Preservation-induced liver injury. Clinical aspects, mechanisms and therapeutic approaches

Implications of liver donor age on ischemia reperfusion injury and clinical outcomes

The aging process causes detrimental changes in a variety of organ systems. These changes include: lesser ability to cope with stress, impaired repair mechanisms and decreased cellular functional reserve capacity. Not surprisingly, aging has been associated with increased susceptibility of donor heart and kidneys grafts to ischemia reperfusion injury (IRI). In the context of liver transplantation, however, the effect of donor age seems to be less influential in predisposing the graft to IRI. In fact, a widely comprehensive understanding of IRI in the aged liver has yet to be agreed upon in the literature. Nevertheless, there have been many reported implications of increased liver donor age with poor clinical outcomes besides IRI. These other poor outcomes include: earlier HCV recurrence, increased rates of acute rejection and greater resistance to tolerance induction. While these other correlations have been identified, it is important to re-emphasize the fact that a unified consensus in regard to liver donor age and IRI has not yet been reached among researchers in this field. Many researchers have even demonstrated that the extent of IRI in aged livers can be ameliorated by careful donor selection, strict allocation or novel therapeutic modalities to decrease IRI. Thus, the goals of this review paper are twofold: 1) To delineate and summarize the conflicting data in regard to liver donor age and IRI. 2) Suggest that careful donor selection, appropriate allocation and strategic effort to minimize IRI can reduce the frequency of a variety of poor outcomes with aged liver donations.

Attenuation of Ischemia-Reperfusion Injury and Improvement of Survival in Recipients of Steatotic Rat Livers Using CD47 Monoclonal Antibody

BACKGROUND

Despite the efficacy of orthotopic liver transplantation in the treatment of end-stage liver diseases, its therapeutic utility is severely limited by the availability of donor organs. The ability to rehabilitate marginal organs, such as steatotic allografts, has the potential to address some of the supply limitations of available organs for transplantation. Steatotic livers are more susceptible to ischemia-reperfusion injury (IRI), which is exacerbated by the thrombospondin-1/CD47 pathway through inhibition of nitric oxide signaling. We postulated that CD47 blockade with a monoclonal antibody specific to CD47, clone 400 (CD47mAb400) may reduce the extent of IRI in steatotic liver allografts.

METHODS

Orthotopic liver transplantation was performed using steatotic liver grafts from Zucker rats transplanted into lean recipients. Control IgG or the CD47mAb400 was administered to the donor livers at procurement. Serum transaminases, histological changes, and animal survival were assessed. Hepatocellular damage, oxidative and nitrosative stress, and inflammation were also quantified.

RESULTS

Administration of CD47mAb400 to donor livers increased recipient survival and resulted in significant reduction of serum transaminases, bilirubin, triphosphate nick-end labeling staining, caspase-3 activity, oxidative and nitrosative stresses, and proinflammatory cytokine expression of TNF-α, IL-6 and IL-1β.

CONCLUSIONS

We conclude that administration of CD47mAb400 to donor grafts may reduce IRI through CD47 blockade to result in improved function of steatotic liver allografts and increased survival of recipients and represent a novel strategy to allow the use of livers with higher levels of steatosis.

"Resuscitation" of marginal liver allografts for transplantation with machine perfusion technology

As the rate of medically suitable donors remains relatively static worldwide, clinicians have looked to novel methods to meet the ever-growing demand of the liver transplant waiting lists worldwide. Accordingly, the transplant community has explored many strategies to offset this deficit. Advances in technology that target the ex vivo "preservation" period may help increase the donor pool by augmenting the utilization and improving the outcomes of marginal livers. Novel ex vivo techniques such as hypothermic, normothermic, and subnormothermic machine perfusion may be useful to "resuscitate" marginal organs by reducing ischemia/reperfusion injury. Moreover, other preservation techniques such as oxygen persufflation are explored as they may also have a role in improving function of "marginal" liver allografts. Currently, marginal livers are frequently discarded or can relegate the patient to early allograft dysfunction and primary non-function. Bench to bedside advances are rapidly emerging and hold promise for expanding liver transplantation access and improving outcomes.

Reversible mitochondrial uncoupling in the cold phase during liver preservation/reperfusion reduces oxidative injury in the rat model

Reversible uncoupling of the mitochondrial electron-transport chain may be one strategy to prevent intracellular oxidative stress during liver cold preservation/warm reperfusion (CP/WR) injury. 2,4-Dinitrophenol (DNP) is a potent water-soluble uncoupling agent for supplementation of the hepatic CP solution. The aim of this work was to investigate the possible influence of DNP in the CP solution on the isolated rat liver state during CP/WR. Livers were subjected to CP at 4 degrees C in sucrose-phosphate based solution (SPS) for 18 h, followed by WR for 60 min in vitro. The final concentration of DNP was 100 microM. DNP presence during the CP stage led to partial ATP level decrease accompanied by a significant diminution in liver TBARS and a prevention of antioxidant enzyme activity decrease (catalase, GSH-PO, GSH-Red) when compared with livers stored without DNP. After DNP wash-out during WR, an improvement in the mitochondrial functional state (higher respiratory control indices and ATP levels, and a decrease in V(4) respiration rates) were observed. This was concurrent with lower TBARS levels, higher antioxidant enzyme activities and significant increase of bile production. The results suggest that reversible uncoupling may be one way to influence oxidative stress associated with hepatic cold preservation.

