| Background.Despite of recent considerable improvements in supportive therapy, acute hepatic failure (AHF) remains a dramatic disease which is associated with a mortality rate ranging from 50% to 80%. Orthotopic liver transplantation is the ultimate treatment for AHF. However, numerous patients died while waiting for a transplant due to persistent scarcity of donors. In such a situation, an artificial liver may improve survival by bridging patients to liver transplantation or to regeneration of their diseased livers.Currently, several BAL systems based on hollow fiber technology have been in various stages of clinical trials and proved to have some efficacy. In each design, hepatocytes are inoculated into extracapiliary space, either alone or attached to microcarriers, oxygenized plasma flows through the fiber lumina. Since the oxygen-binding ability of plasma is less than one-tenth that of whole blood, the oxygen supply to hepatocytes in these systems is far short of demand. This problem leads to poor hepatocyte activity and insufficient hepatic function, thereby decreasing the efficiency of these devices.To address the issue of adequate oxygenation of hepatocytes, we developed a novel BAL system, which is based on a direct hemoperfusion typed nonwoven fabric bioreactor containing primary porcine hepatocytes. Thewhole BAL system consisted of a roller pump, a heparin pump, and a bioreactor which was placed in an incubator keeping an internal temperature of 37℃. The nonwoven fabric bioreactor was designed with 500 polysulfone-based hollow fibers, which were spirally wound with nonwoven fabric polytetrafluoroethylene. In this study, we evaluated the efficacy of our novel BAL system using a D-galactosamine induced AHF in pigs.Part I. A D-Galactosamine Induced Acute Hepatic Failure Model in Unanesthetic Pigs.Objective.Recent attempts to study the artificial liver and evaluate its efficacy, safety and mechanism of action have created an urgent need for a clinically relevant animal model of AHF. This research was designed to establish a modified acute hepatic failure model in unanesthetic pigs induced with D-galactosamine (D-gal), and probe its feasibility for testing the safety and effects of artificial liver devices.Materials and Methods.D-gal was freshly dissolved in 150 ml of 5% dextrose in water and adjusted to pH 6.8 with 1 N NaOH, then used within 2 h. After a 6-h fasting with free access to water, anesthesia achieved with an intramuscular administration of Ketamine (4mg/kg). With sterile techniques, one 16-gauge catheter was placed into the external jugular vein of each adult male Buloc breeding pigs (weighing form 10 to 17Kg). Twenty-four hours after the cannulation, fifteen unanesthetic pigs (n=5 of each group) received an intravenous administrationof D-gal at a dose of 1.0, 1.25 and 1.5 g/Kg body weight, respectively. Four pigs which received the same volume of 5% dextrose in water served as controls.Clinical data were recorded, and blood samples were collected for measurement of hemoglobin, platelet, prothrombin time, ammonia, creatinine, blood urea nitrogen, total bilirubin, bile acid, aspartate aminotransferase, glucose and lactate, etc. Animal survival was measured up to 168 h to compare survival between four groups. Complete postmortem examinations were performed on all pigs. Tissue samples of liver, brain, kidney, heart, spleen, lung, pancreas and intestine, etc, were fixed in 10% buffered formalin and stained with hematoxylin and eosin for morphological evaluation.Results.Galactosamine-treated animals presented as acute hepatic failure characterized by a progressive increase in liver enzymes, total bilirubin, bile acid, ammonia, lactate and prothrome time, and associated hypoglycemia and coma. Galactosamine 1.25 and 1.5 g/Kg yielded in 100% mortality, whereas only 60% of pigs which received galactosamine at a dose of 1.0 g/Kg died. The survival time is negatively related with the dose of galactosamine. Histopathological examination of postmortem liver specimens from all of the galactosamine-treated animals demonstrated massive hepatic necrosis, with severe hemorraghe. The histology of postmortem liver specimens from control animals was completely normal.Conclusions.We have developed a modified AHF model in unanesthetic Buloc breeding pigs. The model is very closely resemble the human AHF, and appears promising for future evaluation of artificial liver devices due to its high reproducity, potiential reversibility, nearly universal mortality from liver failureand a proper