We measured serial plasma concentrations of thromboxane B2 (TXB2), the stable metabolite of the putative pulmonary vasoconstrictor thromboxane A2 (TXA2), and 6-keto-prostaglandin F1 alpha (6-keto-PGF1 alpha), the stable metabolite of the pulmonary vasodilator prostacyclin (PGI2) by double-antibody radioimmunoassay during partial venovenous bypass in 25 awake sheep. The onset of bypass caused mean pulmonary artery pressure (PAP) to increase from 16 +/- 1 to 28 +/- 2 mmHg at 12 +/- 2 min, due to an increase of pulmonary vascular resistance, followed by a return to control within 45 min. There was no systemic hypoxia. TXB2 increased simultaneously with the onset of pulmonary hypertension (PH) (236 +/- 36 to 700 +/- 120 pg/ml at 0 and 5 min) and peaked at 1,724 +/- 172 pg/ml 10 min after maximum PAP was achieved. Positive pulmonary artery-to-aortic differences of TXB2 were measured. 6-Keto-PGF1 alpha increased from 51 +/- 3 to 842 +/- 367 pg/ml at 35 min. PGF2 alpha was unchanged (130 +/- 45 pg/ml). PH, TXB2, and 6-keto-PGF1 alpha increases were blocked by pretreatment with indomethacin or ibuprofen. PH and TXB2 increases were prevented with an imidazole derivative. PH caused by a continuous infusion of an endoperoxide analog did not induce lung release of TXB2 or PGF2 alpha. We conclude that 1) transient pulmonary vasoconstriction is caused by thromboxane; 2) the lung is the primary site of thromboxane synthesis; and 3) bypass causes selective alterations in arachidonic acid metabolism rather than general activation of the cascade.