Macrolides have antiinflammatory effects that are potentially useful in cystic fibrosis (CF). In this placebo-controlled, randomized, double-blind crossover study, 18 CF patients were randomized to receive either clarithromycin (CM) (Group 1) or placebo (Group 2) for three months. After 15 days, the treatments were crossed over. Bronchoalveolar lavage (BAL) was obtained in the beginning and at the end of each treatment period. There was no significant difference in median cell counts and median cytokine levels at baseline, after CM use and after placebo use between the two groups. In Group 2, the median neutrophil elastase (NE) level decreased with CM. Patients had less acute pulmonary exacerbations and median clinical score decreased with CM in both groups. Median z-scores for weight increased with CM in Group 2. We could not demonstrate a fall in proinflammatory cytokines in BAL; however, some improvement in clinical status could be shown with three-month CM.