Is normal bronchial responsiveness in asthmatics a reliable index for withdrawing inhaled corticosteroid treatment?

STUDY OBJECTIVE Inhaled corticosteroid (ICS) treatment is first-line maintenance therapy in bronchial asthma. However, it is not clear whether and when ICS treatment can be withdrawn. The aim of this open study was to assess whether normalization of bronchial responsiveness could be used as a reliable index to assess the opportunity of ICS treatment withdrawal. DESIGN Open study at two different points in time. SETTING Outpatient pulmonary clinic. PATIENTS Eighteen asthmatic subjects. MEASUREMENTS AND RESULTS ICS therapy was withdrawn in subjects treated with beclomethasone dipropionate, at the maintenance dose of 889+/-246 microg/d for >3 months. Upon recruitment, all subjects were asymptomatic, had FEV1 >70% of predicted value, and were in treatment with beta2-agonists on an as-needed basis. Eight subjects (group 1) had normal bronchial responsiveness (methacholine provocative dose causing a 20% fall in FEV1 [PD20] >2,000 microg) and 10 subjects (group 2) had bronchial hyperresponsiveness (BHR) (PD20 < or = 2,000 microg). After withdrawal of ICS treatment, subjects were followed up for 3 weeks and were asked to record their asthma symptoms (cough, dyspnea, and wheezing) and their beta2-agonist use. At recruitment and at the end of follow-up, subjects underwent spirometry and a methacholine challenge test. Frequency of asthma exacerbation was similar in subjects with normal bronchial responsiveness (NBR) and in subjects with BHR (50% vs 60%), but subjects with NBR tended to remain asymptomatic for longer than those with BHR (mean+/-SD, 10.7+/-4.4 days vs 5.5+/-3.8 days) (p=0.08). None of the subjects reported any condition that could have triggered exacerbation. Asthma exacerbation was associated with a significant decrease in FEV1 (-105+/-107 mL; p<0.05) and in PD20 (-1,332+/-1,020 microg; p<0.001). CONCLUSIONS Our study shows that the likelihood of asthma exacerbation is not reduced if ICS treatment is withdrawn when the subjects have NBR, but the exacerbation could be delayed. Further studies in larger populations of asthmatics are needed to confirm these findings.