Published ahead of print on April 20, 2006, doi:10.1164/rccm.200506-962OC
© 2006 American Thoracic Society doi: 10.1164/rccm.200506-962OC
Measurement of Bronchial and Alveolar Nitric Oxide Production in Normal Children and Children with AsthmaDepartment of Paediatric Respiratory Medicine, Royal Brompton Hospital, London; Department of Thoracic Medicine, National Heart and Lung Institute, Imperial College London, London, United Kingdom Correspondence and requests for reprints should be addressed to Emmanouil Paraskakis, M.D., Ph.D., Department of Paediatric Respiratory Medicine, Royal Brompton Hospital, Sydney Street, London SW3 6NP, UK. E-mail: paraskakis{at}edu.med.uoc.gr
Rationale: Airway inflammation is characteristic of asthma. Distal inflammation may be particularly important. Objective: To calculate alveolar nitric oxide (NO) concentration (Calv) and bronchial flux NO (JNO) in children. Methods: We measured Calv and JNO from the fractional exhaled NO (FeNO50) measured at multiple exhalation flow rates in 132 children (aged 4–18 yr) with known atopic status, medication, and asthma control. Measurements and Main Results: Of participants, 85% (112/132) completed all measurements. In 20 of 112, the result did not fit the linear model. Thus, JNO and Calv were assessed in 92 (70%) subjects. The median (range) values of asthmatic (n = 52), normal (n = 20), and nonasthmatic atopic (n = 20) children were as follows: FeNO50: 28.1 (4.3–190), 10.35 (3.3–29), 21.8 (8.7–69) ppb, respectively; JNO: 1,230 (204–9,236), 480 (196–1,913), 1,225 (486–4,119) pl/s, respectively; Calv: 2.22 (0.44–6.63), 1.63 (0.44–3), 1.21 (0.03–2.85) ppb, respectively. A reproducibility study in 18 other children gave intraclass correlation coefficients (single measures) of 0.99 (JNO) and 0.81 (Calv). JNO and Calv were higher in children with asthma than normal children (p = 0.0004 and p = 0.0002, respectively). Children with poorly controlled asthma (n = 27) had higher FeNO50 measurements than children with good symptom control (n = 25): Calv: mean (± SD), 3.17 ± 1.62 versus 2.26 ± 1.30 ppb, p = 0.03; JNO: mean (± SD), 2,634 ± 2,255 versus 1,193 ± 1,294 pl/s, p = 0.007, respectively. Conclusions: Measurement of JNO and Calv is feasible in 70% of school-age children. FeNO50 and JNO give the same information (r = 0.97, p < 0.0001), Calv is higher in asthmatic children than in normal children and is affected by asthma control, but not by atopy. Calv may possibly reflect alveolar inflammation in asthma.
Key Words: airway monitoring • exhaled • pulmonary Chronic inflammation is considered to be a cardinal characteristic of asthma (1). There is accumulating evidence to suggest that airway inflammation occurs in all parts of the respiratory tract. Moreover, alveolar inflammation appears to be a significant part of the inflammatory process in severe asthma in particular (2, 3). Because emphasis has been placed on treating chronic airway inflammation (4), it is clearly desirable to be able to measure it noninvasively (5). Nitric oxide (NO) is synthesized throughout the respiratory tract by nitric oxide synthases (6, 7). Measurement of exhaled NO has been utilized in an attempt to predict the failure of steroid reduction (8) and asthma exacerbations (9). More sophisticated measurements have allowed the creation of mathematical models of pulmonary NO dynamics to provide the means separately to assess the bronchial flux (JNO) and alveolar concentration (Calv) of NO by measuring the fractional exhaled NO (FeNO50) at multiple exhalation flow rates (10, 11). Increased JNO with normal Calv has been found in adult patients with mild asthma (12, 13), whereas Calv in adults with more severe disease and nocturnal symptoms proved to be increased (14). Moreover, it has been shown that inhaled corticosteroids decrease the bronchial but not the alveolar NO of adults with asthma (15). Those results suggest that the assessment of Calv could be used to detect alveolar inflammation in children with asthma. We hypothesized that JNO and Calv could be measured reproducibly in children, and that Calv, as a marker of alveolar inflammation, would be elevated in children with severe or poorly controlled, but not mild, asthma. Therefore, the aim of this study was, first, to assess the feasibility of measurements of JNO and Calv in children with asthma, nonasthmatic atopic children, and normal nonasthmatic nonatopic children and, second, to determine the relationship of these measurements to FeNO50, atopic status, asthma medication, and control of disease.
