Management of hepatorenal syndrome

Name of journal: World Journal of Nephrology

ESPS Manuscript NO: 12259

Columns: MINIREVIEW
Management of hepatorenal syndrome
Dundar HZ et al. Management of hepatorenal syndrome
Halit Ziya Dundar, Tuncay Yılmazlar
Halit Ziya Dundar, Tuncay Yılmazlar, Department of General Surgery, Faculty of Medicine, Uludag University, Gorukle, 16285 Bursa, Turkey
Author contributions: Both authors contributed to this manuscript.
Conflict-of-interest: The authors declare no conflicts of interest regarding this manuscript.
Open-Access: This article is an open-access article which was selected by an in-house editor and fully peer-reviewed by external reviewers. It is distributed in accordance with the Creative Commons Attribution Non Commercial (CC BY-NC 4.0) license, which permits others to distribute, remix, adapt, build upon this work non-commercially, and license their derivative works on different terms, provided the original work is properly cited and the use is non-commercial. See: http://creativecommons.org/licenses/by-nc/4.0/
Correspondence to: Halit Ziya Dundar, MD, Department of General Surgery, Faculty of Medicine, Uludag University, Gorukle, 16285 Bursa, Turkey. [e-mail]
Telephone: [phone]

Fax: [phone]
Received: June 29, 2014

Peer-review started: June 29, 2014

First decision: August 14, 2014

Revised: December 29, 2014

Accepted: Janurary 30, 2015

Article in press:

Published online:
Abstract

Hepatorenal syndrome (HRS) is defined as development of renal dysfunction in patients with chronic liver diseases due to decreased effective arterial blood volume. It is the most severe complication of cirrhosis because of its very poor prognosis. Inspite of several hypothesis and researches, the pathogenesis of HRS is still poorly understood. The onset of HRS is a progressive process rather than a suddenly arising phenomenon. Since there are no spesific tests for HRS diagnosis, it is diagnosed by exclusion of other causes of acute kidney injury in cirrhotic patients. There are two types of HRS with different characteristics and prognostics. Type 1 HRS is characterized by a sudden onset acute renal failure and a rapid deterioration of other organ functions. It may develop spontaneously or due to some precipitating factors. Type 2 HRS is characterized by slow and progressive worsening of renal functions due to cirrhosis and portal hypertension, and it is accompained by refractory ascites. The only definitive treatment for both Type 1 and Type 2 HRS is liver transplantation. The most suitable bridge treatment or treatment for patients who are not eligible for transplantation is combination of terlipressin and albumin. For the same purpose it is possible to try hemodialysis or renal replacement therapies in form of continuous veno-venous hemofiltration. The artificial hepatic support systems are important for patients who do not respond to medical treatment. Transjugular intrahepatic portosystemic shunt may be considered as a treatment modality for unresponsive patients to medical treatment. The main goal of clinical surveillance in a cirrhotic patient is prevention of HRS before it develops. The aim of this article is to provide an updated review about physiopathology of HRS and its treatment.

Key words: Hepatorenal syndrome; Cirrhosis; Renal failure; Vasoconstrictors; Transplantation
© The Author(s) 2015. Published by Baishideng Publishing Group Inc. All rights reserved.
Core tip: Hepatorenal syndrome (HRS) is a severe complication of chronic liver diseases and is usually assosiated with poor prognosis. It is not a renal disease but a renal dysfunction developing as a result of a systemic condition associated with liver failure. To prevent HRS by taking some preventive measures is possible and although the definitive treatment is liver transplantation, a rapid diagnosis and prompt initiation of the treatment leads to an important improvement in the prognosis. In this review we cover the physiopathology, diagnosis and treatment options of HRS.
Dundar HZ, Yılmazlar T. Management of hepatorenal syndrome. World J Nephrol 2015; In press
INTRODUCTİON

