Artificial Liver Support Potential to Retard Regeneration?

Home , Hepatocyte growth factor, TGF alpha

REVIEW ARTICLE Artificial Liver Support Potential to Retard Regeneration?

Emma J. Mullin, MBChB; Matthew S. Metcalfe, FRCS; Guy J. Maddern, MD

Hypothesis: The concept of an “artificial liver” has been growth-promoting factors from these cultured hepato- in development for over 40 years. Such devices aim to cytes? temporarily assume metabolic and excretory functions of the liver, with removal of potentially hepatotoxic sub- Data Sources, Extraction, and Study Selection: stances, thereby clinically stabilizing patients and pre- Data were obtained using PubMed search for reports in- venting deterioration while awaiting transplantation. If volving liver support, extracorporeal circuits, dialysis, sufficient numbers of viable hepatocytes remain, regen- growth factors, and cytokines. Those reports specifi- eration and subsequent recovery of innate liver func- cally looking at the effect of artificial liver support on cy- tion may occur. However, these devices have not yet be- tokines and growth factors are discussed. come part of routine clinical use. Much less is known regarding the effect such devices have, if any, on circu- Conclusions: There is a paucity of information on the lating cytokines and growth factors and the subsequent key events and substances involved in hepatic regenera- effects on the regenerating liver. If these devices remove tion. In addition, there is a potential impact of liver sup- or reduce factors known to promote regeneration, is the port devices on the regeneration of substances associ- rate of regeneration retarded? Conversely, does the in- ated with hepatic regeneration. Further study is needed. corporation of hepatocytes into bioartificial support systems confer an advantage through the production of Arch Surg. 2004;139:670-677

HE LIVER HAS A REMARK- of the liver, with removal of potentially able capacity for regene- hepatotoxic substances, thereby clini- ration, a role it is able to cally stabilizing patients and preventing de- fulfill while performing terioration while they await transplanta- complex metabolic and ex- tion. If sufficient numbers of viable cretoryT functions. hepatocytes remain, regeneration and sub- Fulminant hepatic failure remains a sequent recovery of innate liver function significant cause of morbidity and mor- may occur. Patients who may benefit from tality. With best medical management such devices are those with fulminant he- alone, mortality approaches 80%, which patic failure, acute-on-chronic (ie, an acute can be improved to 55% to 75% if liver condition verging on being chronic) liver transplantation is a therapeutic option.1 failure, primary liver allograft nonfunc- However, 33% to 50% of patients with ful- tion, and posthepatectomy liver failure. minant hepatic failure will die awaiting a However, the task is complex, as the liver liver transplant.1 affects almost every body system through After resection or injury, the liver will metabolic, endocrine, immune, and physi- regenerate as long as there is enough func- ological processes. tional parenchyma remaining. Up to 75% Although biochemical, hormonal, and of a noncirrhotic liver can be resected safely hematologic changes accompanying liver in humans2; beyond that, increasing prob- failure are well known, development of a lems with hepatic dysfunction and post- support device is complicated by the fact operative morbidity and mortality oc- that many of the pathophysiological re- 3 From the University of Adelaide cur. sponses involved in hepatic failure are in- Department of Surgery, The concept of an “artificial liver” has completely understood. For example, a The Queen Elizabeth Hospital, been in development for more than 40 leading cause of death in fulminant he- Woodville, South Australia, years. Such devices aim to temporarily as- patic failure is brainstem herniation sec- Australia. sume metabolic and excretory functions ondary to glial swelling and cerebral

