In 20 patients undergoing coronary arteriography, the hemodynamic effects of an experimental preparation of i.v. amiodarone 5 mg kg-1 without Tween 80 (N) (10 patients) were compared with those of the commercial form with Tween 80 (A) (10 patients). Analysis of variance demonstrated differences during the 3 min of injection and for 3 min afterwards: left ventricular systolic pressure decreased from 110 + 11 to 86 +/- 11 mmHg (P = 0.001) after A and from 114 +/- 22 to 106 +/- 19 (P = 0.05) after N (comparison P = 0.01) while related tachycardia was also more pronounced after A (comparison P = 0.001). Left ventricular end diastolic pressure transiently decreased after A while continuously increasing after N (P = 0.05). During the following 30 min both A and N caused similar bradycardia, increase in ventricular filling pressure, vascular resistance and decrease in cardiac and contractility indexes. Amiodarone blood levels were similar after A or N. These data document a significant initial short duration vasoplegia, mainly related to Tween 80, after A, when amiodarone itself after producing a similar very slight effect causes bradycardia, and a moderate and progressive negative inotropic effect. It was concluded that while the experimental form would be of interest, the risk of severe hypotension after i.v. Cordarone can be largely avoided by using a slower rate of infusion, especially in patients with hypovolemic status.