INTRODUCTION

TOP ABSTRACT INTRODUCTION METHODS RESULTS DISCUSSION REFERENCES

Inducible nitric oxide synthase is active in asthmatic airways (1) and nitric oxide (NO) can be measured in exhaled air (2- 6). So far, however, there have been few studies (7, 8) of NO levels in childhood asthma and none in a community-based population.

The Wythenshawe Community Asthma Project (WYCAP) is a long-term prospective study of the natural history of asthma in two general practice populations on a housing estate in Manchester, United Kingdom. In 1993 and 1995, postal questionnaires were sent to all adults and children registered with the two practices asking about respiratory symptoms and asthma-related conditions (9). The children's questionnaire was based on the ISAAC questionnaire (12). In order to test a hypothesis that children with three or more positive responses to five key questions might have asthma, detailed clinical assessments have been made on a random sample of respondents to the 1995 questionnaire. These examinations provided the opportunity to examine the relationship between exhaled NO (eNO) levels, nasal NO (nNO) levels, and asthma in a community-based group of children 5 to 15 yr of age.

	METHODS

TOP ABSTRACT INTRODUCTION METHODS RESULTS DISCUSSION REFERENCES

In September 1995 the parents or guardians of 3,290 children were sent a postal questionnaire designed to identify the 1-yr period prevalence of designated respiratory symptoms (APPENDIX 1). Nonrespondents were sent reminders at 4 and at 8 wk after the initial mailing. Completed questionnaires were received back for 2,434 children, which after allowance for an estimated 5.5% of children no longer believed to be living at the mailing address, gave an adjusted response rate of 78%. A stratified weighted random sample of respondents attending one of the general practices was personally invited by their general practitioner to attend The North West Lung Centre, Wythenshawe, Manchester, for clinical review. The sampling method was designed to yield a high number of asthma cases. Children were not excluded if they had a previous diagnosis or were receiving treatment for asthma. The clinical assessment involved a full medical history, physical examination, and investigations including exercise challenge, spirometry with reversibility to ₂-agonists, a 1-wk electronic peak flow diary record, skin prick testing to house dust mite, grass pollen, cockroach, dog, and cat. Atopy was defined as a skin wheal 3 mm greater than that of the negative control. The mean wheal size for each allergen was defined as the mean of the maximal wheal diameters at 90 degrees to each other. Five allergens were studied and the mean wheal diameter was calculated for the five allergens together. A positive exercise test was defined as a fall in FEV₁ > 15% from a preexercise value after 6 min of free running exercise. Exercise tests were performed outdoors in the summer months.

The eNO was measured using a chemiluminescence analyzer (Model LR2000; Logan Research, Rochester, UK), sensitive to NO from 1 to 5,000 parts per billion (ppb) by volume, and with a resolution of 1 ppb, which was designed for on-line recording of exhaled NO concentration. The response time (10 to 90%) was < 0.6 s. In addition to NO, the analyzer also measured CO₂ (resolution, 0.1% CO₂; response time, 200 ms), with sample pressure and volume in real time. The sampling rate of the analyzer was 250 ml/min for all measurements. The analyzer was calibrated daily using a certified NO mixture (114 ppb) in nitrogen (BOC Special Gases, Guilford, UK).

Measurements of eNO were made by using a sidearm sampling technique as described by Kharitonov and colleagues (6, 13). The NO value corresponding to the plateau of the end exhaled CO₂ reading (> 5% CO₂) was recorded.

The nNO was measured with a Teflon tube inserted into one of the nares while the subject held his or her breath, i.e., with no active exhalation for at least 30 s (6, 13). Two readings of exhaled and nasal air were taken, with results being expressed as a mean value.

Measurements of NO were made by the clinical investigator TF who had been trained by AA. Readings from the stored data were made by AA who was unaware of the clinical results or of the expert opinion. The data had been stored with only an identifying case number. The expert consensus opinion had not been obtained at the time the readings were taken.

Three independent consultant pediatricians with an interest in asthma were then supplied by post with all the information from the clinical assessment except the NO findings. After reviewing this the pediatricians were asked to rate into four categories the probability that each child had asthma: > 90% (probable asthma), 50 to 90% (possible asthma), 10 to 50% (asthma unlikely) or < 10% (nonasthmatic). When one of the experts disagreed with his colleagues, the final rating used was the consensus view of the two agreeing physicians. When all three consultants disagreed the child was rated according to the middle opinion.

Statistics

The eNO data followed an approximate log-normal distribution, so they were converted to natural logarithms for analysis. The data were reconverted to the original units for presentation; nNO data followed a normal distribution and were analyzed without transformation. Comparisons between the groups were done using a one-way analysis of variance (one-way ANOVA). Differences indicated by the ANOVAs were analyzed using the Tukey-HSD test. Statistical significance was set at the conventional 5% level (p < 0.05).

The study had ethical approval from the local research ethics committee, and all the children and their parents or guardians gave informed consent.

	RESULTS

TOP ABSTRACT INTRODUCTION METHODS RESULTS DISCUSSION REFERENCES

One hundred fifty-eight children attended for clinical assessment. Nitric oxide measurements were not possible in 61 as the necessary equipment was not available when they attended. Three children were unable to make any satisfactory recordings. Of the 94 remaining children one could not perform the eNO measurement and another refused to allow the nNO measurement to be taken. Two of these 94 children refused a skin prick test; 89 of the remaining 94 children were able to perform an exercise test. The three consultants agreed on the diagnosis in 55 children, two of the three agreed on the diagnosis in 20 children, and in 19 children there was no agreement between the consultants.

