INTRODUCTION

TOP ABSTRACT INTRODUCTION METHODS RESULTS DISCUSSION REFERENCES

Our knowledge about the growth of pulmonary function from childhood into early adulthood is incomplete. Factors having a negative impact on the age-related growth of pulmonary function in children and adolescents may result in a lower maximally attained level of pulmonary function, and perhaps also in an earlier onset of decline of pulmonary function, and may therefore potentially increase the risk for subsequent development of both reversible and nonreversible obstructive pulmonary disease.

Gold and coworkers (1) have reported that cigarette smoking is associated with slowed growth of pulmonary function in adolescents, but, comparable to observations from studies of adults, differences in smoking habits are not likely to be the only important determinant for changes over time in level of pulmonary function. Identification of other markers of an increased risk for impaired age-related growth of pulmonary function in children and adolescents is therefore important. Orie and coworkers (2) have previously suggested that atopy might represent a host characteristic predisposing individuals to the development of both asthma and chronic obstructive pulmonary disease (COPD). Further, findings from a longitudinal study of a population-based sample of children suggest that the degree of airway responsiveness to cold air hyperventilation is associated with specific pulmonary function changes (3). Increased airway responsiveness to stimuli such as histamine and methacholine, and atopy, or an interaction between these factors, might therefore be host characteristics identifying individuals with an increased likelihood for impaired age-related growth of pulmonary function.

To describe the growth of pulmonary function, especially the potential impact of asthma, atopy, and increased airway responsiveness, we have examined a population sample comprising 408 children and adolescents, 7 to 17 yr of age at enrollment, twice with an interval of 6 yr.

	METHODS

TOP ABSTRACT INTRODUCTION METHODS RESULTS DISCUSSION REFERENCES

Subjects

A sample of 983 children and adolescents living in the area surrounding Rigshospitalet in the city of Copenhagen was drawn at random from the civil registration list (4); all subjects were born in the first week of each month, and had a mean age of 12 yr (range, 7 to 17 yr). At both surveys, performed 6 yr apart, all subjects (n = 983) were invited by letter to participate in a study concerning asthma, allergy, and airway hyperresponsiveness.

Of the 983 subjects, 527 (54%) participated in the first survey and 665 (68%) participated in the second survey; 408 subjects (199 males and 209 females) participated in both surveys. Only data for the 408 subjects who participated in both surveys are included in the present analysis.

The question of nonresponder bias has been addressed in a previous article (5). Briefly, apart from a significantly higher number of current smokers among the subjects, who participated only in the second survey, no significant differences were found between the groups (i.e., "stayers," "dropouts," and "newcomers") with respect to anthropometric data, pulmonary function, or prevalences of atopy, airway hyperresponsiveness, and allergic diseases, indicating that the subjects included in the present analysis are representative of the entire sample; for further details see Ulrik and colleagues (6, 7).

Exclusion Criteria

The subjects were asked to abstain from cigarette smoking for at least 2 h before their appointment at the laboratory. In case they were taking medication for asthma or allergy, they were asked not to use theophylline or an antihistamine for at least 24 h, astemizole for 6 wk, an oral ₂-agonist for 18 h, and an inhaled bronchodilator for 6 h before the tests; long-acting ₂-agonists were not available in Denmark at the time of the surveys. They were allowed to continue use of any inhaled or oral corticosteroid they had been taking.

Case History

The participants and, if present, their parents were interviewed by one person about birth weight, diseases (respiratory and nonrespiratory), active and passive smoking, use of medication, and known or suspected allergic disease and allergy.

Furthermore, all participants filled out a questionnaire about asthmatic and allergic symptoms, that is, rhinitis (sneezing, runny or blocked nose not associated with a cold) and eczema (an itchy dry rash on face, arms, or legs), as related to themselves, their siblings, and their parents.

The questionnaire concerning respiratory symptoms and the definition of asthma was adopted from studies by the American Thoracic Society, Division of Lung Disease of the National Heart, Lung, and Blood Institute (8, 9), and by Hopp and coworkers (10, 11). Asthma was defined by questionnaire criteria on the basis of the responses to the following questions:

1. Have you ever had asthma?
2. Does your breathing ever sound wheezy or whistling?
3. Do you have attacks of shortness of breath with wheezing?
4. Do you experience wheezing, chest tightness, cough, breathlessness with any of the following: at rest, with exertion, with emotional stress, with exposure to cold air, with chest infections or head cold?
5. Do you experience wheezing after exposure to: dust, fumes, mold, pollen, food, pets, or drugs?
6. Have you ever been hospitalized or observed and treated by a doctor for asthma?
7. Have you ever received medication for your asthma?
8. What was the medication used?
9. Did it help?
10. How many episodes of wheezing have you had during the last year?
11. Have you ever had attacks of wheezing, shortness of breath, or dry cough at night?