(REPRINTED) ARCH SURG/ VOL 139, JUNE 2004 WWW.ARCHSURG.COM 670

©2004 American Medical Association. All rights reserved. Downloaded From: https://jamanetwork.com/ on 09/30/2021 edema, but the exact processes leading to this are not yet 10% of hepatocytes; however, this figure increased sub- clear. stantially when these substances were infused into rats Much less is known regarding the effect, if any, that undergoing 30% hepatectomy, which suggests that a prim- such devices have on circulating cytokines and growth ing stimulation occurred at hepatectomy.9 Substances that factors and the subsequent effects on the regenerating initiate priming include TNF-␣ and interleukin 6 (IL-6), liver. If these devices remove or reduce factors known and it is during this stage that early genes are induced.10 to promote regeneration, does this retard the rate of re- These genes cover many different classes, including tran- generation? Conversely, does the incorporation of hepa- scription factors, metabolic enzymes, inflammatory re- tocytes into bioartificial support systems confer an ad- sponses, and responses involved in cytoskeletal and ex- vantage through the production of growth-promoting tracellular matrix modification.11 factors from these cultured hepatocytes? The molecular mechanisms of regeneration have per- This review summarizes what is known of the key haps been most extensively investigated in the two- events and substances involved in hepatic regeneration thirds hepatectomy of the rat model, which was de- and examines the potential impact of liver support described in 1931.12 In this model, DNA synthesis begins vices on regeneration and substances associated with it. 12 to 16 hours after hepatectomy, peaking at 24 to 48 hours. Hepatocyte mitosis follows 6 to 8 hours later, with MECHANISMS OF LIVER REGENERATION near restoration of volume at 3 days.13 Proliferation of bile duct epithelium peaks at 48 hours, Kupffer and stel- Cells late cells at 72 hours, and sinusoidal endothelial cells at 96 hours.14 Once the liver has reached optimal size, hepa- Under physiologic conditions, hepatocytes are in the rest- tocytes return to a quiescent G0 phase. ing (G0) phase with low proliferative activity. However, The rate of mitotic activity after liver resection, based after resection, most hepatocytes will undergo at least 1 on indirect measurements, is greatest 4 to 5 days post- round of replication before returning to a resting state. operatively.15 In humans, two thirds of liver regenera- This process is stimulated by a highly controlled, com- tion occurs within 2 weeks after major hepatectomy.16 plex, and incompletely understood sequence of events. Final regenerative volume ranges from 74%16 to 100% The liver is a unique organ, as regeneration does not re- at 1 year after resection.17 Normal livers regenerate twice quire the presence of progenitor stem cells, although they as fast as cirrhotic livers with comparable resection vol- play an as-yet ill-understood role. Instead, proliferation umes.18,19 of all mature hepatic cell types occurs, including hepa- After resection, regenerating hepatocytes are ini- tocytes, biliary epithelial cells, endothelial cells, Kupffer tially arranged in nonvascularized clusters. Once hepa- cells, and stellate cells.4 tocyte proliferation has ceased, stellate cells move into Hepatic oval cells are pleuripotential progenitor cells the clusters and neovascularization occurs. Normal his- capable of differentiation into hepatocytes, bile duct epi- tological structure is seen 8 to 10 days after surgery.20 thelia, intestinal epithelia, and possibly exocrine pan- The modulation of extracellular matrix, collagen, ma- creas. The theory that these cells originate from within trix metalloproteinases, and tissue inhibitors of metal- the biliary system is supported by their phenotypic simi- loproteinases during liver regeneration may help ini- larity to bile duct epithelia, their expansion from the peri- tiate and terminate the progression of quiescent portal area into the mesenchyme, and the reduction in hepatocytes into the cell cycle.21 numbers seen with selective damage to the periportal zones.5 However, some evidence suggests a bone mar- Regulatory Substance row origin for some oval cells.6 Although their exact role is incompletely understood, the liver may regenerate by The process of liver regeneration requires a combina- using oval cells in circumstances where large numbers tion of cytokines and growth factors. Evidence of the role of hepatocytes are damaged or hepatocyte replication is of circulating growth hormones arises from studies show- suppressed by hepatotoxins.7 Oval cells have receptors ing that after partial hepatectomy of the host animal, he- for hepatocyte growth factor (HGF), epidermal growth patic tissue transplanted into extrahepatic sites also un- factor (EGF), and transforming growth factor ␣ (TGF- dergoes DNA synthesis and replication.22 We characterize ␣),8 and it has been shown that HGF promotes hepato- EGF, TGF-␣, and HGF as complete mitogens, ie, each is cyte differentiation in bone marrow stem cells and oval capable of stimulating hepatocyte DNA synthesis in cul- cells.4 However, the mechanism of oval-cell activation, ture independently. Comitogens such as insulin, gluca- recruitment, and differentiation has not been fully elu- gon, epinephrine, and norepinephrine potentiate the ac- cidated, and further work is needed to clarify their role.7 tion of mitogens, but are unable to stimulate DNA synthesis alone23,24 (Figure 1). Proliferation Epidermal growth factor is a complete mitogen. In rats, sialoadenectomy (and the subsequent reduction in In the regenerating liver, hepatocytes require a priming circulating EGF levels) reduces the regenerative re- phase in which the hepatocyte undergoes a series of sponse to that of rats undergoing partial hepatectomy, changes to enter into the cell cycle and become recep- an effect that is reversed after administration of EGF.25 tive to the actions of growth factors. This is supported Exogenous EGF can stimulate regeneration in partially by evidence that infusion of HGF, EGF, or tumor necro- hepatectomized rats,26 but in cirrhotic rats it may need sis factor ␣ (TNF-␣) in rats caused replication in only to be combined with insulin to increase the rate of DNA

(REPRINTED) ARCH SURG/ VOL 139, JUNE 2004 WWW.ARCHSURG.COM 671

TNF-α + + HGF

IL-6 EGF After Hepatectomy + Hepatocyte Proliferation + +

Hepatocyte Proliferation

+ + HGF Production TGF-α Insulin – Glucagon Epinephrine

Norepinephrine IL-6 TGF-β

Figure 1. Outline of action and interactions between mitogenic substances involved in hepatocyte proliferation. TNF-␣ indicates tumor necrosis factor ␣; IL-6, interleukin 6; u-Pa, urokinase-type plasminogen activator; α

HGF, hepatocyte growth factor; TGF, transforming growth factor; TNF- EGF, epidermal growth factor; +, positive effect on regeneration; and −, negative effect on regeneration.

27 synthesis after partial hepatectomy. However, levels are β

elevated by only one third after hepatectomy. In hu- TGF- mans, levels of EGF have been found to be unchanged in subsegmental resections but raised after major resection, reaching a peak 5 to 7 days after surgery28 (Figure 2). ␣ Transforming growth factor is a complete mito- α

gen that has roles in liver development, regeneration, and TGF- carcinogenesis. When TGF-␣ is given in combination with HGF to cultured hepatocytes, the rate of DNA synthesis is higher than when either substance is administered in- 0 1 2 3 4 dividually.29 However, in TGF-␣ knockout mice, liver de- Time After Partial Hepatectomy, d velopment and regeneration after partial hepatectomy are 30 ␣ Figure 2. Schematic representation correlating change in growth factors and unaffected. Because TGF- is a part of the EGF family cytokines after partial hepatectomy in rats with hepatocyte proliferation. that signals via the EGF receptor, this finding may rep- HGF indicates hepatocyte growth factor; IL-6, interleukin 6; TNF-␣, tumor resent an increase in associated ligands, ie, redundancy necrosis factor ␣; and TGF, transforming growth factor. in the system. Transforming growth factor ␤ exists in humans in 12 hours and peak at 20 hours, coinciding with decreas- ␤ ␤ ␤ 34 ␤ 3 isoforms (TGF- 1, TGF- 2, and TGF- 3) and has a rela- ing hepatocyte proliferation. Infusion of TGF- at the tive molecular mass of 25000. It has a molecular struc- time of liver resection has been shown to decrease but ture of 112 amino acids, with 80% homology among the not stop hepatocyte proliferation35 (although dose- 3 isoforms, and is highly conserved between mamma- response curves were not assessed), while increasing syn- lian species.31 Its biological role is diverse, and distur- thesis and deposition of extracellular matrix proteins.32 bances in its production or downstream intracellular sig- Delayed regeneration is seen in transgenic mice that over- naling have been implicated in several diseases such as express TGF-␤.36 atherosclerosis, fibrotic diseases, and cancer.31 However, the postresection increase of levels of ␤ ␤ 37 Transforming growth factor has roles in cellular TGF- 1 in the liver, with a peak at 48 hours, creates a proliferation, differentiation, apoptosis, angiogenesis, and paradox of hepatocyte proliferation in the face of el- tissue repair. Levels are elevated in hepatic fibrosis, and evated TGF-␤ levels. This anomaly may be explained by ␤ competitive blockade of the TGF- 2 receptor can pre- the finding that after liver injury by carbon tetrachlo- 32 ␤ ␤ vent or reverse this process. Within the liver, TGF- 1 ride, receptors for TGF- 1 on hepatocytes are down- messenger RNA is found mainly in sinusoidal endothe- regulated for 72 hours. Recovery of expression after this ␤ lium and Kupffer cells, with a lesser amount of TGF- 2 time enables signal transduction to the nucleus and may ␤ ␤ 38 and TGF- 3. Stellate cells have little TGF- , and hepa- have a role in terminating hepatocyte proliferation. Re- tocytes express virtually no TGF in the resting state. Af- ceptor down-regulation is not seen in hepatic stellate cells, ␤ ␤ ter partial hepatectomy, TGF- levels (mainly TGF- 2) enabling the production of extracellular matrix pro- increase in all cell types, but increased levels in stellate teins around the proliferating hepatocytes.38 cells are seen only after bile duct injury.33 Interleukin 6, with a molecular weight of 21 kDa, ␤ After partial hepatectomy in rats, levels of TGF- 2 activates multiple signaling pathways that culminate in ␤ ␤ and TGF- 3 rise early, and TGF- 1 levels increase after the induction of genes involved in growth and liver me-

