Background. Normothermic ex vivo lung perfusion (EVLP) increases the pool of donor lungs by requalifying marginal lungs refused for transplantation through the recovery of macroscopic and functional properties. However, the cell response and metabolism occurring during EVLP generate a nonphysiological accumulation of electrolytes, metabolites, cytokines, and other cellular byproducts which may have deleterious effects both at the organ and cell levels, with impact on transplantation outcomes. Methods. We analyzed the physiological, metabolic, and genome-wide response of lungs undergoing a 6-h EVLP procedure in a pig model in 4 experimental conditions: without perfusate modification, with partial replacement of fluid, and with adult or pediatric dialysis filters. Results. Adult and pediatric dialysis stabilized the electrolytic and metabolic profiles while maintaining acid-base and gas exchanges. Pediatric dialysis increased the level of IL-10 and IL-6 in the perfusate. Despite leading to modification of the perfusate composition, the 4 EVLP conditions did not affect the gene expression profiles, which were associated in all cases with increased cell survival, cell proliferation, inflammatory response and cell movement, and with inhibition of bleeding. Conclusions. Management of EVLP perfusate by periodic replacement and continuous dialysis has no significant effect on the lung function nor on the gene expression profiles ex vivo. These results suggest that the accumulation of dialyzable cell products does not significantly alter the lung cell response during EVLP, a finding that may have impact on EVLP management in the clinic.

Lung transplantation is the only curative option for end-stage chronic respiratory diseases. However the survival rate is only about 50% at 5 years. Although experimental evidences have shown that innate allo-responses impact on the clinical outcome, the knowledge of the involved mechanisms involved is limited. We established a cross-circulatory platform to monitor the early recruitment and activation of immune cells in an extracorporeal donor lung by coupling blood perfusion to cell mapping with a fluorescent marker in the pig, a commonly-used species for lung transplantation. The perfusing pig cells were easily detectable in lung cell suspensions, in broncho-alveolar lavages and in different areas of lung sections, indicating infiltration of the organ. Myeloid cells (granulocytes and monocytic cells) were the dominant recruited subsets. Between 6 and 10 h of perfusion, recruited monocytic cells presented a strong upregulation of MHC class II and CD80/86 expression, whereas alveolar macrophages and donor monocytic cells showed no significant modulation of expression. This cross-circulation model allowed us to monitor the initial encounter between perfusing cells and the lung graft, in an easy, rapid, and controllable manner, to generate robust information on innate response and test targeted therapies for improvement of lung transplantation outcome.