Protective effects of a carbon monoxide-releasing molecule (CORM-3) during hepatic cold preservation

There is increasing evidence that carbon monoxide (CO), a signaling molecule generated during the degradation of heme by heme oxygenase-1 (HO-1) in biological systems, has a variety of cytoprotective actions, including anti-hypoxic effects at low temperatures. However, during liver cold preservation, a direct effect needs to be established. Here, we designed a study to analyze the role of CO, delivered via a carbon monoxide-releasing molecule (CO-RM) in the maintenance of liver function, and integrity in rats during cold ischemia/reperfusion (CI/R) injury. We used an isolated normothermic perfused liver system (INPL) following a clinically relevant model of ex vivo 48 h cold ischemia stored in a modified University of Wisconsin (UW) solution, to determine the specific effects of CO in a rat model. CO was generated from 50 microM tricarbonylchloro ruthenium-glycinato (CORM-3), a water-soluble transition metal carbonyl that exerts pharmacological activities via the liberation of controlled amounts of CO in biological systems. The physiological effects of CORM-3 were confirmed by the parallel use of a specific inactive compound (iCORM-3), which does not liberate CO in the cellular environment. CORM-3 addition was found to prevent the injury caused by cold storage by improving significantly the perfusion flow during reperfusion (by almost 90%), and by decreasing the intrahepatic resistance (by 88%) when compared with livers cold preserved in UW alone. Also, CORM-3 supplementation preserved good metabolic capacity as indicated by hepatic oxygen consumption, glycogen content, and release of lactate dehydrogenase. Liver histology was also partially preserved by CORM-3 treatment.

CONCLUSIONS

These findings suggest that CO-RM could be utilized as adjuvant therapeutics in UW solutions to limit the injury sustained by donor livers during cold storage prior to transplantation, as has been similarly proposed for the heart, and kidney.

Correlation between the liver temperature employed during machine perfusion and reperfusion damage: role of Ca2+

This study compares the effects of machine perfusion (MP) at different temperatures with simple cold storage. In addition, the role of Ca(2+) levels in the MP medium was evaluated. For MP, rat livers were perfused for 6 hours with Krebs-Henseleit (KH) solution (with 1.25 or 2.5 mM CaCl(2)) at 4 degrees C, 10 degrees C, 20 degrees C, 25 degrees C, 30 degrees C, or 37 degrees C. For cold storage, livers were perfused in situ and preserved with Celsior solution at 4 degrees C for 6 hours. The reperfusion period (2 hours at 37 degrees C) was performed under the same conditions used for MP-preserved and cold storage-preserved livers. Hepatic enzyme release, bile production, adenosine triphosphate (ATP) levels, and morphology were evaluated during MP and reperfusion. MP at 37 degrees C caused marked enzyme release; the same findings were obtained during reperfusion. By contrast, MP temperature lowering induced a significant decrease in liver damage. High levels of biliary gamma-glutamyltransferase and lactate dehydrogenase were found with MP at 4 degrees C and 10 degrees C but not with MP at 20 degrees C. When a KH-1.25 mM CaCl(2) solution was used during MP at 20 degrees C, very low enzyme release was observed and significantly lower hepatic damage was present at the end of the reperfusion period in comparison with cold storage. The same results were obtained when ruthenium red, a calcium uniporter blocker, was added to KH-2.5 mM CaCl(2). ATP levels were higher and morphology was better in liver preserved with KH-1.25 mM CaCl(2). MP at 20 degrees C with KH-1.25 mM CaCl(2) resulted in better quality liver preservation, improving hepatocyte and endothelial biliary cell survival, in comparison with cold storage. This raises the need to reconsider the temperature and calcium levels to be used during liver MP.

New preservation strategies for preventing liver grafts against cold ischemia reperfusion injury

BACKGROUND

In spite of improvements in University of Wisconsin (UW) preservation solution, the injury from grafts during cold storage is an unresolved problem in liver transplantation. The aim of the present study was to evaluate the beneficial effect on ischemia-reperfusion injury associated with liver transplantation of the inversion of K(+) and Na(+) concentrations and the replacement of hydroxyethyl starch (HES) by polyethylene glycol (PEG) in UW preservation solution.

METHODS

Using an orthotopic liver transplantation model, the effects on rat liver preservation of a modified preservation solution (UW-PEG) were evaluated, based on the inversion of K(+) and Na(+) concentration and the replacement of HES by PEG 35 kDa (0.03 mmol/L) in UW preservation solution.

RESULTS

The use of UW-PEG preservation solution ameliorated the biochemical and histological parameters of hepatic damage. Thus, at 24 h after transplantation, transaminase levels were reduced significantly when livers were preserved during 8 h in UW-PEG preservation solution compared with the original UW solution. In addition, histological findings revealed fewer and smaller areas of hepatocyte necrosis. The benefits of UW-PEG solution cannot be explained by modifications in oxidative stress or neutrophil accumulation associated with liver transplantation. However, the results of hepatic and portal blood flow indicated that the benefits of this modified preservation solution, UW-PEG were associated with improvements in the microcirculatory disorders after reperfusion.

CONCLUSIONS

The UW-PEG solution, while retaining all the advantages of UW solution, improved hepatic ischemia-reperfusion injury associated with liver transplantation.

General aspects and pitfalls in liver transplant pathology

Pathology in a liver transplant setting addresses four different topics: establishment of a definite diagnosis of the liver disease before listing for transplantation, evaluation of the donor liver with regard to pre-existing diseases, in particular macrovesicular steatosis and fibrosis, assessment of the hepatectomy specimen, and post-transplant biopsy evaluation. Of these, post-transplant biopsy evaluation is the most challenging and clinically the most relevant issue. It requires fast diagnoses to facilitate specific treatment and it has to incorporate a broad spectrum of differential diagnoses. Precise knowledge about rejection, post-transplant therapy, pathology of immunosuppression, and recurrence of the initially underlying liver disease including the characteristic time peaks and atypical histological presentations (e.g., fibrosing cholestatic hepatitis) is needed to evaluate specific and combined histological pictures of liver damage. For adequate interpretation of post-transplant biopsies the hepatopathologist has to be informed about the essential clinico-anamnestic aspects such as time course, medication, imaging results, and serology.