Hepatorenal syndrome (HRS) is defined as unexplainable progressively increasing serum creatinine in a patient with advanced liver disease. HRS is representative of the end stage of a process associated with progressive decrease in renal blood flow and glomerular filtration rate (GFR). Diagnosis is by exclusion of other causes of renal failure since there is no spesific diagnostic test. In 1956 a special type of acute renal failure which was associated with low urinary output, very low urine sodium excreation without proteinuria had been defined^[1]. In postmortem examination of these patient it was observed that kidney’s histologic structure was conserved. In 1969 kidney, taken from cadaveric donors with HRS did function normally^[2]. So it is possible to conclude that HRS is not a renal disease, but a renal dysfunction occuring as a result of a systemic condition. The definitive treatment is liver transplantation(LT). HRS is an important risk factor since it increases the waiting list mortality and incidence of complications after LT^[3]. And also renal function before LT is a predictive of survival^[4]. Vasoconstrictive agents constitute the main part of pharmalogic treatment providing a bridge to LT. Hemodialysis, renal replacement therapies and artificial liver support systems may also be used as bridge treatment. The goal of treatment in HRS should be early diagnosis of disease, effective and quick treatment but most important of all is to take preventive measures. Despite all treatment options likelihood of failure is still high.

DEFINITION

HRS is one of the most severe complications of cirrhosis and defined as renal insufficiency emerging in chronic liver disease patients when all the other causes of renal failure are excluded^[5]. Renal vasoconstriction which is a result of progressive liver failure is the main underlying reason of renal failure in HRS.

HRS is first classified into two groups as type 1 and type 2 by International Ascites Club in 1994. According to this classification type 1 HRS is associated with doubling of initial serum creatinin to a level more than 2.5 mg/dL or reduction in creatinine clearence because of decreased glomeruler filtration rate to a level less than 20 mL/min in a time period shorter than 2 wk^[5-7]. Type 1 HRS usually occurs following a precipitating factor such as infectious conditions particularly spontaneous bacterial peritonitis (SBP) which is considered as most important factor for HRS^[8-11]. Type 2 HRS is a moderate and steady type of renal failure and serum creatinin level is higher than 1.5 mg/dL and is often associated with sodium retantion^[5,7]. Type 2 HRS usually arises spontaneously as a result of refractory ascites^[5].

In addition to this data it is important to consider that the creatinine levels should not be always increased in case of renal failure in decompansated cirrhosis^[12,13]. It is possible to say that even milder degrees of renal failure may be associated with poor prognosis in cirrhotic patients^[1,14]. According to RIFLE (Risk, Injury, Failure, Loss, End stage renal disease) classification it is shown that even a small increase in serum creatinine level may be assosiated with clinically significant outcomes in patients with cirrhosis^[15-17]. In accordance to this The International Ascites Club and the Acute Dialysis Quality Initiative suggested a new definition for acute kidney injury. According to this new definition an increase in serum creatinine level to 0.3 mg/dL or more in a period less than 48 h or a 50% increase in serum creatinine level compared to the baseline levels recorded in last 6 months period regardless of final serum creatinine levels^[18].

PATHOPHYSIOLOGY

HRS is a sort of renal dysfunction which is generally reversible and occuring because of advanced liver disease. Although it is not completely ravelled, the most characteristic reason underlying renal dysfunction in HRS is renal vasoconstriction^[19].

Four major factors considered to be responsible are: (1) Decreased circulating blood volume and as a result decreased mean arterial blood pressure because of splanchnic vasodilatation; (2) Renal vasoconstriction as a result of activated renin-angiotensin-aldosteron system since sympathetic nervous system has been activated; (3) Cardiac dysfunction due to cirrhosis; and (4) Release of several cytokines and vasoactive mediators which may affect blood flow to kidneys and glomeruler vascular bed^[20,21]

The main pathophysiologic mechanism in HRS is reduction of circulating blood volume due to increased resistance to blood flow in cirrhotic liver resulting in splanchnic blood pooling which is infact a multifactorial process^[1]. Decreased circulating blood flow which means decreased mean arterial blood pressure causes stimulation of baroreseptors in the carotid body and consequently activation of sympathetic nervous system. This is followed by activation of renin-angiotensin-aldosteron system and nonosmotic release of vasopressin which causes furthermore decrease in systemic vascular resistance, hypotension and by the way vasoconstriction in renal vessels and glomeruler vascular bed^[22]. This vasoconstriction cannot be only explained with increased activity of endogenous vasoconstrictor systems. Because of extreme hemodynamic changes in advanced liver diseases renal vasodilatator systems become insufficient creating a vicrous cycle which contributes more and more to renal vasoconstriction^[22-24].

