Diagnostic accuracy of sputum outcomes in chronic stable asthma

Asthma with non-remitting airflow obstruction may not always be differentiated from COPD with airway hyperreactivity. Many attempts have been made to find useful markers for the distinction between these two disorders.

In order to help the finding of a useful marker for the diagnosis of asthma in the population of patients with airway obstruction we analysed the diagnostic accuracy of sputum eosinophils and sputum ECP in 91 patients with asthma, 15 patients with chronic bronchitis, 32 patients with chronic obstructive pulmonary disease (COPD) and 20 controls subjects, by performing ROC analysis.

Sputum eosinophils were above the normal range of our laboratory (0-3.7%) in 48 asthma patients and in six COPD patients, while sputum ECP (normal range < 85 microg/L) was high in 65 asthma patients, in 24 COPD patients and in nine chronic bronchitis patients. The ROC analysis revealed that sputum eosinophils count (AUC = 0.82) was more accurate than both sputum ECP levels (AUC = 0.56) (P < 0.0001) and beta2-reversibility (AUC = 0.53) (P = 0.0001) in differentiating asthmatic from non-asthmatic subjects (COPD, chronic bronchitis patients and normal subjects). The diagnostic accuracy of ECP was similar to that of bronchial reversibility (P = 0.76). When ROC analysis was performed by including only patients with airway obstruction (36 asthmatics with airway obstruction and COPD patients), both eosinophil count (AUC = 0.77) and beta2-reversibility (AUC = 0.66) were more accurate than ECP measurement (AUC = 0.39) in discriminating asthmatics from COPD patients (P < 0.00001 and P = 0.04, respectively).

Sputum eosinophils seem to be valid markers for detecting asthma in a population of patients with airway obstruction. Moreover, the higher diagnostic accuracy of eosinophils in the sputum compared to sputum ECP and bronchial reversibility reinforces the role of cytological analysis of sputum in the diagnosis of chronic stable bronchial asthma.