Hypothermic machine perfusion of the liver: is it more complex than for the kidney?

Hypothermic machine perfusion (HMP) is superior to simple cold storage (SCS) for the preservation of kidney grafts. Whether HMP is superior to SCS for liver preservation is not known. Before a HMP system can be used clinically for the liver, its superiority to SCS needs to be demonstrated in an in vivo large animal transplant model.

The aim was to compare outcomes after liver transplantation (LT) following preservation by SCS or HMP using technology/perfusion conditions similar to those for kidney HMP.

Pig livers were perfused via the hepatic artery and portal vein for 4 hours with nonoxygenated 4°C University of Wisconsin machine perfusion solution. In the SCS group, flushed livers were stored in histidine-tryptophan-ketoglutarate solution. After preservation by SCS (n = 6) or HMP (n = 8) and LT, we assessed graft and recipient survivals, pH and lactate, hepatocellular damage [aspartate aminotransferase (AST)], Kupffer cell activation (β-galactosidase), tumor necrosis factor (TNF) α production, endothelial cell function (hyaluronic acid), and expression of Krüppel-like factor (KLF) 2 and 4, which are mediators of the flow-dependent vasoprotective endothelial phenotype.

No primary graft nonfunction was observed; livers recovered equally well from the postanhepatic metabolic acidosis in both groups. Pig survival was 5/6 (83%) in the SCS versus 2/8 (12.5%) in the HMP group (P = .04). Livers from both groups recovered equally well from the postanhepatic metabolic acidosis. AST in liver rinse-out samples obtained before LT were lower in the HMP than in the SCS group (P < .05). After reperfusion, AST and β-galactosidase were equally increased in both groups (P = .13 and 0.962, respectively); TNF-α and hyaluronic acid levels were higher after HMP versus SCS (P = .001 and 0.043, respectively). KLF-2 and -4 expressions were equally up-regulated after reperfusion in the SCS and HMP groups.

In this in vivo model, liver HMP with subsequent transplantation was feasible. However, we did not demonstrate an advantage of HMP, using perfusion conditions shown to be effective for the kidney, over SCS. Despite similar immediate graft function, TNF-α generation, and endothelial cell dysfunction were more pronounced after HMP.