OBJECTIVE The aim was to study (1) the effects of intracoronary saralasin, an angiotensin II receptor antagonist, on ischaemia induced and reperfusion induced regional cardiac noradrenaline release and ventricular arrhythmias; and (2) the implication of angiotensin II in coronary constriction during myocardial ischaemia. METHODS Eighteen adult mongrel dogs, weight 22.6(SD 1.1) kg, anaesthetised with sodium pentobarbitone, were used for the study. The left anterior descending coronary artery was ligated for 60 min and then reperfused for 30 min. Saralasin (60 micrograms.kg-1, n = 9) or its vehicle (Ringer lactate, n = 9) was injected into the artery at the beginning of the occlusion period. Two epicardial veins, one running parallel to the left anterior descending coronary artery and the other parallel to the circumflex coronary artery, were cannulated for the measurement of their respective blood flows and of noradrenaline, lactate, and creatine kinase release. RESULTS Saralasin decreased the incidence of ventricular fibrillation during coronary occlusion (from 44% in the vehicle treated group to 0% in the saralasin treated group, p = 0.0412). This effect was accompanied by significant vasodilatation in both epicardial veins during myocardial ischaemia. Neither the increases in noradrenaline, lactate, and creatine kinase release nor the incidence and duration of the ventricular arrhythmias following reperfusion were modified by the administration of saralasin. CONCLUSIONS Intracoronary saralasin in the early phase of myocardial ischaemia increases the epicardial venous blood flow significantly, suggesting that angiotensin II is implicated in coronary constriction during ischaemia. This haemodynamic effect is accompanied by a significant decrease in the incidence of ventricular fibrillation. However, the renin-angiotensin system does not appear to be implicated in the reperfusion induced noradrenaline release nor in the incidence of the ventricular arrhythmias.