Thrombomodulin Attenuates Inflammatory Damage Due to Liver Ischemia and Reperfusion Injury in Mice in Toll-Like Receptor 4-Dependent Manner.

Liver ischemia reperfusion injury (IRI) is an important problem in liver transplantation. Thrombomodulin (TM), an effective drug for disseminated intravascular coagulation, is also known to exhibit an anti-inflammatory effect through binding to the high-mobility group box 1 protein (HMGB-1) known as a proinflammatory mediator. We examined the effect of recombinant human TM (rTM) on a partial warm hepatic IRI model in wild-type (WT) and toll-like receptor 4 (TLR-4) KO mice focusing on the HMGB-1/TLR-4 axis. As in vitro experiments, peritoneal macrophages were stimulated with recombinant HMGB-1 protein. The rTM showed a protective effect on liver IRI. The rTM diminished the downstream signals of TLR-4 and also HMGB-1 expression in liver cells, as well as release of HMGB-1 from the liver. Interestingly, neither rTM treatment in vivo nor HMGB-1 treatment in vitro showed any effect on TLR-4 KO mice. Parallel in vitro studies have confirmed that rTM interfered with the interaction between HMGB-1 and TLR-4. Furthermore, the recombinant N-terminal lectin-like domain 1 (D1) subunit of TM (rTMD1) also ameliorated liver IRI to the same extent as whole rTM. Not only rTM but also rTMD1 might be a novel and useful medicine for liver transplantation. This is the first report clarifying that rTM ameliorates inflammation such as IRI in a TLR-4 pathway-dependent manner.

Prevention of initial perfusion failure during xenogeneic ex vivo liver perfusion by selectin inhibition.

BACKGROUND: Endothelial cell activation triggered by xenoreactive antibodies and complement products is the main feature of discordant xenograft rejection. The contribution of early cell-mediated mechanisms to this rejection process is poorly understood, and the function of adhesion molecules in xenogeneic cell interactions in vivo is unclear. The aim of the study was to investigate the role of selectins in mediating cell-dependent initial perfusion failure and functional restrictions in xenoperfused guinea pig (GP) livers. METHODS: Isolated GP livers were hemoperfused in a flow-constant, recirculating perfusion system via the portal vein. Microhemodynamic parameters such as sinusoidal perfusion rate and leukocyte flux were analyzed using intravital fluorescence microscopy. Hepatic oxygen consumption and bile production, as well as liver enzymes, potassium level, and numbers of white blood cells and platelets in the perfusate, were determined. The GP livers were perfused either with GP blood (control perfusion), with unmodified rat blood (xenoperfusion), or with rat blood treated with the selectin-blocking polysaccharide Fucoidin. RESULTS: A significant sinusoidal perfusion failure was observed in the xenoperfusion group, which was accompanied by distinct signs of a functional restriction-like reduced oxygen consumption, bile production, and increased perfusion pressure. However, there were significantly fewer impairments in the Fucoidin group. Furthermore, fewer platelets were trapped and a smaller number of stagnant leukocytes were observed in this group. CONCLUSION: Fucoidin did not suppress complement activation during xenoperfusion. Considering that Fucoidin inhibits the selectin-dependent interactions among white blood cells, platelets, and sulfate-containing proteoglycans on the surface of vascular endothelium, these findings suggest an important role for early cellular interactions in the development of organ failure during xenogeneic rejection.

Inhibitory effect of plasma FKBP12 on immunosuppressive activity of FK506.

To evaluate the roles of extracellular FKBP12, we examined the effect of extracellular FKBP12 on the immunosuppressive activity of FK506 in vitro and clinically. The ability of FK506 to suppress phytohemagglutinin-induced proliferative response of human peripheral blood mononuclear cells was inhibited in the presence of recombinant FKBP12 dose-dependently. We measured plasma levels of FKBP12 using a newly developed enzyme-linked immunosorbent assay system in 34 patients receiving FK506 after liver transplantation. In 7 patients with acute cellular rejection, plasma FKBP12 increased significantly at the onset of rejection compared with 1 week before onset (P < 0.05) and further increased to or remained at more than 250 ng/ml 1 week after onset. In 22 of 27 patients without acute cellular rejection, plasma FKBP12 was less than 70 ng/ml during the 4 weeks after transplantation. In the other 5 of 27 patients without acute cellular rejection, plasma FKBP12 exceeded 250 ng/ml. Rapid increase of plasma FKBP12 was observed in only one of these 5 patients, at the onset of high fever due to a liver abscess. There was no significant difference in whole blood trough levels of FK506 between the patients with or without acute cellular rejection. These results suggest that the rapid increase in plasma levels of FKBP12 may contribute to the occurrence and progress of acute cellular rejection probably by inhibiting the immunosuppressive activity of FK506.

