When to use renal replacement therapy and bioartificial support for renal failure in patients with cirrhosis


  • Vincenzo Morabito M.D.,

    1. Liver Failure Group, Institute of Liver and Digestive Health, University College London Medical School, London, United Kingdom
    Search for more papers by this author
  • Gilnardo Novelli M.D.,

    1. P. Stefanini Department of General Surgery and Organ Transplantation, Sapienza University of Rome, Rome, Italy
    Search for more papers by this author
  • Rajiv Jalan M.D.

    Corresponding author
    1. Liver Failure Group, Institute of Liver and Digestive Health, University College London Medical School, London, United Kingdom
    • Liver Failure Group, Institute of Liver and Digestive Health, University College London Medical School, Rowland Hill Street, Upper Third Floor, London, United Kingdom NW3 2PF
    Search for more papers by this author

  • Potential conflict of interest: Nothing to report.


Watch a video presentation of this article

Watch the interview with the author

Answer questions and earn CME

Liver failure can occur against a background of a normal liver (acute liver failure) or in patients with underlying cirrhosis [acute-on-chronic liver failure (ACLF); Fig. 1 and Table 1].2 In both conditions, the lack of the metabolic and regulatory function of the liver results in life-threatening complications such as bleeding, hepatic encephalopathy, and renal failure and a susceptibility to infections and culminates in multiorgan failure.

Figure 1.

Natural history of cirrhosis. ACLF often develops in patients with previously compensated cirrhosis or recently decompensated cirrhosis (<3 months) and is significantly associated with bacterial infections, renal failure, and active alcoholism.1

Table 1. Key Issues
• ACLF is described as an acute and rapid deterioration of liver function accompanied by subsequent rapidly evolving multiple end-organ failure in a patient with previously well-compensated liver disease due to the effects of a precipitating event.
• The pathophysiology of ACLF is related to intestinal permeability (leading to the translocation of bacteria and their products), an imbalanced immune reaction, and aggravated microcirculatory dysfunction as well as an accumulation of toxins.
• Renal failure is one of the most common causes of mortality in ACLF patients.
• Nonbiological liver support might represent a therapeutic tool in ACLF patients with renal failure for either recompensating or bridging patients up to transplantation.

Renal failure in patients with cirrhosis and portal hypertension is thought to be a functional disorder secondary to hemodynamic dysfunction, an increased synthesis of proinflammatory mediators, or both. Recent studies have highlighted that renal failure of cirrhosis is a heterogeneous condition, and only approximately 15% to 20% patients have the classic form of hepatorenal syndrome (HRS).3 In fact, in the vast majority, the underlying pathophysiology of renal failure is superimposed infection or inflammation. The treatment of HRS in patients with advanced liver cirrhosis is designed to abate the most important pathophysiological factor, splanchnic vasodilation, and the consequent reduction in renal blood flow.

Terlipressin and albumin have made a great impact by improving renal function in these patients, but short- and medium-term mortality rates remain unaltered. In patients with HRS and severe renal failure who do not respond to terlipressin and in patients who have non-HRS renal failure, continuous or repeated dialysis is considered as rescue therapy.4

In patients with decompensated cirrhosis and severe renal failure, survival without renal replacement therapy (RRT) is less than 20% on average at 6 months.5 A study reported that 8 of 30 patients with Child-Pugh C liver cirrhosis and HRS had survived at 30 days with the use of dialysis or continuous venovenous hemodialysis in the intensive care unit.6 Both hemodialysis and hemofiltration are capable of removing toxic substances from the circulation, but the efficacy is low because most toxic substances such as bilirubin, endotoxin, and cytokines are protein-bound or have a high molecular weight.

Traditionally, renal dysfunction attributed to HRS has been suggested to be reversible with transplantation. A recent study by Mitra et al.7 suggests that predicting whether there will be sufficient recovery of native kidney function is sometimes difficult because of the difficulty in clearly differentiating acute tubular necrosis from HRS and when HRS is superimposed on underlying intrinsic renal disease.

Because the liver has enormous potential to regenerate, extracorporeal liver support devices are evolving as novel strategies to assist the remaining functional cell mass by providing specific liver functions with the aim of improvements in end-organ function (e.g., renal failure).

The major objective of liver support is to buy time. This may allow patients with liver failure to be bridged to recovery or to be managed until an organ suitable for transplantation is available.