Factors contributing to the persistence of renal vasoconstriction in spite of vasodilatation of peripheral vasculature are investigated in several studies. Iwao et al^[25] investigated the contributing factors of hyperdynamic circulation in cirrhotic patients and they found that mesenteric blood flow decreases as liver disease worsens. They concluded that splanchnic arterial vasodilatation plays an important role in the pathogenesis of decreased systemic vascular resistance in cirrhotic patients^[25]. In accordance with this data in some other human and animal studies it is shown that splanchnic circulation is the main vascular bed responsible for peripheral vasodilatation^[26-29].

Advanced liver disease due to portal hypertension is characterized by a state of hyperdynamic circulation which is accompanied by increased cardiac output^[30]. It is hard to understand how cardiac output is increased while myocardial function is usually impaired in cirrhotic patients. The heart in cirrhotic patients usually has several structural and functional abnormalities associated with alterations in ventricular wall size, systolic and diastolic function^[31,32]. Although the reason of these alterations are not known clearly, neurohumoral factors and continous mechanical stress may be responsible^[33]. Ventricular function is inhibited due to circulating cytokines such as tumor necrosis factor -alfa, nitric oxide in cirrhotic patients. One of the contributing factors to the ventricular dysfunction is reduced beta adrenergic receptors signal transduction in the myocardium^[34,35].

Whatever the cause is ventricular wall thickness is increased slightly, the diastolic function deteriorates especially increasing with physical stress and by the presence of ascites and systolic dysfunction^[30,34,36].

Sympathetic nervous system activity is shown to be increased in case of portal hypertension as a result of hepatorenal reflex^[37,38]. Hepatorenal reflex activation occurs due to decreased sinusoidal blood flow or increased sinusoidal pressure in the liver which is shown in several animal models^[38,39]. Increased renal sympathetic nervous system tone is held to be responsible for renal vasoconstriction together with thromboxanes, endotoxins, endothelins and neurotransmitters. Together with activation of sympathetic nervous system because of low effective circulating volume stimulating baroreseptors in carotid body and aortic arch, activation of renin-angiotensin-aldosteron system and nonosmotic release of antidiuretic hormon occurs. Although all of these compensatory mechanisms help providing effective circulating volume and relatively normalizing the mean arterial blood pressure, they also have important effects on renal functions.

Vasoactive mediators and cytokines are the other actors of HRS. These are the agents affecting both systemic and renal circulation. The major ones studied include prostaglandins, endothelins, endotoxins, glucagon, nitric oxide and tumor necrosis factor- alfa. Among those nitric oxide has a special role. Primary arterial vasodilatation in splanchnic circulation which is a result of portal hypertension is the mainstay in explaining the development of renal insufficiency in cirrhotic patients^[20]. The major cause of this arterial vasodilatation in splanchnic circulation is increased synthesis and activity of nitric oxide and some other vasoactive agents^[20]. The correlation between increased levels of nitric oxide and high plasma renin-angiotensin-aldosteron system activity and antidiuretic hormon levels accompained by low urinary Na excretion in cirrhotic patients especially with ascites is remarkable^[40,41]. It is thought that in the maintanence of hyperdynamic circulation which is the hallmark of HRS nitric oxide may be the primary factor^[42]. However increased nitric oxide levels are not able to prevent renal vasoconstriction. In the early stages of cirrhosis renal perfusion is provided by increased synthesis and activity of renal vasodilatators especially prostaglandins and kallikreins^[43,44]. Vasodilating prostaglandins are the major actors in supplying glomerular blood flow at the beginning^[45]. But as the liver disease progresses vasoconstrictor systems are furthermore activated and synthesis and activity of renal vasodilating factors progressively decrease. The prostaglandin level in the urine of cirrhotic patients is high when compaired with that of patients with HRS^[46]. The reason of decreased prostaglandin production in HRS is a mystery but it is known that it is not the only factor in development of HRS.
DİAGNOSİS

