To assess the potential role of vasoactive cardiac eicosanoids in the modulation of coronary flow, we measure thromboxane(TX)B2, 6-keto-prostaglandin(PG)F1 alpha, PGE2 and sulphido-peptide leukotrienes (LTC4, D4, E4) in the coronary effluent of isolated perfused rat heart in both baseline and post-ischaemic conditions. Leukotrienes were undetectable. The order of production rate for the other eicosanoids was 6-keto-PGF1 alpha > TXB2 > PGE2. Production of such substances was increased about seven-fold over control after 5 min. global ischaemia; experiments with hypoxia showed that this was due to an actual increase in synthesis and not to a washout effect. A platelet source for TXB2 was excluded by 111In platelet labelling experiments. We assessed relative sensitivity to inhibition of cardiac TX synthesis relative to production of 6-keto-PGF1 alpha to inhibition by aspirin, ibuprofen, diclofenac and the specific thromboxane synthase inhibitor OKY-046. Aspirin, ibuprofen and diclofenac decreased 6-keto-PGF1 alpha production at a concentration always greater than required for a similar extent of TX inhibition. On the other hand a selective inhibition (> 90%) of TX was observed in the presence of OKY-046. Regression analysis of various 6-keto-PGF1 alpha/TXB2 ratios, as obtained in these different conditions, vs coronary flow, showed no correlation in baseline conditions, but a significant positive correlation (r = 0.59, P < 0.01) for post-ischaemic values. These data suggest a role for cardiac eicosanoids, including a non-platelet, cardiac-derived TX, in modulating the hyperaemic response in the isolated rat heart.