The protective role of Kupffer cells in humoral injury of xenoperfused rat livers.

BACKGROUND: The role of Kupffer cells in a hepatic xenograft rejection is still unclear. We investigated the effect of blocking Kupffer cells on xenogeneic humoral injury using rat livers as the xenoperfusion models. METHODS: Rat livers were perfused with fresh human blood after pretreatment either with normal saline (group 1; n = 8) or with gadolinium chloride (GdCl3) solution (group 2; n = 8). Tissue injury was evaluated by alanine aminotransferase release and histological examination. Tumor necrosis factor-alpha (TNF-alpha) production from rat livers was measured by enzyme-linked immunosorbent assay and also examined by immunohistochemistry. In addition, Kupffer cells were isolated after pretreatment either with normal saline or with GdCl3 solution and incubated with human serum. Localization of human C3 and IgM was examined by immunofluorescence. RESULTS: Alanine aminotransferase release in group 2 was significantly higher than in group 1 (P = 0.015). Histological examination revealed more severe tissue injury in group 2. The mean TNF-alpha level was not significantly different between the two groups. In immunohistochemistry, TNF-alpha was positive primarily on vascular endothelial cells in both groups. Immunofluorescence of saline-treated Kupffer cells showed an uptake of human C3 in the cytoplasm, whereas no uptake was observed in GdCl3-treated cells. The uptake of human IgM did not differ between the two groups. CONCLUSIONS: These results suggest that Kupffer cells have a protective role in preventing xenogeneic humoral injury. Their ability to absorb xenogeneic complements may contribute to this protective mechanism.

Hepatic artery thrombosis in living related liver transplantation.

BACKGROUND: Hepatic artery thrombosis (HAT) after orthotopic liver transplantation remains a significant cause of graft loss in pediatric patients. We previously reported that the microsurgical techniques for arterial anastomosis can reduce the incidence of HAT in living related liver transplantation (LRLT). The purpose of this study is to analyze the risk factors for HAT after LRLT. A total of 245 patients received 250 liver transplants. METHODS: Eight arteries in eight patients, reconstructed with the use of loupe magnification (HAT; 1/8, 12.5%), were excluded from this study. We observed HAT in 4 patients of the 242 transplants (1.7%, HAT group). Seventeen factors were compared between the HAT and the control group (those without HAT). RESULTS: HAT occurred in 3 of 33 grafts (9%) from ABO-incompatible donors, whereas it occurred in 1 of 209 grafts (0.5%) from identical or compatible donors (P=0.008). The corrected volume of fresh-frozen plasma intraoperatively transfused in the HAT group (46.9+/-30.3 ml/kg) was significantly (P=0.015) different from that in the control group (10.2+/-1.9 ml/mg). In all four patients with HAT, emergent revisions of the anastomosis were performed. Two patients with ABO-incompatible grafts died of hepatic failure and sepsis. CONCLUSIONS: Although microsurgical techniques can minimize the surgical risk factors for HAT, overtransfusion of fresh-frozen plasma in high-risk patients (ABO incompatible) may be a critical factor in the development of HAT in LRLT.

Humoral injury in porcine livers perfused with human whole blood.

BACKGROUND: We investigated the influence of humoral injury during xenoperfusion of porcine livers by human blood. METHODS: The porcine livers were perfused under physiological conditions for 9 hr. The perfusates consisted of porcine whole blood in group 1, human whole blood in group 2, and human whole blood with soluble complement receptor type 1 (300 microg/ml) in group 3. RESULTS: Liver enzyme release and serum hemoglobin in group 2 increased significantly after 3 hr of xenoperfusion, compared with those in group 1 and group 3 (P<0.05). Severe histological damage with minimal cellular infiltration was observed in group 2 after 6 hr of xenoperfusion, but was present only at trace levels in group 1 and group 3. In group 2, von Willebrand factor, a possible target of natural antibodies, was induced on sinusoidal endothelial cells after 3 hr of xenoperfusion, correlating with diffuse deposition of human IgM and membrane attack complex. In group 3, von Willebrand factor, human IgM, and membrane attack complex staining in the intralobular region were present at trace levels. In group 3, the indocyanine green removal capacity, representing hepatocyte function, was significantly higher than in group 2 (P<0.05). CONCLUSIONS: Based on these results, we suggest that humoral injury is a major cause of liver damage during liver xenoperfusion. The pattern of humoral injury in xenoperfused livers may be attributed to anatomical features of the liver and unique responses of sinusoidal endothelial cells to xenoperfusion.

