Children with severe asthma pose a management dilemma; it is difficult to reduce their treatment while they are symptomatic, but the plateau of the dose response curve for inhaled corticosteroids (ICS) may have been reached. Studies in symptomatic adults and asymptomatic children with asthma suggest measuring induced sputum eosinophils allows better asthma management. We will recruit 80 children with severe asthma (treated with at least 500 mcg/day inhaled fluticasone or equivalent). We will see them every three months for a year, and perform sputum induction and measurements of exhaled nitric oxide (eNO). Half will be randomised to conventional management; half will have a reduction of ICS if there are no sputum eosinophils, or (if a sample cannot be produced) eNO is normal. The trial endpoint is whether there is a significant reduction in the median dose of ICS in the inflammatory markers group, with no increase in total number of asthma exacerbations. This study is also hypothesis generating. We will use our panel of markers (a) to try to improve clinical monitoring; and (b) to determine the molecular mechanisms of relapse of severe asthma.