ACLF, acute-on-chronic liver failure; FPSA, fractionated plasma separation and adsorption; HRS, hepatorenal syndrome; MARS, molecular adsorbent recirculating system; RRT, renal replacement therapy; SMT, standard medical therapy.

Liver Support Devices

Liver support devices are traditionally categorized into two main types: artificial (nonbiological) devices, which are typically acellular (e.g., albumin dialysis and plasma exchange/diafiltration), and bioartificial (biological) devices, which contain hepatocytes from human or animal sources. Both types of devices have demonstrated biological effects such as the removal of toxic substances that accumulate during liver failure as well as effects on the function of individual organs.

Artificial Liver Support Devices

Currently available artificial liver support systems are blood purification or detoxification systems using membranes and adsorbents (Table 2).

Table 2. Artificial Liver Support Devices
DevicePrinciplesClinical Studies
  1. The rational basis for the development of extracorporeal liver support was the toxin theory. It was hypothesized that the removal of vasoactive, neurotoxic, and hepatotoxic toxins could not only lead to the recovery of end-organ dysfunction but also help to reconstitute the liver above the critical threshold of minimal functional liver cell mass by creating an improved intrahepatic environment for regenerative cells.

Single-pass albumin dialysisAlbumin dialysis against 2%-5% albumin with standard continuous RRT equipmentImprovements in biochemical parameters comparable to those with MARS
Only single case studies available
No randomized controlled trials
MARSAlbumin dialysis against 20% albumin and adsorption with anion exchange and activated charcoal adsorbersImprovement in hepatic encephalopathy and renal failure
Improved quality of life
No significant survival benefit
FPSA (Prometheus)Plasma separation and adsorption with a neutral resin and anion adsorbersImprovement in biochemical parameters
No significant survival benefit at 28 days

Single-Pass Albumin Dialysis

This is the simplest form of albumin dialysis and uses the principles of hemodialysis and hemofiltration. It can be performed with a standard continuous RRT device. The patient's blood flows through a standard albumin-impermeable high-flux dialyzer and is dialyzed against an albumin-containing dialysate (2%-5% albumin); this allows the removal of protein-bound molecules that are small enough to pass through the membrane pores as well as water-soluble toxins. No randomized controlled trials have been completed, and apart from single case studies, no improvements in survival have been reported.

Molecular Adsorbent Recirculating System (MARS)

MARS combines hemodialysis against albumin dialysate with a conventional dialysis procedure. Between these two circuits, there is interposed a third circuit, in which the albumin dialysate is recirculated through the shell (extrafiber) side of each dialysis cartridge, the charcoal column, and the anion exchanger. A prospective randomized controlled trial at two centers (n = 8 for the MARS group and n = 5 for the control group) treated patients with acute deterioration of chronic liver disease and HRS. The results showed significant prolongation of survival (25 days in the MARS group versus 4 days in the control group) and improvements in serum bilirubin levels and prothrombin times in the treatment group versus the control group.8 The results of a definitive study (the recompensation of exacerbated liver insufficiency with hiperbilirubinemia and/or encephalopathy and/or renal failure (RELIEF) trial9) were recently published; 189 patients with acute deterioration of cirrhosis and end-organ failure (ACLF) were randomized to either MARS (n = 95) or standard medical therapy (SMT; n = 94). The results showed that there was no evidence of improvement in survival between the two groups, and the improvement of renal failure observed in the previous smaller study could not be confirmed (Table 3).

Table 3. Molecular Adsorbent Recirculating System and Prometheus in ACLF
StudyPts (n)Contr.Improvements*
  • Substantial human data exist only for MARS and Prometheus. The available studies, albeit often small and uncontrolled, clearly suggest biochemical and neurological improvements after MARS and Prometheus therapy and an additional positive hemodynamic effect only with MARS.

  • *

    Biochemical improvements include statistically significant reductions in bilirubin, bile acids, creatinine, and ammonia. Cardiovascular improvements include hemodynamic parameters (mean arterial pressure, heart rate, and vasopressor requirements). Central nervous system improvement indicates a decrease in the hepatic encephalopathy grade (neurological improvement).

