The dynamics of cytokine release during 24 hours continuous normothermic machine perfusion liver preservation: An explorative porcine study.

Abstract

Normothermic machine perfusion (NMP) has been proposed to preserve liver grafts in a less pro-inflammatory environment. However, the effect of NMP on liver inflammation remains unclear. Therefore, we aimed at characterizing the inflammatory response during continuous NMP with a comprehensive investigation of cytokine release during perfusion. Ten porcine livers underwent either 24 h NMP or whole blood-based NMP (WB-NMP) immediately after procurement. WB-NMP was used as a positive control to mimic early post-reperfusion inflammation. High mobility group box-1 (HMGB1), interleukin 1-beta (IL-1beta), tumor necrosis factor-alpha (TNFalpha), interleukin 6 (IL-6), 8 (IL-8), and 10 (IL-10), transforming growth factor-beta (TGFbeta), aspartate transferase (AST), and hyaluronic acid were measured in the perfusate. The area under the curve (AUC) of their perfusate concentration was compared between groups. Median (IQR) is given. The AUC of HMGB1 and IL-1beta was similar between groups. Compared to WB-NMP, NMP inhibited the release of TNFalpha [NMP: 20275 (18402-32 152), WB-NMP: 242100 (203511-244 238); p = 0.01], IL-6 [NMP: 1206 (338.9-1686), WB-NMP: 8444 (7359-10 087); p = 0.03], and IL-8 [NMP: 1635 (106.90-2130), WB-NMP: 3951 (3090-4116); p = 0.008]. The release of TGFbeta remained unchanged but IL-10 release was lower in NMP [1612 (1313-1916), WB-NMP: 5591 (4312-6421); p = 0.01]. The ratios TGFbeta:TNFalpha and IL-10:TNFalpha were significantly higher in the NMP than in the WB-NMP group. Importantly, the AUC of AST was significantly lower during NMP [1960 (1950-2893)] than WB-NMP [6812 (6370-7916); p = 0.02]. Continuous NMP leads to the release of detectable levels of cytokines with a slow, linear increase over time and a shift toward anti-inflammatory signaling.