Subjects A total of 132 children (median age, 13 yr; range, 4–18 yr; 73 males) were recruited. Of these, 81 (49 males) were children with asthma treated with inhaled or systemic corticosteroids with no recent (within 2 wk) respiratory infection or exacerbation; 24 (13 males) were atopic and without asthma, and 27 (11 males) were normal nonatopic, nonasthmatic children. Details of the study group are given in Table 1. None of the subjects was a self-reported current or previous smoker. The patients with asthma were recruited from our pediatric asthma clinic and the nonasthmatic atopic and normal children were recruited from a school study that was conducted at neighboring primary schools to our hospital. The study was approved by the Ethics Committee of Royal Brompton Hospital. Informed, written consent was obtained from the parents of the participating children, and age-appropriate assent was obtained from the children themselves.
Children with Asthma The inclusion criteria for the asthmatic group were as follows: (1) clinical diagnosis of asthma (a history of at least two of the following: cough, shortness of breath, recurrent wheeze, chest tightness); (2) increase in FEV1 after bronchodilator > 15%; (3) no clinical evidence of asthma exacerbation in the preceding 2 wk, and no change in asthma treatment; and (4) exclusion of other diseases mimicking asthma. Children with poorly controlled asthma were defined as needing rescue inhalers more than three times per week for control of symptoms. Children with severe asthma were defined as requiring more than 1,600 µg/d inhaled corticosteroids or systemic steroids for the control of their disease. Children with difficult asthma were defined as having symptoms requiring rescue bronchodilators on 3 or more d/wk, despite receiving more than 1,600 µg/d inhaled corticosteroids and regular long-acting  2-agonist and/or systemic steroids (triamcinolone or prednisolone).
Atopic, Nonasthmatic Subjects (Atopic)
Nonatopic, Nonasthmatic Subjects (Normal)
Reproducibility Study
Study Design
Pulmonary Function
FeNO50 Measurement
Allergy Testing Prick tests for five common allergens were performed in all normal and atopic subjects. A wheal diameter of 3 mm or greater above the negative control was considered a positive result.
Statistics In the absence of previously published data, it was not possible to perform a power calculation for any part of the study.
Feasibility A total of 132 children were tested (median age, 13 yr; range, 4–18 yr; Table 1). Of these children, 85% (112/132) completed all four exhalation flows successfully. The groups who could complete tests are compared in Table 2. The subjects who failed to complete all exhalation flows were younger and had significantly lower absolute and predicted FEV1 and absolute FVC than those who completed all flows. There was no difference in atopic status or asthma control between the two groups.
The Linearity of the Results Measurements of NO at multiple flow rates in 20 of the remaining 112 children did not fit the expected linear mode (Figure 1). The details of those whose data could and could not be fitted to the linear model are given in Table 3. The only difference between the groups was that those children who did not fit the linear model had lower absolute and % predicted FEV1 and a higher FeNO50.
Calculation of Bronchial and Alveolar NO JNO and Calv concentrations were calculated in 92 of 112 (82%) of the subjects who could perform measurements at all flow rates. Of these, 52 were asthmatic, 20 atopic, and 20 nonatopic, normal children.