Asthma was defined on the basis of positive responses to Questions 2, 3, 4, and/or 5. Current asthma was defined as symptoms within the preceding 12 mo. Furthermore, all participants reported whether they were current smokers, ex-smokers, or never smokers, and for the first two categories they reported the duration of smoking. Current and ex-smokers also reported their daily tobacco consumption, and an estimate of their lifetime tobacco exposure was calculated as pack-years [current tobacco consumption (g d¹/20) × duration of smoking (years)]. Smoking history (daily tobacco consumption and duration of smoking) was obtained for both parents (i.e., adult members of the participant's household) and included in the analysis as an estimate of passive tobacco exposure.

Pulmonary Function Tests

The forced expiratory volume in 1 s (FEV₁) and forced vital capacity (FVC) were measured with a 7-L dry wedge spirometer (Vitalograph, Buckingham, UK), which was calibrated weekly. Each measurement consisted of at least three maximal expiratory maneuvers from total lung capacity to residual volume with a variation of less than 5%. The highest FEV₁ and FVC values were used in the analyses.

Data on pulmonary function were also expressed as a percentage of predicted values (%pred), using prediction equations based on age, sex, and height (12). A standard dose of bronchodilator (salbutamol) was given at the first survey to aid recovery after the histamine challenge test and at the second survey (all subjects), also as a test for reversibility, and values obtained after administration of bronchodilator were used in the analyses.

Skin Prick Test

Skin prick tests (SPTs) were performed on the volar surface of the forearm, using standard dilutions (100,000 biological units [BU]/ml) of allergens in 50% glycerol (ALK, Hørsholm, Denmark). The allergens used were birch, grass, mugwort, horse, dog, cat, house dust mite (Dermatophagoides pteronyssinus, at the second survey also D. farinae) and two molds (Alternaria iridis and Cladosporium herbarum). Histamine-HCl (10 mg/ml) in 50% glycerol was used as a positive reference, and a negative reference (50% glycerol) was also included. The reactions were read after 15-20 min. A positive SPT was defined as a positive reaction to at least one of the allergens (13). The reaction to each of the allergens was regarded as positive if the mean wheal diameter [(d1 + d2)/2] was at least 3 mm (13). In case of reaction to the negative reference, a positive result was recorded when the difference between the mean wheal diameter of the reaction to the allergen and to the negative reference exceeded 3 mm. Atopy was defined as a positive skin prick test.

Serum IgE and Blood Eosinophils

The total serum IgE level was determined by paper radioimmunosorbent test (PRIST; Pharmacia, Uppsala, Sweden) (at the first survey only) and the number of eosinophil leukocytes in peripheral blood was counted in a counting chamber (at the second survey only) (14).

Histamine Challenge Test

Airway responsiveness to inhaled histamine was measured using the method described by Cockcroft and colleagues (15). Aerosols of the test solution were generated by a nebulizer (Wright; Aerosol Products, London, UK) operated to give an output of 0.14 ml/min. Each aerosol was inhaled through the mouth by tidal breathing for 2 min. The first aerosol was saline (0.9%), and it was followed at 4-min intervals by twofold increasing concentrations of histamine (0.075-8.0 mg/ml; at the second survey to 16 mg/ml). The response was measured by determining the FEV₁ 1 min after each inhalation. The test was terminated when a > 20% decline in FEV₁ from the post-saline value occurred, or at the end of the dose schedule if such a decline did not occur. For all subjects having at least a 20% decline in FEV₁ by the end of the dose schedule, the concentration of histamine causing a 20% fall in FEV₁ (PC₂₀) was calculated by linear interpolation from the individual log dose-response curve as follows (16):

where C₁ is the second-to-last concentration of histamine (< 20% fall in FEV₁), C₂ is the last concentration of histamine (> 20% fall), R₁ is the percent fall in FEV₁ (%fall FEV₁) after C₁, and R₂ is the percent fall in FEV₁ after C₂. A positive test was defined as a PC₂₀ FEV₁ of < 8.0 mg of histamine per milliliter. To avoid censoring of data, a histamine dose-response slope was calculated for all participants as the percent fall in FEV₁ at the last dose divided by the total dose administered (17). To obtain a positive, normally distributed value for logarithmic conversion a constant of 3 was added to all dose-response slopes (18), i.e., values are reported as the percent fall in FEV₁ per micromole plus 3.

