American Thoracic Society guidelines recommend the use of exhaled NO in management of asthmatics, especially in asthma with eosinophilic inflammation and in predicting response to corticosteroids [9]. However, the role of FENO in asthma at present is limited to diagnosis of eosinophilic airway inflammation, monitoring of airway inflammation, and likelihood of steroid responsiveness. FENO has been studied in acute exacerbation, but has been limited to reproducibility of measurements, emergency department disposition, and response to corticosteroids [10-12]. Most of these studies have been done in Caucasian children and studies from the Indian subcontinent are lacking.

There is a need to evaluate the utility of FENO in acute exacerbations as it may reflect the extent of airway inflammation. We conducted this study to determine the utility of FENO measurements in acute exacerbations.

This study was conducted in the Pediatric Chest Clinic of the All India Institute of Medical Sciences (AIIMS), New Delhi, which is a tertiary care teaching hospital. We are following a cohort of pediatric asthma patients (up to 18 yrs age) since August 2009. Study protocol was approved by Ethics committees of AIIMS, New Delhi and CSIR-IGIB, New Delhi..

Written informed consent was taken from the parents/guardian. The diagnosis and treatment of asthma was based on the Global Initiative for Asthma (GINA) guidelines [13] by a pediatric pulmonologist. The patients were followed up every 3 months, symptom diary was maintained, and control was assessed as per GINA guidelines. On each visit, lung function measurements (spirometry and impulse oscillometry) were performed and FENO levels were obtained. Blood was collected at enrollment for peripheral eosinophil counts. Therapy was modified, if required, on the basis of clinical features and spirometry.

In case of appearance of symptoms of acute exacerbation, they contacted the research team (led by a pediatrician) and a visit was scheduled. Children who were not able to perform spirometry or FENO were not included in the analysis.

The patients with acute exacerbation were evaluated and managed in the Pediatrics department by the study team. Acute exacerbation was defined as recent increase in asthma symptoms requiring hospital visit and treatment with salbutamol and/or steroids [14]. Initial evaluation included history and physical examination (including pulse oximetry). Severity of acute asthma was assessed using pulmonary score [15]. Child was initially evaluated in clinic, and subsequently managed in emergency room in case of moderate and severe acute exacerbation. Child underwent FENO measurement followed by spirometry. Patients were managed according to acute asthma guidelines [14]. Apart from asthma management, compliance and technique was checked and re-emphasized at each visit.

FENO measurement: FENO measurement was done using NIOX MINO (Aerocrine AB, Solna, Sweden) in accordance with ATS guidelines [9]. FENO was measured at the time of enrolment, on each follow up visit every 3 months and on each breakthrough visit that was assessed to be acute exacerbation. In mild to moderate exacerbations, FENO measurement was done in clinic before bronchodilator therapy. In case of severe exacerbation, child was managed in emergency room and FENO was done once child was stable. All measurements were performed on the same equipment and by similarly trained team-members, ensuring minimal technical variability.

Spirometry was performed after FENO measurement was done. Spirometry was done using portable spirometer (Superspiro MK2, Micro Medical Ltd, UK) in all children in accordance with ATS standards [16].

Pulmonary score is a validated measure of asthma severity for children with acute asthma exacerbation [15]. Each parameter is rated on a 0-3 scale, with a maximum total score of 9. Mild, moderate, and severe acute exacerbations were defined as pulmonary score of 0-3, 4-6, and 7-9, respectively.

Skin prick testing (SPT) was done using 12 aeroallergens during the follow up of the cohort. Saline was taken as negative control and histamine was used for positive control. Patients were not on antihistaminics for at least 48 hours preceding the test. The twelve allergens tested were rice grain dust, wheat threshing dust, housefly, female cockroach, dog dander, house dust mite (Dermatophagoides farinae), Curvularia lunata, Aspergillus tamari, Alternaria tenius, Prosopis juliflora, Cynodon dactylon, and Holoptelea integrifolia. Allergens were obtained from All Cure Pharma Pvt Ltd, Bahadurgarh, Haryana. Test was considered positive if wheal in any of the allergens was 3 mm or more than the negative control. Child was considered atopic if he demonstrated positive result to one or more allergen, and non-atopic if he had a negative SPT.

Statistical analysis: Data were entered using Microsoft Access. Statistical analysis was performed using Stata 9.0 statistical software (Stata Corp., College Station, TX, USA). We identified each patient’s minimum FENO value during follow up (personal best). We determined the difference in FENO measured during acute exacerbation as compared to the follow up values and also as compared to the personal best for each patient. The distribution of FENO values was skewed to the left, so it was reported as median (IQR). The differences between FENO at various time points were analyzed using Wilcoxon signed-rank test. The difference in FENO across more than two groups was analyzed using Kruskal Wallis test. We constructed receiver operating characteristics (ROC) curves to assess the ability of FENO to predict acute exacerbation and we hypothesized FENO of 20 ppb as a cut off for acute exacerbation.

We divided patients on the basis of baseline FENO into three sub-groups (low, intermediate, and high) i.e., <20 ppb, 20 to 35 ppb, and >35 ppb using ATS guidelines [9]. We then assessed the exacerbation rates and also various FENO values in these three categories. Correlation between FENO and pulmonary score was done using Spearman correlation test. A P value of <0.05 was considered significant.

