The objective of this study was to assess the effect of a novel regimen of antibiotic prophylaxis on the incidence of lower respiratory tract infection in patients requiring prolonged (at least five days) mechanical ventilation. The design was a controlled, prospective, randomized trial, with blinded comparison of the groups regarding the incidence of respiratory tract infection in an intensive care unit of a university hospital. After determination of the APACHE II score for severity of disease, 88 patients were randomly divided in three groups. Twenty-four of these patients did not complete five days of mechanical ventilation, and eight were withdrawn for other reasons. Fifty-six patients (18 in group 1, 21 in group 2, 17 in group 3) completed the study. Patients in both control groups 1 and 2 did not receive antibiotic prophylaxis, but the two groups differed in the antibiotic policy in case of infection. Patients in group 3 received antibiotic prophylaxis consisting of norfloxacin, polymyxin E, and amphotericin B, applied topically in oropharynx and stomach from time of ICU admission until extubation, and intravenous cefotaxime 500 mg three times a day during the first five days of admission. In both control groups, about 90 percent of the patients acquired microbial colonization of oropharynx or stomach. In group 3, only 12 percent and 24 percent of the patients acquired colonization of oropharynx and stomach, respectively (p less than 0.001). This resulted in a reduction of the incidence of lower respiratory tract infection (78 percent in group 1, 62 percent in group 2, 6 percent in group 3 [p = 0.0001]). The regimen of antibiotic prophylaxis studied prevented respiratory tract infection in mechanically ventilated patients. Antibiotic prophylaxis should be considered in all patients expected to require prolonged mechanical ventilation.