Background: The use of lung grafts from non-heart-beating donors (NHBD) is one way of solving the donor organ shortage problem. In this experiment, we studied the effect of retrograde flush (RF) from the left atrium before harvest, inhaled nitric oxide (NO), and gabexate mesilate (FOY), a protease inhibitor, in the lung grafts from NHBD., Methods: Forty-eight Lee-Sung, small-ear, miniature pigs (15-20 kg) were divided into 24 pairs (donor and recipient) and four groups. The donor lungs were flushed and harvested 90 min after cardiac arrest. No i.v. heparin was administered until the time before flush and harvest. Left single lung transplantation was undertaken, and the recipients were observed for 18 hr. The grafts warm and cold ischemia times were 90 (controlled) and 183+/-23.4 min. Group 1 (untreated control, UC, n=6) had core perfusion through a Swan-Ganz catheter followed by a single, antegrade flush with modified Euro-Collin's solution containing heparin, urokinase, and PGE1. Group 2 (RF group, n=6) had the same as group 1, except that one additive retrograde flush through the left atrium was administered. Group 3 (NO group, n=6) had the same as group 1, except that 20 parts per million (ppm) inhaled NO was administered for the cadaver donors before the graft harvest, and for the recipients after the grafts reperfusion. Group 4 (FOY group, n=6) had the same as group 1, except that the recipients received FOY i.v. infusion from the beginning of the recipient's operation and continuously throughout the experiments., Results: Compared with the group 1 (control), group 2 (RF) had significantly (P<0.05) lower mean pulmonary artery pressure, pulmonary vascular resistance (PVR), lung wet/dry ratio, histological lung injury score, and higher PaO2/FiO2 and pulmonary dynamic compliance. Group 3 (NO) had significantly lower mean pulmonary arterial pressure, PVR, lung injury score, degree of tissue neutrophils infiltration (histological and myeloperoxidase assay), bronchoalveolar lavage fluid protein content and neutrophils (PMNs) percentage, and higher PaO2/FiO2 and pulmonary dynamic compliance. Group 4 (FOY) had significantly lower PMNs infiltration, lung injury score, wet/dry ratio, bronchoalveolar lavage fluid protein and PMNs percentage, and higher PaO2/FiO2. Group 2 (RF) revealed better gas exchange (PaO2/FiO2) than the control (group 1) at earlier reperfusion periods (1st and 5th hr). On the contrary, group 4 (FOY) had higher PaO2/FiO2 than group 1 only at later period (18th hr). Pathologically, retrograde flush (group 2, RF) inhibited the intravascular thrombi formation more effectively than the NO or FOY treatment. However, the NO or FOY treatment inhibited the neutrophil infiltration more effectively than did the retrograde flush., Conclusion: The retrograde flush, inhaled NO and FOY infusion are beneficial to the protection of the NHBD lung grafts at an early reperfusion period, through different mechanisms. The use of these treatments in combination might help us to find a better way to protect the NHBD grafts against the preservation and reperfusion injury.