Statistical Methods

Unless stated otherwise, the prevalence figures given are the cumulative prevalence rates from birth to the time of the surveys. Separate cross-sectional analyses were performed on the data from the two surveys. Changes in the proportion of subjects with a positive test/symptoms were assessed by means of McNemar's ² test. Linear regression analysis was used to assess putative risk factors at the first survey in relation to outcome at the second survey, primarily the level of FEV₁, and nonsignificant variables were deleted by backward elimination. A p value < 0.05 was considered significant.

It is debated whether the initial FEV₁ should be considered as a confounder of the relationship between exposure and subsequent change in FEV₁ (19). As reviewed by Irwig and coworkers (19), several methods for statistical adjustment for the initial value of FEV₁ have been used previously. Some investigators have included the initial value of FEV₁ as a confounder in the regression model, others have included the mean value of FEV₁ between start and end of the observation period, and still others have chosen not to include FEV₁ at all as an explanatory variable. It may be argued that adjustment for the initial level of FEV₁ would remove some of the effects under study, if the exposure of interest (e.g., the presence of persistent asthma) has been present for several years. However, we finally decided to adjust for the initial FEV₁ because most of the exposures of interest (e.g. new asthma versus patients without asthma and new atopy to house dust mite [HDM] versus patients without atopy) commenced after the initial measurement of lung function. Thus, in the present study we considered the initial FEV₁ not to reflect the same exposure as that studied during the observation period.

Including height-adjusted FEV₁, age, and sex, as opposed to FEV₁ %pred, were considered as an alternative method of modeling FEV₁. However, as the findings using this approach were consistent with those of the analyses including FEV₁ %pred, and, furthermore, did not reveal significant interactions between the host characteristics of interest (e.g., airway responsiveness) and age and sex and outcome (i.e., level of FEV₁), we decided to present only the results of the analyses including FEV₁ %pred.

	RESULTS

TOP ABSTRACT INTRODUCTION METHODS RESULTS DISCUSSION REFERENCES

Complete data were available for 408 subjects, 199 male and 209 female; characteristics of the examined subjects are displayed in Table 1. The point prevalence of a positive histamine challenge test declined from 20% (n = 80) at the first survey to 6% (n = 24) at the second survey (Table 2). The point prevalence of a positive SPT increased from 26 to 44%, and so likewise did the point prevalence of atopy to HDM increase from 13 to 26% (Table 2).

Preliminary analyses of the data showed that atopy to HDM, but not overall skin test reactivity or atopy to each of the other aeroallergens used, was associated with the level of FEV₁, including change over time in FEV₁, and the former was therefore included in the final models.

The cross-sectional analysis of data from the first survey showed that current asthma (p < 0.02), atopy to HDM (p = 0.03), and an increased degree of airway responsivness (AR) (p < 0.002) were associated with a lower level of FEV₁ %pred, whereas no significant association could be demonstrated between total serum IgE and passive smoking, and level of FEV₁ %pred.

Table 3 shows the results of the cross-sectional regression analysis concerning the level of FEV₁ %pred at the second survey. Apart from asthma, atopy to HDM, and an increased degree of AR, a higher number of blood eosinophils was also associated with a lower level of FEV₁ %pred. The proportion of current smokers among both males and, especially, females was disturbingly high at the second survey. However, it was not possible to demonstrate a significant association between active smoking and level of pulmonary function.

The results of the multiple regression analysis predicting FEV₁ %pred at the second survey as a function of variables measured at the first survey and changes in these variables during the observation period are shown in Table 4. The substantial impact not only of asthma but also of atopy to HDM on pulmonary function was confirmed in the longitudinal analysis. Furthermore, both new asthma, new atopy to HDM, and an increase in the degree of AR from the first to the second survey predicted a lower level of FEV₁ %pred at the second survey; more than 50% of the participants suffering from asthma were sensitized to HDM. The degree of AR, expressed either as the dose-response slope or as a dichotomous variable (subjects with hyperresponsiveness as opposed to subjects with a negative histamine challenge test), at the first survey was not significantly associated with the level of FEV₁ at the second survey. The longitudinal multiple regression analysis did not reveal a significant association between sex, birth weight, tobacco exposure (active or passive), a history of wheezy bronchitis before the age of 2 yr, or symptoms of rhinitis and/or eczema, and level of FEV₁ %pred at the second survey. The interaction terms between sex, and asthma, atopy, and AR in the multiple regression model were nonsignificant.

