Exhaled nitric oxide analysis and applications

INTRODUCTION — Nitric oxide (NO) is a gaseous molecule initially considered to have a health-related role only in the context of its formation from the combustion of fossil fuels and its contribution to air pollution. However, this view has been greatly modified since the 1987 discovery that the free radical NO is the previously uncharacterized endothelial-derived relaxing factor. It is now clear that NO plays an important role in most human organ systems.

Within the respiratory system, NO regulates vascular and bronchial tone (promoting dilation of both vessels and airways), facilitates the coordinated beating of ciliated epithelial cells, and acts as an important neurotransmitter for non-adrenergic, non-cholinergic neurons that run in the bronchial wall [1-8]. This molecule can be detected in exhaled gas as the fractional exhaled NO (FE_NO), which varies in health and disease.

The biology of NO, techniques available for measuring this gas in exhaled breath, and potential uses of these measurements in clinical practice will be reviewed here. The therapeutic use of inhaled NO is discussed separately. (See "Inhaled nitric oxide in adults: Biology and indications for use" and "Acute respiratory distress syndrome: Investigational or ineffective therapies in adults" and "Primary lung graft dysfunction", section on 'Treatment'.)

FORMATION OF NO — In biological systems, nitric oxide (NO) is formed by the action of one of the isoforms of the enzyme nitric oxide synthase (NOS). Three such isoforms have been identified and are termed type I or neuronal cell NOS (nNOS), type II or inducible NOS (iNOS), and type III or endothelial cell NOS (eNOS) (table 1) [9,10]. Although these enzymes are distinct proteins encoded by genes on disparate chromosomes, all three catalyze the adduction of the guanidino nitrogen of the amino acid arginine to molecular oxygen, yielding NO and water (figure 1) [11].

Regulation of NOS — While the three isoforms of NOS catalyze the same reaction, regulation of the activity of these isozymes occurs through distinct processes. Both nNOS and eNOS are usually constitutively active and produce low amounts of NO, with output varying with changes in the intracellular calcium concentration. In contrast, iNOS binds calcium so avidly that its function is not influenced by calcium fluxes within the physiologic range. While not constitutively active in most settings, iNOS is constitutively expressed in the airway epithelium of normal and asthmatic subjects [12,13]. iNOS also has the capacity to generate large quantities of NO when transcriptionally upregulated by inflammatory cytokines, such as tumor necrosis factor (TNF)-alpha, interleukin 1 (IL-1)-beta, interferon (IFN)-gamma, IL-4, and IL-13 [9,10,14]. In vitro evidence suggests that this upregulation can be abolished by glucocorticoids in vitro [15] and in vivo [12].

Location of NOS — The cellular location of the various NOS isoforms in the respiratory system has been investigated using immunohistochemical and DNA hybridization techniques. In the human lung, nNOS localizes to airway epithelium and likely to the submucosal nerves [16,17], iNOS has been documented in airway epithelial cells [12], and eNOS exists in the vascular endothelium [18].

The observation that iNOS is inducible by inflammatory cytokines and is inhibited by glucocorticoids provides circumstantial evidence that iNOS activity within airway epithelial cells is the source of the increased NO encountered in asthma.

Source of exhaled NO — The precise anatomic source of exhaled NO was not initially clear, because nitric oxide is formed by multiple mechanisms and in various types of cells. Early studies suggested that the NO in orally exhaled air was likely to represent contamination with NO derived from the sinuses, as the nasal sinuses contain very high levels of NO (>1000 ppb) [19-21]. However, subsequent studies using bronchoscopic techniques to isolate the lower airway from nasal gas have demonstrated that the concentrations of NO in gas derived exclusively from the lower airway are comparable to those obtained in the orally exhaled breath [22-24]. Thus, it is now accepted that the majority of exhaled NO originates from the lower airways. Mathematical models using simultaneous high resolution recordings of FE_NO and flow rate suggest that the majority of NO is excreted from the larger central airways and that central airway NO derives in large part from airway epithelial cells [25,26].