(REPRINTED) ARCH SURG/ VOL 139, JUNE 2004 WWW.ARCHSURG.COM 672

©2004 American Medical Association. All rights reserved. Downloaded From: https://jamanetwork.com/ on 09/30/2021 tabolism. After liver resection, levels of IL-6 and TNF-␣ have no cellular component (nonbiological). Nonbio- are raised within the first few minutes and peak at 24 logical devices remove circulating toxins by means of fil- hours.39 Both of these substances are important in the ini- tration or adsorption using charcoal columns or albu- tiation of regeneration and hepatocyte priming. In IL-6– min dialysis. Biological devices incorporate cultured deficient mice, hepatic DNA synthesis is significantly re- hepatocytes within a bioreactor through which the blood duced, although recovery of normal liver weight is or the plasma of the patient is perfused. Biological de- eventually achieved. Such mice have a 15% reduction in vices aim to not only remove circulating toxins but also the number of hepatocytes that undergo DNA synthesis replace the metabolic and synthetic functions of the liver. after partial hepatectomy.40 Administration of IL-6 be- The effects of liver support devices on liver func- fore liver resection in these mice restores hepatocyte re- tion and survival, both clinically and in animal models, generative capacity. Tumor necrosis factor ␣ receptor are reviewed elsewhere and therefore not considered as knockout mice show a similarly reduced regenerative re- part of this review.48-52 sponse after hepatectomy, but also have reduced levels The effect of liver support on circulating cytokines of IL-6.41 Again, this effect can be restored after injec- and growth hormones remains to be elucidated. Do sup- tions of IL-6.42 These studies suggest that IL-6 release is port devices remove or sequester these substances? If so, initiated by TNF-␣. does this affect outcomes or regenerative rates? Do these Hepatocyte growth factor is found in extracellular devices affect complementary systems that may result in matrix of the liver, spleen, and kidney and bound to hepa- hemodynamic instability? Also, do biological support de- ran sulfate, thrombospondin, and collagen types I, III, vices infer an advantage to regeneration through the pro- IV, V, and VI.43 It is a potent mitogen and causes prolif- duction and secretion of mitogenic factors by the incor- eration in cultured hepatocytes in the absence of any other porated hepatocytes that may assist regeneration of the growth factors or cytokines.4 Homozygous gene dele- host liver? Is there a role for these devices in the re- tions are associated with embryonic fatality partly due moval of inhibitory factors, eg, TGF-␤? To address this to arrested hepatic growth.4 In rats after liver resection, question, it is necessary to look at evidence from trials levels of HGF are elevated within 1 hour and remain el- with liver support devices, as well as those with other evated for 72 hours.44 Infusion of HGF in normal rats extracorporeal systems such as renal dialysis. causes an increase in hepatocyte proliferation, but when preceded by a collagenase infusion, the rate of DNA syn- Hemodialysis, Hemofiltration, Plasma Exchange, thesis is increased, implying that degradation of matrix and Extracorporeal Circuits is an important early part of regeneration.45 Activated urokinase-type plasminogen activator is Plasma exchange, where plasma is removed by means of present in the liver within 1 minute of partial hepatec- plasmapheresis and replaced by donor plasma, has been tomy. Urokinase-type plasminogen activator activates used in patients with fulminant hepatic failure and has HGF from inactive single-chain form to the biologically shown improvements in hemodynamic stability53 and ce- active 2-chain form. It also acts as a proteolytic enzyme rebral function. However, this method may have seri- to convert plasminogen to plasmin, thus contributing to ous adverse effects such as toxic effects, infections, and the breakdown of extracellular matrix, which may prime pulmonary or neurological complications. hepatocytes for replication and release sequestered growth For patients with acute liver failure, plasma ex- factors.14 change has variably been shown to reduce levels of TNF-␣ After liver resection, levels of norepinephrine are el- and IL-654 or to make no difference to levels unless com- evated within 1 hour.24 This elevation may help stimu- bined with continuous hemodiafiltration.55 late production of EGF from the Brunner gland in the Dialysis membranes may be made from natural ma- duodenum46 and enhance the effects of EGF and HGF terials such as cellulose or synthetic materials such as via adrenergic receptors. This elevation also may be in- polysulfone, polymethylmethacrylate, or polyacryloni- volved in the early phase of DNA synthesis and has been trile with filtration of molecules with molecular weight suggested to have a role in resistance to TGF-␤.47 of up to 40 kDa. Synthetic materials tend to be more bio- As described previously, administration of insulin compatible than the older cellulose-based membranes (ie, in combination with EGF enhances regeneration in cir- reduced production of proinflammatory cytokines and rhotic rat livers. Although insulin is not directly mito- complement activation).56 genic, the diversion of insulin from the pancreas by por- It is known that dialysis circuits induce a range of tocaval shunt results in liver atrophy that can be corrected biological reactions. Levels of IL-6, TNF-␣, IL-1, leukot- by means of exogenous insulin administration.23 riene, and superoxide become elevated; complement cascade is activated with increased levels of C3a/C5a; and IMPACT OF LIVER SUPPORT DEVICES endogenous pyrogens are elaborated from monocytes.57 ON REGENERATION Some of these, eg, TNF-␣ and IL-1, may be raised to sufficient levels to cause systemic effects, especially in sep- The aim of a liver support device is to temporarily as- tic or unstable patients.58 Levels of complement activa- sume the detoxification and metabolic functions of the tion and production of pyrogens may be reduced with liver while awaiting regeneration of the remaining func- the synthetic membranes. In cardiopulmonary bypass cir- tioning hepatocytes and subsequent recovery of hepatic cuits, similar activation of complement with elevated lev- function. Such devices can be broadly separated into those els of the proinflammatory cytokines TNF-␣, IL-6, and that incorporate hepatocytes (biological) and those that IL-8 is seen.59

