The relationship between atopy and asthma is not straightforward. Historically “atopy” has been used as a poorly defined term to refer to allergic conditions such as hay fever, asthma, and eczema, which cluster in families. As knowledge about the immunological features of atopic diseases increased, a clearer definition evolved. Since the original description of IgE in 1966, knowledge about its role in the pathogenesis of allergic diseases has proliferated.¹ Atopy can be broadly defined as the tendency to generate an IgE response to specific allergens² or the tendency to generate a wheal of greater than 3 mm in response to skin prick testing.³

Evidence now implicates serum IgE in development and expression of the asthma phenotype. Observational and epidemiological studies^4,5 prompted the development of specific monoclonal antibodies to IgE.⁶ The therapeutic success of anti-human IgE antibody in adults^7,8 and children⁹ with asthma supports the importance of IgE in disease expression.

The degree of atopy can also be defined as the number and/or size of positive skin prick tests (SPTs) to a standard panel of allergens.^10–,¹³ In children, increasing wheal size to house dust mite Der p 1 has been associated with increasing asthma symptoms and increased risk of hospital admission.^14,15 However, relationships between atopy and asthma severity are often weak or fail to achieve statistical significance. Even in large studies results have been mixed. Siroux et al failed to show a relationship between the degree of allergic sensitisation measured by the number of positive SPTs or the size of the wheal reactions and asthma severity.¹⁰ They suggested further studies including a sufficiently large number of severe patients to clarify such relationships. We therefore used the data collected as part of the GAIN collaboration (Genetics of Asthma International Network) to address this question in a larger number of children than had been available in previous studies.

METHODS

Study subjects

A total of 232 families with at least one asthmatic child (aged 7–18 years) were recruited via an asthmatic proband from the two centres (North Staffordshire and Sheffield) between January 1999 and December 2001. Inclusion criteria for probands included documented episodes of wheezing in the previous 12 calendar months, physician diagnosis of asthma, and agreement to participate by the natural mother and father. All first-degree relatives of probands were invited to participate in this study, but only those with a previous physician diagnosis of asthma and who were less than 19 years of age were included in the present analysis. Identical twins were not eligible to participate, siblings without a physician diagnosis of asthma were not eligible, and subjects with cardiac, other respiratory, or inflammatory diseases were excluded. The study was approved by the local research ethics committee at each centre and written informed consent was obtained from all participants.

Participants completed a standardised observer administered ISAAC questionnaire¹⁶ and underwent baseline spirometry (Morgan Rolling-seal 232).¹⁷ Lung function results in children and adults were expressed as percentage predicted using previously established reference ranges for UK Caucasian children¹⁸ and adults.¹⁹ Spirometry in both centres was undertaken in a dedicated lung function laboratory in accordance with American Thoracic Society standards.²⁰ Skin prick testing (SPT) and measurement of total serum IgE were performed on all subjects. For measurement of total serum IgE, serum was separated and stored at −80°C within 4 hours of venepuncture. Total serum IgE was measured using a standard Immulite assay (EuroDPC UK Ltd, Gwynedd, Wales). The between-run coefficient of variation of the IgE assay was 5.1–6.7%. As IgE values in excess of 2000 IU/ml were reported as >2000 IU/ml, these were assigned a value of 2001 IU/ml. Skin prick testing was performed against a panel of seven standardised aeroallergens and a standard histamine positive control supplied from a single source (ALK (UK), Berkshire). The panel of allergens and test protocols were identical at both centres and included: D pteronyssinus; D farinae; Grass mix; Cat hair; Dog hair; Cockroach; and Alternaria. Prior to testing, antihistamines were withheld for 72 hours; no participants were taking oral steroids in excess of 10 mg prednisolone/day.

Atopy was defined as a positive response to SPT (at least one wheal ⩾3 mm greater than saline response to a panel of seven aeroallergens) or a total serum IgE ⩾100 IU/ml. Asthma severity was determined based on treatment, lung function, and symptoms (fig 1⇓ and Carroll et al²¹). Participants from Stoke-on-Trent (145 families) completed a supplementary questionnaire which included data on hospital admissions in the previous 12 months.

• Download figure
• Open in new tab
• Download powerpoint

Figure 1

 Algorithm for severity scoring.

