Ketoconazole appears to be an effective prophylactic measure in surgical patients at risk of developing ARDS. The beneficial effects may be caused by thromboxane synthetase inhibition because thromboxane B2 concentrations were decreased by ketoconazole in both studies. Two studies were unable to demonstrate a beneficial effect with the selective thromboxane synthetase inhibitor dazoxiben. Both studies consisted of a small number of subjects with already established ARDS, not prophylaxis in patients at risk of ARDS. Although the effects of ketoconazole on mortality in patients at risk of ARDS are conflicting, there may be reduced mortality in patients with sepsis. Several issues must be considered before ketoconazole is used in this setting. First, the studies to date have excluded patients at risk of hepatotoxicity, which is probably wise considering the potential hepatotoxicity with ketoconazole and the unknown benefit/risk ratio in these patients. Also, therapies that reduce gastric acidity should be avoided to ensure bioavailability. If ketoconazole is administered through a jejunostomy tube, it probably should be given with a dilute acid to enhance absorption. Furthermore, ketoconazole is a known inhibitor of the cytochrome P450 system, which results in a number of drug interactions. If ketoconazole is used, the patient's current drug therapy should be reviewed for potential interacting drugs. In light of the current studies, ketoconazole may be considered for surgical patients at risk of developing ARDS (especially patients with sepsis) with the previously noted considerations. Future research should seek to confirm ketoconazole's role for the prevention of ARDS in all critically ill patients. Additional studies also should clarify the role of various inflammatory mediators in the pathophysiology and therapy of ARDS.