Measures of airway inflammation are increasingly being used as outcome measures in asthma intervention studies. Meaningful interpretation of observed changes in bronchial mucosal cell numbers should depend, in part, on the reproducibility of repeat measures over time. We wanted to investigate the reproducibility of immunopathologic and physiologic parameters after short and long measurement intervals. We therefore performed spirometry, bronchial provocation challenge, and fiberoptic bronchoscopy with endobronchial biopsy (always right upper lobe second-generation bronchus) at baseline, after 2 wk, and again after 8 wk on nine subjects with stable atopic asthma (receiving inhaled placebo and beta-agonist therapy only). Numbers of T cells, memory T cells (CD45Ro(+)), macrophages (CD68(+)), and eosinophils (EG1(+) and EG2(+)) on immunohistochemical stains of bronchial biopsies were quantified by computerized image analysis. Intraclass correlation coefficients (ICCs) of reproducibility were calculated for repeat measures of each parameter and a high ICC (greater than 0.6) was interpreted as "highly reproducible." Repeat measures of FEV(1), FEF(25-75%), and PC(20) were highly reproducible after short (2-wk) and long (8-wk) intervals. Only repeat measures of EG2(+) had an ICC greater than 0.6 after 8 wk. Repeat measures of CD45Ro(+), EG2(+), and T cell numbers (but not CD68(+) and EG1(+) cells) are highly reproducible and reliable parameters of asthmatic airway inflammation after a 2 wk interval.