(815i) Quality Assessment of Ischemic Livers for Transplantation

~4,000 people die every year in the United States while waiting for liver transplantation. ~5,000 marginally damaged livers are discarded due to fear of graft failure. Recent studies on liver preservation focus on new and dynamic modalities such as machine perfusion to recover discarded donor organs to a functional state. Machine-perfusion provides oxygen and nutrition enabling recovery of the organ and is an advantageous alternative to the current gold standard ‘cold storage’; however, machine-perfusion and its key factors need to be thoroughly studied for it to become a reality in clinical settings. Two such key factors are determination of biomarkers for graft viability during perfusion and prediction of the likelihood of graft survival after transplantation.

In humans, cardiac death frequently occurs in uncontrolled environments (uDCD). In the absence of cardiac output, ischemic damage to the organ increases in severity as a function of time. Hence, without objective metrics of ischemic duration and organ viability, uDCD organs cannot be safely transplanted.

The aim of this work is the development of an approach that determines the biomarkers for the assessment of uDCD ischemic damage and liver viability during perfusion, and the prediction of clinical outcome post-transplantation.

Machine-perfusion is a bio-process that allows real-time sampling from the graft. The samples are analyzed for blood gases, amino acids, glucose, lactate, and urea. The liver quality at the end of perfusion can further be evaluated via performance indicators such as liver function tests, cell isolation or ATP assays. For the identification of biomarkers to assess liver viability from a large dataset with dynamic trends during or at the end of perfusion, a suitable methodology is statistical process monitoring (SPM) based on partial least squares (PLS) and its extension for dynamic processes: multi-way PLS (MPLS).

MPLS constructs a statistical model of liver dynamic response to perfusion, which enables evaluation of the dynamic patterns of graft function during perfusion and comparisons of unknown uDCD livers’ qualities to those of known quality and infer about their ischemic conditions in real-time. Furthermore, MPLS’ in-silico predictions based on the liver function trajectory during perfusion, provide an estimate for the likelihood of graft success to the surgeon as a decision support system for transplantation.

This work presents a novel statistical approach for the assessment of ischemic damage in uDCD rat livers and prediction of likelihood of graft survival post transplantation, based on samples taken during perfusions of rat livers for 4 hours and were analyzed with HPLC for amino acids and biochemical assays for lactate, urea, glucose, ALT, and AST.