RESULTS

DISCUSSION

We have shown that the type and intensity of atopic sensitisation is positively associated with clinical and spirometric measures of asthma severity in children with asthma. This study is larger than previous reports which may have failed to demonstrate associations because of limited statistical power.¹⁰ Differences in the statistical methodology are unlikely to account for the differences in results, as a subsequent analysis of our data using the generalised estimating equation,²⁴ produced almost identical levels of significance for each of the reported associations.

Few published studies have examined the relationships of atopic markers with asthma severity in children. Of these, some showed greater asthma severity with sensitisation to aeroallergens^14,25 but others did not.^26,27 In children, there are conflicting results regarding the relationships of asthma severity with serum total IgE.^10,25,28 Even when relationships have been established, the statistical significance achieved has often been marginal (p > 0.01).

This analysis was performed in a large population of well characterised children with asthma. It included children recruited from hospital and community sources, although due to the stringent nature of the initial recruitment criteria to the genetic study, these children represent a highly atopic and asthmatic population. This sample may therefore not be entirely representative of typical cases of childhood asthma. However, the wide spectrum of disease severity seen in the group as a whole and the relatively normal values achieved for lung function variables are somewhat reassuring that our results are applicable at a broader level.

The relationship between total serum IgE and risk of asthma is well established in children and adults.^4,5,29 The risk of hospital admission has been linked to total serum IgE in several paediatric studies.^10,14,25 Although data on hospital admissions were available for slightly less than half of the subjects in our study, we were still able to see a clear association between hospital admission and total serum IgE.

As previous studies have failed to demonstrate associations between asthma symptoms and markers of atopic sensitisation, investigators have concluded that atopy may have a lesser role in determining asthma control than in determining severity.¹⁰ This hypothesis is not supported by our data as both the need for hospitalisation and asthma severity scores were significantly associated with total serum IgE. National and international guidelines on asthma clearly define the goals of asthma therapy and identify need for hospital admission and/or persistent symptoms and lung function abnormalities as evidence of poor asthma control.^30,31

SPT wheal size is another measure of allergic sensitisation and SPTs have been used widely in epidemiological and clinical studies. However, arguments persist about what constitutes an abnormal response. The formation of a wheal is the result of a complex immunological host response culminating in the release of histamine locally within the skin. It is dependent on many factors including season, age, and treatment of individuals with antihistamine medications.^32,33 While it was possible to control for factors such as antihistamine and oral steroid use, data on season of testing were not collected in this study. Despite these potential limitations the use of a summative wheal size as a measure of allergic sensitisation appears to be a useful predictor of morbidity in children with asthma. The discordant nature of the results between total serum IgE and summative wheal size suggests a degree of independence between these two measures. Total serum IgE seems to correlate well with asthma severity; the results of SPTs do not. One explanation for this apparent discordant response may lie in the inclusion of treatment in our asthma severity score (fig 1⇑). There are data to suggest that treatment with leukotriene receptor antagonists may reduce SPT wheal sizes in atopic children.³⁴ This could obscure the perceived relationship between severity and SPTs. SPTs may also be less informative than total serum IgE across a population because they represent the reactions to a chosen panel of only a few allergens. In our study we chose seven aeroallergens, two of which are low prevalence allergens in the UK (Alternaria and Cockroach) and only rarely produced a reaction (table 1⇑). However, from our results we can see that SPTs to even a limited panel are informative and correlate positively with hospitalisation, steroid use, and airways obstruction.

While our results are highly significant, this must be distinguished from clinical significance. The very small p values generated in this study reflect in part its power and the power of the statistical techniques undertaken. The majority of our results remain highly significant even after correction for multiple hypothesis testing. Nonetheless, the coefficients generated by the generalised estimating equation can give some estimates of the relationships between atopy and the asthma phenotype (table 4⇓). A child with a total serum IgE of 200 (log₁₀ = 2.30) will have, on average, an FEV₁ of 4.6% less than an asthmatic child with a total serum IgE of 20 (log₁₀ = 1.30).

This study confirms that increasing atopic sensitisation is clearly associated with increasing disease severity in children with asthma. Both total serum IgE and SPTs confer valuable information about the asthma phenotype. These correlate with each other and with other measures of asthma severity including clinical severity, hospital admission, and measures of lung function. Nonetheless, the relationship between different markers of atopic sensitisation and the asthma phenotype is subtly different within our cohort of asthmatic children, and therefore they should probably be considered separately in clinical and epidemiological studies.