OBJECTIVE The aim of this investigation was to test whether the administration of atrial natriuretic peptide (ANP) has cardioprotective effects against ischaemic and reperfusion injury. METHODS Thoracotomized dogs underwent a 30 min left circumflex coronary artery occlusion and 60 min of reperfusion (control group; n = 16). The ANP group (n = 9) received a 20 micrograms bolus injection of synthetic alpha human ANP (SUN 4936) followed by infusion at a dose of 0.1 microgram/kg/min from the beginning of coronary occlusion to the end of the procedure. RESULTS Administration of exogenous ANP increased plasma ANP immediately and maintained levels at 3000 pg/ml, resulting in an 8-fold increase in plasma cyclic guanosine monophosphate (cGMP) levels. Plasma ANP and plasma cGMP levels did not change at all in controls. There were no significant differences in haemodynamic parameters during ischaemia and reperfusion between the groups. In the ANP group, the prevalence and frequency of ventricular extrasystoles within 10 min after reperfusion decreased markedly [ANP 22% vs. control 100%, P < 0.01, and ANP 1 (1) vs. control 92 (50), P < 0.05, respectively]. No dog in the ANP group had ventricular fibrillation (VF), but the incidence of VF was not statistically significant between the groups [ANP 0% vs. control 25%]. ATP content in the inner layers of the ischaemic myocardium in the ANP group was higher than in controls (P < 0.05) [1.92 (0.28) vs. 1.18 (0.13) mumol/g wet weight]. There was no significant difference in the content of myocardial tissue angiotensin II between the groups. CONCLUSIONS These data show that the infusion of ANP has cardioprotective effects on myocardial ischaemia and reperfusion in this model. These beneficial effects are probably due to direct effects through cGMP rather than haemodynamic changes.