HRS is an important risk factor for renal failure in cirrhotic patients. In a prospective study it is estimated that 1 year probability of HRS in cirrhotic patients is 18% and 5 year probability is 39%^[47]. HRS is observed in 28% of alcoholic hepatitis cases without identifiable cirrhosis^[48]. Major factors precipitating HRS are hyponatremia, high plasma renin-angiotensin-aldosteron system activity, gastrointestinal bleeding, bacterial infections, spontaneous bacterial peritonitis, large volume paracentesis without albumin infusion, some drugs such as diuretics, aminoglicosides, non-steroid anti-inflamatory drugs, angiotensin converting enzyme inhibitors, surgical interventions and cholestasis^[47,49,50]. Also Doppler Ultrasonographie may be helpful to detect increased renal resistive indices indicating renal vasoconstriction^[19].

In chronic liver diseases it may be difficult to diagnose renal failure since reduction in GFR is usually masked. This may be because urea and creatinine production is decreased due to chronic liver disease, the muscle mass is decreased due to chronic disease and the protein intake is decreased due to loss of desire to eat. There are no spesific diagnostic criteria for diagnosis of HRS. The diagnosis is by exclution of other causes of renal failure in cirrhotic patients. The major symptoms of HRS are decreased GFR (< 40 mL/min) and increased serum creatinine (> 1.5 mg/dL). Other symptoms defining functional characteristic of HRS are decreased Na excretion (< 10 mmol/L), higher urine osmolality compared to plasma osmolality, hyponatremia (< 130 mmol/L) and decreased diuresis (< 500 mL).

Most widely accepted diagnostic criteria are developed in 1996 by the International Ascites Club and major and minor criteria are defined. According to this diagnostic criteria diagnosis of HRS requires the inclusion of all major criteria and presence of minor criteria are thought to be promotive for the diagnosis of HRS. Various new concepts have arised since the first publication of criteria for the diagnosis of HRS in 1996. These were modified in 2007 by International Ascites Club^[6]. According to current diagnostic critera minor criteria are omitted and concurrent bacterial infection is now not a factor that should be excluded for the diagnosis of HRS. Another important alteration is using albumin instead of 0.09% NaCl solution for plasma volume expantion.

The new diagnostic criteria defined by International Ascites Club in 2007 are listed in Table 1.

Creatinine clearence is the most important diagnostic tool. It is important to exclude other causes of renal failure before the diagnosis of HRS. These include; hypovolemia, parenchymal renal diseases, use of nephrotoxic drugs and shock. In case of hematuria, severe proteinuria and increased renal size in USG renal parenchymal disease should be considered in differential diagnosis^[51]. In such cases renal biopsy is required so that potential need for combined liver and kidney transplantation could be defined^[51]. If there is an organic cause of renal insufficiency making a urine analysis and to see the urine Na concentration will be clinically important. Since muscle mass and production of creatinine in the liver is decreased in chronic liver diseases serum creatinine levels are not very reliable to evaluate renal functions in liver diseases. Such as creatinine monitoring blood urea level is also insufficient in reflecting the GFR in case of chronic liver disease^[5,6]. So that investigations are being conducted to find out more sensitive and spesific markers. Some of these markers are Cistatin-C, Symetric Dimethyl Arginin, Kidney Injury Molecule-1 and Neutrophil Gelatinase-Assosiated Lipokalin. Cistatin-C is found to be more sensitive than creatinine in defining decreased GFR in cirrhotic patients^[52,53]. Symetric Dimethyl Arginin which is an endogenous inhibitor of nitric oxide synthase is shown to be increased in case of HRS when compared with cirrhotic patients with normal kidney functions^[54]. The investigations about Kidney Injury Molecule-1 and Neutrophil Gelatinase-Assosiated Lipokalin which are very susceptible to ischemia and indicators of renal tubuler injury are still going on^[55,56]. There are few studies about renal tubuler damage markers. β2 microglobulin is one of them. It is especially increased in case of aminoglicoside nephrotoxicity^[57]. To make differential diagnosis of HRS from acute tubuler necrosis gama glutamyl transpeptidase, transaminase, Neutrophil Gelatinase-Assosiated Lipokalin, IL 8, liver type fatty acid binding protein and hepatitis virus cell receptor type 1 are other markers which are of interest nowadays, but their significance has not yet been evaluated^[58].

