Nitric Oxide and Clinical Phenotypes of Childhood Asthma
Nitric Oxide and Clinical Phenotypes of Childhood Asthma
Whether exhaled NO helps to identify a specific phenotype of asthmatic patients remains debated. Our aims were to evaluate whether exhaled NO (FENO0.05) is independently associated (1) with underlying pathophysiological characteristics of asthma such as airway tone (bronchodilator response) and airway inflammation (inhaled corticosteroid [ICS]-dependant inflammation), and (2) with clinical phenotypes of asthma.
We performed multivariate (exhaled NO as dependent variable) and k-means cluster analyses in a population of 169 asthmatic children (age ± SD: 10.5 ± 2.6 years) recruited in a monocenter cohort that was characterized in a cross-sectional design using 28 parameters describing potentially different asthma domains: atopy, environment (tobacco), control, exacerbations, treatment (inhaled corticosteroid and long-acting bronchodilator agonist), and lung function (airway architecture and tone).
Two subject-related characteristics (height and atopy) and two disease-related characteristics (bronchodilator response and ICS dose > 200 μg/d) explained 36% of exhaled NO variance. Nine domains were isolated using principal component analysis. Four clusters were further identified: cluster 1 (47%): boys, unexposed to tobacco, with well-controlled asthma; cluster 2 (26%): girls, unexposed to tobacco, with well-controlled asthma; cluster 3 (6%): girls or boys, unexposed to tobacco, with uncontrolled asthma associated with increased airway tone, and cluster 4 (21%): girls or boys, exposed to parental smoking, with small airway to lung size ratio and uncontrolled asthma. FENO0.05 was not different in these four clusters.
In conclusion, FENO0.05 is independently linked to two pathophysiological characteristics of asthma (ICS-dependant inflammation and bronchomotor tone) but does not help to identify a clinically relevant phenotype of asthmatic children.
Numerous studies have evaluated exhaled nitric oxide (NO) correlates in asthma. For instance, exhaled NO fraction (FENO) has been linked to atopy rather than to asthma per se which could be due to the underlying relationship between FENO and eosinophilic inflammation of airways. We and others have emphasized that FENO is also linked to other intrinsic dimensions of asthma such as airway reactivity/tone and remodeling of airways. All these relationships may explain the complex and still debated relationship between exhaled NO and asthma control/severity. Moreover, extrinsic factors also affect FENO such as tobacco exposure and asthma treatment. Finally, the epithelial surface of airways, which is linked to the height of the subject and possibly to sex, also affects exhaled NO. Despite this considerable background, the usefulness of its assessment in clinical practice remains debated because of its multidimensional nature, precisely. Furthermore, all these intrinsic and extrinsic exhaled NO modifiers seem to contribute for a minor part of exhaled NO variance, which constitutes a main limitation.
The recent study of Dweik and colleagues has shown that FENO may define an asthma phenotype. They demonstrated that their high FENO phenotype (FENO 0.05 > 35 ppb) was characterized by greatest airway reactivity, airflow limitation, hyperinflation, sputum eosinophilia and levels of symptoms. Nevertheless, FENO levels were similar among patients with severe and non-severe asthma in this latter study. One may hypothesize that the establishment of one or several relationships between FENO and asthma characteristics (either clinical or physiological) is not sufficient to demonstrate that exhaled NO identifies a specific phenotype with clinical relevance, accordingly to a more focused definition of phenotype.
The aims of our study were therefore (1) to evaluate the strength of the relationships between exhaled NO and physiopathological asthma characteristics, and (2) to assess whether a specific clinical phenotype can be isolated using exhaled NO measurement. For that purpose, we used two different statistical approaches. The first approach determined the clinical and physiological correlates of exhaled NO, and the degree of FENO variance explained by the correlates. The second approach used a more complex statistical tool, namely cluster analysis, which describes the dimensions of disease without the need for arbitrary a priori assumptions about classification, and was specifically designed to test the hypothesis that FENO is associated with a phenotype of childhood asthma.