EFFECT OF NO IN THE AIRWAY — The exhaled air of asthmatic individuals contains higher levels of nitric oxide (NO) than healthy non-smoking individuals [27,28], but the exact role of NO in asthma is less clear and likely multifaceted. Some studies suggest that increased NO relaxes bronchial smooth muscle leading to bronchodilatation and inhibits pro-inflammatory signaling events [29,30], while others suggest it contributes to airway inflammation and injury through formation of toxic reactive nitrogen species (RNS) [31].

NO is highly reactive and once produced it can rapidly lead to the formation of several nitric oxide related end products with differing downstream effects [23,32]. Increased production of NO or RNS may increase nitration of tyrosine residues in proteins or decrease nitrosylation of thiol residues on proteins, which in turn decreases the bronchodilating S-nitrosothiols (SNO) [23,32]. A model of NO dynamics derived from allergen challenge studies in humans demonstrates that NO may have harmful effects through formation of peroxynitrite (one of the reactive nitrogen species), but may also have an antioxidant role by consuming reactive oxygen species during the immediate asthmatic response [23]. Ultimately, the functional role of NO, as with any molecule, depends on both its concentration and association with other biomolecules and proteins [24].

MEASUREMENT OF EXHALED NO — Chemiluminescence analysis allows determination of nitric oxide (NO) concentration in the gas phase by the reaction of NO in the sample with ozone to produce nitrogen dioxide in an excited state [33]. As nitrogen dioxide moves to a relaxed state, light is emitted in a stoichiometric relationship with the amount of NO present in the gas sample. Most initial studies reported in the literature used ozone-chemiluminescence to measure exhaled NO concentrations, and this technology was used in the first US Food and Drug Administration (FDA) approved devices [34]. Subsequently, devices based on other technologies, including hand held devices utilizing electrochemical methods, have also been approved [35,36] and are more commonly used in clinical settings. Chemiluminescence methods are still the gold standard and are more commonly used in research settings.

Standardized methods for measuring exhaled NO were jointly developed by the American Thoracic Society (ATS) and the European Respiratory society (ERS) in 1999 and revised in 2005 [37,38]. The guidelines recommend the use of the term FE_NO (the fractional exhaled NO concentration) to describe the level of NO in exhaled breath. FE_NO is expressed in parts per billion which is equivalent to nanoliters per liter. Further technical standards have been published by the ERS [36].

Exhaled NO can be measured online, while the subject exhales directly into the analyzer, or offline, by collection of exhaled air in an NO-impervious container and later measurement. Important technical factors include exclusion of nasal NO, use of proper procedures for online or offline measurement, adherence to optimal expiratory flow rates, and monitoring ambient NO levels [36-38]. Key features of the technique are reviewed here.

●Online measurements – For online measurements obtained while the subject exhales at a constant flow directly into the analyzer, the mean FE_NO value over a three-second plateau is recorded, not the early peak. In order to achieve an appropriate plateau, the duration of exhalation must be sufficient (at least four seconds for children <12 years and more than six seconds for children ≥12 years and adults) [37].

●Offline measurements – Specifications include using an apparatus to collect exhaled air without loss of NO; depleting the inspiratory air of NO prior to inhalation; and maintaining an expiratory flow rate of 0.350 L/second (range 0.315 to 0.385 L/second) throughout exhalation. Sometimes the first 150 to 250 mL of exhaled air is discarded but the optimal volume is not known and this step requires special valves. As with on-line measurements, the actual flow rate should be recorded with the test results.