The delay of rearterialization after initial portal reperfusion in living donor liver transplantation significantly determines the development of microvascular graft dysfunction

BACKGROUND/AIMS

Graft reperfusion in liver transplantation is usually performed by initial portal reperfusion (IPR) and delayed rearterialization. Its influence on graft microcirculation is unknown. This study aimed to assess sinusoidal perfusion in dependence to this reperfusion technique during human living-donor liver transplantation.

METHODS

Hepatic microcirculation was measured both in the donor and the recipient (n=14) by using the orthogonal polarization spectral imaging technique. By using initial portal reperfusion, the mean time interval between portal venous and hepatic arterial reperfusion was 27.7+/-13.3 min.

RESULTS

Hepatic nutritive perfusion, as given by the functional sinusoidal density and the volumetric blood flow, was found significantly decreased during portal reperfusion when compared to baseline. Rearterialization resulted in hyperperfusion of individual sinusoids at a decreased density of the sinusoidal network. Interestingly, the time interval between portal venous and hepatic arterial reperfusion significantly correlated with the changes of the liver grafts' microcirculation.

CONCLUSIONS

The study indicates graft microcirculatory dysfunction as a major determinant of postischemic liver injury. Moreover, microvascular impairment was significantly influenced by the interval between portal venous and hepatic arterial reperfusion, which suggests the reinforcement of the pathomechanism of injury involving hypoxia and rapid graft rewarming due to initial portal reperfusion.

Xanthine oxidoreductase and preservation injury in human liver transplantation

BACKGROUND

Preservation injury is a major cause of primary graft dysfunction in liver transplantation (LT). Oxidative damage is considered to be the first event leading to graft damage. Xanthine oxidoreductase (XOR) and neutrophil activation, two sources of reactive oxygen species, could play a role in the development of graft dysfunction.

METHODS

We determined activities of XOR forms, polymorphonuclear elastase (PMN-E), aminotransferases, and hyaluronic acid in plasma of 20 patients undergoing LT. Samples were taken from the radial artery (RA) before the anhepatic phase; from the portal vein (PV) before reperfusion; from graft caval effluent (CE) at reperfusion; and from RA, PV, and the hepatic vein (HV) 10 and 90 min postreperfusion.

RESULTS

The graft, but not recipient bowel, released XOR into blood (XOR in CE, median, 61.2 mU/g protein [range, 1.9-160.4 vs. undetectable in PV before reperfusion). Circulating XOR was transformed from dehydrogenase to reversible oxidase (XOrev) (XOrev-to-XOR ratio, 48.1% in CE and 65.1% in HV 90 min postreperfusion). Neutrophil activation was detected in the recipients before reperfusion, and in liver at early post-reperfusion (median PMN-E was 0.85 microg/g protein [range, 0.01-1.58] in RA before the anhepatic phase; 2.22 microg/g protein [range, 0.20-5.88] in PV prereperfu-sion; and 3.60 microg/g protein [range, 0.48-6.78] in HV 10 min postreperfusion). XOR, but none of the other markers, was higher in the CE of patients with moderate primary graft dysfunction than in those with slight primary graft dysfunction.

CONCLUSIONS

XOR release and neutrophil activation are produced during LT, and they are potentially injurious mechanisms associated with this therapy.

Electron microscopical enzyme histochemistry on unfixed tissues and cells. Bridging the gap between LM and EM enzyme histochemistry

In principle, enzyme histochemistry should be performed on unfixed tissues and cells to avoid inhibition of enzyme activity by chemical fixation. For EM enzyme histochemistry, unfixed tissue specimens include fresh tissue blocks, non-frozen tissue chopper sections, cryostat sections and cell preparations. Studies on localization of enzyme activity at the ultrastructural level in unfixed specimens, be it fresh or frozen, are reviewed here. Preservation of ultrastructural morphology is discussed with special attention to the effects of freezing. It is concluded that unfixed cryostat sections are the best alternative for EM histochemistry of tissues, when interposing a semipermeable membrane in between cryostat section and gelled incubation medium. It is an adequate method to preserve structural integrity of unfixed tissue on the one hand and to avoid inactivation of the enzyme by chemical fixation on the other. For EM cytochemistry on individual cells, a better preservation of ultrastructure may be obtained because freezing can be avoided, but mild pretreatment with a fixative or detergent may be necessary to permeabilize cellular membranes for demonstration of intracellular enzyme activity.

The utility of marginal donors in liver transplantation

The shortage of organs has led centers to expand their criteria for the acceptance of marginal donors. The combination of multiple marginal factors seems to be additive on graft injury. In this review, the utility of various marginal donors in patients requiring liver transplantation will be described, including older donors, steatotic livers, non-heart-beating donors, donors with viral hepatitis, and donors with malignancies. The pathophysiology of the marginal donor will be discussed, along with strategies for minimizing the ischemia reperfusion injury experienced by these organs. Finally, new strategies for improving the function of the marginal/expanded donor liver will be reviewed.

Biliary inorganic phosphate as a tool for assessing cold preservation-reperfusion injury: a study in the isolated perfused rat liver model

Ischemia-reperfusion injury is a major cause of early graft dysfunction after liver transplantation. The bile flow has been suggested as an index of ischemic damage, and severely impaired bile flow seems to be predictive of poor survival in experimental studies. Looking for injury markers, biliary inorganic phosphate has the potential of being a useful endogenous marker of diminished hepatobiliary function because this anion is excreted in the bile by a paracellular pathway and it can detect changes in permeability. The goal of this study was to evaluate the effects of cold preservation-reperfusion of the liver on bile flow and bile inorganic phosphate and their relationship with storage-related graft failure. The isolated and perfused rat liver was used to evaluate the injury for ischemia-reperfusion. The intrahepatic resistance, lactate dehydrogenase release, and potassium and biliary inorganic phosphate concentration were used to estimate viability and function of freshly isolated or cold-preserved livers. The intrahepatic resistance and the bile flow were consistent and significantly decreased throughout the perfusion time in relation to the increment in storage. Inorganic phosphate is more concentrated in bile from preserved livers, showing an alteration in paracellular pathway, confirmed by the biliary excretion of horseradish peroxidase. After preservation, concentration and excretion of the paracellular marker were increased during the first peak. The second peak appears earlier in preserved livers (10 minutes) with a different shape but without changes in concentration. In conclusion, inorganic phosphate in bile shows changes in paracellular permeability as occurs in livers after 48 hours of cold preservation.