Differences among Groups
JNO of the asthmatic group was similar to that of the atopic group (median, 1,230 pl/s [range, 204–9,236] vs. 1,225 pl/s [range, 486–4,119]; p = 0.87) and significantly higher than the normal group (median, 1,230 pl/s [range, 204–9,236] vs. 480 pl/s [range, 196–1,913]; p = 0.0004). Atopic children had higher JNO than the normal children (median, 1,225 pl/s [range, 486–4,119] vs. 480 pl/s [range, 196–1,913]; p = 0.0004; Figure 2). Calv was significantly greater (p = 0.0002) in the asthmatic group treated with steroids (median, 2.22 ppb [range, 0.44–6.63]), than in the atopic subjects (median, 1.21 ppb [range, 0.03–2.85]) and the normal children (median, 1.64 [range, 0.44–3]; p = 0.002; Figure 2). Children with poorly controlled asthma (n = 27), irrespective of prescribed medication (whether inhaled or oral steroids), had higher Calv than patients with good control (n = 25) of their disease (mean ± SD: 3.17 ± 1.62 vs. 2.26 ± 1.30 ppb, p = 0.03). Details are given in Table 4. JNO of children with poorly controlled asthma was also found to be significantly higher than that of the well controlled group (mean ± SD: 2,634 ± 2,255 vs. 1,193 ± 1,294 pl/s, p = 0.007; Figure 3.
Children with asthma with nocturnal symptoms (n = 10) had a similar Calv compared with the rest of the asthmatic group (2.73 ± 1.57 vs. 2.75 ± 1.43, p = 0.98). Results are summarized in Table 5. Although the numbers are small, children with difficult asthma (n = 22) had higher Calv than children (n = 6) who had severe asthma but good control of the disease (mean ± SD: 2.97 ± 1.42 vs. 1.50 ± 1.18 ppb, p = 0.03; Table 6). Interestingly, the difference between the values of JNO (mean ± SD: 2,808 ± 2,415 vs. 1,890 ± 1,018 pl/s, p = 0.38) and FeNO50 (mean ± SD: 55.9 ± 46.5 vs. 36.18 ± 17.24 ppb, p = 0.32) for these two groups was not statistically significant. Moreover, using stepwise linear regression, we found from the independent values of height, age, atopy status, medication, night symptoms, reported use of rescue inhaler, JNO, forced expired flow rate between 25 and 75% of vital capacity (FEF25–75), FEF25–75% predicted, FEV1, FEV1%, FVC, FVC%, and FeNO260, only height and the reported use of a rescue inhaler (number of days per week) predicted Calv (Table 7).
Relationship among JNO, Calv, and Lung Function JNO was strongly correlated with FeNO50 (r = 0.97, p < 0.0001) and FeNO260 (r = 0.52, p = <0.0001; Figure 4), but JNO was not correlated with Calv (r = 0.07, p = 0.53), FVC% (r = 0.032, p = 0.76), or FEV1 (r = 0.04, p = 0.72), but weakly correlated with FEV1% (r = –0.25, p = 0.01), FVC (r = 0.21, p = 0.05), and FEV1/FVC (r = –0.32, p = 0.02).
Calv significantly correlated with FeNO260 (r = 0.34, p = 0.0009), but not with FeNO50 (r = 0.19, p = 0.07; Figure 5). A correlation was also found between Calv and the height of the subjects (r = 0.25, p = 0.02) but not between Calv and lung function (FEF25–75: r = 0.02, p = 0.89; FEF25–75%: r = –0.18, p = 0.12; FEV1: r = 0.11, p = 0.3; and FEV1/FVC: r = –0.09, p = 0.39).
Reproducibility A separate group of 18 children performed the reproducibility study. Group median age was 15.2 yr (range, 10.8–18.1 yr) and FEV1% predicted was 89 (range, 46–110). The data are shown graphically in Figures 6 and 7. The intraclass correlation coefficients (single measures), which consider the contribution of variance between repeated measurements to the total variance, were highly acceptable at 0.992 for JNO and 0.809 for Calv, On the other hand, the 95% limits of agreement for two repeated measures were wide (JNO –356 to 420 ppb; Calv –1.51 to 1.16 ppb).
