INTRODUCTION

TOP ABSTRACT INTRODUCTION METHODS RESULTS DISCUSSION REFERENCES

In epidemiologic studies asthma can be defined as the presence of both airway hyperresponsiveness (AHR) and recent symptoms of wheeze (1). However, this definition leaves a group with recent wheeze, but no AHR, in whom it is unclear whether any significant airway abnormality exists. Airway inflammation is a consistent feature of the pathophysiology of asthma (2). In subjects with a history of wheeze but no AHR, an understanding of the extent to which their symptoms are associated with airway inflammation would help to determine if these people have asthma that needs to be taken seriously or just "trivial wheeze," which may be better left untreated.

Measurement of airway inflammation, particularly in epidemiologic studies, is difficult to achieve. However, recently it has been suggested that the concentration of nitric oxide (NO) in exhaled air may be a useful monitor of airway inflammation (3). Nitric oxide is synthesized by many cell types from an L-arginine substrate in the presence of either inducible or constitutive NO synthase. The expression of inducible NO synthase is increased in airway epithelial cells and some inflammatory cells from asthmatic subjects (3). An association between exhaled NO and airway inflammation is supported by the observation that NO levels are increased in subjects with mild asthma (4, 6), increase after a late reaction to allergen (7), and are lower in subjects receiving inhaled corticosteroids (4, 5). Exhaled NO levels are significantly correlated with the number of eosinophils in induced sputum (8). There have been no studies of the distribution of exhaled NO levels in a general population sample of adults.

The aim of this study was to examine the distribution and reproducibility of exhaled NO levels in a population of young adults and to determine the association between exhaled NO and airway hyperresponsiveness, symptoms, and atopy. In particular, we wished to determine if there was any evidence of elevated NO levels among subjects who reported recent wheezing but have normal spirometric function and normal airway responsiveness.

	METHODS

TOP ABSTRACT INTRODUCTION METHODS RESULTS DISCUSSION REFERENCES

Population

This study was part of a follow-up of a cohort of 718 children originally studied in 1982 in Belmont, New South Wales when they were 8 to 10 yr of age (9). The methods for selection of the original cohort have been described in detail but, briefly, all children in Years 3 and 4 of eight primary schools in the Belmont area were invited to participate in the study. A survey of nonattenders at the time suggested that the attenders were a representative random sample of the population. This cohort has been restudied six times in the period 1982 to 1997, and the subjects are now 23 to 25 yr of age. This study was carried out in 376 subjects who attended a study center in Belmont in June and July 1997. These 376 attenders did not differ from the remainder of the original cohort with respect to the prevalence of AHR, atopy, or recent wheeze at the time of recruitment in 1982.

Study Design

Subjects attended the study center on a single occasion where they completed a questionnaire, underwent skin prick tests and bronchial challenge with histamine, and had expired air collected for measurement of NO and blood collected for estimation of eosinophil numbers. The expired air collection was made 15 to 20 min after the bronchial challenge when the lung function was at prechallenge values. A subgroup of 53 subjects returned 1 to 2 wk later for repeat measurement of NO. All subjects gave written informed consent, and the study was approved by the Human Ethics Committee of the University of Sydney.

Questionnaire

The questionnaire was a modified version of the International Union Against Tuberculosis (IUAT) questionnare (10), and comprised questions of recent and past respiratory symptoms, including occupational symptoms, family history, diagnosed asthma and medication use, and hospital and doctor attendances. Subjects who reported wheeze or wheeze after exercise that had occurred in the previous 12 mo were classified as having "recent symptoms." Subjects who reported attacks of hay fever or nasal allergies during the previous 12 mo were defined as having "rhinitis."

Nitric Oxide Measurement

Mixed expired nitric oxide was measured using a modification of the method of Massaro and colleagues (5). The measurement was performed with the subject standing and without a noseclip. Each subject took three deep breaths from a Douglas bag containing medical air, which had a NO content below 3 ppb. They then exhaled from above FRC for 18 to 30 s into a 3-L Tedlar bag against a 2-mm-diameter resistance to close the soft palate and exclude nasal air (11). Subjects were told to blow as hard as they could against the resistance. Expiratory flow rates were not measured. Expired gas from a single breath was analyzed immediately by chemiluminescent analyzer (Model 42C; Thermo Environmental Instruments, Franklin, MA), which has a lower limit of detection of 1 ppb. Two separate breaths were collected and analyzed from each subject at each visit.

