Drugs and the liver

The liver is a major organ with multiple functions. A number of drugs are metabolized by the liver during phase 1 and 2 reactions which include complex processes involving cytochrome P450 enzymes. Genetic and acquired variability in cytochrome P450 activity may have profound effects on pharmacokinetics. Additionally, drugs can also modify how the liver functions and cause dysfunction or even failure of the organ both by a direct effect on the liver or by alteration in liver blood flow. It is important to recognize the signs and symptoms of liver failure in patients and identify possible causes including drug interactions. Furthermore, once a patient has been recognized to be suffering with liver dysfunction or failure drug choice and dosing regime will need to be rationalized.Paracetamol overdose can have severe and life-threatening consequences for patients due to its effect on liver function. It is the leading cause of acute liver failure in the UK,¹ Correct and early management is crucial and will be discussed within this article.     

Drugs and the liver

The liver is a major organ with multiple functions. A number of drugs are metabolized by the liver during phase 1 and 2 reactions which include complex processes involving cytochrome P450 enzymes. Genetic and acquired variability in cytochrome P450 activity may have profound effects on pharmacokinetics. Additionally, drugs can also modify how the liver functions and cause dysfunction or even failure of the organ both by a direct effect on the liver or by alteration in liver blood flow. It is important to recognize the signs and symptoms of liver failure in patients and identify possible causes including drug interactions. Furthermore, once a patient has been recognized to be suffering with liver dysfunction or failure, drug choice and dosing regime will need to be rationalized. Paracetamol overdose can have severe and life-threatening consequences for patients due to its effect on liver function. It is the leading cause of acute liver failure in the UK. Correct and early management is crucial and will be discussed within this article.     

Hepatocellular metabolic change after orthotopic liver transplantation in rats

The hepatocellular metabolic change after liver transplantation following 2 hr cold ischemia was investigated. Of 55 orthotopic liver transplantations in male Wistar rats, 47 animals were sacrificed at 3 hr, and 1, 2, 7 and 30 days to determine hepatic metabolite levels, in the form of adenine nucleotides, lactate and glycogen. Using the other 8 recipients, biochemical examinations were done at 1, 3, 5, 7, 30 and 60 days and metabolic levels estimated at 60 days. The SGOT and SGPT levels decreased gradually after a remarkable increase on the first postoperative day, while the alkaline phosphatase level revealed a peak value at 30 days. All levels recovered to within the normal range in 60 days. The total adenine nucleotide level reached the normal range within 3 hr following the blood reflow and remained at a normal level thereafter. However, all the metabolic levels apart from total adenine nucleotides deteriorated to reach their worst level at 7 days. The results of this investigation indicate that the posttransplanted deterioration of metabolic levels were possibly caused by the imperfect oxygenation due to cellular edema after blood reflow. However, the levels of these metabolites recovered within 60 days after transplantation.     

体外生物人工肝系统对暴发性肝衰竭兔的支持作用

目的探讨培养肝细胞用于生物人工肝及其作为肝移植辅助支持手段的可能性。方法以培养人肝细胞和中空纤维反应器为主要材料构成体外生物人工肝系统,对Ｄ-氨基半乳糖诱导的暴发性肝衰竭（ＦＨＦ）免进行人工肝支持实验。结果尽管两组实验动物的存活时间没有明显差异,但支持治疗组兔的血清转氨酶、总胆红素和肌酐水平均低于对照组,肝组织病理检查见肝细胞坏死程度明显轻于对照组,实验所用肝细胞保持较好的活力和贴壁能力。结论所用体外生物人工肝支持系统已发挥出培养肝细胞的生物作用,能够部分代偿ＦＨＦ兔的肝脏功能。【关键词】##4人工肝;;支持;;暴发性肝衰竭;;兔     

Acute liver failure

Acute liver failure (ALF) is a rare but life-threatening disease with varying aetiologies worldwide. Drug-induced liver injury, including paracetamol poisoning, is the main cause in Europe and the USA. Whereas in the developing world, viral hepatitis is most common. ALF is a multisystem illness that leads to development of hepatic encephalopathy, cerebral oedema, vasodilatory shock, coagulopathy, hypoglycaemia and multiple-organ failure. Early referral to a specialist liver unit is essential. The core principles of ALF management are to identify/treat the underlying cause, provide supportive care and treat any complications. Optimal management will allow time for spontaneous liver regeneration or liver transplantation, and result in improved survival rates. This article provides an overview of the key concepts in ALF diagnosis and management.     

