To determine the effect of clarithromycin therapy in patients with asthma. Randomized, double blind, placebo-controlled trial. A tertiary referral center. Fifty-five subjects with chronic, stable asthma recruited from the general Denver, CO, community. Patients underwent airway evaluation for Mycoplasma pneumoniae and Chlamydia pneumoniae by polymerase chain reaction (PCR) and culture, followed by treatment with clarithromycin, 500 bid, or placebo for 6 weeks. Outcome variables were lung function, sinusitis as measured by CT, and the inflammatory mediators tumor necrosis factor (TNF)-alpha, interleukin (IL)-2, IL-4, IL-5, and IL-12 messenger RNA (mRNA) measured via in situ hybridization, in airway biopsies, and BAL. Mycoplasma or chlamydia were detected by PCR in 31 of 55 asthmatics. Treatment resulted in a significant improvement in the FEV(1), but only in the PCR-positive subjects (2.50 +/- 0.16 to 2.69 +/- 0.19 L, mean +/- SEM; p = 0.05). This was not appreciated in the PCR-negative subjects (2.59 +/- 0.24 to 2.54 +/- 0.18 L, p = 0.85) or the PCR-positive or PCR-negative subjects who received placebo. Sinus CTs revealed no change in sinusitis with clarithromycin treatment. In situ hybridization revealed no significant difference in baseline airway tissue or BAL-mediator expression between the PCR-positive and PCR-negative subjects. However, the PCR-positive subjects who received clarithromycin demonstrated a reduction in TNF-alpha (p = 0.006), IL-5 (p = 0.007), and IL-12 (p = 0.004) mRNA in BAL and TNF-alpha mRNA in airway tissue (p = 0.0009). The PCR-negative subjects who received clarithromycin only demonstrated a reduction in TNF-alpha (p = 0.01) and IL-12 (p = 0.002) mRNA in BAL and TNF-alpha mRNA in airway tissue (p = 0.004). There were no significant differences in cytokine expression in those subjects who received placebo. These observations support the hypothesis that clarithromycin therapy improves lung function, but only in those subjects with positive PCR findings for M pneumoniae or C pneumoniae.