Elimination of indocyanine green in the perioperative evaluation of donor liver function

IMPLICATIONS

The authors describe the intraoperative use of indocyanine green dye elimination to detect critical reductions in donor liver function.

Vascular endothelial growth factor production by isolated rat hepatocytes after cold ischemia-warm reoxygenation

Inflammatory disturbances in the liver microcirculation have been associated with preservation injury of hepatic grafts. Vascular endothelial growth factor (VEGF), a proinflammatory growth factor released by hepatocytes, acts on sinusoidal endothelial cells, but its implication in graft injury is still unclear. We studied VEGF production by rat hepatocytes after cold ischemia and warm reoxygenation and compared the capacity of University of Wisconsin (UW) and sodium-lactobionate-sucrose (SLS) preservation solutions to maintain this hepatocellular function. Isolated hepatocytes were kept for 0, 24, and 48 hours at 4 degrees C in either solution (cold ischemia), then incubated for 1 to 24 hours at 37 degrees C (warm reoxygenation). We assessed cell viability and production of VEGF messenger RNA (mRNA) and protein. Cell viability decreased in a linear time-dependent fashion by 10% after 48 hours of cold preservation and by an additional 40% after 24 hours of warm culture. Very little VEGF mRNA could be detected after up to 48 hours of simple cold preservation in either solution. However, subsequent warm culture led to a robust and rapid increase in VEGF mRNA expression within the first hour, which declined to close to background levels within 8 to 12 hours in culture. This effect was more important in cells preserved in SLS than UW solution. Similarly, cold preservation alone did not trigger VEGF secretion. VEGF secretion was detected after culturing hepatocytes at 37 degrees C and reached a maximal secretion rate within 12 to 15 hours. However, VEGF production by preserved cells was reduced compared with unstored cells. In conclusion, cold ischemia and warm reoxygenation triggers VEGF mRNA expression by hepatocytes, but subsequent VEGF secretion is partially impaired, suggesting posttranslational defects.

Are parenchymal changes in early post-transplant biopsies related to preservation-reperfusion injury or rejection?

BACKGROUND

The progression of parenchymal changes in liver allograft biopsies due to preservation-reperfusion injury (PRI) and their differentiation from rejection related changes is poorly understood. The aim of this study was to determine which changes in a 1-week posttransplant biopsy could be attributed to PRI and which to acute rejection.

METHODS

One week protocol liver transplant biopsies from patients with mild PRI (day 1 AST<400 IU/L) were compared with those from patients with severe PRI (day 1 AST>2000 IU/L). Parenchymal changes (cholestasis, ballooning, steatosis, necrosis) and rejection-related inflammatory features (portal tract inflammation, bile duct inflammation, portal vein endothelial inflammation, hepatic vein endothelial inflammation, and centrilobular inflammation) were blindly assessed semiquantitatively.

RESULTS

Fat, cholestasis, and hepatocyte ballooning were significantly worse in the severe PRI group, and these features showed no correlation with histological features related to acute rejection. Centrilobular hepatocyte necrosis correlated with hepatic venular endothelial inflammation and centrilobular inflammation but not with rejection related features in portal tracts or with PRI. These findings suggest that centrilobular necrosis is a manifestation of a rejection-related parenchymal injury and may involve different pathogenetic mechanisms to rejection-related features in portal tracts.

CONCLUSIONS

This study indicates that in early posttransplant biopsies, fat, cholestasis, and ballooning can largely be attributed to PRI. By contrast, centrilobular hepatocyte loss should be suspected as a rejection related phenomenon, even if typical portal tract changes are not prominent, and augmentation of immunosuppression should be considered.

Determination of xanthine oxidoreductase forms: influence of reaction conditions

Xanthine oxidoreductase (XOR) has been implicated in ischaemia-reperfusion injury, and increases in this enzyme have been found in plasma of patients with different illnesses. The catalytic concentrations of the XOR forms found in plasma, using various reaction conditions, greatly differ in the related literature. We studied the effect of the assay conditions on the xanthine oxidation rate catalysed by the XOR forms. Our results demonstrate inhibition of XOR by the reaction products and a time-dependent decrease in the reaction rates of XOR forms. Substrate consumption and inhibition by the products did not account for this decrease. Determination at 60 min incubation leads to catalytic concentrations up to 80% lower for the XOR forms than those obtained at 10 min. We conclude that elimination of the reaction products (NADH, H(2)O(2) and O(2)) from the reaction mixture, and short incubation times, are necessary for accurate measurement of the XOR activities.

Antioxidants and gadolinium chloride attenuate hepatic parenchymal and endothelial cell injury induced by low flow ischemia and reperfusion in perfused rat livers

The objective of this study was to determine whether Kupffer cells contribute to parenchymal and endothelial cell damage induced by ischemia-reperfusion in perfused rat livers. Parenchymal and endothelial cell injury were determined by measuring activities of lactate dehydrogenase (LDH) and purine nucleoside phosphorylase (PNP), respectively, in the effluent perfusate of livers subjected to 60 min of low flow ischemia followed by 30 min of reperfusion. Upon reperfusion, LDH and PNP activities increased significantly within the first 10 min of reperfusion and remained elevated over control values throughout the duration of reperfusion. Pretreatment with gadolinium chloride, an inhibitor of Kupffer cell function, significantly decreased LDH and PNP efflux during reperfusion by approximately 60% and 50%, respectively. When Kupffer cells were stimulated by vitamin A pretreatment, PNP efflux was doubled during reperfusion. Vitamin E pretreatment attenuated LDH and PNP release by approximately 70% during reperfusion compared to enzyme release in untreated livers. Moreover, the water-soluble antioxidants superoxide dismutase and desferrioxamine reduced reperfusion injury, whereas catalase had no effect on enzyme release. These results demonstrate that superoxide anions released from Kupffer cells are involved in oxidative damage to endothelial cells as well as hepatocytes during the early stages of hepatic reperfusion.