 Stange et al.15 (1999)13NoYesN/AYes69%
 Schimdt et al.16 (2001)8NoYesYesNo50%
 Jalan et al.17 (2003)8NoYesYesYes50%
 Di Campli et al.18 (2005)13NoYesN/AYes38%
 Mitzner et al.19 (2000)13YesYesYesYes37% vs 0%
 Heemann et al.20 (2002)23YesYesYesYes90% vs 55%
 Sen et al.21 (2004)18YesYesNoYes45%
 Blei et al.22 (2007)70YesN/AN/AYesN/A
 Lalelam et al.23 (2006)18YesYesYesN/A66% vs 33%
 Rifai et al.24 (2003)11NoYesNoNo28%
 Laleman et al.23 (2006)18YesYesNoN/A66% vs 33% vs 33%

Fractionated Plasma Separation and Adsorption (FPSA)

This system uses a plasma separation technique combined with adsorption. It involves a membrane with a cutoff of 250 kDa (AlbuFlow), which allows the separation of blood cells and large proteins from plasma. Albumin with protein-bound toxins crosses the membrane and passes over a neutral resin adsorber and then an anion exchanger. The filtered plasma and blood are then combined and hemodialyzed to remove water-soluble toxins before they are returned to the patient. In the recently published study of Kribber et al.,10 145 patients with ACLF were randomized to FPSA (n = 77) or only SMT (n = 68). The results showed that the therapy was safe and well tolerated but failed to show any improvement in survival for ACLF patients at 28 days in comparison with SMT. In a post hoc analysis, patients with HRS or a high Model for End-Stage Liver Disease score (>30) were shown to have a possible survival advantage if they were treated with FPSA.

Bioartificial Liver Support Devices

Bioartificial or biological liver systems are cell-based dialysis techniques using viable animal or human hepatocytes loaded into bioreactors that are perfused with the patient's blood or plasma to provide some of the functions of the failing liver, including synthetic (proteins and clotting factors), regulatory (hormones), and immunological functions and biotransformation. Primary porcine hepatocytes, immortalized human cells, and cells derived from hepatic tumors are the most common cell sources (Table 4).

Table 4. Bioartificial Liver Support Devices
DevicePrinciple and Cell TypeMain ConcernClinical Studies
  1. Biological approaches rely on the functionality of livers or hepatocytes of xenogeneic or human origin that can be exploited to support the patient's liver. These functions include detoxification, several metabolic functions, and the synthesis of proteins and other molecules. Liver support could be provided by human cross-circulation, but the potential toxicity and adverse reactions in the donor severely limit this approach.

Modular extracorporeal liver systemPlasma separation and then passage of plasma through human hepatocytesLow supplies and function difficult to maintain11 patients successfully bridged to transplantation
Extracorporeal liver assist deviceHuman hepatoblastoma cell line–derived hepatocytesTumorigenicity6 human studies with 150 patients treated; survival benefit in 49 patients in ACLF study
Bioartificial liver support systemPorcine hepatocytesXenozoonosesPhase 1 study in 4 patients with no serious adverse events

Modular Extracorporeal Liver System

The patient's blood is separated; plasma is pumped through a dialysis filter before it is passed through a bioreactor containing human hepatocytes. The treated plasma is then mixed with its cellular components and returned to the patient. So far, 11 patients with ACLF have been successfully bridged to liver transplantation.11

Extracorporeal Liver Assist Device

Blood is separated, and plasma is passed through a bioreactor containing immortalized human liver cell line C3A (derived from a human hepatoblastoma). The treated plasma, recombined with cellular components, is then returned to the patient. Six human studies have been completed, with more than 150 patients treated thus far.12 A recent study of 49 patients with ACLF showed a survival benefit in the treated group.13

Bioartificial Liver Support System

Whole blood is passed through a bioreactor containing porcine hepatocytes before it is returned to the patient. No serious adverse events were observed in the original phase 1 study, but only modest biochemical responses were observed.14 In a large study of patients with acute liver failure in which 171 patients were randomized, no survival benefit was observed. No large studies have been performed in patients with ACLF.


RRT using venovenous hemofiltration should be used in patients as a measure for buying time for recovery from an acute insult or for bridging them to liver transplantation. Because of the shortage of suitable livers for transplantation, a device shown to allow liver regeneration while supporting metabolic demands could potentially have an enormous impact on both survival and treatment in liver failure. Currently, artificial livers offer the greatest potential for bridging patients to liver transplantation, and further trials are awaited.