Atracurium has been reported to have minimal haemodynamic effects in healthy patients. The purpose of this study was to determine its effects in patients with coronary artery disease. Sixteen patients scheduled for elective coronary artery surgery were studied in two equal groups. Group 1 received a bolus injection of atracurium 0.3 mgkg-1 and group 2 0.4 mgkg-1. Under local anaesthesia, radial artery, pulmonary artery thermodilution and central venous catheters were placed and the appropriate vascular pressures continuously monitored, as were leads II and V5 of the electrocardiogram. Sleep was induced with lorazepam and fentanyl while the patients were breathing nitrous oxide in oxygen (50:50). Control measurements of arterial pressure (AP) (mean, systolic, diastolic), CO (thermodilution), CVP, PA, PCW and HR were obtained. Atracurium was administered as a bolus and measurements repeated at 2, 5, and 10 min. In group 1 mean and diastolic arterial pressure decreased significantly at 2 min (73 +/- 2 to 66 +/- 3 mm Hg, P less than 0.05; 58 +/- 3 to 51 +/- 2 mm Hg, P less than 0.05). The changes were not significant at 5 or 10 min. There were no significant changes in CO or SVR. One patient in this group exhibited a typical histamine response with vasodilatation and flushing. In this patient mean arterial pressure decreased from 70 to 55 mm Hg and CO increased from 4.90 to 7.24 litre min-1. Excluding this patient from group 1 eliminated the significance of the haemodynamic changes for the rest of the group (MAP = 73 +/- 2 to 68 +/- 2 mm Hg, n.s.; mean diastolic AP = 58 +/- 3 to 53 +/- 2 mm Hg, n.s.). In group 2 none of the haemodynamic parameters measured showed significant changes. These results demonstrate minimal haemodynamic effects with 0.3- or 0.4-mgkg-1 bolus injections of atracurium in 15 patients with coronary artery disease, but in one patient doses of 0.3 mgkg-1 produced a typical histamine response with marked cardiovascular changes.