(REPRINTED) ARCH SURG/ VOL 139, JUNE 2004 WWW.ARCHSURG.COM 673

©2004 American Medical Association. All rights reserved. Downloaded From: https://jamanetwork.com/ on 09/30/2021 Acute-phase proteins (eg, ␣1-acid glycoprotein) are ture hepatocyte source may arise from the emerging field produced mainly by the liver and are regulated by proin- of hepatocyte stem cell research. flammatory cytokines. They are found in increased quan- Techniques of hepatocyte culture and mainte- tities in end-stage renal failure. Studies have compared nance of viability and function are continually evolving. the effect of adding the ultrafiltrate produced by high- Difficulties arise in the balance between maintaining vi- and low-flux dialysis membranes to cultured hepato- ability and subsequent loss of function. Once estab- cytes. This showed that the ultrafiltrate causes reduced lished, cell lines may quickly lose liver-specific func- metabolic activity [by results of the 3-(4,5-dimethyl- tions. To address this issue, hepatocyte cellular thiazole-2-yl)-2,5-diphenyl tetrazolium bromide MTT morphologic features, bioreactor design, type of cellu- test] and increased cytotoxicity as measured by lactate lar scaffold, and culture medium have all been manipu- dehydrogenase release in hepatocytes. The ultrafiltrates lated, with viable hepatocytes maintained in studies from can also stimulate increased ␣1-acid glycoprotein re- 3 weeks69 to 5 months.70 Techniques of measuring vi- lease from hepatocytes, an effect more pronounced with ability include oxygen consumption, albumin secre- high-flux compared with low-flux membranes. This find- tion, lidocaine metabolism, ammonia removal, and urea ing suggests that dialysis can remove hepatotoxic sub- synthesis. Although synthetic function can be mea- stances, which may be an important factor in liver sup- sured by albumin secretion, evidence that cultured hepa- port systems incorporating it.60 tocytes produce or are able to produce growth factors and Dialysis membranes (polyacrylonitrile membranes cytokines is lacking. This may be difficult to answer ex- more so than polysulfone membranes) may adsorb TNF-␣, perimentally, as cultured hepatocytes within bioreac- with minimal amounts removed by filtration owing to tors are not exposed to the stimulators of regeneration its large molecular weight (50 kDa).61 Interleukin 6 and that are produced in hepatic failure. In addition, several IL-8 may be filtered to some extent, but removal has been of the substances involved in regeneration are secreted postulated to be mainly by adsorption.62 However, little from surrounding hepatic stellate cells, which are not cur- evidence exists of the effects of dialysis on other mito- rently included within hepatocyte-incorporating biore- genic substances such as HGF or EGF. actors, but may be significant in studies using whole- In septic patients with systemic inflammatory re- organ liver perfusion as a liver support. sponse syndrome or multiorgan dysfunction syndrome, In rats that have undergone partial hepatectomy fol- in whom levels of circulating cytokines are known to be lowed by partial devascularization of the remnant, the elevated, continuous hemofiltration can reduce IL-6 and use of a BAL containing microcarrier-attached hepato- IL-8 levels. Reduction of these cytokine levels is associ- cytes compared with sham liver-support devices (extra- ated with improved urine output, blood pressure, and sys- corporeal circuit without hepatocytes) has shown im- temic vascular resistance, with decreasing dependence provement in prothrombin time and reduced alkaline on inotropic support in this group of patients.63 How- phosphatase, urea, and uric acid levels. However, no dif- ever, these reductions have not been universally repro- ference in serum levels of HGF between rats receiving duced, and the role of hemofiltration in sepsis and its ef- BALs and those undergoing sham operation was found, fect on survival have yet to be clarified by randomized with elevated levels seen in both groups. Levels of HGF controlled trials.64 Some evidence suggests that the frac- RNA were also increased in the rats with a BAL but un- tion of inflammatory mediators removed by hemofiltra- detectable in the sham-operation rats.71 Decreased lev- ␤ tion is insubstantial compared with that removed by en- els of TGF- 1 were found in rats treated with BAL sup- dogenous clearance.65 port.71 In humans with acute liver failure, both HGF and ␤ Bioartificial Liver Support Devices TGF- 1 levels are known to be elevated. In patients treated with an extracorporeal liver-assist device (ELAD) incor- To develop a bioartificial liver (BAL), a source of viable, porating human C3a liver tumor cells within hollow fi- functioning hepatocytes is required. At present, the pre- ber filters, levels of HGF were found to be elevated after dominant sources are human and porcine-derived cells. 6 hours of hemoperfusion compared with untreated pa- Primary human cells are a scant resource, as most suit- tients, then decreased back to the levels of the untreated able donor organs are required for transplantation pro- patients after 48 hours. There was no change observed ␤ 72 grams. Human hepatocytes are difficult to replicate in vitro, in plasma TGF- 1 level between the 2 groups. This ob- and daughter cells have not been shown to express suffi- served elevation of HGF levels may be due to the heparin- cient levels of liver-specific functions.66 Hepatocyte cul- binding qualities of HGF, thereby reducing metabolism tures derived from human hepatoblastoma cell lines have and excretion by the remaining liver, as elevation of HGF been used in liver support devices, but have lower levels levels is also seen in patients treated with continuous veno- of ammonia removal and amino acid metabolism com- venous hemodialysis.72 pared with porcine hepatocytes.67 A concern with the use Another study using ELAD for patients with acute of hepatoblastoma and immortalized hepatocytes is the risk liver failure found that levels of TNF-␣ and IL-6 were el- for transmission of potentially tumorigenic cells; there- evated in all patients before treatment. Six hours after com- fore, interest has developed in the incorporation of “sui- mencement of ELAD treatment, levels of TNF-␣ and IL-6 cide genes.”68 Although porcine hepatocytes are easily avail- were further increased when compared with controls, but able, with a large yield per animal, concern still exists then decreased toward baseline within 48 hours. A small regarding immunogenic complications and potential trans- elevation seen in C-reactive protein levels may reflect the fer of porcine endogenous retroviruses. However, the fu- damaged liver’s inability to mount an acute phase re-