●Exclusion of nasal NO – Exclusion of nasal NO is important as these levels are high relative to lower respiratory tract levels [37]. In general, a nose clip is not used as this may allow nasal NO to accumulate and leak into the posterior pharynx. On the other hand, if the patient is not able to avoid nasal inspiration, a nose clip may be used. After inserting the mouthpiece, the patient inhales through the mouth to total lung capacity and immediately exhales completely to avoid breath holding. In addition, having the subject maintain a positive mouthpiece pressure of about 10 cm H₂O (but not above 20 cm H₂O) during exhalation helps to close the velopharyngeal opening and further minimize nasal NO leakage.

●Expiratory flow rate – Exhaled NO levels are dependent on the expiratory flow rate [39], so a constant exhalation rate of 0.050 L/second is preferred for online measurements and a rate of 0.35 L/second for offline measurements (as noted above) [37]. Acceptable methods to achieve a constant flow rate include a visual display of the target flow rate while the patient exhales against a fixed resistance, use of a dynamic resistor, and operator controlled flow rates [37]. In any case, the flow rate should be recorded with the test results.

The ambient NO level should be recorded at the time of testing, as this level can also influence the test results.

CLINICAL USE OF FENO IN ASTHMA — A large body of research has examined the potential role of FE_NO as a noninvasive biomarker in asthma. It has been hoped that using FE_NO, rather than just symptoms or airflow limitation, would help in asthma diagnosis, characterization of asthma in individual patients (eg, eosinophilic or noneosinophilic), guidance regarding selection and adjustment of asthma therapy, and advancement of our understanding of the effects of medications on airway inflammation [40-51].

Interpretation of exhaled NO in asthma — Initial studies of FE_NO in asthma used reference ranges to define normal and abnormal values. However, it is difficult to determine appropriate reference ranges as healthy individuals sometimes fall outside the “normal” range and individuals with asthma occasionally fall within the range [52-56].

●Suggested cut-points for FE_NO interpretation – Rather than develop reference ranges based on the varied combinations of individual characteristics, a simpler method of using cut-points has been proposed in the American Thoracic Society Clinical Practice Guideline for interpretation of FE_NO [40,57], as follows:

•A FE_NO less than 25 ppb in adults and less than 20 ppb in children younger than12 years of age implies the absence of eosinophilic airway inflammation.

•A FE_NO greater than 50 ppb in adults or greater than 35 ppb in children suggests eosinophilic airway inflammation.

•Values of FE_NO between 25 and 50 ppb in adults (20 to 35 ppb in children) should be interpreted cautiously with reference to the clinical situation.

•A rising FE_NO with a greater than 20 percent change and more than 25 ppb (20 ppb in children) from a previously stable level suggests increasing eosinophilic airway inflammation, but there are wide inter-individual differences.

•A decrease in FE_NO greater than 20 percent for values over 50 ppb or more than 10 ppb for values less than 50 ppb may be clinically important.

Once the clinical setting is taken into consideration, certain patterns begin to emerge. FE_{NO l}evels above 50 ppb correlate with glucocorticoid responsiveness [58], while levels below 25 ppb may predict a lack of responsiveness to addition of inhaled glucocorticoids [59]. In a glucocorticoid naive individual with symptoms, FE_{NO l}evels above 20 to 25 ppb may support the presence of asthma, while lower levels are less likely to be associated with eosinophilic airway inflammation or respond to addition of inhaled glucocorticoids [60-62].

●Other factors that affect FE_NO – Several factors other than asthmatic inflammation, such as age, sex, height, atopy, and cigarette smoking, affect FE_NO values to a variable extent (table 2) [53,63,64]. The two main factors that need to be taken into consideration when interpreting FE_NO measurements are smoking (associated with lower FE_NO levels) and atopy (associated with higher FE_NO levels). There are no specific guidelines, however, regarding the exact adjustments to use.