Lung Function and Airway Hyperresponsiveness

A bronchial challenge test with histamine was administered to all subjects with baseline FEV₁  60% predicted using the rapid method (12). Lung function was recorded by Mijnhardt dry rolling seal spirometers (Mijnhardt BV, Bunnick, Holland) connected to IBM-PC computers running scientific and medical software for immediate data acquisition. Forced expiratory maneuvers were repeated until two readings of FEV₁ and FVC within 100 ml were obtained, the largest of which was used in analyses. Values for FEV₁ and FVC were recorded as a percentage of the predicted values of Knudson and colleagues (13). Subjects who had used short-acting -agonist or anticholinergic aerosols within 6 h of presenting were asked to withhold medication before returning for later testing. None of the subjects were receiving long-acting -agonist aerosols. Histamine diphosphate (ICN Pharmaceuticals Inc, Costa Mesa, CA) was administered, using DeVilbiss No. 45 handheld nebulizers (DeVilbiss Health Care Inc, Somerset, PA), in doubling doses from 0.03 to 3.9 µmol. The test was stopped if the FEV₁ fell by 20% or more. Salbutamol aerosol was administered to aid recovery when necessary. The dose of histamine that provoked a 20% fall in FEV₁ (PD₂₀FEV₁) was estimated by interpolation. Dose-response ratio (DRR) was calculated for all subjects as the percent fall in FEV₁ at the last dose, divided by the total dose administered (14, 15).

Subjects with baseline FEV₁ < 60% predicted were given 200 µg salbutamol by metered-dose inhaler through a spacer device, and the change in FEV₁ was measured after 10 min as a percentage of the baseline FEV₁. A positive response was defined as an increase in FEV₁ of 20% or greater.

AHR was defined as PD₂₀FEV₁  3.9 µmol histamine or a positive response to bronchodilator.

Allergic Sensitization

Sensitization to common allergens was measured by skin prick test reactions to the forearm (16). The eight allergens tested were house dust, house dust mites (Dermatophagoides pteronyssinus and D. farinae), cat dander, ryegrass, plantain, Alternaria tenuis, and cockroach. Histamine and glycerol were used as positive and negative controls. After 15 min, wheal size was recorded as the long axis and its perpendicular; mean wheal size was used in analyses. A reaction was regarded as positive if the wheal size was 4 mm or greater. Subjects were considered atopic if they had a positive reaction to any of the allergens tested.

Subject Classification

Subjects were divided into four groups on the basis of airway responsiveness and wheeze or wheeze after exercise in the previous 12 mo. "Current Asthma" was defined as both recent symptoms and AHR, "Wheeze Only" as recent symptoms but no AHR, "Asymptomatic AHR" as AHR but no recent symptoms, and "Normal" as neither recent symptoms nor AHR.

Data Analysis

Repeatability of NO measurements was estimated, firstly, by calculating the coefficient of repeatability as twice the standard deviation of the differences between repeat tests (16) and, secondly, by calculating the intraclass correlation coefficient (ICC) (17).

The distribution of the NO values was tested for normality using the Shapiro-Wilk test. Because this test showed that NO was log-normally distributed, all subsequent analyses were carried out on log-transformed data. The mean of two NO values was calculated for each subject and used in analyses. Summary values for both NO and DRR are geometric means with their 95% confidence intervals. Summary values for all other parameters are arithmetic means and 95% confidence intervals. Differences between groups were assessed by single factor analysis of variance or by t test. Significance was accepted at the 5% level.

	RESULTS

TOP ABSTRACT INTRODUCTION METHODS RESULTS DISCUSSION REFERENCES

Of the 376 people who attended the study, 309 had exhaled NO measured. Of these, 303 had a histamine challenge and three with poor lung function had a positive bronchodilator response. Data from these 306 subjects were analyzed for this report. Of the 67 subjects without NO measurement, nine completed a questionnaire but refused further tests, and 58 had a histamine challenge but did not have NO measured because the analyzer was not available at the time of testing. The prevalence of AHR in these 58 subjects was 12.0%, and it did not differ significantly from the 11.1% in the 306 tested.

The features of the population sample are shown in Table 1. The asthma group had significantly higher values for DRR, atopy, and peripheral blood eosinophils than did the normal or wheeze-only groups. The prevalence of atopy (41%) and recent symptoms (35%) in this sample were not significantly different from that observed in the last follow-up of this cohort in 1992 (41.5 and 32%). Few of the asthma group or the wheeze-only group were receiving inhaled corticosteroids at the time of the study.