Role of sodium nitroprusside in the improvement of rat liver preservation in University of Wisconsin solution: A study in the isolated perfused liver model

BACKGROUND

Long-term liver preservation is needed to transform liver transplantation from an emergency operation into an elective procedure and, therefore, to improve the results of liver transplantation.

AIMS

We have studied the possibility of extending the period of cold ischemia of the rat liver, maintaining good hemodynamics and functional conditions, by adding the NO donor sodium nitroprusside (NPNa) to the preservation solution.

MATERIALS AND METHODS

Rat livers were preserved for 24, 48, and 72 h in University of Wisconsin solution (UW) at 4 degrees C (groups I, II, and III) or UW to which 500 microM NPNa was added (groups IV, V, and VI). Following the preservation time, liver viability was assessed using the isolated perfused liver model. Lactate dehydrogenase (LDH) and K(+) release, bile flow, and portal resistance were evaluated in each group and compared with those of liver controls (group VII) excised and perfused without preservation.

RESULTS

Some deleterious effects can be seen during cold storage conditions as assessed by an increment in intrahepatic resistance and a diminution in the capacity of the organ to produce bile. On histological observation, we see vacuolated hepatocytes and free endothelial cells (detached) in the sinusoidal lumen. Addition of 500 microM NPNa to UW significantly moderates these injuries, with an improvement in intrahepatic circulation (less intrahepatic resistance), an increment in bile production, and better histological appearance of the organ. We were also able to determine the capacity of the UW + NPNa to produce NO.

CONCLUSION

We assume that the beneficial vascular effects of NPNa are mediated by NO production.

Direct cytotoxicity of hypoxia-reoxygenation towards sinusoidal endothelial cells in the rat

AIMS/BACKGROUND

Sinusoidal endothelial cells are the primary target of ischemia-reperfusion injury following liver preservation. The present study was undertaken to examine the susceptibility of sinusoidal endothelial cells to hypoxia-reoxygenation and the potential role of oxygen free radicals in the induction of cell injury.

METHODS

Sinusoidal endothelial cells were isolated from rat liver. After 2 3 days of primary culture, the cells were exposed to hypoxia (N2/CO2 95/5) for 120 min and reoxygenation (O2/CO2 95/5) for 90 min. Control cells were exposed to hypoxia alone, to 95% O2 alone or were maintained under normoxic conditions. Human umbilical vein endothelial cells were used as a model of vascular endothelial cells and submitted to the same protocol. Cell viability and lipid peroxidation were assessed by LDH leakage and malondialdehyde production, respectively. In order to test the potential role of xanthine oxidase and mitochondrial dysfunction in cell injury, the cells were treated with allopurinol and potassium cyanide (KCN) respectively.

RESULTS

The different gaseous treatments did not affect LDH leakage in human umbilical vein endothelial cells. In sinusoidal endothelial cells, the sequential hypoxia-reoxygenation caused a significant increase in LDH release, malondialdehyde production and xanthine oxidase activity while hypoxia alone had no effect except on xanthine oxidase activity. Allopurinol inhibited xanthine oxidase without preventing cell injury or lipid peroxidation in this latter cell type.

CONCLUSIONS

The results suggest that sinusoidal endothelial cells, as opposed to vascular endothelial cells, are susceptible to a direct cytotoxic effect of hypoxia-reoxygenation. This effect occurs in combination with an increase in xanthine oxidase activity and lipid peroxidation, although cell injury is mediated at least in part by mechanisms independent of xanthine oxidase such as mitochondrial dysfunction.

Indocyanine green clearance reflects reperfusion injury following liver transplantation and is an early predictor of graft function

BACKGROUND/AIMS

Primary graft dysfunction is difficult to predict. We have previously shown that indocyanine green clearance measured at 24 h following orthotopic liver transplantation predicts graft survival and outcome. We prospectively evaluated the use of indocyanine green clearance (with a cut-off value of 200 ml/min) as a marker of graft function following orthotopic liver transplantation and investigated its relationship with the markers of reperfusion injury during orthotopic liver transplantation.

METHODS

In all patients indocyanine green clearance was measured at 24 h. Repeated blood samples were taken before, during the anhepatic and reperfusion phase and up to 12 h following orthotopic liver transplantation to measure the levels of neutrophil elastase and reactive oxygen intermediates. All patients studied had normal hepatic arterial pulse on Doppler-ultrasound post orthotopic liver transplantation.

RESULTS

All patients with indocyanine green clearance >200 ml/min recovered following orthotopic liver transplantation and remained well up to 3 months of follow up. Four patients had an indocyanine green clearance <200 ml/min; three were re-transplanted for graft failure within 3 days of the transplant, while one survived after prolonged intensive support and hospitalization. Indocyanine green clearance significantly correlated with reactive oxygen intermediates production and neutrophil elastase during orthotopic liver transplantation (r=-0.61, p<0.002 and r=-0.66, p<0.0009, respectively). Indocyanine green clearance was also significantly correlated with alanine aminotransferase and prothrombin time at 24 h post-transplantation (r=-0.35, p<0.02 and r=-0.4, p<0.0077, respectively).