(REPRINTED) ARCH SURG/ VOL 139, JUNE 2004 WWW.ARCHSURG.COM 674

©2004 American Medical Association. All rights reserved. Downloaded From: https://jamanetwork.com/ on 09/30/2021 sponse to elevated cytokine levels. The level of anti- els of HGF, IL-6, and TNF-␣ occur in patients with acute thrombin III (a marker of coagulation system activa- liver failure treated with BAL support, but conflicting tion) was elevated in all patients and had a small but not changes occur in IL-6 and TNF-␣ levels for those treated statistically significant rise after treatment with ELAD.73 with nonbiological adsorption-based devices. Levels of ␤ None of these cytokine changes had hemodynamic ef- TGF- 1 are unchanged in humans treated with BAL sup- fects on the patients. port, but levels of TGF-␤ in rats may be reduced. Levels of TNF-␣ in hepatectomized pigs74 were found The significance of these findings, if any, is un- to be elevated when the animals were placed on an ex- clear. Studies of the effects of liver-support devices and tracorporeal circuit containing hepatocytes compared with growth factors are limited and do not address the issue when they were perfused with an extracorporeal circuit of the impact of these devices on regeneration. Al- alone, indicating that the presence of hepatocytes may though we know that reductions in IL-6 and TNF-␣ lev- cause activation of leukocytes and release of TNF-␣. els do not prevent regeneration, the fact the levels of these substances are affected suggests that levels of other cy- Adsorption Columns/Albumin Dialysis tokines, eg, HGF and EGF, will also be affected. Future studies of liver support devices need to incorporate evalu- Although evidence suggests that the dialysis membrane ation of the effects of the device not only on liver func- itself may remove TNF-␣ by adsorption of molecules to tion and survival but also on the rate of regeneration. Com- the membrane, rather than through filtration or siev- puted tomographic volumetric scanning or indirect ing,75 techniques using adsorptive columns may in- measurements of DNA synthesis such as serum thymi- crease removal rates. dine kinase levels may assess rates of regeneration. If liver Hemoperfusion with albumin-rich dialysate has been support retards rates of regeneration, would there be a shown to reduce circulating levels of IL-6 and TNF-␣ in proportional increase in the number of oval cells seen in addition to other substances with a high albumin affinity regenerating livers? such as bilirubin, bile acids, benzodiazepines, middle chain Comparison between the different causes of he- fatty acids, and arachidonic acids76 in patients with acute patic failure and liver-support devices may show differ- liver failure. This system uses an albumin-impregnated hol- ing effects on growth factors and regeneration. This may low-fiber dialysis membrane with high-albumin dialy- be a reflection on the differing pathogenesis of liver fail- sate. The albumin removes protein-bound substances into ure between acetaminophen-induced failure and autoim- the dialysate, which is then cleaned by means of charcoal mune causes, for example. and resin adsorption columns, and the dialysate is re- If liver support devices are shown to retard rates of cycled. This reduction of IL-6 and TNF-␣ levels is en- regeneration, then the period of use required in patients hanced with albumin dialysis compared with hemodiafil- may be longer than expected. However, if promotion of tration with isotonic sodium chloride dialysate.77 regeneration is demonstrated, then the potential impli- The BioLogic-DT system (HemoCleanse, Lafayette, cations for earlier use in an extended group of patients Ind) uses a parallel-plate cellulose membrane dialyzing with hepatic failure may be a possibility. against powdered charcoal to increase surface area. The system has been found to cause elevations in levels of Accepted for publication October 31, 2003. TNF-␣ in patients with acute severe alcoholic hepatitis.78 Corresponding author: Guy J. Maddern, MD, Univer- Reductions in TNF-␣ and IL-6 levels have also been sity of Adelaide Department of Surgery, Woodville Road, found in patients with systemic inflammatory response Woodville, South Australia 5011, Australia (e-mail: syndrome and organ failure undergoing a single treat- [email protected]). ment with powdered charcoal and silica.79 Initial use of sorbents such as charcoal and ion exchangers that al- REFERENCES lowed direct contact between blood and the adsorption column caused drops in fibrinogen levels and activation 76 1. Riegler JL, Lake JR. Fulminant hepatic failure. Med Clin North Am. 1993;77: of the complement cascade. Therefore, systems now re- 1057-1083. quire either coated charcoal, which reduces adsorptive 2. Abdalla EK, Hicks ME, Vauthey JN. Portal vein embolisation: rationale, tech- efficacy, or a plasma separation step. nique and future prospects. Br J Surg. 2001;88:165-175. 3. Shoup M, Gonen M, D’Angelica M, et al. Volumetric analysis predicts hepatic COMMENT dysfunction in patients undergoing major liver resection. J Gastrointest Surg. 2003;7:325-330. 4. Michalopoulos GK, DeFrances MC. Liver regeneration. Science. 1997;276:60-66. The process of regeneration is incompletely under- 5. Paku S, Schnur J, Nagy P, Thorgeirsson SS. Origin and structural evolution of stood, incorporating a complex interplay among numer- the early proliferating oval cells in rat liver. Am J Pathol. 2001;158:1313-1323. ous factors. Although some factors (eg, HGF) are more 6. Petersen BE, Bowen WC, Patrene KD, et al. Bone marrow as a potential source of hepatic oval cells. Science. 1999;284:1168-1170. potent stimulators of regeneration than others, it is clear 7. Oh SH, Hatch HM, Petersen BE. Hepatic oval “stem” cell in liver regeneration. that no one substance is singularly responsible for con- Semin Cell Dev Biol. 2002;13:405-409. trolling this process. The evolving role of liver-support 8. Evarts RP, Hu Z, Fujio K, Marsden ER, Thorgeirsson SS. Activation of hepatic devices is, as yet, incompletely established and has yet stem cell compartment in the rat: role of transforming growth factor-alpha, hepa- to drive major inroads into routine clinical practice. The tocyte growth factor and acidic fibroblast growth factor in early proliferation. Cell Growth Differ. 1993;4:555-561. relationship between these devices and growth factors and 9. Webber EM, Godowski PJ, Fausto N. In vivo response of hepatocytes to growth its subsequent impact on regeneration remains to be fully factors requires an initial priming stimulus. Hepatology. 1994;19:489-497. elucidated. Some evidence suggests that elevations in lev- 10. Fausto N. Liver regeneration. J Hepatol. 2000;32:19-31.