Diagnosis and characterization of asthma — While FE_NO levels correlate with the presence of asthma and with eosinophilic airway inflammation and increase with exposure to asthma triggers, the exact role of FE_NO measurement in the diagnosis of asthma has not been defined [65,66]. The Global Initiative for Asthma advises against the use of FE_NO for the diagnosis of asthma, as it may not be elevated in noneosinophilic asthma and may be elevated in diseases other than asthma, such as eosinophilic bronchitis or allergic rhinitis [66]. The following observations have been reported:

●Patients with asthma have higher concentrations of nitric oxide (NO) in their exhaled air than do nonasthmatic subjects [67,68]. In addition, exhaled NO levels rise in association with acute airway inflammation, sputum eosinophilia, viral upper respiratory infections, and other clinical parameters associated with deteriorating asthma control [68-73].

●Inhalation of inflammatory stimuli, including allergens, plicatic acid (the etiologic agent of Western red cedar asthma), isocyanates, and hyperventilation of cold-dry air, increases FE_NO [74-78].

●Levels of FE_NO correlate with bronchial reactivity, the number of eosinophils in induced sputum in patients with stable asthma not using glucocorticoids, and the percentage of eosinophils in bronchoalveolar lavage [79,80]. Intermediate or high FE_NO values are associated with asthma, wheeze, and asthma exacerbations [81].

Ongoing research should provide guidance on whether characterization of asthma phenotypes by assessment of biomarkers (eg, FE_NO, sputum or blood eosinophilia) will improve asthma management.

As a guide to therapy — National and international guidelines suggest using FE_NO levels, in addition to other assessments (eg, clinical care, questionnaires) to guide initiation and adjustment of asthma controller therapy [40,57,66]. However, usual clinical care without FE_NO measurement is also reasonable. Of note, guidelines caution that FE_NO should not be used to decide against inhaled glucocorticoid therapy in patients with a diagnosis of asthma or suspected diagnosis of asthma [66].

●Predicting response to inhaled glucocorticoids – FE_NO levels generally predict which patients will respond to inhaled glucocorticoid therapy [42,48,82,83], but systematic reviews have found mixed evidence regarding the use of FE_NO to guide asthma therapy [84-88]. As an example, a systematic review (16 studies) compared trials of asthma therapy guided by FE_NO with those in which guidance was based on standard measures (symptoms with or without spirometry/peak flow) [87]. The analysis found that asthma management based on FE_NO was associated with reduced odds ratio for asthma exacerbations in adults 0.60 (95% CI 0.43-0.84) and children 0.58 (95% CI 0.45-0.75), but did not improve measures of asthma control or lung function. The review noted heterogeneity between studies in the definition of asthma exacerbations, algorithms for adjustment of medication, and cut-off values for increasing or decreasing therapy. An earlier meta-analysis noted that the rate of exacerbations was significantly reduced in a FE_NO-based asthma management algorithm compared with a clinically-based algorithm [14].

For centers in which FE_NO levels are incorporated into asthma management, the interpretation of FE_NO levels should consider the context in which the measurement is being obtained (eg, presence/absence of symptoms, use of glucocorticoids, change from prior values, cigarette smoking, viral infection), as described in the table (table 3) [40,57] and above. (See 'Interpretation of exhaled NO in asthma' above.)

Based on the ATS guidelines, some common clinical scenarios where FE_NO is helpful in the clinic include the following [40,57]:

•Low FE_NO (<25 ppb) in adult asthmatics with persistent symptoms suggests other etiologies for these symptoms and a lower likelihood of benefit from adding or increasing inhaled glucocorticoids.

•High FE_NO (>50 ppb) in adult asthmatics even with atypical symptoms suggests glucocorticoid responsiveness.

•High FE_NO (>50 ppb) can help identify poor adherence or uncontrolled inflammation in asthma patients with otherwise seemingly “controlled” asthma.

•In children, the cut-points for FE_NO are slightly different: a low FE_NO value is <20 ppb, and an elevated FE_NO value that would predict responsiveness to inhaled glucocorticoids is >35 ppb.

