UNLABELLED We tested the hypothesis that in vitro coronary and myocardial effects of propofol (10-300 microM) should be significantly modified in an isolated and erythrocyte-perfused rabbit heart model in the absence (PaO(2) = 137 +/- 16 mm Hg, n = 12) or in the presence (PaO(2) = 541 +/- 138 mm Hg, n = 12) of hyperoxia. The induction of hyperoxia provoked a significant coronary vasoconstriction (-13% +/- 7%). Propofol induced increased coronary vasodilation in the presence of hyperoxia. Because high oxygen tension has been reported to induce a coronary vasoconstriction mediated by the closure of adenosine triphosphate-sensitive potassium channels, we studied the effects of propofol in 2 additional groups of hearts (n = 6 in each group) pretreated by glibenclamide (0.6 microM) and cromakalim (0.5 microM) in the absence and presence of hyperoxia, respectively. The pretreatment by glibenclamide induced a coronary vasoconstriction (-16% +/- 7%) which did not affect propofol coronary vasodilation. The pretreatment by cromakalim abolished the amplification of propofol coronary vasodilation in the presence of hyperoxia. Propofol induced a significant decrease in myocardial performance for a concentration >100 micro M both in the absence and presence of hyperoxia. We conclude that propofol coronary vasodilation is amplified in the presence of hyperoxia. This phenomenon is not explained by the previous coronary vasoconstriction induced by glibenclamide. However, the pretreatment of hearts by cromakalim abolished the amplification of propofol coronary vasodilation in the presence of hyperoxia. The myocardial effects of propofol were not affected by the presence of hyperoxia. IMPLICATIONS Propofol induced a coronary vasodilation that was amplified in the presence of hyperoxia. This phenomenon does not seem to be related to previous coronary vasoconstriction. The myocardial effects of propofol were not significantly modified in the presence of hyperoxia.