Logo

Recent Developments in
Transplantation Medicine

Liver Transplantation

Return to Table of Contents

Extracorporeal Hepatic Support

John J. Brems
Mathew Brunson
Daniel R. Salomon

Contents

Tables and Figures

  • Figure 1. Bioartificial liver with porcine hepatocytes attached to microcarriers and placed in the extra-fiber compartment of hollow-fiber modules.
  • Figure 2. Schematic representation of blood flow from the patient, through the plasma separator, then the bioreactor, and back to the patient.
  • Table 1. Neurological improvement and patient survival with hemodialysis/hemoperfusion

Introduction

More than 27,000 deaths occur from liver failure each year in the United States.1 Since about one in ten of these people has an acute, potentially reversible form of liver failure, many could survive if a temporary extracorporeal hepatic support system were available. Can we foresee the availability and use of such devices anytime soon?

The development of an artificial liver has been complicated by the complex array of functions that the normal liver provides — synthesis of coagulation factors, opsonins, albumin, and prohormones (eg, renin substrate); gluconeogenesis; amino acid and lipid metabolism; and (via Kupffer cells and endothelium) reticuloendothelial clearance of bacteria and small particle debris. The most difficult hepatic function to replace exogenously is detoxification, which is accomplished primarily through biotransformation and/or excretion of various toxic metabolites and exogenous agents. Although synthetic deficits may be ameliorated by the administration of fresh frozen plasma and nutrient preparations such as hyperalimentation, an effective artificial liver must be able to detoxify. Unfortunately, our present understanding of the molecular nature of these toxic metabolites is very incomplete. Moreover, the complexity of human metabolism is so great that the most efficient way to address this challenge may be the development of a biological system.

Early on, two fundamental methods used to clear toxins from the blood of patients with liver failure were plasma exchange by apheresis2-4 and whole blood exchange transfusion.5,6 In these procedures, toxins were removed in addition to a percentage of the blood volume. These proved to be cumbersome and inefficient procedures for removing unspecified toxins from the bloodstream. In the face of continuous toxin production, an exchange procedure lasting only a few hours and affecting only a proportion of the total blood volume was limited by the relatively small daily clearance that was ultimately achieved.

It has also been noted that certain toxins present in patients with acute, fulminant liver failure can be cleared from the total blood volume across artificial membranes.7 In the late 1950s, Kiley et al8 reported the use of hemodialysis in the treatment of five patients with hepatic encephalopathy. Four of the five improved neurologically, but all died from liver failure within one month of treatment. In 1976 Opolon and colleagues9 developed the polyacrylonitrile (PAN) membrane. In comparison to standard dialysis, this membrane allowed removal of higher molecular-weight substances, in the range of 5,000 daltons, and further improved hemofiltration as a therapy for liver failure. In the initial study 24 patients were treated and 13 improved neurologically.9 However, only five survived. These patients died from liver failure, since liver transplantation was not available at that time.

Other hemoperfusion resins and absorbents were subsequently introduced in an attempt to augment the efficacy of hemodialysis.10-12 One of the most popular innovations was charcoal hemoperfusion to remove large, protein-bound molecules.11 This technique demonstrated a modest improvement in survival when compared to those utilized previously (Table 1).2,8-13 Nonetheless, substantial room for improvement remained. It should be noted, however, that liver transplantation was not available at the time of these early studies, and temporary support could only change outcome if native liver function recovered.

Table 1. Neurological improvement and patient survival with hemodialysis/hemoperfusion

Failure of all of the filtration or exchange techniques resulted largely from their inability to match or even approach the efficiency of the liver in clearance function and led investigators to explore cross-hemodialysis methods using living animals.14,15 The first use of this approach was in 1959, when Kimoto14 used cross-hemodialysis between human and dog to support a patient with liver failure. The device was made up of two blood-filled circuits, each containing an overlapping segment of semipermeable cellophane tubing. The patient's blood circulated through one circuit, while the dog's blood circulated through the other circuit. Dialyzable toxins from the patient purportedly crossed the cellophane tubing and were metabolized by the dog's liver. This device was used on a 20-year-old man with severe hepatic encephalopathy. After 55 minutes of cross-hemodialysis, he regained consciousness and became lucid, but he died seven days later of cardiac failure. This case demonstrated that cross-circulation with a xenogeneic liver could provide short-term hepatic support.

To avoid the complications that occur when utilizing live animals, subsequent trials investigated the applicability of whole liver preparations for detoxification. This experience with ex vivo perfusion through human,16 primate,17 and pig18 livers demonstrated that whole liver preparations could detoxify, but the unavailability of whole organs and their limited viability make this approach generally impractical.