TREATMENT

The main principle in treatment of HRS is to bring back the renal function until the patients undergo LT. So all the therapeutic interventions for HRS are sort of bridge therapy. During the treatment of HRS etiology oriented treatment of liver disease such as antiviral drug treatment should not be impeded. Choice of medical treatment depends upon several factors including availability of drugs which is variable according to countries even regions, whether the patient is admitted to intensive care unit and the patient is a candidate for LT. Cirrhotic patients with gastrointestinal system (GIS) bleeding, ascites, infections, arterial hypotension and dilutional hyponatremia should be monitored closely because of increased risk of HRS.

Complications like GIS bleeding and spontaneous bacterial peritonitis should be prevented and urgently treated. Large volume paracentesis with plasma volume expension decreases the incidence of HRS. Diuretic treatment may trigger HRS because of intravascular volume depletion so diuretic treatment should be stopped and electrolyte imbalance such as hyponatremia and hypocalcemia should be corrected. NSAIDs should also be stopped and appropiate infection treatment should be planned.

Effective circulating volume should be increased. Infusion of 0.9% NaCl and synthetic plasma expanders even by monitoring central venous pressure is not found to be helpful. It is proved that albumin is the most useful of all volume expanders. After albumin use incidence of type 1 HRS has been shown to be decreased. When albumin is used concomitantly with other agents, it is observed that the effectiveness of these agents were also increased^[23,51].

Prostaglandins, dopamin and endothelin receptor blockers are the first renal vasodilatators used in HRS treatment. Oral prostaglandin-E₁ analogue misoprostol or IV prostaglandin infusion did not provide a significant improvement in HRS^[59,60]. Intravenous dopamin infusion has also been investigated in several studies but any improvements aren’t observed in renal functions^[61,62].

There is no spesific vasoconstrictive agent used to increase the systemic vascular resistance. Several vasoconstrictive agents such as norepinephrine, angiotensin 2 and vasopressin are used for this purpose but alone they were not found to be effective. Vasoconstrictive agents especially when used together with plasma expanders are the most helpful pharmacologic agents in the management of HRS^[63,64]. Development of synthetic vasopressin analogues provide an important progression in the HRS treatment. Ornipressin and terlipressin are vasoconstrictive agents that are effective on mesenteric circulation rather than renal and other vascular systems. Ornipressin is not being used because of its severe ischemic side-effects.