Successful long-term xenoperfusion of the pig liver: continuous administration of prostaglandin E1 and insulin.

For clinical utilization of extracorporeal liver perfusion as an artificial liver assist device, we examined the possibility of long-term xenoperfusion of the pig liver by the continuous administration of prostaglandin E1 (PGE1) and insulin. After a 3-hr perfusion period, pig livers that were xenoperfused with human blood exhibited a drastic decrease in the perfusate volume, a progressive elevation of the hepatic artery pressure, a gradual deterioration of bile production, and a marked increase in the release of creatine kinase-BB component. The continuous administration of PGE1 (25 microg/hr) and insulin (1 U/hr) significantly improved these derangements (P<0.05) and allowed stable perfusion for up to 9 hr. This manipulation also inhibited leukocyte aggregation in the graft, the characteristic perfusate hemolysis, and acceleration of ketogenesis. Histological examination revealed that the interlobular edema and hemorrhage, characteristics of tissue injuries in xenogeneic hyperacute rejection, were markedly alleviated in the PGE1 and insulin-treated group. This study clarifies the finding that the combined administration of PGE1 and insulin is effective for long-term xenogeneic extracorporeal liver perfusion, with the graft viability well maintained.

Evaluation of ammonia and lidocaine clearance, and galactose elimination capacity of xenoperfused pig livers using a pharmacokinetic analysis.

BACKGROUND: We introduced the pharmacokinetic method into the functional evaluation of xenogeneic extracorporeal liver perfusion as an artificial liver assist device, and examined the influence of xenogeneic humoral injury on the metabolic function of xenoperfused pig livers. METHODS: Isolated pig livers were perfused with fresh porcine blood (group 1; n=5) or fresh human blood (group 2; n=5) for 9 hr. Clearance (CL) of ammonia and lidocaine, and galactose elimination capacity (Vmax) were determined at three points during the perfusion using a one-compartment pharmacokinetic model. RESULTS: Concentrations of ammonia and lidocaine decreased exponentially and those of galactose decreased linearly after a bolus injection in both groups. A one-compartment model provided satisfactory curve fittings for these test substances. No decreases of ammonia CL, lidocaine CL, or galactose Vmax were observed until 9 hr in either group. No differences were observed between the two groups with respect to these metabolic functions. In group 1, only slight interlobular edema was observed at 9 hr. In group 2, membrane attack complex was diffusely deposited at 3 hr and severe interlobular damage was histologically observed at 9 hr, although hepatocellular damage was minimal even at 9 hr. Alpha glutathione S-transferase and mitochondrial aspartate aminotransferase were comparable between the two groups. CONCLUSIONS: Pharmacokinetic analysis allowed the evaluation of ammonia CL, lidocaine CL, and galactose Vmax of the perfused pig livers. Despite xenogeneic humoral injuries, the xenoperfused livers maintained these metabolic functions at the same levels as the alloperfused livers for 9 hr.

Reduction of hepatic microcirculatory failure caused by normothermic ischemia/reperfusion-induced injury by means of heat shock preconditioning.

Transient sublethal hyperthermia and the recovery from this exposure to heat (heat shock preconditioning) provides a cytoprotective effect on oxidative insults through an intracellular protective response, heat shock response. The impact of heat shock preconditioning on hepatic microvascular failure, which is a causative determinant of ischemia/reperfusion-induced injury of the liver, was investigated by using intravital fluorescence microscopy. In Sprague-Dawley rats, normothermic ischemia was induced by totally clamping the hepatoduodenal ligament for 20 min, followed by 120 min of reperfusion. Heat shock preconditioning was performed by whole-body hyperthermia (42 degrees C for 15 min) and subsequent 48 h recovery. In accordance with the prominent induction of heat shock protein 70 in the liver tissue, the postischemic decrease in sinusoidal perfusion rate and sinusoidal diameter, and the postischemic increase in the number of stagnant leukocytes in sinusoids and adherent leukocytes in postsinusoidal venules were significantly attenuated in the heat shock-treated animals. Furthermore, liver enzyme release (glutamate pyruvate transaminase and alpha-glutathione S-transferase) was significantly reduced and postischemic deterioration of bile production was attenuated. The 7-day survival rate after 20-minute ischemia was significantly improved from 50% to 80% (heat shock-nontreated group vs. heat shock-treated group, P < 0.05). These results indicate that heat shock preconditioning attenuates ischemia/reperfusion-induced hepatic injury by preventing postischemic microvascular disturbances, and that its protective effect is circumstantially associated with the concomitant induction of heat shock protein 70.

