Abstract
Background and objective:
The American Thoracic society (ATS) suggests fractional exhaled nitric oxide (FeNO) as a marker of eosinophilic airway inflammation. FeNO<25ppb indicates that eosinophilia is unlikely, 25-49ppb indicates possible eosinophilia, and FeNO>50 indicates that eosinophilia is likely. The present study aimed to validate these cut-off values in a non-selected clinical population.
Methods:
All patients referred consecutively over a 12-months period for specialist assessment of asthma at the Respiratory Outpatient Clinic at Bispebjerg Hospital in Copenhagen were examined with FeNO and induced sputum, among others. Eosinophilia was defined as >3% eosinophils in sputum. Sensitivity, specificity and predictive values were assessed for the cut-off values of FeNO. A ROC curve was constructed.
Results:
Of 155 subjects with asthma, 107 were able to produce induced sputum (62% females, mean age 31.7(±14.3). 28 asthmatics had >3% sputum eosinophils. The PPV for FeNO <25ppb, 25-50ppb and >50ppb was 15.6%, 33.3% and 61.5% respectively. The NPV was 84.4%, 66.8% and 28.5% in the corresponding FeNO groups. The sensitivity of the 50 ppb cut-off was 28.6%, increasing to 64.3% at the 25ppb cut-off, whereas the specificity was 93.7% at the 50 ppb cut-off, decreasing to 68.4% at the 25 ppb cut-off. The area under the ROC curve was 0.73.
Conclusion:
Increasing values of PPV was found with increasing cut-off values, indicating the higher FeNO, the greater is the likelihood of eosinophilic inflammation. This supports the proposed cut-points. However, the low sensitivity of the 50ppb cut-off indicates that the 25ppb cut-off should be used when a high sensitivity is required clinically.
- © 2014 ERS