The development of hollow-fiber technology allowed for viable human or animal hepatocytes to become an integral part of an artificial liver device and potentially increase clearance efficiency of an ex vivo method. These devices contain the hepatocytes in a separate chamber, isolated by a semipermeable membrane, around which blood or plasma circulates in a manner similar to hemodialysis13,19 (Figure 1).

Flow through the artificial liver has four distinct steps: separation of plasma from whole blood by filtration through a microporous membrane; removal of toxins from the separated plasma as it passes by the functional hepatocytes in the hollow fibers; reconstitution of the plasma with the blood in the extracorporeal circuit; and, finally, reinfusion into the patient.20 (Figure 2).

The separate chamber design allows replacement with fresh hepatocytes as dictated by viability and functional parameters. Further, by separating the hepatocytes from the bloodstream with a semipermeable membrane, the potential for immune responses to cell surface alloantigens or xenoantigens is theoretically reduced. Nonetheless, the potential exists for soluble antigens or antigenic peptides to pass through the semipermeable barrier and initiate an immune response. So far, this last issue has been more theoretical than real; the gravely ill patients who have undergone perfusion through whole animal livers have never produced anti-xenotypic antibodies.17 Presumably, the likelihood of an immune response would increase as the duration of liver support lengthened or in patients less critically ill, and this possibility would have to be considered if such support were used as a bridge to transplantation.

Figure 1

Figure 1. Bioartificial liver with porcine hepatocytes attached to microcarriers and placed in the extra-fiber compartment of hollow-fiber modules. Adapted from Digest Dis Sci 1991, vol. 36, number 9.

Go to top

As of this writing, at least six different prototype bioartificial livers are entering various phases of preclinical and clinical testing. These devices all use viable hepatocytes within an artificial dialysis-type cartridge. In 1992 Neuzil et al21 reported using a bioartificial liver in a patient with acute liver insufficiency. Their system employed porcine hepatocytes attached to microcarriers placed on the outer surfaces of hollow fibers. In initial studies with this device, the hepatocytes were viable for only five to six hours. One of the reasons for the short life span was presumably the high level of hepatotoxic metabolites such as ammonia in the patient's plasma. In an effort to reduce toxicity to the hepatocytes, Demetriou has modified his artificial liver. The new hybrid system consists of plasma perfusion through a charcoal column preceding plasma passage through the hepatocyte cartridge.22 It is hoped that the charcoal column will clear the plasma of low- to mid-molecular weight toxic metabolites and allow the hepatocytes to remain viable for a longer period. However, the viability of hepatocytes in this system reportedly is still only six to seven hours.

Figure 2

Figure 2. Schematic representation of blood flow from the patient, through the plasma separator, then the bioreactor, and back to the patient.

Go to top

Sussman et al23 have developed a similar artificial liver using human hepatocytes immortalized by viral transformation as the cell source. This bioartificial liver device has been used with success in one patient with syncytial giant-cell hepatitis24 and with clinical improvement in a second patient suffering from fulminant hepatitis.25 The use of transformed hepatocytes23 is an innovative solution to the problem of finding an adequate source of human hepatocytes, but it raises safety issues which need further study. In mice, hepatocytes transformed with a closely related virus causes tumors.26 Although transformed hepatocytes are unlikely to enter into the recipient's bloodstream, viruses such as those used to transform the hepatocytes can cross this semipermeable membrane. To date, Sussman's device has been used on more than 10 patients without any evidence of tumors (personal communication), but longer follow-up is needed.

In an effort to improve the viability and function of the hepatocytes within the bioartificial liver, Nyberg et al27 have developed a method of entrapping the hepatocytes on gel within a hollow fiber. Gel entrapment allows larger numbers of hepatocytes to be maintained in the bioartificial liver. This technique also facilitates closer cell-cell contacts between the hepatocytes, more closely resembling the anatomy of the liver, and may enhance hepatocyte function. Closer cell-cell contacts might be particularly important for those functions which depend on a basoapical orientation of the epithelium. With the use of gel-entrapped xenohepatocytes in an anhepatic rabbit model, Nyberg and associates demonstrated albumin production and cytochrome P-450 activity as well as improved biochemical function.28

Advanced Tissue Sciences Inc. (La Jolla, California) has developed a technique to grow normal human liver tissue on a three-dimensional framework of polymer mesh (personal communication). Hepatocytes adherent to the framework divide and differentiate. This liver tissue can be placed in hollow chambers, separated from the plasma or blood by a semipermeable membrane. The use of this system currently is limited by the ability to obtain human hepatocytes and grow them to a sufficient mass to support hepatic function.