Experiences of 120 microsurgical reconstructions of hepatic artery in living related liver transplantation.

BACKGROUND: We reviewed 120 microsurgical reconstructions of a hepatic artery in living related liver transplantation and discussed the problems encountered. METHODS: From January 1991 to July 1994 we performed a series of 105 living related liver transplantations on children with end-stage liver disease. Arterial reconstruction was performed under the optical field of a continuous zoom magnification of approximately 10 times with an operating microscope. RESULTS: Twenty-six percent of the graft arteries were less than 2 mm in diameter. The time required for an arterial reconstruction was 49.5 +/- 1.8 minutes. In 15 of the 31 cases in which there were two graft arteries, two arterial reconstructions were required. The caliber differences between the graft artery and the recipient artery in 30 instances was dealt with by cutting an undersized artery obliquely (17 instances), by fish-mouth method (10 instances), by end-to-side anastomosis (1 instance), or by funnelization method (2 instances). In one case we performed an intimal dissection of a recipient hepatic artery and substituted a splenic artery. Consequently, hepatic arterial thrombosis occurred in only two cases (1.7%). CONCLUSIONS: Microsurgical technique has overcome the high risk of hepatic arterial thrombosis in cases of fine graft arteries, enabled the reconstruction of arteries with caliber difference, and decreased arterial complications with its delicate manipulation.

In situ pedicle resection in left trisegmentectomy of the liver combined with reconstruction of the right hepatic vein to an inferior vena caval segment transpositioned from the infrahepatic portion.

A combination of an in situ pedicle resection of the liver and a hepatic vein reconstruction using a cranially transpositioned segment of the IVC after implantation of an ePTFE graft at the missing IVC was useful in treating a patient who suffered from a huge liver tumor involving all of the hepatic venous confluence and the IVC. Although early tumor recurrence remains an unresolved problem for such patients, a surgical approach is feasible. This technique can be justified as a therapeutic modality, not only because it improves quality of life, but also because it provides patients with an opportunity for additional treatment.

Dermatomyositis with splenic and renal infarctions during corticosteroid therapy.

A 60-year-old woman was admitted to our hospital with complaints of muscle weakness and erythema on her extremities. Gottron's sign, heliotrope rash, elevation of serum myogenic enzymes, electromyography and magnetic resonance imaging findings established a diagnosis of dermatomyositis (DM). She was treated with 60 mg of daily prednisolone. One week later, she suddenly developed splenic and renal infarctions, which were considered to have resulted from vasculopathy associated with DM. Cyclophosphamide and anticoagulants along with increasing the dosage of corticosteroid were effective. This is the first report describing splenic and renal infarctions in a patient with adult-onset DM.

[Resection and reconstruction of the inferior vena cava for major hepatic resection].

The inferior vena cava (IVC) is partially or segmentally resected in major hepatic resection for malignant hepatic tumors in case of possible direct invasion to the IVC wall or IVC tumor thrombosis. The reconstruction methods of the IVC are divided into three categories depending on the degree of IVC resection: simple suture; patch repair; and segmental replacement. In segmental replacement, a synthetic material such as a cylindrical expanded polytetrafluoroethylene (ePTFE) grafts is widely utilized as a substitute. The total hepatic vascular exclusion technique is usually necessary in concomitant resection of the suprahepatic IVC. When a longer duration of hepatic vascular exclusion is required to resect and reconstruct the suprahepatic IVC and hepatic vein confluence, in situ hypothermic perfusion, the ante situm technique, or ex vivo bench surgery must be applied. When an ePTFE graft is replaced in the resected IVC, a Carrel patch of the IVC is used for the hepatic vein orifice to maintain anastomotic patency. Alternatively, the hepatic vein can be reanastomosed to an inferior vena caval segment transpositioned from the intact infrahepatic IVC portion by replacing the resected infrahepatic IVC with an ePTFE graft.