Relationship of deranged energy metabolism in liver and kidney to arterial ketone body ratio following liver ischemia in rats

Changes in energy metabolism in the liver and kidney in liver ischemia induced in rats were simultaneously studied, in terms of energy charge (EC) and mitochondrial oxidoreduction state. Mean arterial blood pressure, glucose and lactate, total ketone bodies (acetoacetate+β-hydroxybutyrate) and the ketone body ratio in arterial blood (KBR) were also investigated. During and after liver ischemia, both organs showed similar patterns of reversibility, and KBR, which reflects the mitochondrial oxidoreduction state, correlated well with EC, in both organs. Referring to the mortality and changes in substrates above mentioned, KBR is a pertinent parameter for detection of viability following induced liver ischemia. It was also suggested that KBR may indicate a regulation role by the liver, in kidney energy metabolism.     

Artificial liver: Review and Cedars-Sinai experience

During the past decade, whole organ transplantation has become the only clinically effective method of treating fulminant hepatic failure and chronic liver failure due to specific genetic, hepatocellular, and anatomic defects of liver function. However, wider application of liver transplantation is restricted by shortage of organ donors, high cost, relatively high morbidity, and need for life-long immunosuppression. As a result, investigators have attempted to develop alternative methods to treat liver insufficiency. These ranged from use of plasma exchange to utilization of detoxification columns and extracorporeal devices loaded with various liver tissue preparations. Recently, advances in hepatocyte isolation and culture techniques, improved understanding of hepatocyte-matrix interactions, availability of new biomaterials, improved hollow-fiber technology, and better understanding of flow and mass transport across semipermeable membranes have resulted in the development of a new generation of liver assist devices. Some of these devices, including the one developed by the authors, are currently being tested in the clinical setting. In this paper, the past experience with liver support systems is reviewed, the present status of the field is critically examined, and the results of a phase I clinical trial with the bioartificial liver, utilizing primary porcine hepatocytes, are summarized. Received for publication on Feb. 2, 1998; accepted on Feb. 2, 1998     

Fulminant hepatic failure post liver transplantation: clinical syndromes,correlations and outcomes

This paper reports the clinical syndrome of fulminant hepatic failure (FHF) following liver transplantation. FHF was defined as the sudden onset of liver failure [encephalopathy and prolonged International Normalised Ratio (INR)] without arterial thrombosis in the setting of a liver allograft. FHf post-transplant was seen in 8/154 (5.2%) adult patients undergoing transplantation. These eight patients developed a clinical syndrome characterised by: (a) a rapid rise in ALT levels to above 1000 U/l (mean maximum 1600 U/l), (b) a sudden increase in the INR to above 5 (mean maximum 5.6), (c) the development of high fever, (d) the persistence of thrombocytopenia (mean nadir 40×10⁹/dl), (e) a progressive rise in the bilirubin (mean maximum 400 mol/l) and (f) the development of hepatic encephalopathy. In seven cases this syndrome occurred following good initial graft function at day 6 post (mean)-transplant. In one case the above syndrome developed immediately after liver transplantation. Four of the eight patients developed multiorgan failure associated with systemic acidosis (mean pH 6.84). All of these patients died (mean day 11). Four patients developed systemic alkalosis. Two of these four patients underwent successful retransplantation (on days 12 and 13) and remain alive at a mean of 11 months post-transplant. Six of the eight patients received OKT3 therapy without any apparent affect on clinical outcome. Compared to a control group of patients (n=28), 2/8 versus 2/28 had a positive crossmatch with donor lymphocytes (P=NS), 1/8 versus 7/28 were ABO-non-identical (P=NS), 3/8 versus 10/21 had total MHC mismatches (P=NS) and 5/7 versus 6/16 had UW ischemic times above 10 h (P=NS). No patients had main hepatic artery thrombosis on angiography although four patients had evidence of intrahepatic microthrombi or arterial necrosis at autopsy. In all cases the histology showed massive haemorrhagic necrosis. Three cases had evidence of veno-occlusive lesions whilst foam cell arteriopathy was seen in two cases. Immunofluorescence was performed in three cases. In two cases there was evidence of immunoglobulin, complement and fibrin deposition in blood vessels. In conclusion, we describe an uncommon clinical syndrome occurring post liver transplant. This syndrome represents humorally mediated allograft rejection but there seems to be no relationship with tissue matching (antibody, ABO, MHC) or donor ischaemic times. If recognised earlier in the absence of multiorgan failure, urgent retransplantation seems to be the only effective therapy.     

