Diagnostic accuracy of fractional exhaled nitric oxide in predicting cough variant asthma and eosinophilic bronchitis in adult patients with chronic cough: A systematic review and meta-analysis

Cough is an important reflex we need to remove irritants from the airways, but for many people, a hypersensitive cough reflex can negatively affect quality of life.  A major trigger of chronic cough is airway inflammation from immune cells including type 2 helper T-cells (T_H2), but conventional tests required for diagnosis are technically challenging and often require specialist expertise.  Fortunately, measurement of the fractional exhaled nitric oxide (FENO), a potential marker of T_H2 airway inflammation, has become much more common in allergy and pulmonary practices.  In this month’s issue of JACI, Song and colleagues review the literature on the use of FENO to diagnose Cough-Variant Asthma (CVA) and Eosinophilic Bronchitis (EB), two major causes of T_H2-mediated chronic cough (J Allergy Clin Immunol 2017; 140(3): 701-709).

They looked at thousands of articles from multiple databases in order to answer the question “What is the diagnostic accuracy of FENO for CVA and/or EB in patients with chronic cough?”   After an exhaustive search, they found 15 studies with 2187 adult patients.  The authors then collected and compared the data to determine the accuracy.  Overall, when looking at either CVA or EB, the pooled sensitivity and specificity were 0.73 and 0.89.  For diagnosing CVA, they found moderate diagnostic accuracy, suggesting that the FENO test alone is not sufficient to diagnose CVA.  However, its high specificity means that it may be more useful as a rule-in test than as a rule-out test.  In contrast, results for EB suggested that FENO testing may not be precise enough for prediction. 

This article provides guidance on how to further research on how best to use FENO testing in patients with chronic cough. However, there remain many unanswered questions because of limitations of the review, including the limited number of studies, generalizability of studies which were mostly conducted in Asia, and the imprecision of current diagnostic criteria for CVA.