CONCLUSION

Indocyanine green reflects the degree of reperfusion injury and is a good early marker of primary graft function. Indocyanine green clearance over 200 ml/min is associated with favorable outcome.

Hypothermic oscillating liver perfusion stimulates ATP synthesis prior to transplantation

BACKGROUND

ATP and glycogen depletion often have been demonstrated during cold storage of the liver prior to transplantation. Suppression of events that lead to metabolic depression and to lipid peroxidation could contribute to improvement of liver preservation. A new method of liver preservation for transplantation is therefore suggested, an oscillating oxygenated hypothermic liver perfusion.

METHODS

Biochemical analysis of liver tissue samples and perfusate after 10 h of perfusion by the presented oscillating perfusion model were compared with results after continuous liver perfusion for 10 h as well as with data derived from cold-stored livers over a period of 10 h. Particular reference was made to nucleotide metabolites, glycogen content, lipid peroxidation, glutathione content, glycolytic metabolites, and enzyme release before and after preservation.

RESULTS

Glycogen depletion occurred to the same degree in hypothermic storage and machine perfusion (oscillating as well as continuous perfusion), but the energy charge was significantly increased after oxygenated perfusion, whereas cold storage resulted in a significant energy charge depletion. In addition, perfusion by an oscillating technique yielded superior energy charge loading compared to the continuous perfusion technique and diminished the other hand lipid peroxidation.

CONCLUSIONS

Hypothermic oscillating oxygenated perfusion could be important for the improvement of the quality of energy-depleted organs prior to transplantation.

Prevention of reoxygenation injury by sodium salicylate in isolated-perfused rat liver

Sodium salicylate can be used as a chemical trap for hydroxyl radicals, the most damaging reactive oxygen species. Because reactive oxygen species are involved in the pathogenesis of hepatic hypoxia/reoxygenation injury, the goal of this study was to determine if trapping hydroxyl radicals with salicylate would prevent or at least ameliorate such injury. Isolated rat livers, continuously perfused with Krebs-Henseleit bicarbonate buffer in the presence or absence of salicylate (2 mM), were exposed, after 30 min of recovery, to 60 min of hypoxia, followed by 30 min of reoxygenation. During reoxygenation, control livers experienced a sharp increase in the rate of lactic dehydrogenase release, taken as index of cell injury, protein carbonyl content, and malondialdehyde, taken as index of protein oxidation and lipid peroxidation, respectively. The presence of salicylate in the solution perfusion significantly reduced the rate of lactic dehydrogenase release, protein carbonyl content, and malondialdehyde production during reoxygenation. Hepatic histology documented a significantly reduced cell injury in salicylate-perfused livers compared to control livers. These data suggest that the hydroxyl radical chemical trap sodium salicylate, acting as an antioxidant, may represents an effective agent to reduce liver injury due to hypoxia/reoxygenation in a model of isolated-perfused rat liver.

Hepatocellular defence against acidosis is preserved after cold storage

BACKGROUND

Primary non-function of liver allografts is related to preservation time, during which hypoxia leads to intracellular accumulation of acid. Preservation-induced failure of hepatocellular pH regulation may play a role in the pathogenesis of primary graft non-function.

METHODS

Using cultured/suspended rat hepatocytes and fluorimetric determination of intracellular pH, we determined whether preservation in University of Wisconsin solution (4 degrees C) impairs hepatocellular defence mechanisms against acidosis.

RESULTS

In non-preserved, 24-h-preserved and 48-h-preserved hepatocytes acidified to pH 6.7-6.8, initial Na+/H+ antiport-mediated H+ fluxes averaged 12 +/- 5, 9 +/- 5 and 12 +/- 5 nmol microL-1 min-1 and initial Na+/HCO3- symport-mediated HCO3- fluxes 7 +/- 2, 7 +/- 3 and 6 +/- 2 nmol microL-1 min respectively (P = NS). Preservation did not affect the inverse relationship between Na+/H+ antiport activity and intracellular pH. Thus, hepatocellular defence against intracellular acidosis is maintained during up to 48 h in University of Wisconsin solution.

CONCLUSION

Altered pHi homeostasis is unlikely to play a role in the pathogenesis of primary non-function of liver allografts.

Glutathione movements during cold preservation of rat hepatocytes

In this study we have examined the movements of glutathione (GSH) during cold preservation of rat hepatocytes in University of Wisconsin solution. During the preservation process at a low temperature (4 degrees C), with a high extracellular potassium concentration, an extracellular nondiffusible anion (lactobionate), and a Cl(-)-free medium, there is a depletion of metabolites and the development of a time-dependent injury. Also, there is a loss of GSH that is not compensated by transport or synthesis and is basically due to increased catabolic processes. This sensitizes the cells to different forms of oxidative injury, which can play a negative role during transplantation. The addition of GSH improves liver cell preservation but the mechanism is unclear. To elucidate this process we have isolated hepatocytes and preserved them under different conditions: with or without GSH: in the presence of DL-buthionine-[S,R]-sulfoximine, an inhibitor of glutathione synthetase, and acivicine to inhibit the ectoactivity of cellular gammaglutamyl transpeptidase; or by obtaining hepatocytes from rats depleted of GSH by an injection of diethyl maleate. Under all these conditions we evaluated the GSH content of the cells during cold storage. We also report the time course of accumulation of [glycine-2-3H]GSH. Our results show that during hypothermic storage in University of Wisconsin solution, hepatocytes are permeable to GSH, and the mechanism involved is a rapid nonsaturable process, with linear dependence of the extracellular GSH concentration. This finding may have valuable applications in the improvement of the delivery of compounds to cells.

Significance of early aminotransferase elevation after liver transplantation

BACKGROUND

We retrospectively reviewed 213 consecutive patients who received their first liver allograft between January 1 and December 31, 1993, in order to study the impact of ischemia/preservation/reperfusion injury (IPRI) on patient and graft outcome.