With a contrasting strategy, Li et al29 improved the function and viability of primary hepatocytes within a bioartificial liver by avoiding attachment to the interface. They allow the hepatocytes to circulate through the bioreactor, which is perfused with oxygenated nutrient medium, at a rate optimal for collision and aggregation to occur. The hepatocyte aggregates are subsequently entrapped in a packed bed of glass beads. By removing the hepatocytes from the interface and "recharging" them periodically, Li's group has maintained viability for up to three weeks in culture. Hepatocytes nurtured this way resembled in vivo hepatocytes, with cuboidal shapes and intercellular contacts. Hepatocytes cultured for 15 days in this bioreactor manifested interconnecting three-dimensional structures resembling the sinusoidal plate. Electron microscopy revealed plentiful mitochondria, rough and smooth endoplasmic reticulum, glycogen granules, and desmosomes.

Because hepatocytes are not fixed to the plasma interface in Li's bioreactor, an efficient countercurrent flow can be incorporated, greatly expanding the effective plasma/hepatocyte contact. This, and the ability to extract "exhausted" cells from the circuit, offer a potential advantage, possibly allowing artificial liver support to be maintained continuously for an extended period. The technique has been tried in one individual. The patient, a 10-year-old female with hepatitis C, awakened from hepatic coma and recovered without liver transplantation. As with other support devices, further human trials are eagerly awaited.

The SYBIOLTM system being developed by Xenogenics, Inc. (San Diego, California) is the latest application of hepatocyte aggregates in an artificial liver. Although hepatocytes in suspension lose viability rapidly, data from Li's laboratory (personal communication) show that circulating hepatocytes have prolonged viability both in vitro (in culture media) and in vivo (connected to pigs). Circulating hepatocyte aggregates were viable for at least six hours, a commonly used duration of time for artificial liver perfusion. The advantages demonstrated by the SYBIOLTM system are the ability to monitor hepatocyte viability easily by sampling the aggregate suspension, and the ability to replace hepatocytes easily if viability decreases.

Currently, the impetus for developing a bioartificial liver is to serve as a bridge to liver transplantation in patients with fulminant hepatic failure. It is clear that survival posttransplant is improved if patients are transplanted before they require hospitalization for treatment of complications of their disease. This approach is cost-effective as well. Therefore, if the use of a bioartificial liver improved a patient's condition and allowed recovery from some of the complications of chronic liver disease prior to transplantation, the bioartificial liver might well prove both medically effective and economical. It might allow patients to leave the hospital and wait at home for elective liver transplantation. It would also end the practice of transplanting the sickest patients first, a strategy which is very costly and results in a high incidence of retransplantation.

In order to encourage manufacturers to develop a successful bioartificial liver, its potential use must be expanded to include the majority of patients with chronic end-stage liver disease. As noted by Sussman and Kelly,30 the real need for these complicated devices in patients with fulminant hepatic failure is to prevent or reverse the potentially fatal cerebral edema that occurs in this setting. In fact, bioartificial livers will be tested initially in patients with fulminant hepatic failure from acute hepatitis, primary nonfunction, or acute exacerbations of chronic injury. Patient survival will be the ultimate outcome parameter for these studies of the artificial liver.

As the indications for liver transplantation have expanded and as donor organs have failed to keep pace, the need for an artificial liver has become more critical. Given that the regenerative capacity of the liver is practically unlimited, the liver can usually recover sufficient function to sustain life if function can be supported for a few weeks. If an artificial liver were routinely available, many patients with liver failure might recover without liver transplantation. In the end, bioartificial livers will probably be used as a bridge to transplantation, to support patients with acute processes through the recovery phase, and to improve quality of life for patients with chronic liver disease.

