Liver transplantation is currently the only established treatment for end-stage liver disease, but is limited by a severe shortage of viable donor organs; roughly 3,000 patients per year perish due to lack of transplantable donor organs. Livers obtained from donors after cardiac death (DCD) are currently an untapped donor source for liver transplantation and could increase the supply of donor livers by an estimated 6,000 per year. However, preservation of DCD livers by conventional methods, namely simple cold storage (SCS) in University of Wisconsin (UW) solution, is associated with a higher risk of primary non-function and delayed graft failure. We recently demonstrated that it is possible to resuscitate such livers using extracorporeal perfusion approaches. As compared to the current gold standard in organ preservation, cold storage, these methods provide the liver with oxygen and nutrition during storage; hence evidently enable a certain degree of repair by the organ. However, the state of the perfused liver is far from well understood, or characterized, hence a rational basis for further engineering the perfusion system is not available. Further, considering the expectable variability in these damaged livers, it is a critical necessity to evaluate their condition prior to transplantation, and optimize the repair during perfusion in order to maximize the probability of graft and patient survival.

Herein we describe the development of a metabolic flux model of ischemic reconditioning, with the ultimate goal engineering the perfusion systems in order to render marginal donor organs transplantable. Rat livers are subjected are harvested and subjected to 0, 60 or 90 minutes of warm ischemia in warm saline solution, as the model of donor after cardiac death. The livers are perfused for 6 hours in an extracorporeal perfusion system with erythrocyte supplemented WilliamsE medium. Perfusate samples obtained hourly from the inlet and outlet of the liver are analyzed for oxygen, amino acids and other metabolites to construct a metabolic flux model (46 metabolites and 72 fluxes). The differences between healthy livers (6/6 post-trasplant survival), ischemic livers reconditioned successfully (60 min, transplantation 11//11) and unsuccessful transplants (0/5) are analyzed, which reveals the dynamic progression of the metabolic processes during reconditioning, as well as pathways that cannot be adequately recovered with extensive ischemic damage. In addition, the results (fluxes and concentrations) are used to develop a viability score to predict the organ transplantability as a function of several metabolites during the perfusion, which can be used to quantitatively predict the probability of survival for a reconditioned liver. The use of the developed models and analysis for designing a second generation, portable perfusion system with a model-based metabolic feedback control is discussed.