Title: Tolerance of livers to warm ischemia in a preclinical model of liver transplantation from non-heart-beating donors
Other Titles: Tolerantie aan warme ischemie van levers afkomstig van hartdode donoren in een preklinisch model van levertransplantatie
Authors: Monbaliu, Diethard
Issue Date: 26-Jun-2007
Abstract: T OLERANCE OF L IVERS TO WARM I SCHEMIA IN A PRECLINICAL M ODEL OF L IVER TRANSPLANTATION FROM NON - HEART - BEATING D ONORS : S UMMARY Subject and rationale of the research project As liver transplantation (LTx) has become the preferred method of treating end-stage liver disease, the number of patients listed for LTx exacerbated the shortage of organs from brain dead donors. Moreover, an increasing number of patients die while waiting for a suitable transplant liver. An alternative supply of organs is represented by Non-Heart-Beating Donors (NHBD) who are declared dead based on definite cardiac arrest (“dead by cardiac death”) rather than neurological criteria (“dead by braindeath”). Contrary to brain dead donors, there is a period without cardiac or respiratory activity, defined as Warm Ischemia (WI), during which organs are subjected to normothermic hypoxia. Unlike in kidney transplantation, the use of NHBD for LTx has remained limited. The most important reasons are the high risk of primary non-function (PNF), a complication that - unlike for the kidney - causes death in the absence of a rapid re-transplantation and the high incidence of biliary complications. The subject of this research project was to investigate the tolerance of livers exposed to WI in a preclinical model of LTx from NHBD and to elucidate some aspects pertaining to NHBD LTx:(i) the exact tolerance of the liver to WI is not known in a biologically unmodifiedenvironment;(ii) no reproducible techniques are currently available - contrary to the kidney - toevaluate WI damage and to predict liver graft viability ex vivo prior to implantation;(iii) finally, it is unknown whether it is possible to increase the resistance of the liver to WI. The aims of this study were 1. To develop a reproducible and preclinical large animal model of liver transplantationfrom non-heart-beating donors.2. To study the maximal duration of warm ischemia that a liver graft can tolerate withoutprecluding graft function3. To study the influence of prolonging cold ischemia after exposure towarm ischemiaon the outcome of LTx from NHBD.4. To understand some of the mechanisms that lead to failure of non-heart-beating donor livers5. To develop reliable and objective methods to predict viability of these livers prior orearly after their transplantation.6. To cautiously apply LTx from NHBD clinically. Summary of the results 1. A reliable and reproducible model of LTx from NHBD could beestablished. Important risk factors for recipient survival such as low-weight animals, preoperative hypothermia, blood loss exceeding 400 ml and pre-existing porcine specific diseases were identified and prevented accordingly. In the final study groups, 6 animals were available for final analysis ineach group. Amongst all groups, duration of the anhepatic phase, total operating time and peri-operative blood loss were comparable. 2. The maximal duration of warm ischemia that a porcine liver graft can tolerate without precluding graft function following LTx. Using incremental WI periods -prior to a short cold ischemic period- exposure of graft to 15 min WI is well tolerated, 30 and 45 min WI induce an unacceptable high rate of Primary graft Non-Function (PNF) (50 %) and60 min WI induces 100 % PNF. 3. The influence of cold ischemia after exposure to warm ischemia could be defined. Substantially prolonging the period of Cold Ischemia (CI) in groups previously not at risk to develop PNF was poorly tolerated whenever grafts were exposed to (very short) periods of WI. This resulted in an increased incidence of early graft dysfunction of these grafts and a poor recipient survival at day 4. 4. Mechanisms leading to failure of non-heart-beating donor liver grafts are multifactorial in nature. In grafts destined to fail, WI - on its own - causes hepatocellular damage that is immediately present after reperfusion and that is independent of the ischemia/reperfusion injury that ensues. The extent of this damage is well reflected by the peak of transaminases, such as AST and the timing of this peak. In groups with no or minimal WI, AST culminated on post-LTx day 1, whereas in groups exposed to > 30 min WI, AST culminatedvery early after reperfusion. In accordance, liver specimens taken immediately after prolonged WIalready showed substantial hepatocyte necrosisand vacuolization.WI by its own effect activates Kupffer cells, prior to cold storage andreperfusion. In grafts destined to fail, exposure to WI induced an increase in plasma levels of ß-galactosidase (a surrogate serum marker for Kupffer cell activation). This increase is the highest in recipients of grafts exposed to the longest WI and occurs very rapidly (15 min) after reperfusion. ß-galactosidase levels in serum also discriminate between recipients with fatal injury (PNF) from non-fatal liver injury (non-PNF). Unequivocal signs of Kupffer cell activation (cellular enlargement, presence of many large heterogeneous lysosomes, and of fragments of phagocytized red blood cells) were also found on electromicroscopical evaluationof biopsy specimens taken immediately after WI. The presence of large lysosomal granules already prior to reperfusion explains why ß-galactosidase – a lysosomal enzyme - can be released so early after reperfusion. In PNF recipients, the early activation of Kupffer cells was followed (at3 hrs after reperfusion) by higher TNF-α and IL-6 levels compared to non-PNF recipients. This late production of pro-inflammatory cytokines may have been initiated earlier during WI as biopsies taken before reperfusion showed higher expression of TNF-α mRNA (by RTPCR) in livers exposed to