ObjectiveTo compare the effects of surfactant replacement by aerosol inhalation and bolus instillation on acute lung injury caused by the intratracheal injection of endotoxin in rats.DesignProspective, randomized study.SettingUniversity laboratory.SubjectsMale Wistar rats weighing 368 plus minus 31 (SD) g.InterventionsEscherichia coli endotoxin (57 plus minus 20 mg/kg) was injected into the tracheas of 36 anesthetized and mechanically ventilated rats (FIO2of 1.0). When the PaO2had decreased to less than 200 torr (less than 26.7 kPa), the rats were randomly assigned to one of three groups: a control group (n equals 12) given no material; a bolus group (n equals 12) given a modified natural surfactant suspension (100 mg/kg in 2.0 mL/kg saline) by bolus instillation into the trachea; and an aerosol group (n equals 12) given surfactant aerosolized with an ultrasonic nebulizer for 60 mins.Measurements and Main ResultsBolus instillation transiently decreased the mean blood pressure by approximate 30%. However, aerosol inhalation did not. The PaO2values of the control group remained less than 90 torr (less than 12.0 kPa) until the end of the experiment (180 mins). In contrast, the PaO2of the bolus group increased to 387 plus minus 134 torr (51.6 plus minus 17.9 kPa; p less than .05 vs. other groups) 15 mins after surfactant replacement, and remained at approximate 400 torr (approximate 53.3 kPa) throughout the experiment. The PaO2values of the aerosol group increased slowly, peaked at 240 plus minus 109 torr (32.0 plus minus 14.5 kPa; p less than .05 vs. the control group) 60 mins after the start of surfactant replacement, and remained at approximate 200 torr (approximate 26.7 kPa).ConclusionsBolus instillation was superior to aerosol inhalation concerning maximum efficacy, the rapid onset of therapeutic effects, and the necessary dose of surfactant. However, aerosol inhalation that does not cause hypotension may be of use in the treatment of adult respiratory distress syndrome in patients with circulatory instability.(Crit Care Med 1996; 24:488-494)