Predictors of Loss of Asthma Control Induced by Corticosteroid Withdrawal

Abstract

Asthma guidelines recommend reducing the dose of inhaled corticosteroids after establishing control. To identify predictors of loss of control and the kinetics of symptoms, and inflammatory and physiological measurements when inhaled corticosteroids are reduced in patients with stable asthma. In a single-blind study, the daily dose of inhaled corticosteroid was reduced by one-half at intervals of 20+/-2 days in 17 adults with controlled asthma until loss of asthma control occurred or until the corticosteroid was replaced with placebo for 20 days. The patients recorded symptoms and peak expiratory flow each day, and forced expiratory volume in 1 s (FEV1), the provocative concentration of methacholine causing a 20% fall in FEV1 (PC20), exhaled nitric oxide, and eosinophils in sputum and blood were measured every 10 days. A loss of asthma control was defined as a worsening of the symptoms score of at least 20%, and either a decrease in FEV1 of at least 15% or a decrease in PC20 of at least fourfold. Two patients had a respiratory infection and were withdrawn from the study. In eight patients, asthma became uncontrolled after a mean of 33 days (range 13 to 48 days). This was accurately reflected by a worsening of all parameters. The first parameter to change was the sputum eosinophil percentage (20 days before the loss of asthma control). Significant changes in exhaled nitric oxide, FEV1 and methacholine PC20 were observed only when the symptoms became uncontrolled. A high blood eosinophil count at baseline (risk ratio of 2.5, 95% CI 1.0 to 6.5) and an increase in sputum eosinophil count after the reduction of corticosteroids were predictors of loss of asthma control. In patients whose asthma is controlled on inhaled corticosteroid, it is prudent not to reduce the dose further if the blood eosinophils are increased or if the sputum eosinophils increase by as little as 1% after the reduction of corticosteroids.