(REPRINTED) ARCH SURG/ VOL 139, JUNE 2004 WWW.ARCHSURG.COM 675

©2004 American Medical Association. All rights reserved. Downloaded From: https://jamanetwork.com/ on 09/30/2021 11. Su AI, Guidotti LG, Pezacki JP, Chisari FV, Schultz PG. Gene expression during 38. Date M, Matsuzaki K, Matsushita M, Tahashi Y, Furukawa F, Inoue K. Modula- the priming phase of liver regeneration after partial hepatectomy in mice. Proc tion of transforming growth factor ␤ function in hepatocytes and hepatic stel- Natl Acad Sci U S A. 2002;99:11181-11186. late cells in rat liver injury. Gut. 2000;46:719-724. 12. Higgins GM, Anderson RM. Experimental pathology of the liver, I: restoration of 39. Selzner M, Clavien PA. Failure of regeneration of the steatotic rat liver: disruption liver white rat following partial surgical removal. Arch Pathol. 1931;12:186-202. at two different levels in the regeneration pathway. Hepatology. 2000;31:35-42. 13. Michalopoulos GK. Liver regeneration: molecular mechanisms of growth con- 40. Sakamoto T, Liu Z, Murase N, et al. Mitosis and apoptosis in the liver of interleukin- trol. FASEB J. 1990;4:176-187. 6–deficient mice after partial hepatectomy. Hepatology. 1999;29:403-411. 14. Stolz DB, Mars WM, Petersen BE, Kim TH, Michalopoulos GK. Growth factor sig- 41. Yamada Y, Kirillova I, Peschon JJ, Fausto N. Initiation of liver growth by tumor nal transduction immediately after two-thirds partial hepatectomy in the rat. Can- necrosis factor: deficient liver regeneraion in mice lacking type I tumor necrosis cer Res. 1999;59:3954-3960. factor receptor. Proc Natl Acad Sci U S A. 1997;94:1441-1446. 15. Francavilla A, Panella C, Polimeno L, et al. Hormonal and enzymatic parameters 42. Cressman DE, Greenbaum LE, DeAngelis RA, et al. Liver failure and defective of hepatic regeneration in patients undergoing major liver resections. Hepatol- hepatocyte regeneration in interleukin-6–deficient mice. Science. 1996;274: ogy. 1990;12:1134-1138. 1379-1383. 16. Nagino M, Ando M, Kamiya J, Uesaka K, Sano T, Nimura Y. Liver regeneration 43. Schuppan D, Schmid M, Somasundaram R, et al. Collagens in the liver extracellu- after major hepatectomy for biliary cancer. Br J Surg. 2001;88:1084-1091. lar matrix bind hepatocyte growth factor. Gastroenterology. 1998;114:139-152. 17. Nagasue N, Yukaya H, Ogawa Y, Kohno H, Nakamura T. Human liver regenera- 44. Lindroos PM, Zarnegar R, Michalopoulos GK. Hepatocyte growth factor (hepato- tion after major hepatic resection: a study of normal liver and livers with chronic poietin A) rapidly increases in plasma before DNA synthesis and liver regenera- hepatitis and cirrhosis. Ann Surg. 1987;206:30-39. tion stimulated by partial hepatectomy and carbon tetrachloride administration. 18. Yamanaka N, Okamoto E, Kawamura E, et al. Dynamics of normal and injured Hepatology. 1991;13:743-750. human liver regeneration after hepatectomy as assessed on the basis of com- 45. Liu ML, Mars WM, Zarnegar R, Michalopoulous GK. Collagenase pretreatment puted tomography and liver function. Hepatology. 1993;18:79-85. and the mitogenic effects of hepatocyte growth factor and transforming growth 19. Mann DV, Lam WW, Hjelm NM, et al. Human liver regeneration: hepatic energy factor-alpha in adult rat liver. Hepatology. 1994;19:1521-1527. economy is less efficient when the organ is diseased. Hepatology. 2001;34:557- 46. Olsen PS, Poulsen SS, Kirkegaard P. Adrenergic effects on secretion of epider- 565. mal growth factor from Brunner’s glands. Gut. 1985;26:920-927. 20. Martinez-Hernandez A, Amenta PS. The extracellular matrix in hepatic regenera- 47. Houck KA, Michalopoulos GK. Altered responses of regenerating hepatocytes to tion. FASEB J. 1995;9:1401-1410. norepinephrine and transforming growth factor type beta. J Cell Physiol. 1989; 21. Rudolph KL, Trautwein C, Kubicka S, et al. Differential regulation of extracellular 141:503-509. matrix synthesis during liver regeneration after partial hepatectomy in rats. Hepa- 48. Riordan SM, Williams R. Extracorporeal support and hepatocyte transplantation tology. 1999;30:1159-1166. in acute liver failure and cirrhosis. J Gastroenterol Hepatol. 1999;14:757-770. 22. Jirtle RL, Michalopoulos G. Effects of partial hepatectomy on transplanted hepa- 49. Mitzner SR, Stange J, Klammt S, Peszynski P, Schmidt R, Noldge-Schomburg tocytes. Cancer Res. 1982;42:3000-3004. G. Extracorporeal detoxification using the molecular adsorbent recirculation sys- 23. Starzl TE, Watanabe K, Porter KA, Putnam CW. Effects of insulin, glucagon, and tem for critically ill patients with liver failure. J Am Soc Nephrol. 2001;12(suppl insulin/glucagon infusions on liver morphology and cell division after complete 17):S75-S82. portacaval shunt in dogs. Lancet. 1976;1:821-825. 50. Kjaergard LL, Liu J, Als-Nielsen B, Gluud C. Artificial and bioartificial support sys- 24. Cruise JL, Knechtle SJ, Bollinger RR, Kuhn C, Michalopoulos G. Alpha-1- adr- tems for acute and acute-on-chronic liver failure: a systematic review. JAMA. energic effects and liver regeneration. Hepatology. 