therapeutic window that is long enough to enable the treatment to be initiated and its effects to be assessed.Part II: Evaluation of a Newly Established Bioartificial Liver in Acute Hepatic Failure Pigs.Objective.Bioartificial liver (BAL) is anticipated to be an effective treatment of choice for patients with acute hepatic failure (AHF). To address the issue of adequate oxygenation of hepatocytes, we developed a novel BAL system, which is based on a direct hemoperfusion typed nonwoven fabric bioreactor containing porcine hepatocytes. In this study, we evaluated the efficacy of our BAL using a D-gal induced AHF in pigs.Materials and Methods.The whole BAL consisted of a roller pump, a heparin pump, and a nonwoven fabric bioreactor which was placed in an incubator keeping an internal temperature of 37 "C. The volume of entire system was approximately 120 ml. The nonwoven fabric bioreactor was designed with 500 polysulfone-based hollow fibers, which were spirally wound with nonwoven fabric polytetrafluoroethylene. The three-dimensional nonwoven fabric provides matrix which allows for hepatocytes immobilization and aggregationat high density (5 * 107 cells/ml). Whole blood flows through the intra-luminal space of hollow fibers into direct contact with hepatocytes for effective oxygen supply, CO2 removal and material exchange.AHF was induced with intravenous administration of D-gal at a dose of 1.3 g/kg in each unanesthetic Buloc breeding pigs (weighing from 20 to 25 Kg). 12 h post D- galactosamine injection, fifteen pigs were divided into: a BAL group (n = 5), in which pigs received the BAL treatment with hepatocytes for 6 h, a sham BAL group (n = 5), in which pigs received the BAL treatment without hepatocytes, and a AHF group (n = 5), in which pigs only received intensive care for purposes of establishing the basic parameters of AHF model. Hepatocytes were isolated from Chinese experimental miniature pigs using a modified four-step Dispase-collagenase perfusion method. 20ml of DMEM medium containing 1.0 to 1.3-109 porcine hepatocytes was inoculated into the extra-fiber space of the bioreactor, whole blood perfused through the intra-luminal space at a rate of 40 ml/min for a duration of 6 h. Blood samples were collected for measurement of hemoglobin, platelet, prothrombin time, ammonia, amino acid, creatinine, urea nitrogen, total bilirubin, aspartate aminotransferase, glucose and lactate, et al. Animal survival was measured up to 168 h to compare survival among three groups. Tissue samples of liver, brain, kidney, heart, spleen, lung, pancreas and intestine, etc, were fixed in 10% buffered formalin and stained with hematoxylin and eosin for morphological evaluation.Results.All pigs tolerated the BAL treatment well. No obvious coagulation was observed in the treatment period. All pigs except two in the BAL group subsequently developed grade IV hepatic encephalopathy. The onset of hepatic encephalopathy was significantly delayed in the BAL group (61±8h) as compared with both control groups (AHF group: 44±14h;sham BAL group: 40±15h, P<0.05).All pigs died, except two in the BAL group survived and were sacrificed 30 and 90 days later, respectively. Survival time was significantly prolonged in the BAL group (106±23h) as compared with both control groups (AHF group: 48±7h;sham BAL group: 43±8 h, P < 0.05).All pigs showed marked increases in prothrombin time, plasma ammonia and lactate acid levels, and a significant decrease in plasma glucose level 12 h post D-gal administration. However, blood ammonia (P < 0.05) and plasma lactate (P < 0.01) levels significantly decreased after BAL treatment, these levels were lower than those in the other two groups, whereas serum glucose levels (P < 0.01) and Fischer index (P < 0.05) were higher in the BAL group than those in the other two groups at the end of BAL treatment.In addition, the hepatocytes viability was 90-99% before BAL treatment, and varied between 75 and 79% at the end of the treatments, as determined by try pan blue exclusion test.Conclusions.We therefore conclude that (1) the newly designed direct hemoperfusion typed nonwoven fabric bioreactor facilitated porcine hepatocytes keep their viability and function in vitro, and (2) the BAL based on a nonwoven fabric bioreactor with porcine hepatocytes appears to be effective in treatment of AHF in pigs, with improvements in plasma ammonia, serum glucose, lactate acid, and a prolonged survival time. The safety and efficacy of this novel system seems to be promising for human clinical care. |