Terlipressin and albumin infusion is the most important choice of treatment in type 1 HRS^[65]. It is observed that terlipressin is effective in 40%-60% of patients with type 1 HRS. Clinical response to the terlipressin treatment is slow but the reduction in serum creatinine level is continuous^[65,66]. To reverse HRS may take a long time and it is observed that in 50% of patients recurred. In case of recurrence usually same treatment regimen is found to be succesful^[7]. When terlipressin and albumin treatment is succesful arterial blood pressure, urine amount and serum Na level increase. Systemic circulation improves and plasma renin and norepinephrin levels decrease significantly. Time required for recovery usually changes depending on the initial serum creatinine level, but mean recovery time is 7 d. If the initial serum creatinine level is low the recovery will be faster^{[23,51,67,68]}. Terlipressin therapy is suggested to be in combination with albumin. Terlipressin therapy may be given as IV bolus (0.5-1 mg/4-6 h) or IV continous infusion with an initial dose of 2 mg/d. During the follow-up period if 25% decrease is not observed in serum creatinine level, IV bolus dose may be increased up to 2 mg/4 h or IV continous infusion dose may be increased up to maximum 12 mg/d. Monitoring CVP is essential and albumin infusion is required to retain CVP at a level of 10-15 cmH₂O. Albumin is given for 2 d in form of IV bolus theraphy with an initial dose of 1 g/kg (maximum 100 g/d) and the maintenance albumin dose should be 25-50 g/d until terlipressin therapy is ceased and serum creatinine level becomes normal^[69]. There are some studies that showed nearly 75% improvement in HRS patients by using IV continuous infusion of terlipressin. In these studies how terlipressin is given also found to be important^[70-72]. The information about treatment of type 2 HRS by albumin and vasoconstrictive agents is limited. When albumin and vasoconstrictive agents are used in type 2 HRS treatment improvement in renal functions is observed but there is a 50% recurrence rate after the cessation of therapy^[23,73,74]. The most common side effects of terlipressin treatment are cardiovascular and ischemic ones and they are reported nearly 12% of patients treated^[20,75]. According to the 2012 cochrane meta-analysis during terlipressin therapy GIS and infectious side effects did not increase significantly whereas cardiovascular side effects increased remarkably^[76].

Other vasoconstrictive agents currently being used in HRS treatment are somatostatin analogues (ocreotide), α-adrenergic agonists, midodrine and norepinephrine. Their effectiveness is studied in several studies, some found they are less effective than terlipressin^[23,77,78] and some found their effectiveness similar with terlipressin^[79,80]. Midodrine is an orally available α-adrenergic agonist. Its effect is systemic vasoconstiction. When it is used in combination with ocreotide and albumin systemic and renal hemodynamic status is improved^[81]. Midodrine is given orally with an initial dose of 7.5 mg/8 h (maximum: 15 g/8 h) and ocreotide may be given either as continuous infusion with a dose of 50 mcg/h or subcutaneously with a dose of 100-200 mcg/8 h. In combination with midodrine and ocreotide, albumin will be given as IV bolus with an initial dose of 1 g/kg (maximum: 100 g) and the maintanence dose of albumin will be 25-50 g/d. Using midodrine and ocreotide incombination has been shown to decrease mortality^[82]. Nevertheless number of patients reported using this therapy is not enough^[78,83]. So more trials are required for a more accurate conclusion.

Norepinephrine is a vasoconstrictive agent generally used in intensive care units since it isn’t convenient to use in general medical wards. It is given as intravenous continuous infusion with a dose of 0.5-3 mg/h. The effectiveness of norepinephrine and terlipressin were shown to be similar while norepinephrine was cheaper^[80,84].

Since dopamin is known to decrease renal vascular resistance and increase renal blood flow, low doses of dopamin was tried in the past. Neither alone nor in combination with ornipressin its clinical effectiveness could not be proved and results are controversial^[23,78].

According to several studies increasing mean arterial blood pressure is suggested to have favorable effects on treatment process. Mostly used predictors of favorable treatment response are a serum bilirubin concentration of < 10 mg/dL and an increase in mean arterial blood pressure ≥ 5 mmHg on the third day of treatment^[68].

Principle is that the earlier the treatment has been started the better the results are. If serum creatinine level is < 5 mg/dL when the treatment is started probability of favorable response is increased.

In a study by Nazar et al^[68] in patients with both decreased bilirubin level and increased mean arterial blood pressure treatment success was 100%. In patients with only decreased bilirubin level the success rate was found to be 53% and it was 25% in patients with only increased mean arterial blood pressure. In patient group neither bilirubin levels decreased nor mean arterial blood pressures increased the success rate was 10%^[68]. If there is treatment unresponsiveness underlying renal disease other than HRS should be considered.

The goal of all vasopressor treatments is to achieve a 10-15 mmHg increase in the mean arterial blood pressure. Increased mean arterial blood pressure is usually associated with decreased serum creatinine levels^[85].