Abstract and Introduction
Abstract
Whether exhaled NO helps to identify a specific phenotype of asthmatic patients remains debated. Our aims were to evaluate whether exhaled NO (FENO0.05) is independently associated (1) with underlying pathophysiological characteristics of asthma such as airway tone (bronchodilator response) and airway inflammation (inhaled corticosteroid [ICS]-dependant inflammation), and (2) with clinical phenotypes of asthma.
We performed multivariate (exhaled NO as dependent variable) and k-means cluster analyses in a population of 169 asthmatic children (age ± SD: 10.5 ± 2.6 years) recruited in a monocenter cohort that was characterized in a cross-sectional design using 28 parameters describing potentially different asthma domains: atopy, environment (tobacco), control, exacerbations, treatment (inhaled corticosteroid and long-acting bronchodilator agonist), and lung function (airway architecture and tone).
Two subject-related characteristics (height and atopy) and two disease-related characteristics (bronchodilator response and ICS dose > 200 μg/d) explained 36% of exhaled NO variance. Nine domains were isolated using principal component analysis. Four clusters were further identified: cluster 1 (47%): boys, unexposed to tobacco, with well-controlled asthma; cluster 2 (26%): girls, unexposed to tobacco, with well-controlled asthma; cluster 3 (6%): girls or boys, unexposed to tobacco, with uncontrolled asthma associated with increased airway tone, and cluster 4 (21%): girls or boys, exposed to parental smoking, with small airway to lung size ratio and uncontrolled asthma. FENO0.05 was not different in these four clusters.
In conclusion, FENO0.05 is independently linked to two pathophysiological characteristics of asthma (ICS-dependant inflammation and bronchomotor tone) but does not help to identify a clinically relevant phenotype of asthmatic children.
Introduction
Numerous studies have evaluated exhaled nitric oxide (NO) correlates in asthma. For instance, exhaled NO fraction (FENO) has been linked to atopy rather than to asthma per se which could be due to the underlying relationship between FENO and eosinophilic inflammation of airways. We and others have emphasized that FENO is also linked to other intrinsic dimensions of asthma such as airway reactivity/tone and remodeling of airways. All these relationships may explain the complex and still debated relationship between exhaled NO and asthma control/severity. Moreover, extrinsic factors also affect FENO such as tobacco exposure and asthma treatment. Finally, the epithelial surface of airways, which is linked to the height of the subject and possibly to sex, also affects exhaled NO. Despite this considerable background, the usefulness of its assessment in clinical practice remains debated because of its multidimensional nature, precisely. Furthermore, all these intrinsic and extrinsic exhaled NO modifiers seem to contribute for a minor part of exhaled NO variance, which constitutes a main limitation.
The recent study of Dweik and colleagues has shown that FENO may define an asthma phenotype. They demonstrated that their high FENO phenotype (FENO 0.05 > 35 ppb) was characterized by greatest airway reactivity, airflow limitation, hyperinflation, sputum eosinophilia and levels of symptoms. Nevertheless, FENO levels were similar among patients with severe and non-severe asthma in this latter study. One may hypothesize that the establishment of one or several relationships between FENO and asthma characteristics (either clinical or physiological) is not sufficient to demonstrate that exhaled NO identifies a specific phenotype with clinical relevance, accordingly to a more focused definition of phenotype.
The aims of our study were therefore (1) to evaluate the strength of the relationships between exhaled NO and physiopathological asthma characteristics, and (2) to assess whether a specific clinical phenotype can be isolated using exhaled NO measurement. For that purpose, we used two different statistical approaches. The first approach determined the clinical and physiological correlates of exhaled NO, and the degree of FENO variance explained by the correlates. The second approach used a more complex statistical tool, namely cluster analysis, which describes the dimensions of disease without the need for arbitrary a priori assumptions about classification, and was specifically designed to test the hypothesis that FENO is associated with a phenotype of childhood asthma.