Changes in liver regenerative factors in a case of living-related liver transplantation

Liver regeneration in a patient with fulminant hepatic failure (FHF) who underwent living-related partial liver transplantation (LRLT) was investigated regarding hepatic growth factors. The patient was a 16-yr-old Japanese male who developed severe subacute FHF. LRLT was performed using an extended left lobe of the ABO matched patient's mother. In the recipient, the pre-transplant levels of both plasma hepatocyte growth factor (HGF) and transforming growth factor (TGF)-beta were extremely high and rapidly decreased following the liver replacement. The liver volume evaluated using a CAT scan increased 195% after 2 wk in graft liver and 110% after 2 wk in the hepatectomized donor. The explanted liver (FHF liver), the liver from donor (normal liver), and the graft liver [the 3rd post-transplant day (POD 3)] were all investigated immunohistochemically. FHF liver: No liver regeneration was observed [proliferative cell nuclear antigen (PCNA) labeling index (L.I.): 0%]. In the liver, both HGF in the hepatocytes and c-met on the membrane of the hepatocytes were positive. TGF-beta was positive in the hepatocytes and no apoptosis was detected by the TUNEL method. Donor liver (POD 0): Few PCNA stained hepatocytes were detected. No HGF was detected but c-met was clearly detected on the cell membrane of the hepatocytes. Neither TGF-beta nor apoptosis was detected. Graft liver (POD 3): The PCNA L.I. was conspicuous at 40%. HGF was positive in non-parenchymal cells and c-met was positive in the cytoplasm of the hepatocytes. TGF-beta was negative while apoptosis was positive in the zone 3 hepatocytes. In conclusion, these findings suggested that the liver of the patient with FHF did not respond to liver regenerative stimulus, in part, through involvement of inhibitor TGF-beta. On POD 3, the transplanted graft was in a vigorous regenerative status in comparison to that in the hepatectomized donor. The HGF/c-met system is thought to be involved in the mechanism of regeneration. Intrahepatic apoptosis was detected in the graft on the 3rd post-transplant day probably due to transient ischemia in the liver, which was not related to the Fas/Fas-ligand system.     

Recovery of hepatic function determined by cytochrome P450-dependent drug metabolism lags after compensatory hepatic volume changes after portal vein ligation in rats

BACKGROUND: Clinically, portal vein embolization has been proven to be useful as a preoperative treatment for major hepatic surgeries with impaired liver function. However, its effects on the metabolism and elimination of various drugs after portal vein embolization or ligation remain to be elucidated. MATERIALS AND METHODS: A portal vein branch that perfuses the central and left lobes of the liver of male Wistar rat was ligated, and changes in the weights of ligated and nonligated lobules as well as hepatic levels and activities of cytochrome P450 (CYP) isoforms, such as CYP3A2 and CYP2C11, were determined. To evaluate in vivo the effect of PVL on hepatic drug metabolism, the narcotic activity (sleep time) of midazolam, a specific substrate for CYP3A2, was measured. RESULTS: Although plasma levels of alanine aminotransferase and hepatic weight returned to basal levels at day 7 after the portal vein ligation, hepatic activities of CYP3A2 and CYP2C11 still remained low (53% and 54% of control levels, respectively), and returned to their initial levels after about day 14. The metabolism of midazolam was prolonged by approximately three times at day 7 after ligation and returned to basal levels at day 14. CONCLUSIONS: Because hepatic CYP-dependent drug metabolism by CYP isoforms recovered more slowly than the apparent recovery of hepatic volume and plasma alanine aminotransferase levels, the therapeutics of drugs metabolized by the CYP isoforms should be used carefully in patients who receive major hepatectomy with portal vein branch embolization.     

