Accumulating evidence strongly suggests that ventilatory strategy has an important impact on development of lung injury and patient outcome. Adverse ventilatory strategies have been shown to cause release of pulmonary-derived cytokines and may permit bacterial translocation from the lung to the systemic circulation. Because endotoxin is a potent and clinically important stimulant of cytokine-mediated systemic inflammatory responses that can lead to multiorgan failure, we investigated the effects of ventilatory strategy on lung-to-systemic translocation of endotoxin. We studied the effects of protective (tidal volume [VT] 5 ml. kg(-)(1), positive end-expiratory pressure [PEEP] 10 to 12.5 cm H(2)O) versus nonprotective (VT 12 ml. kg(-)(1), PEEP zero) ventilatory strategy on translocation of endotracheally instilled endotoxin. Anesthetized New Zealand White rabbits were subjected to saline lung lavage, and 32 were randomized to one of four groups: PS (protective ventilation + instilled saline); PE (protective ventilation + instilled endotoxin); NS (nonprotective ventilation + instilled saline); NE (nonprotective ventilation + instilled endotoxin), and ventilated for 3 h. Plasma endotoxin levels increased significantly in the NE group, and remained low and unchanged in the other groups. Peak levels of plasma tumor necrosis factor-alpha (TNF-alpha) were higher in NE versus other groups. Pa(O(2)) and mean arterial pressure (Pa) were lowest, and requirement for pressor and bicarbonate support greatest, in the NE group. Finally, plasma endotoxin levels were significantly greater in eventual nonsurvivors than survivors. These data provide convincing evidence for pulmonary translocation of lung-derived endotoxin. This translocation depends on ventilatory strategy, and suggests a pathophysiologic link between ventilatory strategy and outcome.