Safety and application of induced sputum analysis in childhood asthma

Background: The value of sputum induction in pediatric asthma lies in its potential to directly and noninvasively assess airway inflammation in children, because bronchoscopy and biopsy carry some risk. The Childhood Asthma Management Program (CAMP) study was designed to evaluate the long-term effects of budesonide and nedocromil compared with placebo in children with mild to moderate asthma across 8 centers. Objective: At the Denver CAMP site, we sought to evaluate the safety of sputum induction, to determine differences in airway inflammation between treatment groups by using induced sputum analysis, and to examine correlations between other biomarkers and sputum eosinophils. Methods: Sputum induction was performed, and exhaled nitric oxide, circulating eosinophil counts, and serum eosinophil cationic protein were obtained at treatment discontinuation and after washout. Spirometry and a methacholine challenge were also performed according to the CAMP protocol. Results: Ninety of 117 children provided an adequate sputum sample for analysis. In 9 subjects (3 nedocromil and 6 placebo), sputum induction resulted in bronchospasm. These subjects had greater disease severity, as measured by a lower median prebronchodilator FEV1 percentage predicted (85.0% vs 96.0%; P = .024) and FEV1/FVC ratio (70.0% vs 79.0%; P = .0008); greater bronchodilator reversibility (16.5% vs 6.8%; P = .004); higher serum IgE (1390.0 vs 495.0 ng/mL; P = .017) and circulating eosinophil count (757.0 vs 282.0/mm(3); P = .04); greater use of prednisone (1.9 vs 0.9 courses per 100 person-years; P = .05); and greater supplemental inhaled steroid doses (85.3 vs 0 mg; P = .016). At treatment discontinuation, budesonide-treated patients had a lower median (1st, 3rd quartile) sputum percentage eosinophil (SPEos) (0.2% [0%, 1.2%] vs 0.8% [0.2%, 4.6%]; P = .03) compared with those treated with placebo; no significant difference was noted between nedocromil- and placebo-treated patients. Higher SPEos at the time of treatment discontinuation was associated with asthma worsening that required rescue prednisone (n = 23) during the washout period compared with patients who remained stable (3.6% [0.4%, 6.4%] vs 0.6% [0.2%, 3.2%] SPEos; P = .023). Finally, greater SPEos was associated with atopy, higher bronchodilator reversibility, lower FEV1/FVC ratio, higher exhaled nitric oxide levels, circulating eosinophils, sputum and serum eosinophil cationic protein, more prednisone courses during the treatment period, and greater asthma severity. Conclusions: Sputum induction is a relatively noninvasive and safe procedure that can provide information on eosinophilic inflammation and treatment response and is also associated with several measures of asthma control. However, this procedure still remains a research tool in asthma because of its requirements for technical expertise.