Bioartificial liver support devices: historical perspectives

Fulminant hepatic failure (FHF) is an important cause of death worldwide. Despite significant improvements in critical care therapy there has been little impact on survival with mortality rates approaching 80%. In many patients the cause of the liver failure is reversible and if short-term hepatic support is provided, the liver may regenerate. Survivors recover full liver function and a normal life expectancy. For many years the only curative treatment for this condition has been liver transplantation, subjecting many patients to replacement of a potentially self-regenerating organ, with the lifetime danger of immunosuppression and its attendant complications, such as malignancy. Because of the shortage of livers available for transplantation, many patients die before a transplant can be performed, or are too ill for operation by the time a liver becomes available. Many patients with hepatic failure do not qualify for liver transplantation because of concomitant infection, metastatic cancer, active alcoholism or concurrent medical problems. The survival of patients excluded from liver transplantation or those with potentially reversible acute hepatitis might be improved with temporary artificial liver support. With a view to this, bioartificial liver support devices have been developed which replace the synthetic, metabolic and detoxification functions of the liver. Some such devices have been evaluated in clinical trials. During the last decade, improvements in bioengineering techniques have been used to refine the membranes and hepatocyte attachment systems used in these devices, in the hope of improving function. The present article reviews the history of liver support systems, the attendant problems encountered, and summarizes the main systems that are currently under evaluation.     

Immediate early genes and p21 regulation in liver of rats with acute hepatic failure

It has been observed that liver regeneration in acute hepatic failure (AHF) is suppressed [Eguchi et al. Hepatology 1996;24(6):1452-9]. The molecular mechanism regulating this inhibition is not known. We previously reported that in AHF rats, hepatocyte proliferation was significantly impaired with elevation in serum IL-6, TGF-beta1, and HGF [Kamohara et al. Biochem Biophys Res Commun 2000;273(1):129-35]. Following either 70% partial hepatectomy (PH) or liver injury, quiescent mature hepatocytes are "primed" to re-enter the cell cycle. The process of "priming" appears to be triggered by extracellular cytokines (IL-6 and TNF-alpha) and is characterized by expression of immediate early genes. Under the stimulation of growth factors such as HGF, "primed" hepatocytes exit the G1 phase of the cell cycle. G1-associated cyclins and their inhibitors play a pivotal role in G1/S cell cycle transition. Here, we demonstrate that immediate early gene (i.e. c-myc, c-fos) expression and AP-1 activity are preserved in AHF rat livers despite absence of hepatocyte proliferation. In contrast, p21 mRNA and protein are both over-expressed in AHF livers compared to livers from rats undergoing PH; this elevation leads to inhibition in Cdk2 activity, resulting in G1 cell cycle arrest and inhibition of regeneration.     

Continuous peritransplant assessment of consciousness using bispectral index monitoring for patients with fulminant hepatic failure undergoing urgent liver transplantation

Hwang S, Lee SG, Park JI, Song GW, Ryu JH, Jung DH, Hwang GS, Jeong SM, Song JG, Hong SK, Lim YS, Kim KM. Continuous peritransplant assessment of consciousness using bispectral index monitoring for patients with fulminant hepatic failure undergoing urgent liver transplantation.
Clin Transplant 2010: 24: 91–97. © 2009 John Wiley & Sons A/S. Abstract: Background: Rapid deterioration of consciousness is a critical situation for patients with fulminant hepatic failure (FHF). Bispectral (BIS) index was derived from electroencephalography parameters, primarily to monitor the depth of unconsciousness. Aim: To assess the usability of peritransplant BIS monitoring in patients with FHF. Methods: A prospective study using peritransplant BIS monitoring was performed in 26 patients with FHF undergoing urgent liver transplantation (LT). Results: Pre‐transplant Child‐Pugh score was 12.2 ± 1.0; model for end‐stage liver disease score was 32.4 ± 4.4; Glasgow coma score (GCS) was 9.9 ± 1.3; and BIS index was 44.0 ± 6.7. Pre‐transplant sedation significantly decreased BIS index. After LT, all patients having endotracheal intubation recovered consciousness within one to three d and showed progressive increase in BIS index, which appeared slightly earlier and was more evident than the increase in derived GCS score. There was a significant correlation between BIS index and derived GCS scores (r² = 0.648). Timing of eye opening to voice was matched with BIS index of 66.3 ± 10.4 and occurred 12.7 ± 8.3 h after passing BIS index of 50. Conclusion: These results suggest that BIS monitoring is a non‐invasive, simple, easy‐to‐interpret method, which is useful in assessing peritransplant state of consciousness. BIS monitoring may therefore be a useful tool during peritransplant intensive care for patients with FHF showing hepatic encephalopathy.     