Although we have included established cutoff values for ease of clinical use, different cutoffs have been found useful in different populations, with the usual diagnostic tradeoffs between sensitivity and specificity [67]. Overzealous use of specific cutoff values without additional context may lead to suboptimal clinical decision-making [57].

●Predicting response to biologic agents – FE_NO has been used in a number of clinical trials of biologic therapies for asthma to evaluate the presence of Type 2 inflammation [89-91]. In one trial of omalizumab for severe asthma, a FE_NO ≥19.5 ppb was associated with a greater reduction in exacerbation frequency [89], although FE_NO levels were not predictive of a response in a separate trial [91]. In a trial of dupilumab versus placebo, baseline FE_NO levels ≥50 ppb or ≥25 to <50 ppb predicted a better response to dupilumab in reducing asthma exacerbations than that achieved in patients with a FE_NO <25 ppb [92]. (See "Treatment of severe asthma in adolescents and adults", section on 'Persistently uncontrolled asthma'.)

Use in clinical research — Exhaled nitric oxide has an important role in clinical research and will likely help in expanding our understanding of asthma, such as the factors responsible for asthma exacerbations [81] and the sites and mechanisms of action of medications for asthma [93]. In addition, it is used to characterize the asthma phenotype of subjects in clinical trials [50]. As an example of how FE_NO levels might improve our understanding of asthma therapy, FE_NO levels, which are thought to reflect airway inflammation mediated by interleukin (IL)-4 and IL-13, appear to be a good marker for response to inhaled glucocorticoids, lebrikizumab (anti-IL-13), and omalizumab (anti-IgE), consistent with IL-4 and IL-13 mediating the effects of these agents [89,93,94]. On the other hand, FE_NO levels do not correlate with responses to mepolizumab (anti-IL-5), which is thought to affect systemic (rather than local airway) eosinophil production and mobilization [95]. (See "Investigational agents for asthma", section on 'Biologic agents'.)

USE IN OTHER RESPIRATORY DISEASES — Diseases other than asthma are associated with altered levels of exhaled nitric oxide (NO) [96]. The role of measuring the fractional exhaled NO (FE_NO) in these diseases is not established, although with further study FE_NO may be useful as a screening test in chronic cough due to nonasthmatic eosinophilic bronchitis.

Bronchiectasis and cystic fibrosis — Children with cystic fibrosis (CF) have lower FE_NO levels than appropriately matched controls [97,98]. This may be due to elevated arginase activity; arginase competes for L-arginine, the substrate of nitric oxide synthesis [99]. In contrast, one study found that patients with non-CF bronchiectasis had elevated levels of FE_NO, and these levels were correlated with the degree of abnormality apparent on chest CT [100]. Whether this discrepancy relates to disease activity, especially coexistent infection, is not yet clear.

Interstitial lung disease and sarcoidosis — A few studies have assessed exhaled NO values in interstitial lung disease related to scleroderma and sarcoidosis. In general, the differences in values are within the range of variability of the test. In a study of patients with scleroderma, a higher exhaled NO was noted among patients with interstitial lung disease (ILD) compared with those without ILD [101], while the opposite was found in another study [102].

Variable results have also been reported in sarcoidosis. In a study of 52 patients with sarcoidosis, the mean FE_NO value was 6.8 ppb, which is substantially less than the cut-point of 25 ppb used to denote asthma inflammation [103]. A separate study found a mean FE_NO of 6.7 ppb among patients with sarcoidosis compared with 5.17 ppb in healthy controls [104]. However, the FE_NO values did not correlate with disease activity [104].

Chronic obstructive pulmonary disease — FE_NO levels are minimally elevated in stable COPD, but may increase with more severe disease and during exacerbations [105-108]. Current smokers have approximately 70 percent lower levels of FE_NO [63]. In patients with COPD, FE_NO levels may be useful in establishing the presence of reversible airflow obstruction and determining glucocorticoid responsiveness, although this has not been assessed in large randomized trials [109].