Go to top

References

  1. National center for health statistics. Births, marriages, divorces, and deaths for 1990. Monthly Vital Statistics Report 1991;39:1-14.
  2. Knell AJ, Dukes DC. Dialysis procedures in acute liver coma. Lancet 1976;2:402-403.
  3. Yamazaki F, Inoue N. Hepatic assist device, using membrane plasma separator and dialyzer. Med Prog Technol 1987;12:17-24.
  4. Munoz JJ, Ballas SK, Moritz MJ, Martinez J, Friedman LS, Jarrell BE, et al. Perioperative management of fulminant and subfulminant hepatic failure with therapeutic plasmapheresis. Transplant Proc 1989; 21:3535-3536.
  5. Lee C, Tink A. Exchange transfusion in hepatic coma: report of a case. Med J Aust 1958;1:40-42.
  6. Trey MB, Burns DC, Saunders JJ. Treatment of hepatic coma by exchange blood transfusion. N Engl J Med 1966;274:473-481.
  7. Hager JC, Carman R, Stoller R, Panol G, Leduc EH, Thayer WR, et al. A prototype for a hybrid artificial liver. Trans Am Soc Artif Intern Organs 1978;24:250-253.
  8. Kiley JE, Welch HF, Perder JC, Welch CS. Removal of blood ammonia by hemodialysis. Proc Soc Exp Biol Med 1956;91:489-490.
  9. Opolon P, Rapin JR, Huguet C, Granger A, Delome ML, Boschat M, et al. Hepatic failure coma (HFC) treated by polyacrylonitrile membrane (PAN) hemodialysis (HD). Trans Am Soc Artif Intern Organs 1976;22:701-710.
  10. Schechter DC, Nealon TF, Gibbon JH. A simple extracorporeal device for reducing elevated blood ammonia levels. Surgery 1958;44:892-897.
  11. Yatzidis H. A convenient hemoperfusion microapparatus over charcoal for the treatment of endogenous and exogenous intoxications: its use as an effective artificial kidney. Proc Eur Dial Transplant Assn 1964;1:83-87.
  12. Ton HY, Hughes RD, Silk DBA, Williams R. Albumin coated Amberlite XAD-7 resin for hemoperfusion in acute liver failure. Part I: Adsorption studies. Artif Organs 1979:3:20-22.
  13. Takahashi, T, Malchesky PS, Nose Y. Artificial liver: state of the art. Dig Dis Sci 1991;36(9):1327-1340.
  14. Kimoto J. The artificial liver experiments and clinical applications. ASAIO Trans 1959;5:102-110.
  15. Burnell JM, Dawborn JK, Epstein MD, Gutman RA, Leinbach GE, Thomas ED, et al. Acute hepatic coma treated by cross-circulation or exchange transfusion. N Engl J Med 1967;276:935-943.
  16. Fox IJ, Langnas AN, Fristoe LW, Schaefer MS, Vogel JE, Antonson DL, et al. Successful application of extracorporeal liver perfusion: A technology whose time has come. Am J Gastroenterol 1993;88(11):1876-1881.
  17. Abouna GM, Serrou B, Boehnig HG, Amemiya H, Martineau G. Long-term hepatic support by intermittent multi-species liver perfusions. Lancet 1970;2:391-396.
  18. Eiseman B, Liem DS, Rafucci F. Heterologous liver perfusion in treatment of hepatic failure. Ann Surg 1965;162:329-345.
  19. Demetriou AA, Whiting J, Levenson SM, Chowdhury NR, Schechner R, Michalski S, et al. New method of hepatocyte transplantation and extracorporeal liver support. Ann Surg 1986;204:259-271.
  20. Malchesky PS, Ouchi K, Piatkiewicz W, Nose Y. Membrane plasma filtration systems with multiple reactors for hepatic support. Artif Organs 1978;2(suppl X):265-268.
  21. Neuzil DF, Rozga J, Moscioni AD, Man-Soo R, Hakim R, Arnaout WS, et al. Use of a novel bioartificial liver in a patient with acute liver insufficiency. Surgery 1993; 113(3):340-343.
  22. Rozga J, Holzman MD, Man-Soo R, Griffin DW, Neuzil DF, Giorgio T, et al. Development of a hybrid bioartificial liver. Ann Surg 1993;217(5):502-511.
  23. Sussman NL, Kelly JM. Artificial liver: a forthcoming attraction. Hepatology 1993; 17(6):1163-1164.
  24. Sussman NL, Finegold MJ, Kelly JH. Recovery of syncytial giant cell (SGCH) hepatitis following treatment with an extracorporeal liver assist device (ELAD) [abstract]. Hepatology 1992;16:51A.
  25. Sussman NL, Kelly JH. Improved liver function following treatment with an extracorporeal liver assist device. Artif Organs 1993;17:27-30.
  26. Muller TE, Jauregui HO. Letter to the editor. Artif Organs 1993;17:44-45.
  27. Nyberg SL, Peshwa MV, Payne WD, Hu WS, Cerra FB. Evolution of the bioartificial liver: the need for randomized clinical trials. Am J Surg 1993;166:512-521.
  28. Nyberg SL, Payne WD, Amiot B, Shirabe K, Remmel RP, Hu WS, et al. Demonstration of biochemical function by extracorporeal xenohepatocytes in an anhepatic animal model. Trans Proc 1993;25(2):1944-1945.
  29. Li AP, Barker G, Beck D, Colburn J, Monsell R, Pellegrin C. Culturing of primary hepatocytes as entrapped aggregates in a packed bed bioreactor: a potential bioartificial liver. In Vitro Cell Dev Biol Anim 1993;29A(3 Pt 1):249-254.
  30. Sussman NL, Kelly JH. Liver assist devices (LADS) will not be used to treat fulminant hepatic failure (FHF), but its consequences, namely hepatic encephalopathy (HE) [letter]. Artif Organs 1993;17(1):43-45.

Go to top · Return to Table of Contents

Please be aware that medical advice, diagnoses and physician references cannot be obtained from this site.