Transient sputum eosinophilia may occur over time in non-eosinophilic asthma and this is not prevented by salmeterol.

Background And Objective: Symptomatic, steroid-naïve asthmatic patients may have low sputum eosinophil numbers. The aim of the study was to determine whether low sputum eosinophil numbers persisted over time, during treatment with salmeterol monotherapy.

Methods: Forty steroid-naïve, symptomatic asthmatic patients, with sputum eosinophils <3%, were randomized to receive open-label salmeterol (50 µg twice a day, n = 30) or fluticasone (125 µg twice a day, n = 10) and were then assessed at 1, 3 and 6 months. All patients underwent spirometry, a methacholine challenge test and sputum induction at each visit. Symptom scores and peak expiratory flow were recorded throughout the study. Patients were permitted to withdraw from the study at any time, if they experienced exacerbations or deterioration of symptoms.

Results: The average sputum eosinophil percentage remained normal (≤1.9%) in both groups over the study period. The eosinophil percentages were ≤1.9% in 65 of the 80 samples obtained from salmeterol-treated patients throughout the study period. Eight patients had an asthma exacerbation or deterioration, during which one developed sputum eosinophilia. Twelve patients, 11 of whom were randomized to salmeterol and one to fluticasone, developed transient sputum eosinophilia at least once during the study. This was not associated with asthma exacerbation (except for one patient). Sputum neutrophil percentage did not change in either group.

Conclusions: Low sputum eosinophil numbers persisted over 6 months in a majority of patients with non-eosinophilic asthma who received salmeterol monotherapy. However, transient sputum eosinophilia occurred in 40% indicating that non-eosinophilic asthma may not be a stable phenotype.