Comparative study of bioartificial liver support and plasma exchange for treatment of pigs with fulminant hepatic failure

Recently, bioartificial liver (BAL) treatment was reported to provide beneficial effects for patients with fulminant hepatic failure (FHF). Some success in experimental or clinical trials has been reported; however, the evaluation of BAL efficacy remains unclear, especially in comparison with other treatments for FHF. The purpose of this study was to compare the efficacy between BAL and plasma exchange (PE) in experimentally induced FHF in pigs. Pigs undergoing hepatic devascularization (HD) were placed into the following groups: no treatment (control; n = 6), BAL treatment (BAL; n = 5), and plasma exchange (PE; n = 5). Each treatment was initiated 6 h after HD and lasted for 4 h. BAL treatment significantly improved liver functions in FHF pigs. The decrease in cerebral perfusion pressure was also significantly suppressed in the pigs with BAL, and their survival time was prolonged compared with the results in pigs with PE. The effects of BAL outperform those of PE in the treatment of experimental FHF model.     

The renal cytochrome P-450 arachidonic acid system

In addition to cyclooxygenase and lipoxygenase, arachidonic acid (AA) is metabolized by the cytochrome P-450 monooxygenase system. The kidney is one of the major extrahepatic tissues that display cytochrome P-450 enzyme activities, in particular the cortex, specifically the proximal tubule demonstrate the highest concentration. AA is metabolized by the renal cytochrome P-450 epoxygenase and /-1 hydroxylases to epoxyeicosatrienoic acids and /-1 alcohols (20- and 19-mono-hydroxyeicosatetraenoic acids), respectively. These metabolites possess a broad spectrum of biological and renal effects which include: vasodilation, vasoconstriction, inhibition and stimulation of Na⁺–K⁺-ATPase, inhibition of ion transport mechanisms, natriuresis, inhibition of renin release and stimulation of cell growth. These metabolites are endogenous constituents of the kidney and are present in urine with increasing concentration under pathological conditions such as pregnancy-induced hypertension. The cytochrome P-450-dependent metabolism of AA is specifically localized to the proximal tubule and exhibits developmental changes, i.e., renal production of metabolites is very low in the fetus, newborn and up to 3 weeks of age, after which a remarkable increase in enzyme activities is observed. These characteristics call attention to the importance of this enzyme system in producing cellular mediators for regulating renal function in normal and diseased states.     

Compressed spectral arrays of patients with fulminant hepatic failure in hepatic coma undergoing liver transplantation

Assessing the coma status of patients with fulminant hepatic failure (FHF) is important for determining the reversibility of brain damage and for properly timing liver transplantation. The compressed spectral array (CSA) method is a frequency analysis technique that processes electroencephalogram signals by computer to facilitate on-line interpretation. This method has been used to monitor the consciousness levels of neurointensive care unit patients. In this study, we determined whether CSA could be used to assess the coma status of patients with FHF, and whether CSA provided information that was useful in deciding when to proceed with liver transplantation. CSA recording was carried out in 17 FHF patients with encephalopathy (coma grade III-IV) who underwent living-related liver transplantation between August 1997 and May 1999. Recording was performed with a Neuromonitor OEE-72044 (NIHON KOHDEN, Osaka, Japan) every 24 h before and after transplantation, until the patients regained consciousness. The CSAs of healthy controls were distributed almost equally between 0 and 16 Hz. The CSAs of FHF patients in hepatic coma were classified into three patterns. Eight of the 17 patients showed very prominent slow waves of about 2 Hz (group A), and seven patients showed strongly suppressed rapid waves between 8 and 16 Hz (group B). The remaining two patients showed CSA patterns that were similar to those of healthy controls, even though these patients were comatose (group C). Abnormal CSA patterns were observed in 15 of the 17 patients (88%). Group B patients seemed to have higher coma grades than did group A patients. Sixteen patients underwent liver transplantation, completely recovered from hepatic encephalopathy, and subsequently showed CSA patterns similar to those of healthy controls. One patient died without regaining consciousness. These results suggest that CSA is useful in assessing the coma status of FHF patients and in evaluating electrophysiological recovery from hepatic coma after liver transplantation.     