Cough variant asthma — FE_NO has moderate diagnostic accuracy in predicting a diagnosis of cough variant asthma (CVA) in patients with chronic cough. In a systematic review of 13 studies (2019 patients), the optimal cut-off range for FE_NO was 30 to 40 ppb (although lower values were noted in two studies), and the summary area under the curve was 0.87 (95% CI, 0.83-0.89) [110]. Specificity was higher and more consistent than sensitivity (0.85 [95% CI 0.81-0.88] and 0.72 [95% CI 0.61-0.81], respectively), although heterogeneity among studies was high. The interstudy variability was attributed, at least in part, to differences in the diagnostic criteria for CVA, the type of device used to measure FE_NO, and variability in the requirement for a normal or near normal chest radiograph prior to study entry.

The same review assessed the ability of FE_NO to identify patients with CVA OR nonasthmatic eosinophilic bronchitis. Among five studies (529 patients), the sensitivity and specificity were 0.72 (95% CI 0.61-0.81) and 0.85 (95% CI 0.81-0.88) [110].

Nonasthmatic eosinophilic bronchitis — In patients with nonasthmatic eosinophilic bronchitis (NAEB), sputum eosinophils and FE_NO are increased in a range similar to patients with asthma [111]. In a systematic review of four studies (390 patients) in patients with chronic cough due to NAEB, optimal FE_NO cut-off levels were 22.5 to 31.7 ppb [110]. The estimated sensitivity was 0.72 (95% CI 0.62-0.80) and estimated specificity was 0.83 (95% CI 0.73-0.90). Thus, FE_NO is more useful to confirm NAEB, than to exclude it. (See "Causes and epidemiology of subacute and chronic cough in adults", section on 'Nonasthmatic eosinophilic bronchitis'.)

Upper respiratory infections — In one study of patients without underlying pulmonary disease, viral upper respiratory infections resulted in increased FE_NO [112]. These levels were dramatically reduced when rechecked three weeks later.

Pulmonary hypertension — NO is well-recognized as a pathophysiologic mediator in pulmonary arterial hypertension (PAH) [113,114]. In addition to vasodilation, NO regulates endothelial cell proliferation and angiogenesis, and maintains overall vascular health [115]. Interestingly, patients with PAH have low FE_NO values [115]. Individuals with PAH also have a lower than normal concentration of NO reaction products in the bronchoalveolar lavage fluid, which is inversely related to the degree of pulmonary hypertension [116]. Replacement of NO seems to work well in treating this disease, although PAH is a far more complex issue than the simple lack of a vasodilator [114,117]. (See "Inhaled nitric oxide in adults: Biology and indications for use".)

Therapies that target other steps in the NO pathway have revolutionized the treatment of PAH, including the widely-used phosphodiesterase type 5 (PDE5) inhibitors, which prevent the breakdown of the NO effector molecule 3’, 5’-cyclic guanosine monophosphate (cGMP), thus prolonging NO effects on tissues [117,118]. FE_NO seems to also have a prognostic significance, with improved survival in patients who have a rise in FE_NO level with therapy (calcium channel blockers, epoprostenol, treprostinil) compared to those who do not [119]. Thus, the low FE_NO levels in patients with PAH and the improvement with effective therapies suggest it may be a promising biomarker for this disease. More studies are needed to determine whether serial monitoring of FE_NO levels are clinically useful in evaluating the response (or lack of it) to medical therapy in PAH [47].

Primary ciliary dysfunction — Nasal NO is very low or absent in patients with primary ciliary dysfunction (PCD). The use of nasal NO to screen for PCD in patients with a clinical suspicion of PCD is discussed separately. (See "Primary ciliary dyskinesia (immotile-cilia syndrome)", section on 'Nasal nitric oxide'.)