To investigate further the association between atopy to HDM and level of FEV₁, the data were reanalyzed after exclusion of subjects with asthma and/or evidence of airway hyperresponsiveness (AHR). This analysis revealed that subjects with either persistent or new atopy to HDM on average had a 5.5 (SEM 2.6; p = 0.37) and 4.3 (SEM 1.9; p = 0.23) %pred, respectively, lower FEV₁ at the second survey compared with those subjects who had a persistent negative skin reaction to HDM.

	DISCUSSION

TOP ABSTRACT INTRODUCTION METHODS RESULTS DISCUSSION REFERENCES

The present population study suggests that symptomatic asthma, an increase in the degree of airway responsiveness, and atopy to house dust mites have a negative impact on the growth of ventilatory function in children and adolescents. Furthermore, sensitization to house dust mites may also be a marker of impaired growth of pulmonary function in children and adolescents without respiratory symptoms and/or evidence of airway hyperresponsiveness. An inverse association between sensitization to house dust mites and growth of ventilatory function in asymptomatic, nonhyperresponsive subjects has, to our knowledge, not been reported previously.

A number of follow-up studies of selected groups of children with asthma have consistently shown that more severe respiratory symptoms in childhood predict a lower level of lung function in early adulthood (22). Our finding of an association between symptomatic asthma and pulmonary function impairment is therefore in keeping with previous reports, although the subjects enrolled in the present population study may be anticipated to have, in general, milder disease than hospital-based samples of patients with asthma. Weiss and co-workers (25) have previously reported from a population-based study of 602 children that persistent active asthma has a progressive negative impact on the annual change in FEV₁ in females, whereas active asthma was not a significant predictor for change in FEV₁ in males. However, in the present study a sex difference in the impact of asthma on lung function could not be demonstrated, possibly owing to the design of the study with only two observations.

Increased airway responsiveness is associated with a reduced level of lung function in both children and adults (3, 26), and data from a longitudinal epidemiological study of adolescents and young adults suggest that airway hyperresponsiveness, especially persistent hyperresponsiveness, is associated with a lower maximally attained level of lung function (27). Our findings therefore appear to be in line with previous reports. However, the considerable long-term variability in airway responsiveness (28) and the well-known occurrence of transient airway hyperresponsiveness in asymptomatic individuals might explain the lack of significant association between the degree of airway responsiveness at enrollment and the level of lung function at the second survey. But, on the basis of the substantial impact of an increase over time in airway responsiveness on the level of FEV₁ at follow-up, our findings suggest that an increase in the degree of airway responsiveness may be an important marker of impaired growth of lung function in children and adolescents.

Cross-sectional studies have consistently shown that lung function in both children and adults with asthma is lower than predicted; and, furthermore, patients with asthma have a higher prevalence of atopy than does the general population. Inclusion of patients with asthma will therefore increase the likelihood of a significant association between atopy and reduced level of lung function in the present cross-sectional analyses. However, confining the cross-sectional analyses to nonasthmatic subjects (n = 347, data not shown) did not substantially change our findings (data not shown).

The present longitudinal analysis suggests that atopy to house dust mite is associated with impaired growth of ventilatory capacity in children and adolescents, an association not driven by inclusion of subjects with respiratory symptoms (and evidence of airway hyperresponsiveness). Gottlieb and co-workers (29) have reported that cutaneous hypersensitivity to common allergens is a significant independent predictor of subsequent decline of lung function among men who are middle-aged and older, and nonasthmatic. However, Annesi and colleagues (30) found no relationship between decline of FEV₁ and atopy, including atopy to house dust mite, in a cohort of 308 working men who were tested for atopy at the end of the study. In patients with asthma, an association might be expected between atopy and outcome, but longitudinal studies of both children and adults with known asthma have consistently shown no independent effect of atopy on annual changes in lung function (31). In the present study both asthma and new atopy to house dust mite, but not a persistent positive skin reaction to house dust mite, predicted a lower level of FEV₁ %pred in early adulthood. Sensitization to house dust mite is a known marker of an increased risk for subsequent development of symptomatic asthma, and more than 50% of the patients with asthma in the present study were atopic to HDM. From these pieces of information one might infer that the ongoing inflammatory reaction in the airways of patients with established obstructive lung disease might be independent of the patients' atopic status, whereas atopy in nonasthmatic, nonhyperresponsive children and adolescents may be an independent predictor of impaired age- related growth of lung function.

The findings of the present study lend support to the hypothesis that atopy may be a host characteristic that predisposes susceptible individuals to the development of pulmonary function impairment and obstructive pulmonary disease. Furthermore, apart from the probable deleterious effects of atopy to house dust mite on lung function development, this study also showed that asthma and an increase in the degree of airway responsiveness are likely to be markers of an increased risk for impaired growth of pulmonary function in children and adolescents.