FeNO in the diagnosis of bronchial asthma
Chronic airway inflammation is an important feature in the development and progression of bronchial asthma [
10,
33‐
36]. FeNO levels are increased in asthmatic patients [
37‐
39] as a result of induction of NOS2 by proinflammatory cytokines [
3,
40‐
43]. There is some degree of correlation between elevated FeNO levels and increased eosinophils in blood, bronchoalveolar lavage fluid [
28], bronchial biopsies [
44], sputum [
45], this indicating that FeNO reflects eosinophilic inflammation [
28,
33,
46,
47]. However, asthma is probably not a single disease since at least 3 adult phenotypes of airway inflammation have been identified on the basis of predominant eosinophilic, neutrophilic, or pauci-granulocitic cellular pattern [
48]. It is therefore not surprising that the difference in FeNO levels between symptomatic and asymptomatic children is relatively small, and a large overlap in the distribution of FeNO levels from subjects with and without asthma has been reported [
49‐
58] (Table
2). Another limitation is that even though FeNO is increased in children with allergic asthma [
59‐
62], the obtained levels cannot discriminate among schoolchildren with non-allergic asthma, or those with allergy without asthma, or the healthy population [
63]. Moreover, levels of FeNO appear increased in asthmatics with pollen allergy during the season also in the absence of symptoms of lung function impairment [
64]. Children with atopic eczema exhibit high levels of FeNO even though they do not have asthma [
65], but the mechanism is unclear.
Table 2
Diagnostic accuracy of FeNO measurement in patients with asthma
| Children, 3-7 years | 83 (25) | > 9.7 ppb | 86 | 92 |
| Adults | 160 (66) | > 16 ppb | 49 | 90 |
| Adults | 40 (47) | > 7 ppb | 82 | 89 |
| Children, 12-18 years | 106 (70) | > 19 ppb | 86 | 89 |
| Children, 10-12 years | 60 (45) | > 20.4 ppb | 41 | 97 |
| Children, 6-14 years | 57 (40) | > 19 ppb | 91 | 87 |
| Children, 5-18 years | 70 (4.5) | > 28 ppb | 64 | 70 |
| Adults, Children | 42 (30) | > 27 ppb | 78 | 92 |
| Children, 8-16 years | 167 (68) | > 22 ppb | 57 | 87 |
Obesity is another confounding factor in the assessment of FeNO levels. A consistent body of data now indicates that asthma is closely related to obesity, and obese patients with asthma usually have more severe symptoms than non-obese asthmatics [
66]. Indeed, obese asthma may be a unique phenotype that is characterized by more severe symptoms for a given degree of lung function impairment, destabilization or lack of asthma control, worse quality of life, lack of eosinophilic inflammation and a different response to controller medication [
67]. At any age, obesity can adversely impact on lung function, and obesity duration is a determinant of lower pulmonary function [
68]. FeNO levels have been extensively investigated in pediatric excess adiposity [
9,
69]. Some pediatric studies showed that FeNO is negatively associated with body mass index, waist-to-hip ratio, and percent body fat [
70], while others found no relationship between FeNO levels and adiposity measures [
69,
71]. Furthermore, FeNO levels do not differ between obese and normal weight subjects with asthma [
9]. A possible explanation for this results might be a selection bias, in particular overdiagnosis of asthma attributable to non specific obesity-related respiratory symptoms among obese children. Indeed, recent meta-analyses have pointed out that some obese patients with “asthma” may have respiratory symptoms caused by obesity without objective physiological criteria for asthma, or an exaggerated symptom perception [
72].
As far as the discrimination between children with asthma and healthy subjects, FeNO measurement provides significantly higher diagnostic accuracy than lung function tests [
58,
73], and has diagnostic value comparable to that of conventional bronchial challenge tests [
57]. Several studies showed an inverse correlation between FeNO levels and bronchial hyperreactivity tests in children [
58,
70‐
76], and only one group did not confirm this finding [
77]. Nevertheless, elevated FeNO levels increase the probability of exercise-induced bronchoconstriction in asthmatic school-age children [
78]. These conflicting results may be explained with the different methods used to measure FeNO as well as with the heterogeneity of the study populations, in particular regarding the presence of atopy and the use of steroids.
Cut-points rather than reference values have been proposed to interpret FeNO levels [
33,
34,
72,
79]. In children, FENO values less than 20 ppb indicate that eosinophilic inflammation is less likely, or that in patients presenting with non specific respiratory symptoms alternative diagnoses to asthma should be considered [
79]. High FeNO concentration (>35 ppb in children) strongly suggests significant airway eosinophilia [
79]. At high expiratory flows, ranging from 200 to 280 mL/s, the negative and positive predictive values for FeNO >25 ppb as predictor of asthma rise to 80% and 100%, respectively [
80]. Nonetheless, intermediate FeNO levels (20-35 ppb in children) indicate that cautious interpretation in the etiology of the airway disorder is required.