1987;7:1189-1194. 2003;289:217-222. 25. Jones DE Jr, Tran-Patterson R, Cui DM, Davin D, Estell KP, Miller DM. Epider- 51. Allen JW, Hassanein T, Bhatia SN. Advances in bioartificial liver devices. Hepa- mal growth factor secreted from the salivary gland is necessary for liver regen- tology. 2001;34:447-455. eration. Am J Physiol. 1995;268:G872-G878. 52. Bertani H, Gelmini R, Del Buono MG, et al. Literature overview on artificial liver 26. Rasmussen TN, Jorgensen PE, Almdal T, Kirkegaard P, Olsen PS. Stimulatory support in fulminant hepatic failure: a methodological approach. Int J Artif Or- effect of epidermal growth factor on liver regeneration after partial hepatectomy gans. 2002;25:903-910. in rats. Scand J Gastroenterol. 1992;27:372-374. 53. Kondrup J, Almdal T, Vilstrup H, Tygstrup N. High volume plasma exchange in 27. Hashimoto M, Kothary PC, Eckhauser FE, Raper SE. Treatment of cirrhotic rats fulminant hepatic failure. Int J Artif Organs. 1992;15:669-676. with epidermal growth factor and insulin accelerates liver DNA synthesis after 54. Iwai H, Nagaki M, Naito T, et al. Removal of endotoxin and cytokines by plasma partial hepatectomy. J Gastroenterol Hepatol. 1998;13:1259-1265. exchange in patients with acute hepatic failure. Crit Care Med. 1998;26:873- 28. Yamada A, Kawata S, Tamura S, et al. Plasma heparin-binding EGF-like growth 876. factor levels in patients after partial hepatectomy as determined with an enzyme- 55. Nakae H, Asanuma Y, Tajimi K. Cytokine removal by plasma exchange with con- linked immunosorbent assay. Biochem Biophys Res Commun. 1998;246:783- tinuous hemodiafiltration in critically ill patients. Ther Apher. 2002;6:419-424. 787. 56. Horl WH. Haemodialysis membranes: interleukins, biocompatibility, and middle 29. Webber EM, Fitz-Gerald MJ, Brown PI, Bartlett MH, Fausto N. Transforming TGF- molecules. J Am Soc Nephrol. 2002;13(suppl 1):S62-S71. alpha expression during liver regeneration after partial hepatectomy and toxic 57. Davison AM, Cameron JS, Grunfeld JP, Kerr DNS, Ritz E, Winerals CG, eds. Ox- injury, and potential interactions between TGF-alpha and HGF. Hepatology. 1993; ford Textbook of Clinical Nephrology. Vol. 2. 2nd ed. New York, NY: Oxford Uni- 18:1422-1431. versity Press Inc; 1998:1403-1417. 30. Russell WE, Kaufmann WK, Sitaric S, Luetteke NC, Lee DC. Liver regeneration 58. Memoli B, Libetta C, Rampino T, et al. Hemodialysis related induction of inter- and hepatocarcinogenesis in transforming growth factor-alpha–targeted mice. leukin-6 production by peripheral mononuclear cells. Kidney Int. 1992;42:320- Mol Carcinog. 1996;15:183-189. 326. 31. Takenoshita S, Fukushima T, Kumamoto K, Iwadate M. The role of TGF-␤ in di- 59. Wan S, LeClerc JL, Vincent JL. Cytokine responses to cardiopulmonary bypass: gestive organ disease. J Gastroenterol. 2002;37:991-999. lessons learned from cardiac transplantation. Ann Thorac Surg. 1997;63:269- 32. Nakamura T, Sakata R, Ueno T, Sata M, Ueno H. Inhibition of transforming growth 276. factor ␤ prevents progression of liver fibrosis and enhances hepatocyte regen- 60. Riegel W, Ulrich C, Sauernheimer S, Deppisch RM, Kohler H. Hepatotoxic sub- eration in dimethylnirosamine-treated rats. Hepatology. 2000;32:247-255. stance(s) removed by high-flux membranes enhances the positive acute phase 33. Bissell DM, Roulot D, George J. Transforming growth factor ␤ and the liver. Hepa- response. Kidney Int Suppl. 2001;78:S308-S314. tology. 2001;34:859-867. 61. Silvester W. Mediator removal with CRRT: complement and cytokines. Am J Kid- 34. Bissell DM, Wang SS, Jarnagin WR, Roll FJ. Cell-specific expression of trans- ney Dis. 1997;30(5, suppl 4):S38-S43. forming growth factor-beta in rat liver: evidence for autocrine regulation of hepa- 62. Bouman CS, van Olden RW, Stoutenbeek CP. Cytokine filtration and adsorption tocyte proliferation. J Clin Invest. 1995;96:447-455. during pre- and postdilution hemofiltration in four different membranes. Blood 35. Russell WE, Coffey RJ Jr, Ouellette AJ, Moses HL. Type ␤ transforming growth Purif. 1998;16:261-268. factor reversibly inhibits the early proliferative response to partial hepatectomy 63. Wakabayashi Y, Kamijou Y, Soma K, Ohwada T. Removal of circulating cytok- in the rat. Proc Natl Acad Sci U S A. 1988;85:5126-5130. ines by continuous haemofiltration in patients with systemic inflammatory re- 36. Sanderson N, Factor V, Nagy P, et al. Hepatic expression of mature transform- sponse syndrome or multiple organ dysfunction syndrome. Br J Surg. 1996;83: ing growth factor ␤1 in transgenic mice results in multiple tissue lesions. Proc 393-394. Natl Acad Sci U S A. 1995;92:2572-2576. 64. Rogiers P. Hemofiltration treatment for sepsis: is it time for controlled trials? 37. Armendariz-Borunda J, Seyer JM, Kang AH, Raghow R. Regulation of TGF beta Kidney Int Suppl. 1999;72:S99-S103. gene expression in rat liver intoxicated with carbon tetrachloride. FASEB J. 1990; 65. Sieberth HG, Kierdorf HP. Is cytokine removal by continuous hemofiltration fea- 4:215-221. sible? Kidney Int Suppl. 1999;72:S79-S83.

