In this study, we first tested the hypothesis that the previously demonstrated circulatory failure and thrombocytopenia induced by intracaval administration of thromboxane A2 (TxA2) analogues in nonpregnant (NP) rabbits [G. Losonczy, I. Mucha, J. DiPirro, J. Sweeney, G. Brown, J. Brentjens, and R. Venuto. Am. J. Physiol. 265 (Regulatory Integrative Comp. Physiol. 34): R772-R780, 1993] could be avoided if the compounds were given instead into the aortic arch. Conscious New Zealand White rabbits received bolus injections of U-46619 (5-20 micrograms) through a previously implanted catheter threaded into the aortic arch. Indeed, mean arterial pressure (MAP) rose modestly, and thrombocytopenia did not develop. Next, we compared the blood pressure responses of pregnant (P) rabbits with those of NP rabbits to intra-aortic U-46619 and I-BOP, because they had been found to be resistant to both the hypotensive and platelet aggregatory effects of intracaval U-46619. Resting blood pressure was lower in P than in NP rabbits (74 +/- 3 vs. 95 +/- 2 mmHg), but showed a greater increase in response to U-46619. For example, following a 20-micrograms dose blood pressure rose 20 +/- 0.3 mmHg in P vs. 12 +/- 2.1 mmHg in NP rabbits (P < 0.02). Similar results were obtained with the second TxA2 analogue I-BOP. Pregnancy-induced enhancement of blood pressure elevation may be the consequence of peripheral vasoconstriction, which was not seen in NP rabbits. Thus the actions of TxA2 analogues U-46619 and I-BOP are markedly influenced by the route of administration.(ABSTRACT TRUNCATED AT 250 WORDS)