In epidemiologic studies of asthma there is a group with recent wheeze, but with no airway hyperresponsiveness (AHR), in whom it is unclear whether any significant airway abnormality exists. Exhaled nitric oxide (NO) has been proposed as a measure of airway inflammation. We measured exhaled NO in a population sample of 306 young adults who also underwent bronchial challenge with histamine or a bronchodilator test. Subjects blew into a 3-L Tedlar bag against a 2-mm-diameter resistance to close the soft palate and exclude nasal air. The NO content of expired gas from a single breath was analyzed by chemiluminescent analyzer. Exhaled NO was log-normally distributed in the population sample and duplicate measurements were highly reproducible (intraclass correlation coefficient = 0.98). Exhaled NO correlated significantly with airway responsiveness, measured as the dose-response ratio to histamine (r = 0.39, p < 0.001) and with peripheral blood eosinophils (r = 0.35, p < 0.001). Exhaled NO was significantly greater in asthmatic subjects (geometric mean, 22.2; 95% confidence intervals, 16.1 to 30.7 ppb) than in normal subjects (7.8, 7.1 to 8.4, p < 0.001) or in subjects with wheeze but no AHR (8.8, 7.5 to 10.3, p < 0.001). We conclude that exhaled NO is log-normally distributed, is highly reproducible and discriminates well among subjects, suggesting that it is both a feasible and useful measurement for epidemiologic studies of asthma. The findings suggest that wheeze in the absence of AHR is unlikely to be associated with airway inflammation.

In the future, the important longitudinal studies will be those that divide the broad spectrum of asthma into phenotypic groups in order to provide more precise information about mechanisms and outcomes. Asthmatics who present in a clinical setting often have a combination of two or three different phenotypes, and may have more severe illness as a result, but subjects who are tested in epidemiological studies are frequently characterized by only one phenotype. Although the definition of asthma as a combination of symptoms plus either AHR or reduced lung function has greater predictive isolation for assessing the prognosis of respiratory illness that occurs in childhood (27,40,55,66-68), studies that are able to classify subjects according to their different phenotypes are more likely to lead to a better understanding of the factors that influence prognosis. More importantly, by using this approach, future research studies should be better able to separate the mechanisms whereby therapeutic or environmental interventions influence the prognosis of each phenotype. The potential benefits will be a better understanding and evaluation of the factors that make an important contribution to respiratory health.

To explore the natural history of asthma and its relation to allergic responses, we examined the relation between total serum IgE in early adulthood and a history of respiratory symptoms, airway hyperresponsiveness (AHR), and atopy during childhood. We studied 180 subjects aged 18-20 years who had been studied since the age of 8-10 years. We measured wheeze in the previous year by questionnaire, AHR by histamine inhalation test, atopy by skin prick tests, and serum IgE levels by immunoassay. Subjects with AHR in early adulthood had higher IgE levels (mean 257.0 IU/ml) than subjects with past AHR (mean 93.3 IU/ml) or with lifelong normal responsiveness (mean 67.6 IU/ml) (P < 0.001). Subjects who had symptoms had higher IgE levels (mean 125.9 IU/ml) than those who were lifelong asymptomatic (mean 63.1 IU/ml) (P < 0.001). Recent wheeze, AHR, and allergic sensitization all had a positive relation to serum IgE, but IgE was not more predictive of AHR than skin prick tests. The finding that young adults who are sensitized to common allergens are highly likely to have AHR even in the absence of symptoms is further evidence of the fundamental role of IgE-mediated responses in the natural history of AHR throughout childhood and into adulthood.

In recent years, airway responsiveness has commonly been measured in epidemiological studies using one of two methods. In one method, histamine is administered via a handheld DeVilbiss nebulizer and in the other, methacholine is administered via a dosimeter. Allergic sensitivity has commonly been measured by either the allergen droplet method or by Phazet. We wanted to assess the comparability of airway responsiveness and of allergic sensitivity measured by both methods. A total of 48 volunteers, including normal and asthmatic subjects, participated in the study. Subjects first underwent one of the two tests of airway responsiveness and allergic sensitivity, and then returned within 10 days to undergo tests using the second protocol. Commencement protocol was allocated in random order. There was good agreement between both methods for assessing airway responsiveness and for assessing allergic sensitivity. The difference for dose response ratio (DRR) between histamine and methacholine was a 1.19 (95% (CI) 0.78, 1.82) fold changes, which was not statistically significant. Agreement between allergic sensitivity methods was perfect for Alternaria tenuis, good for rye-grass (kappa = 0.71) and moderate for cat and Dermatophagoides pteronyssinus (kappa approximately 0.5). It is possible to compare data from epidemiological studies which use these methods.