Other conditions — In addition to pulmonary hypertension, other conditions associated with low FE_NO levels include hypothermia, and bronchopulmonary dysplasia, as well as the use of alcohol, tobacco, caffeine, and other drugs (table 3) [18,63,120-125].

SOCIETY GUIDELINE LINKS — Links to society and government-sponsored guidelines from selected countries and regions around the world are provided separately. (See "Society guideline links: Pulmonary function testing".)

SUMMARY AND RECOMMENDATIONS

●Role of nitric oxide in lungs – Within the respiratory system, nitric oxide (NO) regulates vascular and bronchial tone (promoting dilation of both vessels and airways), facilitates the coordinated beating of ciliated epithelial cells, and acts as an important neurotransmitter for non-adrenergic, non-cholinergic neurons. (See 'Formation of NO' above and 'Effect of NO in the airway' above.)

●Measurement of exhaled NO – NO can be detected in exhaled gas by chemiluminescence and other technologies. The level of exhaled NO is expressed as the fractional exhaled NO (FE_NO) in parts per billion (ppb), which is equivalent to nanoliters per liter. Accurate measurement depends upon factors such as the rate of exhalation (generally 0.05 L/sec for online measurement, 0.35 L/second for offline measurement), ambient value of NO, and the accuracy of the device. (See 'Measurement of exhaled NO' above.)

Measurement of FE_NO is noninvasive, can be performed repeatedly, and can be used in children and patients with severe airflow obstruction where other techniques are difficult or not possible to perform. FE_NO may also be more sensitive in detecting eosinophilic airway inflammation than other noninvasive tests. (See 'Clinical use of FENO in asthma' above.)

●FE_NO as biomarker – Patients with asthma generally have high levels of FE_NO in their exhaled breath that returns to normal after treatment with glucocorticoids. It is believed that eosinophilic inflammation in the airways stimulates airway epithelial cells to produce NO, thus making FE_NO a potentially useful marker to monitor airway inflammation in asthma. (See 'Formation of NO' above and 'Clinical use of FENO in asthma' above.)

A number of factors other than asthma and eosinophilic inflammation affect FE_NO levels, such as atopy, age, sex, height, smoking status, and medications. (See 'Interpretation of exhaled NO in asthma' above.)

●Interpretation of FE_NO – The use of the following cut-points, rather than normative values, is suggested for the interpretation of FE_NO levels in asthma. (See 'Interpretation of exhaled NO in asthma' above.)

•A FE_NO less than 25 ppb in adults and less than 20 ppb in children younger than 12 years of age implies noneosinophilic airway inflammation or the absence of airway inflammation.

•A FE_NO greater than 50 ppb in adults or greater than 35 ppb in children suggests eosinophilic airway inflammation.

•Values of FE_NO between 25 and 50 ppb in adults (20 to 35 ppb in children) should be interpreted cautiously with reference to the clinical situation (eg, symptomatic, on or off therapy, current smoking).

•A rising FE_NO with a greater than 20 percent change and more than 25 ppb (20 ppb in children) from a previously stable level suggests increasing eosinophilic airway inflammation, but there are wide inter-individual differences.

•A decrease in FE_NO greater than 20 percent for values over 50 ppb or more than 10 ppb for values less than 50 ppb may be clinically important.

●FE_NO in other respiratory diseases – Several other diseases are associated with altered levels of exhaled NO: low levels of FE_NO have been noted in cystic fibrosis, current smoking, pulmonary hypertension, hypothermia, primary ciliary dyskinesia, and bronchopulmonary dysplasia. Elevated FE_NO has been noted in atopy, nonasthmatic eosinophilic bronchitis, COPD exacerbations, noncystic fibrosis bronchiectasis, and viral upper respiratory infections. (See 'Use in other respiratory diseases' above.)

ACKNOWLEDGMENTS — The UpToDate editorial staff acknowledges Aaron Deykin, MD, and Anthony Massaro, MD who contributed to earlier versions of this topic review.