METHODS

The extent of IPRI was assessed by the peak value of aspartate aminotransferase (ASTmax) observed within the first 72 hr after transplant. For the purpose of univariate analysis, categorical classification of recipients was done based upon ASTmax as follows: group 1, ASTmax<600 U/L (n=46); group 2, ASTmax=600-2000 U/L (n=97); group 3, ASTmax>2000-5000 U/L (n=50), and group 4, ASTmax>5000 U/L (n=17). For multivariate analysis, stepwise Cox regression was performed with age, ASTmax, and United Network for Organ Sharing (UNOS) status as covariates.

RESULTS

Groups were comparable with respect to age, UNOS status at the time of transplantation, and diagnostic case mix. Median follow-up was 644 days. The overall incidence of primary graft nonfunction (PNF) was 7.6%. PNF incidence was significantly correlated with the severity of IPRI (0%, 4%, 10%, and 41% for groups 1 to 4, respectively, P < 0.0001), but this impact was confined to the respective rates of retransplantation as early patient survival was unaffected. The 1-year survival of patients whose initial grafts manifested extreme IPRI (group 4) was significantly inferior to recipients in the three other groups (77%, 71%, 73%, and 52% for groups 1 to 4, respectively, P=0.03). This increased mortality was confined to patients who never achieved discharge from their initial hospitalization, with no significant differences between groups being detected in the survival of those patients who were discharged (84%, 80%, 85%, and 81% for groups 1 to 4, respectively, P=NS). Although overall 1-year graft survival was strongly correlated with the extent of IPRI (77%, 67%, 62%, and 41% for groups 1 to 4, respectively, P=0.001), this correlation was abolished when survival of grafts not lost to PNF was examined at 1 and 2 years. Stepwise Cox regression analysis confirmed the independent association between ASTmax and patient and graft survival. The long-term quality of allograft function as well as the incidence of chronic rejection and biliary complications were unrelated to the extent of IPRI.

CONCLUSIONS

We conclude that: (1) patient survival is influenced by IPRI only when it is extreme (ASTmax>5000 U/L), provided parameters of graft function are used in conjunction with aminotransferase values to assess the need for prompt retransplantation; (2) short-term graft survival is proportional to the extent of IPRI, but grafts that are not lost to PNF have equivalent 1- and 2-year survival irrespective of the magnitude of IPRI; (3) 40% of grafts with extreme IPRI are lost to PNF, but the same proportion also provide long-term function; and (4) for surviving grafts, long-term biochemical function as well as the incidence of biliary complications and of chronic rejection are unrelated to the extent of IPRI.

Neutrophil elastase: a determinant of endothelial damage and reperfusion injury after liver transplantation?

Reperfusion injury has been implicated in the development of primary graft dysfunction (PGD) after liver transplantation. Neutrophil migration and activation may be involved in the pathogenesis of this injury. We studied neutrophil activation and its role in the etiology of PGD by measuring neutrophil elastase by radioimmunoassay, in serial blood samples of 19 patients before, during, and for 24 hr after transplantation. In a subgroup of patients, we also measured soluble thrombomodulin at the same time points as a marker of endothelial damage. The pretransplant elastase level was significantly raised (40.13+/-4.84 ng/ml, mean+/-SEM) compared with levels of healthy controls (18.7+/-5.6 ng/ml, P<0.05). A marked increase in elastase activity followed reperfusion, with a peak at 2 hr (370+/-50.5 ng/ml, P<0.01). Thereafter, there was a decline, but elastase remained elevated at 24 hr (186+/-60.94 ng/ml). The mean increase in neutrophil elastase after reperfusion correlated significantly with markers of graft function (P<0.05) and with the mean rise in soluble thrombomodulin (P=0.042), which increased from a pretransplant level of 81.2+/-11.32 to 186+/-50.4 ng/ml, 6 hr after reperfusion (P<0.05). The results of this study indicate that marked neutrophil activation and endothelial cell damage occurs after graft reperfusion during orthotopic liver transplantation, and the degree of activation correlates with markers of graft function, which may suggest a role in the etiology of PGD.

Serum levels of endothelial injury markers creatine kinase-BB and soluble thrombomodulin during human liver transplantation

Endothelial damage within the sinusoids of the liver probably plays a key role in primary liver dysfunction following transplantation. The aim of this work was to study the serum levels of two potential markers of endothelial damage, creatine kinase-BB and soluble thrombomodulin, during human graft revascularization. Thirteen human liver grafts were preserved in UW solution (mean time: 13.8 h). Creatine kinase-BB and transaminase activities and soluble thrombomodulin levels were measured: 1) in effluent and 2) in serum samples sequentially collected before revascularization, then during the first 120 min of revascularization and first post-operative week. No correlation was observed between serum values (peak) and effluent values. In serum, pre-operative creatine kinase-BB activities were correlated with soluble thrombomodulin levels (p = 0.01). Both increased significantly during the first minutes of the revascularization, then decreased markedly. In contrast, AST activity was maximal at day 1. This detectable and early release of creatine kinase-BB and soluble thrombomodulin in blood is in keeping with the early occurence of endothelial damage. Together with previous data, these findings suggest that serum determination of these two markers may be a useful tool in the assessment of endothelial injury in liver transplantation.

