender and allergic sensitization status. Melén et al24 found a significant interaction between male gender and parental asthma and/or hay fever on the risk for several wheezing phenotypes until 4 years of age. As was suggested by the authors, this effect could diminish with age as the importance of male gender decreases. In our univariate analysis, there was no significant difference by gender of the affected parent (mother versus father: P = .51 for asthma, P = .27 for allergy). In multivariate analysis, there was a tendency toward a stronger association between paternal as compared with maternal asthma and current asthma in the child. When stratified by child's gender, the association with maternal asthma tended to be stronger in girls than in boys.
There are several plausible mechanisms for parent-of-origin effects. Maternal asthma has a larger number of possible mechanisms because of the intensive exposure in utero and during breastfeeding.26 The higher OR of maternal compared with paternal asthma as a risk factor for asthma in girls could indicate a predominantly hormonal effect. It could also be the case that maternal asthma genes exert a greater effect in girls, known as genetic imprinting. Withers et al10 showed an association between maternal asthma and late-onset persistent wheeze in girls but not in boys and that wheeze in girls tended to persist more often. We found no gender difference in the remission of asthma from ages 7 to 8 to ages 11 to 12 years.7 Therefore, it seems that in our cohort, despite the link with maternal asthma, a higher incidence rather than lower remission probability accounts for the higher prevalence of asthma in adolescent girls. In young children, asthma is more common among boys, and observations in our cohort are no exception. The earlier onset of asthma in boys could hypothetically suggest a greater importance of hereditary mechanisms in boys. However, our findings suggest that the effect of parental asthma does not seem to be modified by the gender of the child.
Åberg et al27 observed that a higher prevalence of atopic disease in mothers biased the association with atopic disease in the child upward. In our cohort, the prevalence of parental atopy was
30% and significantly higher in mothers, whereas that of asthma was 9% in both genders. In our cohort, questionnaires were more commonly completed by the mother.28 This could lead to a higher reported prevalence in mothers in that only the more severe cases of paternal disease are reported. Both of these are possible sources of bias. Also, it is crucial which factors are included in the multivariate analysis with regard to statistical power and the risk for making type 2 errors. The magnitude of the association is important to consider as well as the range of the CI. For childhood asthma, a lack of statistical significance in the association with asthma in 1 parent compared with a significant association with asthma in the other parent should not necessarily be interpreted as a statistically significant difference by parental gender, as described by Gelman and Stern.29 We acknowledge that this subject is prone to bias and methodologic pitfalls. Nonetheless, our large, population-based study with exceptional response rate does not confirm previous findings of major different impacts of paternal and maternal asthma.
Parental History of Atopy and Asthma in the Child
The association between parental atopy and current asthma in the child was significant but weak, which conforms to previous studies.14 Two parents with atopy had an OR similar to that of 1 parent with asthma. This underlines the considerable difference in the risk associated with parental asthma as compared with that of parental rhinitis or eczema. Therefore, parental asthma, rhinitis, and eczema should not together be considered "a family history of atopic disease" when it comes to calculation of asthma risk.
Parental history of atopy increases the risk for allergic sensitization in the child that starts the "atopic march," which in childhood may lead to asthma.30 Therefore, we expected parental atopy to be associated with asthma in the SPT-positive 52%. However, there was no major difference in the effect of parental atopy between sensitized and nonsensitized children (OR: 1.3 [95% CI: 0.7–2.3] vs 1.7 [95% CI: 1.0–3.1]).
The proportions of sensitized and nonsensitized children with asthma were equal, and parental asthma was included in the multivariate model to correct for comorbidity. The relationships between inheritance of asthma and allergic diseases are complex. Crestani et al13 found that parental asthma was associated with atopy in the child regardless of parental atopic status. The local immune response in the airway epithelium is altered in both allergic and nonallergic individuals with asthma,31 and this could be related to a predisposition for immunologic alteration inherited from allergic parents, regardless of parental asthma or allergy in the child. These findings taken together indicate that parental allergic disease could influence asthma development in the child through other paths of causality than allergic sensitization and subsequent progression of the atopic march. This could be either on the genetic level (separate genes for asthma and rhinitis/eczema) or at the cellular level with different pathogenetic mechanisms. Also, subclinical bronchial hyperreactivity not subjectively recognized as asthma in some of the allergic parents may be inherited by the children and result in symptomatic asthma.
| CONCLUSIONS |
|---|
|
|
|---|
| FOOTNOTES |
|---|
The authors have indicated they have no financial relationships relevant to this article to disclose.
| REFERENCES |
|---|
|
|
|---|
This article has been cited by other articles:
![]() |
K. Larsson Research at the Unit of Lung and Allergy research at Karolinksa Intitutet, Stockholm, Sweden Eur. Respir. Rev., April 1, 2008; 17(107): 15 - 17. [Full Text] [PDF] |
||||
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||