Abstract
INTRODUCTION: Hypothermic machine perfusion (HMP) is increasingly investigated as a means to assess liver quality, but data on viability markers is inconsistent and the effects of different perfusion routes and oxygenation on perfusion biomarkers are unclear. METHODS: This is a single-centre, randomised, multi-arm, parallel study using discarded human livers for evaluation of HMP using arterial, oxygen-supplemented venous and non-oxygen-supplemented venous perfusion. The study included 2 stages: in the first stage, 25 livers were randomised into static cold storage (n = 7), hepatic artery HMP (n = 10), and non-oxygen-supplemented portal vein HMP (n = 8). In the second stage, 20 livers were randomised into oxygen-supplemented and non-oxygen-supplemented portal vein HMP (n = 11 and 9, respectively). Changes in dynamic, biochemical, and morphologic parameters during 4-hour preservation were compared between perfusion groups, and between potentially transplantable and non-transplantable livers. RESULTS: During arterial perfusion, resistance was higher and flow was lower than venous perfusion (p = 0.001 and 0.01, respectively); this was associated with higher perfusate markers during arterial perfusion (p>0.05). Supplementary oxygen did not cause a significant alteration in the studied parameters. Morphology was similar between static and dynamic preservation groups. Perfusate markers were 2 fold higher in non-transplantable livers (p>0.05). CONCLUSIONS: Arterial only perfusion might not be adequate for graft perfusion. Hepatocellular injury markers are accessible and easy to perform and could offer insight into graft quality, but large randomised trials are needed to identify reliable quality assessment biomarkers.