The hypothesis tested was that release of endogenous endothelin plays a role in events associated with or leading to myocardial ischaemia and/or post-ischaemic reperfusion damage. Release of endogenous endothelin into the coronary perfusate of isolated perfused rat hearts during ischaemia and reperfusion was measured with a sensitive radioimmunoassay using a polyclonal antibody with 100% cross-reactivity for all three endothelin isomers. Basal endothelin release was 0.69 +/- 0.02 pg/min/g wet heart weight (n = 35) and was constant up to 180 min. During low-flow hypoxic ischaemia for 180 min (PO2 approximately 250 mmHg) and in the presence of 1% foetal calf serum, the release rate was reduced to below 10% of controls (P < 0.01) and increased four-fold on reperfusion (P = 0.05). The influence of endothelin on vascular and myocardial reperfusion damage was studied with exogenous endothelin-2. After 1 h of low-flow ischaemia, endothelin-2 increased the coronary perfusion pressure to a similar extent as in non-ischaemic hearts, but with a 30-times higher potency. The threshold dose for the constrictive effect was approximately 100 to 300 pg per heart, about ten times more than was recovered in the coronary effluent upon reperfusion. The influence of endothelin on myocardial reperfusion mechanical function (stunning) was assessed with 100 ng endothelin-2, a dose some 3500-fold higher than the amount released during 30 min reperfusion. This dose, given at the onset of reperfusion, improved post-ischaemic aortic output recovery during the first 20 min of reperfusion, but worsened it thereafter (up to 40 min). These data indicate that, in the isolated perfused rat heart model, (1) endothelin is released in measurable amounts into the coronary circulation, (2) the release is much reduced during ischaemia and increased on early reperfusion following prolonged ischaemia, (3) based on the amounts released and the post-ischaemic sensitization of the coronary vasculature to endothelin, the peptide could contribute to reperfusion vascular damage, and (4) endothelin is unlikely to influence stunning owing to the extremely high dose necessary to alter reperfusion mechanical function.