Incorporating dynamics for predicting poor outcome in acute liver failure patients

Acute liver failure (ALF), also known as fulminant hepatic failure (FHF), is a devastating clinical syndrome with a high mortality of 60%-90%. An early and exact assessment of the severity of ALF together with prediction of its further development is critical in order to determine the further management of the patient. A number of prognostic models have been used for outcome prediction in ALF patients but they are mostly based on the variables measured at one time point, mostly at admission. ALF patients rarely show a static state: rapid progress to a life threatening situation occurs in many patients. Since ALF is a dynamic process, admission values of prognostic variables change over time during the clinical course of the patient. Kumar et al developed a prognostic model [ALF early dynamic (ALFED)] based on early changes in values of variables which predicted outcome. ALFED is a model which seems to be worthwhile to test in ALF patients in other parts of the world with different aetiologies. Since the exact pathophysiology of ALF is not fully known and is certainly complex, we believe that adding promising variables involved in the pathophysiology of ALF to the dynamic approach might even further improve prognostic performance. We agree with Kumar et al that an improved dynamic prognostic model should be based on simplicity (easily to be performed at the bedside) and accuracy. Our comments presented in this paper may be considered as recommendations for future optimization of ALF prediction models.     

Experimental monitoring of hepatic glucose,lactate, and glutamate metabolism by microdialysis during surgical preparation of the liver hilus

Mechanical liver manipulation can lead to hepatic microcirculation (MC) impairment. The pathobiochemical relevance of this phenomenon is not fully understood. Microdialysis (MD) allows a quantification of metabolic products in interstitial fluid, thus enabling analysis of the hepatic metabolic state during changes of liver perfusion. The aim of the study was to quantify the functional effects of standardized surgical liver preparation both on liver metabolism and microperfusion. Two groups of animals (pigs, n = 25) were formed: In the trial group (TG; n = 13) the liver was mobilized, followed by hilar preparation. In the control group (CG; n = 12) mobilization of the liver without hilar dissection was performed. Surgical manipulation was followed by an observation in both groups. Hepatic interstitial glucose, lactate, and glutamate concentrations were detected by MD and liver MC by thermodiffusion. During liver mobilization MC decreased significantly in both groups (TG; 86.7 +/- 2.0 to 73.4 +/- 2.3 ml/100 g min; and CG; 88.3 +/- 3.1 to 71.9 +/- 2.2 ml/100 g/min). In the trial group levels decreased further during hilar preparation reaching minimal values of 65.6 +/- 2.8. After preparation MC recovered to baseline. Glucose, lactate, and glutamate concentrations increased significantly during liver mobilization in the trial (glucose; 0.52 +/- 0.13 to 0.88 +/- 0.19 mmol/L; lactate; 0.34 +/- 0.07 to 0.54 +/- 0.07 mmol/L; glutamate; 34.5 +/- 3.6 to 52.6 +/- 8.0 micromol/L) and control group (glucose; 0.58 +/- 0.06 to 0.95 +/- 0.13 mmol/L; lactate; 0.30 +/- 0.06 to 0.49 +/- 0.07 mmol/L; glutamate; 32.9 +/- 2.36 to 56.1 +/- 5.12 micromol/L). Throughout hilus preparation maximum values could be measured in TG (glucose; 1.69 +/- 0.34; lactate; 0.90 +/- 0.18; glutamate; 63.5 +/- 7.2). After termination of mobilization or preparation baseline concentrations were reached again. MD allows monitoring of metabolic changes in hepatic parenchyma. Surgical liver preparation leads to changes of intrahepatic glucose, lactate, and glutamate levels (without alterations of parameters in systemic plasma) along with hepatic MC impairment. Reconstitution of hepatic MC was accompanied by rapid normalization of metabolic parameters. By measuring specific parameters, MD could prove to be of use for functional assessment of metabolic effects due to MC disturbances.     

The Effects of Homologous Cross-Circulation and In Situ Liver Perfusion on Fulrninant Hepatic Failure Rats

A galactosamine-induced fulminant hepatic failure (FHF) rat model was used to study the effects of homologous cross-circulation and in situ liver perfusion. Cross-circulation with homologous donors did not significantly improve the survival time or recovery rate of grade II hepatic coma rats. Homologous in situ liver perfusion significantly improved the survival time and recovery in FHF only when started in grade II coma; it has no effect in the later stage of coma.