We evaluated the utility of FENO in children with acute exacerbation of asthma in this cohort study. This is the first study to have compared FENO levels during follow up to FENO levels during acute exacerbation. The personal best FENO was significantly lower than FENO during acute exacerbation.

Poor asthma control can lead to asthma exacerbation, so we expected increase in FENO during exacerbation from personal best levels. While in around 68% children, the FENO during acute exacerbation was at least 5 ppb higher than the personal best FENO, the difference was either less than 5 ppb or even negative (i.e., best personal FENO less than acute exacerbation FENO) in the remaining one-third. Thus, FENO’s utility in predicting the presence of acute exacerbation is limited.

FENO is influenced by a number of factors. Atopy [17], viral infections [18], allergen exposure [19, 20] and concomitant rhinitis [21] are known to increase FENO. FENO is known to decrease with smoke exposure [22], post-spirometry [23], and corticosteroid treatment [4,5]. In our study, almost all children were on anti-inflammatory treatment. There is some evidence that height correlates positively with FENO [24]. Smoke exposure and air pollution are particularly important to our patients in Delhi, India, where air quality is notoriously poor. It is likely that at one point in time there are multiple factors which play role in precipitating an acute exacerbation, and these factors have a complex effect on airway NO metabolism. Thus, in a pragmatic clinical setting, analogous to that seen by respiratory physicians in developing nations like India, FENO has probably only limited clinical utility such as predicting response to inhaled steroids.

The ROC curve for FENO as a marker for acute exacerbation had an AUC of 0.59. Cut of off 20 ppb had poor sensitivity and specificity for diagnosis of acute exacerbation. An FENO value of 16 ppb had a sensitivity of 56.6% and a specificity of 58.3%, which was the best combination of sensitivity and specificity. No study has so far evaluated FENO cut offs for acute exacerbation. It is evident from our study that FENO cut off values have poor sensitivity and specificity in predicting acute exacerbations.

Atopy is associated with high FENO, airway hyperresponsiveness, and deterioration with response to allergen exposure. We were interested in knowing whether atopic children demonstrate a higher increase in FENO during acute exacerbation than non-atopic children, which was not evident in the results (P=0.10). The possible explanation being that most children were on anti-inflammatory medications during acute exacerbations, and the complex interplay of NO metabolism affected by various factors.

We investigated the association between severity of acute exacerbation and FENO concentrations. We used pulmonary score [14] to assess the severity of acute exacerbation. Majority of the acute exacerbations were of mild-moderate severity. Only one severe exacerbation was seen in our cohort. The reason behind this could be rigorous follow up, telephonic interactions, and home visits to ensure adequate adherence to the controller treatment regimen which may have prevented severe exacerbations.

We did not find any association between FENO and severity of acute exacerbation. Kwok et al measured FENO during acute exacerbation in children aged 2-18 years of age, and reported no difference in their median FENO concentration, regardless of their severity of acute asthma [10].

It is logical to think that the more severe the exacerbation, the more inflammation should be evident. Airway caliber also affects FENO concentrations, and decrease in airway caliber has been shown to decrease FENO [25, 26]. This is one of the reasons which possibly could negate the increase in FENO in moderate-severe exacerbations, apart from the other confounders like, age, height, viral infections, allergen exposure etc.

Not much work has been done on evaluation of FENO in the setting of acute exacerbation. Kwok et al found measurement of FENO difficult for a large proportion of children with acute asthma. In their study FENO measurement could be obtained in only 68% children [10]. Gill, et al. found poor reproducibility of FENO measurements obtained in emergency department patients with acute asthma exacerbations [11]. However, Baptist, et al. showed acceptable intraclass correlation coefficient and coefficient of variation values (0.98 and 9.42%, respectively) for reproducibility [12]. With the advent of portable hand held NO analyzer, it is possible to measure FENO not only in clinic setting but also in the emergency department, but the utility of FENO in acute exacerbation setting is probably limited.

Asthma is now increasingly being recognized to be a heterogeneous disease constituting of several inflammatory phenotypes. FENO is a surrogate marker of eosinophilic inflammation in asthmatic children. Other phenotypes like neutrophilic and pauci-granulocytic are unlikely have high FENO. Our study enrolled heterogeneous children with asthma and assessed the utility of FENO in acute exacerbations. In our study, it is possible that children with non-eosinophilic inflammation had near normal FENO even during an inflammation. The heterogeneity of patients (steroid naïve, intermittent, mild persistent, moderate persistent, and severe persistent) could be one of the reasons of inability of FENO to predict exacerbation. We did not evaluate viral infection as a cause for acute exacerbation as it is known to increase FENO. The personal best FENO value was taken as minimum of the follow up values. The observed personal best FENO value may not have been true for children with fewer visits.

To conclude, FENO levels during acute exacerbation increase from their minimum follow up levels. However, no appropriate cut off could be identified which could help diagnosing acute exacerbation. The FENO values did not correlate with the severity of acute exacerbation. It appears that FENO measurement may add little to the diagnosis of acute exacerbation of asthma in children. Diagnosis of acute exacerbation should be based on history, clinical examination, and spirometry and rising FENO (from personal best to acute exacerbation) may be taken as supportive evidence.