Children with probable asthma had significantly higher geometric mean eNO levels (8.3 ppb) (Table 1) than did those who were nonasthmatic (3.6 ppb) (p < 0.05, Tukey-HSD test). The mean nNO levels for children with probable asthma were also greater than those in the other groups, but the differences did not reach statistical significance. Atopic children had significantly higher geometric mean eNO levels (7.8 ppb) than did nonatopic children (3.4 ppb) (p < 0.05, one-way ANOVA); they also had higher mean nNO (996 and 658 ppb, respectively) (p < 0.05, one-way ANOVA). Children with positive exercise test results had significantly higher geometric mean eNO levels (8.5 ppb) than did those with negative exercise test results (4.4 ppb) (p < 0.05, one-way ANOVA) and higher mean nNO levels (1,031 and 793 ppb, respectively) (p < 0.05, one-way ANOVA) (Table 1).

The atopic probable asthmatics had significantly higher geometric mean eNO levels (12.5 ppb) than did the nonatopic probable asthmatics (3.2 ppb), the atopic nonasthmatics (3.8 ppb), and the nonatopic nonasthmatics (3.4 ppb) (p < 0.05, Tukey-HSD test) (Table 2 and Figure 1). Atopic probable asthmatics also had significantly higher mean nNO levels (1,108 ppb) than did nonatopic nonasthmatics (671 ppb) (p < 0.05, Tukey-HSD test).

View larger version (13K): [in this window] [in a new window]   Figure 1.   Scatter plot of tidal nitric oxide with relation to atopy and likelihood of asthma.

Atopic children with positive exercise test results had higher geometric mean eNO levels (12.22 ppb) than did atopic children with negative exercise test results (5.69 ppb), nonatopic children with positive exercise test results (3.17 ppb), or nonatopic children with negative exercise test results (3.47 ppb) (p < 0.05, Tukey-HSD test) (Figure 2). Atopic children with positive exercise test results also tended to have higher mean nNO levels (1,143 ppb) than did nonatopic children with positive exercise test results (631 ppb) and nonatopic children with negative exercise test results (696 ppb) (p < 0.05, Tukey-HSD test) (Table 3).

View larger version (17K): [in this window] [in a new window]   Figure 2.   Scatter plot of tidal nitric oxide with relation to atopy and result of exercise test.

There were no significant differences in eNO between male and female subjects or between different age groups.

The mean wheal size for the five allergens tested were added together and the combined mean wheal size was compared with the exhaled eNO levels using bivariate analysis. Pearson's correlation coefficient was calculated at 0.66 (p < 0.05), showing good correlation between mean wheal size and exhaled NO.

	DISCUSSION

TOP ABSTRACT INTRODUCTION METHODS RESULTS DISCUSSION REFERENCES

Our results show that eNO levels in children with probable asthma are higher than those found in children less likely to have the condition, in those with positive exercise test results compared with those with negative exercise test results, and in atopic children compared with nonatopic children. The higher eNO levels in asthmatic children, however, were mainly due to associated atopy. These results suggest that atopic asthmatic children produce greater amounts of eNO than do the nonatopic asthmatic children, perhaps because of different mechanisms operating in the two groups. This theory is supported by Tang and colleagues (14) who found differences in cells in bronchoalveolar lavage samples taken from atopic asthmatics exposed to house dust mite compared with those found in nonatopic asthmatics. On the other hand, Humbert and colleagues (15) found some evidence that both atopic and nonatopic asthmatic patients had infiltration of the bronchial mucosa with cells expressing Th2 type cytokines, suggesting similarities in the immunopathogenesis of these clinically distinct types of asthma.

We found that atopic children had higher eNO than did nonatopic children, but these differences were explained by each child's asthma status since nonasthmatic children had similar eNO levels whether atopic or nonatopic.

Children with positive exercise test results had higher eNO than did those with negative exercise test results, but the important cofactor was atopic status, with the highest values being seen in atopic asthmatics.

Children were not asked to stop their medication prior to the review, and this could have affected our findings. If preventative asthma treatment lowered eNO levels, as suggested by other workers (16, 17), we might expect the differences between asthmatic and nonasthmatic children to be less marked. We found that atopic asthmatic children, whether receiving preventative asthma medication or not, had higher eNO levels than did nonatopic asthmatic children (Figure 3) and that, overall, children with probable asthma receiving preventative treatment had higher eNO levels than did those who were not. This observation may reflect differences in the severity and diagnosis of asthma in the community, children with more severe asthma perhaps being more likely to have their asthma diagnosed and therefore given preventative treatment than those with less severe asthma. Alternatively, it may reflect inadequate compliance with the preventative treatment in the asthmatics studied in this population. There were no nonasthmatic children receiving asthma medication and only one atopic nonasthmatic child was receiving nasal steroids.

View larger version (19K): [in this window] [in a new window]   Figure 3.   Scatter plot of tidal nitric oxide levels in children with asthma in relation to their atopic status and the treatment they were receiving.

We used a consensus diagnosis of asthma based on three consultant pediatricians reviewing all the information, except the NO levels, available after a clinical review. By this approach we tried to replicate the type of diagnostic information that would be available to a consultant when seeing a child referred by their general practitioner for diagnostic opinion. In the study, however, the children were more likely to be asymptomatic at the time of their clinical assessment than if they were attending a new patient clinic. The clinical information made available to our consultants, therefore, is more likely to have been normal, and this may have reduced the chances of them diagnosing asthma.

Although this study showed that there may be differences in the amounts of eNO produced by atopic and nonatopic asthmatics, further work is needed to confirm the finding in children who have not yet been started on inhaled steroid therapy.