Lung function and sputum characteristics of patients with severe asthma during an induced exacerbation by double-blind steroid withdrawal

Some patients with severe asthma are difficult to control and suffer from frequent exacerbations, whereas others remain stable with anti-inflammatory therapy. To investigate mechanisms of exacerbations, we compared 13 patients 20 to 51 yr of age (11 female, two male) with difficult-to-control asthma (two or more exacerbations during the previous year) and 15 patients 20 to 47 yr of age (13 female, two male) with severe but stable asthma (no exacerbations) after matching for sex, age, atopy, lung function, airway responsiveness, and medication. Exacerbations were induced by double-blind, controlled tapering of inhaled corticosteroids (fluticasone propionate) at weekly intervals. FEV1, airway responsiveness for methacholine (PC20MCh) and hypertonic saline (HYP slope), eosinophils and soluble markers (ECP, albumin, IL-6, IL-8) in induced sputum were assessed at baseline and during exacerbation (peak flow <60% of personal best), or after 5 wk if no exacerbation occurred. Steroid tapering caused a decrease (mean +/- SEM) in FEV1 (12.1 +/- 3.1% pred; p = 0.045), PC20MCh (2.1 +/- 0.4 doubling dose; p = 0.004) and HYP slope (1.7 +/- 0.3 doubling dose; p = 0.001), and an increase in sputum eosinophils (10 +/- 3%; p = 0.008) and soluble markers for the two groups combined, without significant differences between the groups. Patients with difficult-to-control asthma had more exacerbations than did the stable asthmatics during both steroid tapering (7 versus 2; p = 0.022) and corticosteroid treatment (6 versus 0; p = 0.003). Exacerbations during steroid treatment in the patients with difficult-to-control asthma were associated with a decrease in FEV1 and PC20MCh, but not in HYP slope or increase in sputum eosinophils. We conclude that tapering of inhaled corticosteroids induces a rapid, reversible flare-up of eosinophilic airway inflammation. Patients with difficult-to-control asthma may develop exacerbations despite treatment with inhaled corticosteroids, which appear to have an eosinophil-independent mechanism. This implies that assessment of the nature of exacerbations may contribute to improved treatment for these patients