We evaluated the dose response to a stable thromboxane (Tx) A2 analogue (sTxA2; 0.3-30 micrograms) in the pulmonary circulation and its effect on the distribution of pressure gradients determined by the occlusion technique in isolated nonblood perfused newborn lamb lungs. The total pulmonary pressure gradient (delta Pt) was partitioned into pressure drops across the relatively indistensible arteries and veins (delta Pv) and relatively compliant vessels. We also evaluated the effects of prostacyclin (PGI2) and a Tx receptor antagonist (ONO 3708) on the sTxA2-induced pulmonary responses. Injection of sTxA2 caused a dose-related increase in the pulmonary arterial pressure, with the primary component of the increase in delta Pt (4.1 +/- 0.8 to 13.9 +/- 0.4 Torr) at 30 micrograms derived from the prominent rise in delta Pv (1.8 +/- 0.3 to 9.8 +/- 0.9 Torr). Infusion of PGI2 (0.4 microgram.kg-1.min-1) reduced the response to sTxA2 mainly by attenuating the delta Pv elevation. Infusion of ONO 3708 (100 micrograms.kg-1.min-1) completely abolished the sTxA2-induced pulmonary hypertension. Injection of sTxA2 resulted in pulmonary edema characterized by a significant increase in wet-to-dry lung weight ratio (9.13 +/- 0.35 vs. 7.15 +/- 0.41 in control lungs). The sTxA2-induced pulmonary edema was increased by PGI2 and inhibited by ONO 3708. We conclude that thromboxane-induced pulmonary hypertension is primarily produced by venoconstriction and prostacyclin may worsen the edema induced by thromboxane.

We have previously shown that the pulmonary venoconstriction produced by a stable thromboxane A 2 analogue (STAJ is attenuated by prostacyclin (PGIi), but PGI, increases the STA 2 -induced edema. The present study was designed to determine the effects of STA 2 and PGI 2 on the fluid balance in isolated blood-perfused newborn lamb lungs. Vascular permeability was evaluated by use of the fluid filtration coefficient (Kf) and the osmotic reflection coefficient for total proteins (cr, hematocrit-protein double indicator technique), and pulmonary capillary pressure (Pc) was estimated by the double occlusion technique. All lungs had a period of hydrostatic stress induced by elevation of the left atriai pressure from 5 to 20 mm Hg to promote fluid filtration, and the rate of lung weight gain (AW/AT) during this period was determined. Studies were made in four groups; before the hydrostatic stress, lungs were given 1) STA 2 (50 /*g, n=6), 2) PGI 2 (0.4 /ig/kg/min, n = 6), 3) both PGI 2 and STA 2 (n=6), or 4) vehicles (control, n=5). Measurements of Kf were made at the baseline period and after the hydrostatic stress. T hromboxane A 2 (TxA 2 ) is a potent pulmonary vasoconstrictor that may be involved in the early phase of endotoxin-induced pulmonary hypertension and edema. This pulmonary hypertensive response has been suggested to be the result of TxA 2 -mediated pulmonary venoconstriction since the protein-poor lung lymph flow is increased in this early phase.12 By use of the arterial and venous occlusion technique in isolated newborn lamb lungs, we have recently demonstrated that a stable TxA 2 analogue, 9,11-epithioll,12-methano-TxA 2 (STA 2 ), produced pulmonary hypertension primarily by causing pulmonary Received March 27, 1989; accepted June 16, 1989. venoconstriction. 3 Although exogenous prostacyclin (PGI 2 ) attenuated this response to STA 2 by decreasing the pulmonary venoconstriction, PGI 2 administration significantly increased the STA 2 -induced pulmonary edema.3 These observations were interpreted as being consistent with an increase in microvascular permeability by PGI 2 that potentiated the STA 2 -induced pulmonary edema, despite the attenuation of the elevation of the pulmonary microvascular pressure by PGI 2 . However, the possibility that PGI 2 increased lung vascular surface area rather than permeability was not completely excluded in this study because vascular permeability was not directly evaluated.Pulmonary microvascular membrane permeability may be estimated by changes in filtration coefficient (Kf) in isolated perfused lungs. 4 However, when changes in the number of perfused microvessels occur, as with microvascular embolism, vasodilation, or vasoconstriction, Kf may not be a reliable parameter of changes in permeability because Kf depends on membrane surface area.