For patients with HRS who are not admitted to intensive care unit combination therapy with terlipressin and albumin is suggested. If terlipressin is not avaliable combination therapy with midodrine, ocreotide and albumin should be used in patients who are not in the intensive care unit. After two weeks of medical treatment if there weren’t any improvements in renal functions medical treatment would have been considered to be useless.

Transjuguler interahepatic portosystemic shunts (TIPS) have been reported to improve renal function in patient with type 1 HRS^[86-88], and it is also being used for the treatment of refractory ascites in patients with type 2 HRS^[89-91]. However TIPS therapy is possible under limited circumstences because of its contraindications and complications. Major complications associated with TIPS are hepatic encephalopathy which is a common and treatable condition, worsening of hepatic functions, bleeding due to procedure and acute kidney injury because of intravenous contrast injection during the procedure^[92]. The underlying mechanism explaining how renal functions are improved after TIPS is not known completely. TIPS provides portal decompression in cirrhotic patients, portal pressure decreases and blood pooling in the splancnic vascular bed returns to the systemic circulation. As a result of this RAAS and SNS activity is supressed and renal vasoconstriction improves. In a study investigating renal functions after TIPS in seven patients with type 1 HRS a significant decrease in serum creatinine level and increase in urine volume was observed in six of the patients in one month period. These are accompained by significant improvement in renal blood flow and GFR^[86]. However amelioration of renal functions may take as long as six months in some cases after TIPS^[89]. Also effect of TIPS on survival in type 1 HRS patients is appreciable. HRS is improved in nearly 50% of patients and survival increased more than three months after TIPS^[78,87]. In case of type 2 HRS when TIPS is applied to take ascites under control it is found to be succesful and 70% of patients survived in the first following year^[87,93]. The average survival after TIPS was approximately five months and this was longer than the expected survival for such patients^[86]. Unfortunately mostly it is too late for patients with HRS to undergo TIPS and so it is suggested as a choice of treatment for only a selected group of patients. Before the decision to apply TIPS high incidence of complications and especially encephalopathy should be considered.

Although LT is the most effective and definitive treatment of liver failure and HRS, until LT supportive treatment modalities are required. Non biologic liver support systems are developed fort his purpose. The mechanism of action is provided by detoxification through a semi-permeable membrane in those non biologic support systems. During the course of liver failure according to the dominant clinical presentation (HRS, hepatopulmonary syndrome, hyperbilirubinemia) among different types of support systems each with different prominent features the most appopriate one will be chosen. Whereas in HRS veno-venous hemodiafiltration constitutes the first choise of treatment in case of treatment unresponsiveness molecular adsorbents recirculation system should be considered. High-flux dialysis provides effective elimination of water soluble substances such as ammoniac and lactate. But it is insufficient in eliminating substances binding to proteins such as bile acids. Plasma exchange is not being used nomore. It is too risky since large volumes are required to be exchanged in this procedure. For that reason nowadays continue veno-venous hemodiafiltration may be useful in patients with HRS if there is a reversible precipitating factor such as infections^[71]. Hemodialysis or continuous hemofiltration is used in treatment of acute renal failure in cirrhotic patients^[94,95]. In a study by Witzke et al^[96] thirty day survival was reported as 50% after renal replacement theraphy. But it is obvious that long term survival is usually poor^[96]. According to Acute Dialysis Quality Initiative Group renal support therapies will be suggested for patients who are candidates for LT^[71]. There are several studies reporting albumin dialysis has been beneficial in HRS^[23,97]. In a randomized controlled trial in management of type 3-4 hepatic encephalopathy molecular adsorbents recirculation system was observed to be more effective and safer when compared to standart medical treatment^[23,97]. However the data about this subject in the literature is limited.

Renal transplantation is the best treatment choice for both type 1 and type 2 HRS patients^[98]. If liver transplantation is performed after the HRS is improved, posttransplantation morbidity and mortality is decreased. Three year survival after liver transplantation in patients with HRS is 60%, while it is 70%-80% if it is performed before HRS has been developed^[98,99].