more prolonged WI. PNF recipients are also characterized bythe release of large quantities of redox-active iron. We also demonstrated an independent role for secretory Phospholipase A2 (sPLA2), which degrades cell membrane phospholipids and plays an important role in the synthesis of pro-inflammatory lipid mediators. In PNF recipients, sPLA2 activity increased and peaked significantly at 1 hour after reperfusion, prior to the peak of the pro-inflammatory cytokines )TNF-α andIL-6). Together with the activation of Kupffer cells and the generationof reactive oxygen species early during the ischemia/reperfusion injury, some anti-oxidant mechanisms (Reduced Glutathione and α-Tocopherol in particular) fail in PNF recipients. Finally, failing grafts displayprogressive microcirculatory disturbances and establishment of a progressive “no flow” phenomenon, as documented by a decreased Hyaluronic acidclearance (suggestive of sinusoidal endothelial cell dysfunction) and decreasing blood flow into the failing grafts Better understanding of themechanisms involved in failure of NHBD liver grafts allowed the design of a multifactorial “cocktail of drugs” to alleviate this injury. Preliminary results of such a study have already shown to improve tolerance toWI. 5. Reliable and objective methods to predict viability of these livers prior or early after their transplantation were developed. To predict viability of liver grafts prior to LTx, liver biopsies taken immediately after exposure to increasing lengths of WI were reviewed retrospectively. After a first retrospective review, hepatocellular vacuolization (a possible sign of hepatocellular suffering, present on biopsies immediately after exposure to WI) seemed to be associated with longer WI and PNF. In a second retrospective assessment, all biopsies were scored blindly using 3 different methods: (i) a semiquantitative scoring manner by an independent pathologist, (ii) by stereological point counting, (iii) and by digital image analysis. Vacuolization was found a risk factor to develop PNF and this risk varies according to the technique used to assess vacuolization: pathologist’s semi-quantative score (p=0.058 ; Odds ratio=1.065), stereological point counting (p=0.027; Odds ratio=1.528), digital image analysis score (p=0.043; Odds ratio=1.003). To predict viability as early as possible after LTx, a serum marker of hepatocellular damage: Liver-Fatty Binding Protein was examined. This small protein involved in the intracellular transport of fatty acids, can discriminate - as early as 15 min after graft reperfusion - groups without risk of developing PNF (exposure to 15 min or less WI) from groups with a substantial risk for developing PNF (exposure to 30 and 45 min of WI) and groups witha constant deleterious outcome (exposure to 60 min or more WI). 6. A clinical NHBD LTx program was launched at our Abdominal Transplant Surgery Department and at other Belgian LTx centers following the experimental evidence that NHBD LTx is feasible, provided that WIand CI are kept short. Between January 1, 2003 and December 31, 2005, 15 out of 584 LTx in Belgium were done with NHBD grafts. Mean interval from stop therapy/ventilation to cardiac arrest was 20 min and from cardiac arrest to liver cold perfusion 9.1 min. The mean cold ischemia time was 7 hrs 22 min, a relatively short period in clinical LTx. As expected by the relatively short WI periods, no PNF requiring re-transplantation was seen. Mean peak AST post- Tx was higher than normally observed (2209 IU/L). Graft/patient survivals were 65% and 80% respectively. Major biliary complications required re-LTx in 2 patients. Ongoing research and future plans: A better understanding of the mechanisms involved in failure of NHBD liver grafts allowed designing strategies to alleviate this injury. Two interventions are possible: intervention by pharmacological modulation during the ischemia/reperfusion injury, or an intervention during the preservation of livers by machine perfusion. Biological interventions aimed at improving this injury consist of several strategies, acting at different moments of WI injury (after exposure to WI, prior to and during ischemia/reperfusion injury). Such a multifactorial “cocktail” strategy focuses on modulating of Kupffer cell activation, inhibiting the generation of pro-inflammatory cytokines, optimization of anti-oxidant mechanisms, chelating circulating redox-activeiron, and maintaining adequate flow and viability to the graft. Currently, the effect of such a cocktail on livers exposed to 45 min WI has been investigated in our previously described model. Preliminary results are promising as they revealed excellent graft function in 6 out of 6 livers exposed to 45 min of WI, with the omission of the expected Kupffer cell activation and generation of pro-inflammatory cytokines. The use of machine perfusion to preserve grafts prior to transplantation represents a second strategy to alleviate the warm ischemic injury in NHBD LTx. Over the last decades, the use of hypothermic machine perfusion instead of cold storage to preserve NHBD kidneys has improved the outcome after transplantation. The advantage of hypothermic machine perfusion may be the continuous, complete perfusion of the liver. Moreover, during hypothermic machine perfusion the graft can be monitored and its viability tested, and different drugs administered. Therefore, this technology will be further developed andapplied to LTx from NHBD in an attempt to ameliorate tolerance of the liver to WI and to design a testing method that could allow discriminating viable and non-viable liver grafts.
Publication status: published
KU Leuven publication type: TH
Appears in Collections:Laboratory of Abdominal Transplantation
Translational Cell & Tissue Research

Files in This Item:

There are no files associated with this item.

Request a copy


All items in Lirias are protected by copyright, with all rights reserved.