BACKGROUND In experimental pneumonia, nebulization of antibiotics provides high lung tissue concentrations and rapid bacterial killing. OBJECTIVE To assess the efficacy and safety of nebulized ceftazidime and amikacin in ventilator-associated pneumonia caused by Pseudomonas aeruginosa. METHODS Forty patients with ventilator-associated pneumonia caused by Pseudomonas aeruginosa were included in a randomized comparative phase II trial. Twenty patients infected with susceptible or intermediate strains received nebulized ceftazidime (15 mg·kg(-1)·3 h(-1)) and amikacin (25 mg·kg(-1)·d(-1)). Seventeen patients infected with susceptible strains received intravenous ceftazidime (90 mg·kg(-1)·d(-1), continuous administration) and amikacin (15 mg·kg(-1)·d(-1)). In three patients infected with intermediate strains, amikacin was replaced by ciprofloxacin (400 mg·12 h(-1)). RESULTS After 8 days of antibiotic administration, aerosol and intravenous groups were similar in terms of successful treatment (70 vs. 55%), treatment failure (15 vs. 30%), and superinfection with other microorganisms (15 vs. 15%). Antibiotic-induced changes in lung aeration, determined by computed tomography, were not different between groups (increase in gas volume, 159 ± 460 vs. 251 ± 583 ml; decrease in tissue volume, -58 [-77, 25] vs. -89 [-139, 5] ml). Acquisition of per-treatment antibiotic resistance was observed exclusively in the intravenous group. In the aerosol group, four patients infected with intermediate strains were successfully treated. Nebulization induced an obstruction of the expiratory filter in three patients. The obstruction caused cardiac arrest in one patient, who fully recovered after brief cardiopulmonary resuscitation. CONCLUSIONS Nebulization and intravenous infusion of ceftazidime and amikacin provide similar efficiency for treating ventilator-associated pneumonia caused by Pseudomonas aeruginosa. Nebulization is efficient against intermediate strains and may prevent per-treatment acquisition of antibiotic resistance.