Analysis of induced sputum before and after withdrawal of treatment with inhaled corticosteroids in asthmatic patients

Objective To assess whether sputum eosinophilia predicts the recurrence of asthma symptoms after withdrawal of therapy in moderate stable asthmatics o n low-dose inhaled corticosteroids. Methods and design Randomized, double-blind, placebo-controlled study invol ving 30 subjects with stable asthma, asymptomatic, with low PEF variability measured over two run-in weeks, on treatment with low-dose inhaled beclome thasone dipropionate (BDP, 250 mug b.i.d. in the last 3 months). At the end of the run-in, all patients underwent a methacholine challenge test and sp utum induction (T1). They then stopped therapy and received either placebo (20 subjects, study group) or BDP at the same dose as in the previous 3 mon ths (10 subjects, control group). They continued to monitor PEF and symptom score for 3 months, or until asthma symptoms recurred (diurnal and nocturn al symptom score greater than or equal to2 on two consecutive days). At the end of the study (T2), i.e., either within 5 days from the beginning of as thma symptoms or after 3 months in subjects without recurrence of asthma sy mptoms, all subjects repeated the methacholine challenge test and sputum in duction. Results In the placebo-treated group, sputum eosinophils at T1 were signifi cantly higher in subjects who subsequently developed recurrence of asthma s ymptoms (n = 7) after cessation of treatment than in subjects who remained asymptomatic for 3 months (8.2% [0-56.6] vs 0.9% [0-11], P < 0.05). At the time of recurrence of asthma symptoms, sputum eosinophil percentages signif icantly increased (from 8.2% [0-56.6] to 16.6% [5.8-73.6], P < 0.05). The p ositive predictive value of sputum eosinophils for the recurrence of asthma symptoms was 71%, while the negative predicting value was 84%. In the BDP- treated control group, none of the subjects experienced recurrence of asthm a symptoms, and sputum eosinophil percentages measured at the beginning (T1 ) and at the end (T2) of the study were similar. Conclusion Sputum eosinophil percentages may vary over a wide range in asth matic subjects, although regularly treated and apparently well controlled. However, high sputum eosinophil percentages are related to early recurrence of asthma symptoms after cessation of inhaled corticosteroids.