As we hypothesized at the start of the study, JNO and Calv could be measured reproducibly in children, and Calv was elevated in children with severe or poorly controlled, but not mild, asthma. However, the major and unexpected finding was that Calv, unlike FeNO50 and JNO, was elevated in children with asthma but not in atopic, nonasthmatic children. There was no difference in Calv between atopic and nonatopic, nonasthmatic (normal) children. This is the first report of the feasibility and the results of variable flow NO measurements to calculate Calv and JNO in such a large group of children. We found that we could not calculate these parameters using the linear model in 30% of children, either because they could not manage all the maneuvers (15%) or because the results did not fit the linear model (15%). We also report for the first time that, although expected, JNO and FeNO50 were higher in asthmatic and atopic children than in healthy nonatopic children, Calv was not different between atopic and nonatopic children without asthma. Previous studies have shown that FeNO50 is increased in atopic people (18–21). The present study extends these observations by showing for the first time that atopic, nonasthmatic children, although having similar FeNO50 and JNO values as children with asthma, did not have an increased Calv. Calv was higher in the children with asthma than in nonasthmatic children, whether atopic or not, and higher in the subgroup of children whose asthma was poorly controlled than in those whose asthma was well controlled. This suggests that alveolar inflammation may be contributing to asthma control. Although we believe that Calv represents distal eosinophilic inflammation, in other contexts there is an imperfect correlation between NO levels and eosinophilic inflammation, and we do not have transbronchial biopsy data to validate our assumption. Such an invasive procedure is not likely to be acceptable for research in children (24). It should also be noted that some of the subgroup analyses had small numbers of patients, so the conclusions should be interpreted cautiously. Furthermore, we were unable to perform a power calculation to inform the size of the study, because of the lack of previous data. Our own results could serve as a basis for such calculations in future studies. The asthmatic group was limited to a fairly tight phenotype, namely those with documented bronchodilator reversibility and atopy, and the conclusions may have been different in other groups of children with asthma. We did not have access to a sufficient number of nonatopic, asthmatic children, and we did not include those with fixed obstruction for fear of contaminating the study group with children who had obliterative bronchiolitis, and not asthma. There are many different ways of assessing poor asthma control. For simplicity, and in accord with international guidelines, we chose bronchodilator usage as our sole criterion. This is used as a measure to adjust prophylactic medications in routine clinical practice. Were this simple approach to have misclassified patients, it would have biased us against finding any differences between the well- and poorly controlled groups. This is the largest study of its kind, and the first in children to explore the effect of atopy on JNO and Calv. The fact that measurements could only be made on 70% of the children studied is a weakness. We used visual cues, and coached the children in the flow measurements in detail to try to improve our success rate. However, even in those in whom all flows could be accomplished, 15% of the data could not be fitted to a linear model. Interestingly, the children whose data did not fit the linear model had significantly lower FEV1% and no difference in age, sex, atopy status, asthma control, or absolute values of FEV1 and FVC compared with children whose measurements did fit the model. It is possible that models, which are valid in normal individuals and individuals with modest degrees of airflow obstruction, may be less valid if airflow obstruction becomes more profound. In particular, the implicit assumption of a single compartment airway–alveolus model breaks down if there is a range of time constants because of patchy airflow obstruction, giving rise to a multicompartment model of much greater mathematical complexity (22). It has also been shown that there are other factors that might increase or decrease NO levels, which might thus decrease the accuracy of the models. These include rhinovirus infection, Th-1 cytokines, airway pH, the extent of endogenous nitrite stores, and airway thiol binding (22, 23). Another weakness of the study is the lack of longitudinal data, which might have enabled us to determine whether increasing the dose of inhaled steroids in the children whose asthma was poorly controlled would improve either or both of asthma control and Calv. Such data would have enabled us to establish more clearly the relationship (if any) between the two. A previous study (25) reported variable flow NO measurements in children. In contrast to our results, they were much less successful in obtaining measurements at the highest (success rate 20/60 participants) and lowest (25/45 asthmatic participants) flow rates, despite studying subjects with asthma of a similar age group and generally lesser severity. We cannot account for this discrepancy. This group also used another equation (23) to attempt to calculate JNO and Calv in the children in whom the data were either incomplete or did not fit the linear model. We elected not do this: first, because it was not part of our prospective data analysis plan; and second, because although there is good agreement between the two calculations in the children