Contribution of true cold and rewarming ischemia times to factors determining outcome after orthotopic liver transplantation

The role of true cold ischemia times (CIT) and rewarming ischemia times (WIT) in determining outcome after liver transplantation was investigated in 230 adult recipients. Using multivariate analysis, WIT (time from the start of implantation until restoration of arterial and portal blood supply) and donor intensive care stay (P = .04 and .0004, respectively) but not CIT (the time from donor portal vein flushing until the graft was removed from University of Wisconsin solution; P > .30) emerged as independent determinants of graft survival. In the small number of patients with a WIT of greater than 180 minutes, there were reductions in graft survival (58% v 80% for WIT greater than 180 minutes) but these just failed to reach significance (P = .055). CIT had no influence on graft survival using cut-offs of 12 or 18 hours. A WIT of greater than 180 minutes was associated with an increased median area under the curve of day 1 through 7 serum bilirubin (1,370 v 915 mumol/L.day; P = .048) and trends towards an increased incidence of primary graft nonfunction or dysfunction (22.2% v 6.2% for WIT of less than 180 minutes; P = .065) and the day 1 through 7 area under the curve of serum aspartate aminotransferase (3,310 v 1,440 IU/L.day; P = .092). A prolonged CIT (greater than 18 hours) led to a prolonged hospital stay (69 v 31 days; P = .03), an increased area under the curve of day 8 through 14 serum bilirubin (2,500 v 995 mumol/L.day; P = .003), and a trend towards an increased incidence of initial poor graft function (33.3% v 6.3% for less than 18 hours; P = .092). The incidence of acute rejection increased (to 64.3% from 53.4%; P = .04) in patients with preservation injury (serum aspartate aminotransferase greater than 1,500 IU/L during the first 2 postoperative days). True CIT and WIT are important determinants of outcome after liver transplantation.

Liver cell necrosis: cellular mechanisms and clinical implications

Based on our current understanding, we have developed a provisional model for hepatocyte necrosis that may be applicable to cell necrosis in general (Figure 6). Damage to mitochondria appears to be a key early event in the progression to necrosis. At least two pathways may be involved. In the first, inhibition of oxidative phosphorylation in the absence of the MMPT leads to ATP depletion, ion dysregulation, and enhanced degradative hydrolase activity. If oxygen is present, toxic oxygen species may be generated and lipid peroxidation can occur. Subsequent cytoskeleton and plasma membrane damage result in plasma membrane bleb formation. These steps are reversible if the insult to the cell is removed. However, if injury continues, bleb rupture and cell lysis occur. In the second pathway, mitochondrial damage results in an MMPT. This step is irreversible and leads to cell death by as yet uncertain mechanisms. It is important to note that MMPT may occur secondary to changes in the first pathway (e.g. oxidative stress, increased Cai2+, and ATP depletion) and that all the "downstream events" occurring in the first pathway may result from MMPT (e.g., ATP depletion, ion dysregulation, or hydrolase activation). Proof of this model's applicability to cell necrosis in general awaits further validation. In this review, we have attempted to highlight the advances in our understanding of the cellular mechanisms of necrotic injury. Recent advances in this understanding have allowed scientists and clinicians a better comprehension of liver pathophysiology. This knowledge has provided new avenues of therapy and played a key role in the practice of hepatology as evidenced by advances in organ preservation. Understanding the early reversible events leading to cellular and subcellular damage will be key to prevention and treatment of liver disease. Hopefully, disease and injury specific preventive or pharmacological strategies can be developed based on this expanding data base.

Expression of cytokine-dependent adhesion molecules in postreperfusion biopsy specimens of liver allografts

BACKGROUND/AIMS

Allogeneic recognition of donor cells by host T lymphocytes requires the expression of cytokine-dependent molecules, such as class II major histocompatibility antigens, intracellular adhesion molecule 1 (ICAM-1), and lymphocyte function-associated antigen 3 (LFA-3). In the liver, activation of Kupffer cells after ischemic injury during the transplantation procedure may result in an early induction of cytokine-dependent molecules.

METHODS

The pattern of induction of ICAM-1, HLA-DR, HLA-DQ, and LFA-3 was investigated in 30 postreperfusion surgical biopsy specimens of liver allografts by an immunohistochemical technique.

RESULTS

Two patterns of induction were observed: focal or diffuse. On hepatocytes, ICAM-1 was induced in 22 cases (11 focal, 11 diffuse), HLA-DR in 18 cases (13 focal, 5 diffuse), HLA-DQ in 13 cases (3 focal, 10 diffuse), and LFA-3 in 1 case (focal). On bile duct cells, HLA-DR was expressed in 19 cases, associated with HLA-DQ in 7 cases. No induction of ICAM-1 and LFA-3 was detected. Compared with the other patients, the group of patients with diffuse postoperative hepatocellular induction of ICAM-1 was characterized by higher preharvesting serum transaminase levels in the donor (P < or = 0.001), suggestive of preoperative ischemic injury, and increased incidence of acute graft rejection (P = 0.04).

CONCLUSIONS

Preoperative warm ischemia may modify the immunogenicity of liver allografts.

Expression of complement-regulatory proteins in normal and UW-preserved human liver

BACKGROUND/AIMS

Somatic cells are protected against complement-mediated injury by specialized membrane proteins, known as complement-regulatory proteins (CRP). The knowledge of the pattern of CRP expression in the liver is important to evaluate the role of complement-mediated injury in graft rejection.

METHODS

We determined the distribution of four main CRP: membrane cofactor protein (MCP), decay accelerating factor (DAF), protectin, and complement receptor 1 (CR1) in 30 histologically normal livers, 13 samples from University of Wisconsin cold-storage solution (UW)-preserved tissue and 17 postoperative biopsies of UW-preserved allografts.

RESULTS

In normal liver, hepatocytes expressed only MCP. Bile duct cells were reactive for MCP and protectin. Sinusoidal endothelial cells expressed MCP and protectin but displayed no or faint expression of DAF. Endothelial cells of portal vessels and centrilobular veins expressed high levels of DAF, MCP, and protectin. No expression of CR1 was observed. No change in CRP expression was usually detected after UW preservation, except for protectin, induced on hepatocytes in 9 samples of UW-preserved liver tissue and in 9 allografts.

CONCLUSIONS

Hepatocytes and sinusoidal endothelial cells, which have a defective expression of CRP, might be at risk for complement-mediated injury. However, this risk is not aggravated after UW preservation.