(REPRINTED) ARCH SURG/ VOL 139, JUNE 2004 WWW.ARCHSURG.COM 676

©2004 American Medical Association. All rights reserved. Downloaded From: https://jamanetwork.com/ on 09/30/2021 66. Tilles AW, Berthiaume F, Yarmush ML, Tompkins RG, Toner M. Bioengineering tracorporeal liver assist device in acute liver failure. Artif Organs. 1998;22:854- of liver assist devices. J Hepatobiliary Pancreat Surg. 2002;9:686-696. 858. 67. Wang L, Sun J, Li L, Mears D, Horvat M, Sheil AG. Comparison of porcine hepa- 74. Gerlach J, Jorres A, Trost O, et al. Side effects of hybrid liver support therapy: tocytes with human hepatoma (C3A) cells for use in a bioartificial liver support TNF-alpha liberation in pigs, associated with extracorporeal bioreactors. Int J Ar- system. Cell Transplant. 1998;7:459-468. tif Organs. 1993;16:604-608. 68. Kobayashi N, Noguchi H, Totsugawa T, et al. Insertion of a suicide gene into an 75. Lonnemann G, Linnenweber S, Burg M, Koch K. Transfer of endogenous pyro- immortalized human hepatocyte cell line. Cell Transplant. 2001;10:373-376. gens across artificial membranes? Kidney Int Suppl. 1998;66:S43-S46. 69. Gerlach JC, Encke J, Hole O, Muller C, Ryan CJ, Neuhaus P. Bioreactor for a larger 76. Stange J, Hassanein TI, Mehta R, Mitzner SR, Bartlett RH. The molecular adsor- scale hepatocyte in vitro perfusion. Transplantation. 1994;58:984-988. bents recycling system as a liver support system based on albumin dialysis: a sum- 70. Funatsu K, Ijima H, Nakazawa K, Yamashita Y, Shimada M, Sugimachi K. Hybrid mary of preclinical investigations, prospective, randomized, controlled clinical trial, artificial liver using hepatocyte organoid culture. Artif Organs. 2001;25:194- and clinical experience from 19 centers. Artif Organs. 2002;26:103-110. 200. 77. Awad SS, Sawada S, Soldes OS, et al. Can the clearance of tumor necrosis fac- 71. Suh KS, Lilja H, Kamohara Y, et al. Bioartificial liver treatment in rats with ful- tor alpha and interleukin 6 be enhanced using an albumin dialysate hemodiafil- minant hepatic failure: effect on DNA-binding activity of liver-enriched and growth- tration system. ASAIO J. 1999;45:47-49. associated transcription factors. J Surg Res. 1999;85:243-250. 78. Ellis AJ, Hughes RD, Nicholl D, et al. Temporary extracorporeal liver support for 72. Miwa Y, Ellis AJ, Hughes RD, Langley PG, Wendon JA, Williams R. Effect of ELAD severe acute alcoholic hepatitis using the BioLogic-DT. Int J Artif Organs. 1999; liver support on plasma HGF and TGF-beta 1 in acute liver failure. Int J Artif Or- 22:27-34. gans. 1996;19:240-244. 79. Levy H, Ash SR, Knab W, Steczko J, Carr DJ, Blake DE. Systemic inflammatory 73. Hughes RD, Nicolaou N, Langley PG, Ellis AJ, Wendon JA, Williams R. Plasma response syndrome treatment by powdered sorbent pheresis: The Biologic- cytokine levels and coagulation and complement activation during use of the ex- Detoxification Plasma Filtration System. ASAIO J. 1998;44:M659-M665.

IN OTHER AMA JOURNALS

ARCHIVES OF INTERNAL MEDICINE Prevalence and Risk Factors for Carriage of Methicillin-Resistant Staphylococcus aureus at Admission to the Intensive Care Unit: Results of a Multicenter Study Jean-Christophe Lucet, MD, MPH; Sylvie Chevret, MD, PhD; Isabelle Durand-Zaleski, MD, PhD; Claude Chastang, MD, PhD; Bernard Re´gnier, MD; for the Multicenter Study Group Background: Detection of methicillin-resistant Staphylococcus aureus (MRSA) carriers on admission to the intensive care unit (ICU) is an important component of strategies for controlling the spread of MRSA. Methods: A prospective multicenter study was conducted in 14 French ICUs for 6 months. All patients were screened within 24 hours after admission, using nasal and cutaneous swabs. In addition, clinical samples were obtained. Patient data col- lected on ICU admission included presence of immunosuppression; history of hospital stay, surgery, antimicrobial treat- ments, or previous colonization with MRSA; chronic health evaluation and McCabe scores; reason for admission; whether the patient was transferred from another ward; severity of illness; presence of skin lesions; and invasive procedures. Risk factors for MRSA carriage at ICU admission were estimated, and significantly associated variables were used to develop a predictive score for MRSA carriage. A cost-benefit analysis was then performed. Results: Of the 2347 admissions with MRSA screening, 162 (6.9%; range, 3.7%-20.0% among ICUs) were positive for MRSA, of whom 54.3% were detected through screening specimens only. Of the 2310 first admissions (vs repeat admissions) to the ICU, 96 were newly identified MRSA carriers. Factors associated with MRSA carriage in the multivariate analysis were age older than 60 years, prolonged hospital stay in transferred patients, history of hospitalization or surgery, and presence of open skin lesions in directly admitted patients. Only universal screening detected MRSA carriage with acceptable sensitivity. A cost-benefit analysis confirmed that universal screening and preventive isolation were beneficial. Conclusions: The prevalence of MRSA carriage on admission to the ICU is high in this endemic setting. Screening for MRSA on admission is useful to identify the imported cases and should be performed in all ICU-admitted patients. (2003;163:181-188) Corresponding author and reprints: Jean-Christophe Lucet, MD, MPH, Unite´ d’Hygie`ne et de Lutte contre l’Infection Nosocomiale, Hoˆpital Bichat–Claude Bernard, 75877 Paris CEDEX 18, France (e-mail: [email protected]).

(REPRINTED) ARCH SURG/ VOL 139, JUNE 2004 WWW.ARCHSURG.COM 677