who accomplished all flows, we cannot assume that this would be so in those whose data did not fit the model or could not be obtained. Our raw data are given in the online supplement, so the calculations can be performed by other investigators if they wish. This group (25), like ourselves, reported a higher Calv in children whose asthma was poorly controlled, but did not have data to comment on any differences between atopic and nonatopic control subjects. From the data reported here, it is clear that Calv gives different information to that obtained from JNO and FeNO50. However, and not unexpectedly, JNO is closely correlated with FeNO50 (r = 0.97, p < 0.0001) and therefore provides little if any novel information. Thus, if only bronchial NO production is of interest, there is no advantage to making measurements at multiple expiratory flows. By contrast, the calculation of Calv might provide useful additional information to that obtained from FeNO50 about the inflammatory status of the asthmatic lung. Airway inflammation is increasingly reported as a cardinal characteristic of severe asthma in adults and children. Previous studies, mostly of subjects with mild, steroid-naive asthma, have shown that the alveolar Calv was normal (15). In our study, in which the majority of children had severe asthma (36/52), we found that the Calv was significantly higher than that of normal children. Moreover, a number of studies reported that loss of control of asthma may be due to an increase in inflammation (26), which is more profound in the alveoli of adults with severe asthma (27). In addition, we found that the children with severe asthma and good control of their disease (expressed as the d/wk need of rescue inhaler) had significantly lower Calv than the children with the same asthma severity and poor control of their disease (difficult asthma). Using a stepwise linear regression model with the variables being atopy, age, sex, medication, spirometric parameters, and height, we found that the control of the disease is the major factor affecting the peripheral airway NO concentration. It is important not to overinterpret the utility of measurements of Calv. We believe that the differences in Calv are useful in studying mechanisms that may differ among groups, but are unlikely to be of use in monitoring individuals in the clinic. There is considerable overlap between groups of patients, and, whereas the intraclass correlation coefficient for Calv was highly acceptable, suggesting that the variation between subjects was greater than that within subjects, the limits of agreement for a pair of measurements for an individual were wide. Furthermore, longitudinal studies are now needed to determine whether associations are causal, and whether there is any clinical utility, as opposed to scientific interest, in the measurements. We also need to find ways of improving the numbers of children in whom the measurements can be made. In summary, this study suggests that the calculations of Calv and JNO are feasible in around 70% of school-age children. Calv, unlike FeNO50 and JNO, is elevated in atopic, asthmatic, but not in atopic, nonasthmatic children, and there is no difference in Calv between atopic and nonatopic, nonasthmatic children. Although FeNO50 and JNO give essentially the same information, Calv is higher in asthmatic children than in normal children, and atopy does not affect this measurement. The study highlights the relationship between poor control of asthma and Calv (which we assume to be a marker of alveolar inflammation) but further work is needed to confirm the relevance of this.
This article has an online supplement, which is accessible from this issue's table of contents at www.atsjournals.org Originally Published in Press as DOI: 10.1164/rccm.200506-962OC on April 20, 2006 Conflict of Interest Statement: E.P. does not have a financial relationship with a commercial entity that has an interest in the subject of this manuscript. C.B. does not have a financial relationship with a commercial entity that has an interest in the subject of this manuscript. L.F. does not have a financial relationship with a commercial entity that has an interest in the subject of this manuscript. R.K. does not have a financial relationship with a commercial entity that has an interest in the subject of this manuscript. S.A.K. is a member of the scientific advisory board for Aerocrine. He received lecture fees from Aerocrine, AstraZeneca, and Merck Sharp and Dohme (MSD), and research grants from GlaxoSmithKline, AstraZeneca, Duska, MSD, and Aerocrine. N.M.W. does not have a financial relationship with a commercial entity that has an interest in the subject of this manuscript. P.J.B. received research funding, lecture fees, and has served on scientific advisory boards for GlaxoSmithKline, AstraZeneca, Boehringer Ingelheim, Novartis, Altana, Pfizer, and Scios. A.B. received 30 portable NO machines from Aerocrine for a study funded by the British Lung Foundation. Received in original form June 22, 2005; accepted in final form April 12, 2006
|
||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
TOP
ABSTRACT
METHODS
E), expressed in ml/s, JNO was calculated as JNO = FENO x 
2 method was used to test differences between percentages. The Mann-Whitney and Kruskal-Wallis tests were used to compare nonparametric data. Differences were considered significant at a value of p < 0.05. The reproducibility of two repeated measurements of Calv and JNO was calculated in two ways: (1) as the 95% limits of agreement (mean ± 1.96 SDdiff) and (2) as the intraclass correlation coefficient (single measurements) expressed on a 0–1 scale where 0 represents no agreement and 1 represents perfect agreement. 
