The effects of adenosine (ADO) on pulmonary vascular resistance (PVR) distribution, vascular compliance (C), and permeability were determined in normal and PMA-injured isolated rabbit lungs perfused with a 1:1 mixture of 6% albumin in Krebs-Henseleit buffer and autologous blood. ADO or vehicle was continuously infused into the reservoir at 1, 4, or 5-mu-mol/min after a 1-mu-mol bolus of ADO or vehicle. The capillary filtration coefficient (K(f)) and arterial, venous, and double occlusion pressures were measured at baseline and 30 min after phorbol myristate acetate (PMA; 4 X 10(-8) M) or vehicle. Perfustate differential and total leukocyte counts as well as adenine nucleotides, 6-ketoprostaglandin F1-alpha (6-keto-PGF1-alpha), and thromboxane B2 (TxB2) concentrations were determined at each measurement period. ADO was recovered as hypoxanthine and inosine in the perfusate. ADO alone did not alter PVR, C, K(f), or TxB2 but reduced 6-keto-PGF1-alpha levels. PMA induced an increase in K(f) (0.024 +/- 0.002 to 0.040 +/- 0.006 g.cmH2O-1.min-1, P < 0.05) that was completely blocked by 4 or 5-mu-mol/min ADO. PVR increased by 63 +/- 11% after PMA, primarily in the arteries and arterial and venous microvessels. The postcapillary resistance increase was blunted by 4-mu-mol/min ADO; 5-mu-mol/min ADO prevented the PVR increase in all segments. ADO did not affect the initial adherence of neutrophils in the lung or the PMA-induced 87 +/- 2% decrease in circulating leukocytes (> 98% lymphocytes) or threefold increase in TxB2 levels. These results suggest that protection by ADO is not mediated by the altering of cyclooxygenase products or by leukocyte adherence.