An additional, novel, potential and attractive application of FeNO in asthma is the “prediction” of asthma onset. In the absence of symptoms, increased FeNO levels may reflect subclinical airway inflammation that may be predictive of “early asthma”, especially in allergic subjects [
81,
82]. This could be explained by an enhanced Th2 cytokine-driven airway response in allergic individuals that may precede the clinical presentation. Furthermore, in asymptomatic adolescents, increased FeNO may predict the development of rhinitis symptoms within a follow-up period of 4 years [
83], suggesting that FeNO may be a sensitive biomarker of the “allergic march”. These findings have potential clinical and therapeutic implications, also because studies in animal models seem to show a possibility to block induction of Th2 responses, thus preventing the development of future asthma [
84].
FeNO in the follow-up of bronchial asthma
The goal of asthma long-term treatment is to reduce inflammation for controlling symptoms. Treatment options are usually guided by symptoms and lung function. However, these factors do not reflect chronic airway inflammation. This is also shown by contrasting results of studies on the relationship between FeNO levels and both symptoms (including recent symptoms, symptom frequency, symptom scores), symptom control, or rescue ß2-agonist use [
77,
85‐
94] and pulmonary function test results [
25,
85‐
87,
89,
91,
93‐
100].
Inhaled corticosteroids (ICS) are the first choice for asthma maintenance treatment. Interestingly, inhaled or systemic corticosteroid administration results in dose dependent reductions of FeNO levels [
27,
100]. Moreover, in corticosteroid-naïve patients with suspected asthma, the baseline FeNO value may predict an ICS response in terms of improved lung function and reduced airway reactivity [
73]. Therefore, FeNO seems a suitable biomarker for modifying ICS dose in order to obtain better asthma control [
101]. However, in children, daily monitoring of FeNO at home [
102], as well as measurement of FeNO levels every 3 months for 1 year [
103], or 5 times in 6 weeks [
104] do not provide any advantage in improving the symptom score. In adolescents and adults, Szefler
et al showed that the addition of FeNO measurement as an indicator of asthma control resulted in higher doses of ICS and long-acting β2 agonists than did standard guideline-based treatment, and did not determine improvements in asthma symptoms or lung function [
105]. A recent meta-analysis concluded that the number of asthma exacerbations is not significantly reduced in adults and children when ICS was tailored based on FeNO [
106].
These findings may be explained by the fact that day-to-day variations of FeNO are common and do not correlate with changes in symptom score [
107]. Furthermore, some atopic asthmatic patients showed a lack of FeNO responsiveness to ICS [
108,
109], or may have increased FeNO levels despite high dose ICS [
110].
Another issue is whether the change in FeNO values may be a better predictor than absolute levels. FeNO levels quickly decrease in response to ICS and therefore they may be useful to ascertain that ICS is regularly taken. Finally, baseline FeNO seems a worse predictor of asthma improvement than the change in FeNO after 80 days of ICS [
111].
According to the Clinical Practice Guideline of the American Thoracic Society (ATS) [
79] it has been suggested to consider as significant the increase in FeNO greater than 20% for values over 50 ppb, or more than 10 ppb for values lower than 50 ppb from one visit to the next. The second one recommends to use a reduction of at least 20% in FeNO for values over 50 ppb (or more than 10 ppb for values lower than 50 ppb) as the cut point to indicate as significant the response to antiinflammatory therapy.
The results of two recent studies indicate new possible clinical applications of FeNO measurement in pediatric asthma. Pifferi
et al assessed the value of spirometry and FeNO measurements, alone or in combination, in models developed by a machine learning approach for the objective classification of asthma control [
112]. The combined use of spirometry parameters and FeNO levels modeled by a soft computing learning approach applied to spirometry could discriminate the level of asthma control. Van der Valk
et al found that FeNO measured daily by a hand-held device started to increase approximately 10 days before moderate exacerbations occurred, this suggesting that regular FeNO measurements in the home setting could help to detect and even to prevent the loss of asthma control [
113]. Apart from ICS, other established controller therapies, such as leukotriene modifiers or anti-IgE therapy with omalizumab, have been demonstrated to reduce FeNO in children, alone or combined with ICS [
114‐
117].
On the basis of the studies that have provided evidence regarding the applications of NO measurements in clinical practice, ATS recently indicated the rationale for FeNO measurement in asthma, even in the pediatric population [
79], highlighting the following situations:
-
Diagnosis of eosinophilic airway inflammation
-
Support of asthma diagnosis when objective evidence is needed
-
Baseline evaluation and follow-up monitoring of airway inflammation
-
Assessment of potential response or failure to respond to inhaled corticosteroids
-
Evaluation of adherence to antiinflammatory medications
-
Guide for dose changes in antiinflammatory medications