The increasing application of this technique has led to the recognition that inflammation in asthma patients is more heterogeneous than was previously thought with the identification of noneosinophilic asthma.2–4 This phenotype is common, accounting for 25 to 55% of patients with corticosteroid-naive asthma, and is repeatable, with few subjects with noneosinophilic asthma developing airway eosinophilia during follow-up for 1 year.5 This asthma phenotype is not reserved to patients with severe asthma, nor is it a consequence of asthma therapy, but it is present across the range of asthma severity.4 Its identification is important as it is related to a poor response to corticosteroids.2 This view is supported by a study reported in this edition of CHEST (see page 565) by Bacci et al,,6 who studied a group of 67 symptomatic asthmatic patients before and 2 and 4 weeks after receiving therapy with inhaled beclomethasone dipropionate, 500 μg twice daily. A total of 17 patients (25%) had noneosinophilic asthma. Consistent with previous studies,2,7 they found that a sputum eosinophil count, but not other markers of inflammation in sputum or peripheral blood, correlated with changes in lung function after corticosteroid treatment. One shortcoming of this study was its open-label design. However, in support of the findings of Bacci et al,6a placebo-controlled cross-over trial7 of inhaled mometasone, 400 μg once daily for 8 weeks, in subjects who were characterized as having either eosinophilic or noneosinophilic asthma prior to study entry confirmed that a favorable response to corticosteroids is reserved for asthmatic patients with sputum eosinophilia.