A randomized trial of ceftazidime versus placebo was conducted in patients with cystic fibrosis hospitalized for acute respiratory exacerbations. Patients 12 years of age or older were included if they had mild to moderately severe illness according to the following criteria: erythrocyte sedimentation rate less than or equal to 50 mm/hr and less than three other abnormalities (leukocyte count greater than or equal to 15,000/microliter, pulse greater than or equal to 100 beats/min, respirations greater than or equal to 30/min, or temperature greater than or equal to 38.5 degrees C). In all 16 episodes treated with ceftazidime, the patients were rated improved in comparison with 10 of 12 patients treated with placebo. Three placebo-treated patients dropped out of the study within 3 to 5 days because they wanted antibiotic therapy. None of the 15 placebo-treated patients showed clinical deterioration. There were no significant differences in rate of improvement of symptom score, weight gain, or pulmonary function between the two treatment groups. There was no difference in the course during the 6 to 24 months after the study period. Intravenous antibiotics are not essential in the management of all acute respiratory exacerbations of mild to moderate severity in patients with cystic fibrosis.