BACKGROUND: To evaluate the efficacy of two methods of obtaining lung recruitment to reduce ventilator-induced lung injury(VILI). METHODS: Fifteen New-Zealand white rabbits were ventilated in the pressure-controlled mode maintaining constant tidal volume(10 ml/kg) and fixed respiration rate. Lung injury was induced by repeated saline lavage (PaO2 < 100 mmHg) and pressure-volume curve was drawn to obtain Pflex. Then the animals were randomly assigned to three groups and ventilated for 4 hours. In the control group(n=5), positive end-expiratory pressure(PEEP) was applied at a level less than Pflex by 3 mmHg throughout the study. In the recruitment maneuver(RM) group(n=5), RM(CPAP of 22.5 mmHg, for 45 seconds) was performed every 15 minutes in addition to PEEP level less than Pflex by 3 mmHg. In the Pflex group, PEEP of Pflex was given without RM. Parameters of gas exchange, lung mechanics, and hemodynamics as well as pathology were examined. RESULTS: 1) Both the control and RM groups showed decreasing tendency in PaO2 with time to show significantly decreased PaO2 at 4 hr compared to 1hr(p<0.05). But in the Pflex group, PaO2 did not decrease with time(p<0.05 vs other groups at 3, 4 hr). PaCO2 did not show significant difference between the three groups. 2) There was no significant difference in static compliance and plateau pressure. Mean blood pressure and heart rate also did not show any significant difference in the three groups. 3) In the pathologic exam, Pflex group had significantly less neutrophil infiltration than the control group(p<0.05). The difference in hyaline membrane score also showed borderline significance among groups(p=0.0532). CONCLUSION: Recruiting the injured lung may be important in decreasing VILI. Recruitment maneuver alone, however, may not be enough to minimize VILI.