Exhaled NO values were log-normally distributed (Figure 1) (Shapiro-Wilk statistic = 0.98, p = 0.45). The normal range for exhaled NO was estimated by calculating 1.96 × SD of values from the normal, nonatopic subjects as 2.3 to 21.9 ppb, with a mean of 7.79 ppb. The difference between repeated measurements, plotted against their mean, in 294 subjects with duplicate measurements made on the same day is shown in Figure 2, where the upper panel shows untransformed data and the lower panel shows log-transformed data. Using the untransformed data the ICC was 0.98, and the coefficient of repeatability, representing the range within which 95% of pairs of repeated measurements would be expected to lie, was 4.58 ppb. Using the log data the ICC was 0.96 and the coefficient of repeatability was 0.21 log units. In 53 subjects who had the measurements repeated 1 to 2 wk later, the ICC was 0.94 and the coefficient of repeatability was 11.0 ppb using untransformed data, and 0.91 and 0.36 log units, respectively, using the log data.

View larger version (28K): [in this window] [in a new window]   Figure 1.   Frequency distribution of the log of exhaled NO values in 306 young adults.

View larger version (17K): [in this window] [in a new window]   Figure 2.   Difference between repeated exhaled NO measurements, plotted against their mean, in 294 subjects with duplicate measurements made on the same day. The upper panel (A) shows untransformed data and the lower panel (B) shows log-transformed data.

Geometric mean exhaled NO values in each of the four groups are shown in Figure 3. In the group with current asthma, exhaled NO was 22.2 ppb (16.1 to 30.7 ppb), significantly higher than in subjects with recent symptoms and no AHR (wheeze) (8.8 ppb and 7.5 to 10.3 ppb) (p < 0.0001) or subjects without symptoms or AHR (normal) (7.8, 7.1 to 8.4 ppb) (p < 0.0001). The wheeze and normal groups did not differ significantly (p = 0.16). In the group with asymptomatic AHR, exhaled NO was 11.03 (7.3 to 16.6), although the small number of subjects in the group (n = 7) precludes any meaningful statistical comparisons.

View larger version (11K): [in this window] [in a new window]   Figure 3.   Mean (± 95% confidence intervals) exhaled NO in 192 normal subjects (no AHR, no recent symptoms), 80 subjects with wheeze only (no AHR), seven subjects with asymptomatic AHR, and 27 subjects with current asthma (AHR and recent symptoms). ***p < 0.0001 for comparison with asthmatic group.

In the sample as a whole, exhaled NO was significantly correlated with the dose-response ratio to histamine (r = 0.39, p < 0.001) (Figure 4) and with the number of eosinophils in the peripheral blood (r = 0.35, p < 0.01). When the sample was divided on the basis of AHR, the correlation between exhaled NO and DRR remained significant in the hyperresponsive group (r = 0.33, p < 0.05), but it was not significant in the normoresponsive group (r = 0.12).

View larger version (17K): [in this window] [in a new window]   Figure 4.   Relationship between exhaled NO and dose response ratio to histamine in the population sample of 192 normal subjects (closed circles), 80 subjects with recent wheeze (open squares), seven subjects with asymptomatic AHR (inverse open triangles), and 27 subjects with current asthma (closed triangles).

In subjects without AHR, exhaled NO was significantly higher in atopic than in nonatopic subjects, both in the wheeze group (11.7, 9.1 to 15.0 ppb versus 7.3, 6.1 to 8.7 ppb, p < 0.01) and in the normal group (9.4, 8.1 to 11.0 ppb versus 7.1, 6.4 to 7.8 ppb, p < 0.01). After adjusting for atopic status, rhinitis was not associated with significant increases in exhaled NO. In atopic subjects exhaled NO in rhinitic and nonrhinitic subjects was 13.2 (11.0 to 15.6) and 10 (8.4 to 12), respectively (p = 0.06). In nonatopic subjects the values in rhinitic and nonrhinitic subjects were 7.98 (6.6 to 9.5) and 6.9 (6.3 to 7.6) (p = 0.14).

	DISCUSSION

TOP ABSTRACT INTRODUCTION METHODS RESULTS DISCUSSION REFERENCES

This study has shown that exhaled NO is log-normally distributed in a population sample and, using the methods described here, is a highly repeatable measurement. These measurement characteristics of exhaled NO have not previously been reported for a population sample. The data also show that exhaled NO levels are increased in subjects with current asthma, and are correlated with factors known to be associated with asthma. In addition, we have shown that exhaled NO is not increased in subjects with wheeze but no AHR, suggesting that they are unlikely to have airway inflammation.

The study sample is likely to be representative of the population, albeit within a very narrow age range. We compared the characteristics at the time of recruitment in 1982 and found no differences between attenders and nonattenders in 1997 in the prevalence of atopy, recent wheeze, or AHR. Furthermore, the prevalence of atopy and recent wheeze in this sample did not differ significantly from that measured at the last follow-up of the cohort, in 1992, or from that of other adult populations in New South Wales (18).

