The specific benzodiazepine antagonist flumazenil is currently under intense study. Despite much clinical experience, no detailed invasive hemodynamic studies of its use in cardiac patients have been published. In the present study, hemodynamic and respiratory variables were measured in 10 cardiac patients undergoing catheterization of the right and left sides of the heart, before and after sedation with intravenous diazepam, and after reversal of sedation with flumazenil. A sleep dose of diazepam (12.2 +/- 5.1 mg, mean +/- SD) caused only slight decreases in mean arterial pressure (103 +/- 12 to 98 +/- 14 mm Hg; P less than 0.05), pulmonary capillary wedge pressure (13.2 +/- 6.3 to 11.7 +/- 6.6 mm Hg; P less than 0.05), and left ventricular end-diastolic pressure (20.8 +/- 7.5 to 17.3 +/- 10.0 mm Hg; P less than 0.05), with no significant changes in respiratory gas homeostasis. Intravenous flumazenil (0.22 +/- 0.07 mg) resulted in spontaneous awakening and return to full orientation, yet caused no significant alteration in either hemodynamic or respiratory variables measured. Reversal of diazepam-induced sedation by flumazenil in cardiac patients appears safe and effective.