Abstract-The contribution of prostacyclin (PGI2) to the coronary vaso dilating action of arachidonic acid (AA) and bradykinin (BK) was examined in isolated perfused guinea-pig hearts. The injection of AA (100 to 1,000 ng) and BK (1 to 100 ng) into the heart resulted in a dose-dependent increase in the total amount of coronary flow and a release of 6-keto prostaglandin Fl., a stable metabolite of PGI2. Both AA and BK showed weak positive chronotropic effects. In addition, higher doses (300 and 1,000 ng) of AA caused a transient reduction in the coronary flow rate, left ventricular systolic pressure, and left ventricular dp/dt.The changes in coronary flow, release of PGI2, and all cardiodynamic parameters induced by AA were abolished by pretreatment of the preparation with diclofenac Na. On the other hand, the BK-induced increase in coronary flow rate was only partially reduced by diclofenac-Na when the release of 6-keto prostaglandin Fla was completely inhibited. It is concluded that in isolated perfused guinea-pig hearts, BK has both PGI2-independent and PGI2 dependent coronary vasodilating actions; the latter action is less than 25%, and the coronary vasodilating action of AA is mainly mediated via PGI2.Arachidonic acid (AA) and bradykinin (BK) have potent coronary vasodilating actions in isolated mammalian hearts (1, 2). The vasodilating action of AA is attributed to a metabolic product of AA, prostacyclin (PG12) (3), which has been described as being a prostaglandin-like substance (PLS). On the other hand, the mechanism for the coronary vasodilating action of BK remains obscure. Needleman et al. (2) suggested that the BK-induced coronary vasodilation seen in isolated rabbit heart preparations was mediated via the release of prostaglandins. Later, it was argued that PGI2 might play a main role in the coronary action of BK in isolated rabbit hearts (4). Recently, however, Schror et al. provided the evidence for an additional PG12-independent mechanism of coronary vasodilation, as induced by BK in guinea-pig hearts (5). In these studies, the release of PGI2-like substance was estimated by bioassay, and a quantitative relation between coronary vasodilation and PGI2 release following injection of BK was not studied in detail. We now report our findings of the quanti tative relation between coronary vasodilation and release of PGI2 as induced by AA or BK. PGI2 was measured by the radioimmunoassay of 6-keto-prostaglandin F,. (6-keto-PGF,.), a stable derivative of PG 12. The cardio hemodynamic effects of both compounds