BACKGROUND: Acute lung injury is a common complication in critically ill patients. Several studies suggest that volatile anesthetics have immunomodulating effects. The aim of the current study was to assess possible postconditioning with sevoflurane in an in vivo model of endotoxin-induced lung injury. METHODS: Rats were anesthetized, tracheotomized, and mechanically ventilated. Lipopolysaccharide (saline as control) was administered intratracheally. Upon injury after 2 h of propofol anesthesia, general anesthesia was continued with either sevoflurane or propofol for 4 h. Arterial blood gases were measured every 2 h. After 6 h of injury, bronchoalveolar lavage was performed and lungs were collected. Total cell count, albumin content, concentrations of the cytokines cytokine-induced neutrophil chemoattractant-1 and monocyte chemoattractant protein-1, and phospholipids were analyzed in bronchoalveolar lavage fluid. Expression of messenger RNA for the two cytokines and for surfactant protein B was determined in lung tissue. Histopathologic examination of the lung was performed. RESULTS: Significant improvement of the ratio of oxygen tension to inspired oxygen fraction was shown with sevoflurane (mean + or - SD: 243 + or - 94 mmHg [32.4 kPa]) compared with propofol (88 + or - 19 mmHg [11.7 kPa]). Total cell count representing effector cell recruitment as well as albumin content as a measure of lung permeability were significantly decreased in the sevoflurane-lipopolysaccharide group compared with the propofol-lipopolysaccharide group in bronchoalveolar lavage fluid. Expression of the cytokines protein in bronchoalveolar lavage fluid as well as messenger RNA in lung tissue was significantly lower in the sevoflurane-lipopolysaccharide group compared with the propofol-lipopolysaccharide group. CONCLUSIONS: Postconditioning with sevoflurane attenuates lung damage and preserves lung function in an in vivo model of acute lung injury.