The exhaled NO levels found in normal subjects in this study are similar to levels measured directly in the lower airways (19, 20), and to values reported elsewhere for mixed expired levels (19, 21) and end-expiratory plateau levels (22). Because differences in methods of measurement may lead to differences in the measured values, the European Respiratory Society (ERS) Task Force (23) on exhaled NO has recommended that expired gas be collected during a slow exhalation against a low resistance. In this study subjects were asked to blow as hard as they could against the resistance provided by the 2-mm-diameter valve on the Tedlar bags. This was intended to close the soft palate and thus exclude nasal air from the mixed expired gas. Contamination of the lower airway gas with nasal air, which is very high in NO, can increase the NO content of mixed expired gas. However, expiration against a resistance is extremely effective in reducing nasal contamination (11).

Exhaled NO was measured after the subjects had completed the histamine challenge test, at a time when the FEV₁ had returned to normal. It is unlikely that the exhaled NO values were affected by either the bronchial challenge test (7, 24) or by any bronchodilator administered (24, 25), but it has been suggested that repeated spirometry may lower NO levels (26).

Exhaled NO levels are dependent on expiratory flow rate, an effect that is most pronounced at low flow rates (27). The ERS Task Force recommends flow rates of 10 to 15 L/min. Expiratory flow was not measured in the present study, although it is likely that the 2-mm valve on the Tedlar collection bag tended to reduce flow rates. To estimate the effect of the valve on flow rates, 10 normal subjects in the laboratory had expiratory flow rate estimated while blowing into the Tedlar bag using identical methods to those described above. The flow rates varied from 6 to 10 L/min. The data of Silkoff and colleagues (27) suggest that exhaled NO values are relatively stable above 6 L/min, and that flow dependency is minimal at these higher flow rates. Our data support this suggestion since the high degree of reproducibility of exhaled NO in our study clearly establishes that any variation in flow rate, or indeed the effects of any nasal contamination or dilution by dead space gas, have no significant effect on within-subject variation in the measurements using this method. The clear separation between clinically distinct groups (i.e., asthma, normal, atopic) further supports the contention that between-subject variation in these technical factors does not degrade the validity of the test.

Asthma has been defined as an inflammatory disorder of the airways associated with both symptoms and AHR (28). In the present study the subject groups were classified on the basis of AHR and symptoms, and the group with current asthma had high exhaled NO levels that clearly distinguished it from the normal group, which probably reflects a high level of airway inflammation. This definition of asthma, as the presence of AHR and recent symptoms, has been shown to detect subjects with relatively severe, ongoing abnormality (1, 29). By contrast, in subjects with wheeze, but with no objective evidence of AHR, exhaled NO was not increased, suggesting that airway inflammation is not present in these subjects. This is in accord with data showing little evidence of any significant functional impairment in subjects with wheeze, but no AHR (1, 29). The cause of their symptoms is not known, but it seems likely that these subjects do not have asthma.

Three subjects in the wheeze group were receiving ICS at the time of the study, and it is possible that these three were, in fact, asthmatic since ICS can reduce both exhaled NO (30) and AHR (31) to normal levels. However, drug therapy is unlikely to have influenced the findings in the remainder of the subjects in the wheeze group. In the asthmatic group, six subjects were receiving ICS at the time of the study, and they had a mean exhaled NO value of 22.0 (13.5 to 36.1), similar to that of the rest of the asthmatic group. Although they might have been expected to have lower values than asthmatics not receiving ICS, the effect of steroids is likely to be dose-dependent (32). There is no information available about the dose or frequency of medication use in this community of young adults, but our anecdotal evidence suggests that they used their medication infrequently.

The high intraclass correlation coefficients obtained show that not only are exhaled NO measurements highly reproducible but that they also discriminate very well among subjects. This study has shown that a part of these between-subject differences relate to AHR and allergy. Exhaled NO levels were significantly, but weakly, correlated with AHR and peripheral blood eosinophils. Both AHR and blood eosinophilia are characteristic features of asthma, and the presence of significant correlations with exhaled NO lend support to the concept that exhaled NO may be useful for monitoring asthma. However, the weakness of the correlations suggests that these parameters reflect different aspects of the inflammatory response. Exhaled NO was greater in atopic than in nonatopic subjects, both in the present and previous (33) studies, suggesting that mild airway inflammation is present in atopic, nonasthmatic subjects. Evidence from previous bronchoscopic studies showing mild eosinophilia in biopsies from atopic nonasthmatic subjects (34) support this suggestion.

In conclusion, this study suggests that measurement of exhaled NO is both feasible and useful for epidemiologic studies of asthma. A simple, rapid, noninvasive measure of airway inflammation allows important hypotheses about the role of airway inflammation in asthma to be tested. This study has tested one such hypothesis showing that wheeze, in the absence of any objective measurement of airway hyperresponsiveness, is unlikely to be associated with significant airway inflammation.