Exogenous prostacyclin is effective in reducing pulmonary vascular resistance in some forms of human pulmonary hypertension (PH). To explore whether endogenous prostaglandins played a similar role in pulmonary hypertension, we examined the effect of deleting cyclooxygenase (COX)-gene isoforms in a chronic hypoxia model of PH. Pulmonary hypertension, examined by direct measurement of right ventricular end systolic pressure (RVESP), right ventricular hypertrophy (n = 8), and hematocrit (n = 3), was induced by 3 weeks of hypobaric-hypoxia in wild-type and COX-knockout (KO) mice. RVESP was increased in wild-type hypoxic mice compared with normoxic controls (24.4 +/- 1.4 versus 13.8 +/- 1.9 mm Hg; n = 8; p < 0.05). COX-2 KO mice showed a greater increase in RVESP following hypoxia (36.8 +/- 2.7 mm Hg; p < 0.05). Urinary thromboxane (TX)B(2) excretion increased following hypoxia (44.6 +/- 11.1 versus 14.7 +/- 1.8 ng/ml; n = 6; p < 0.05), an effect that was exacerbated by COX-2 gene disruption (54.5 +/- 10.8 ng/ml; n = 6). In contrast, the increase in 6-keto-prostacyclin(1alpha) excretion following hypoxia was reduced by COX-2 gene disruption (29 +/- 3 versus 52 +/- 4.6 ng/ml; p < 0.01). Tail cut bleed times were lower following hypoxia, and there was evidence of intravascular thrombosis in lung vessels that was exacerbated by disruption of COX-2 and reduced by deletion of COX-1. The TXA(2)/endoperoxide receptor antagonist ifetroban (50 mg/kg/day) offset the effect of deleting the COX-2 gene, attenuating the hypoxia-induced rise in RVESP and intravascular thrombosis. COX-2 gene deletion exacerbates pulmonary hypertension, enhances sensitivity to TXA(2), and induces intravascular thrombosis in response to hypoxia. The data provide evidence that endogenous prostaglandins modulate the pulmonary response to hypoxia.