PRiMeR

­­Hepatorenal syndrome
Pere Ginès1,2,3*, Elsa Solà1,2,3, Paolo Angeli4, Florence Wong5, Mitra K. Nadim6
and Patrick Kamath7
Abstract | Hepatorenal syndrome (HRS) is a form of kidney function impairment that
characteristically occurs in cirrhosis. Recent changes in terminology have led to acute HRS
being referred to as acute kidney injury (AKI)-HRS and chronic HRS as chronic kidney disease
(CKD)-HRS. AKI-​HRS is characterized by a severe impairment of kidney function owing to
vasoconstriction of the renal arteries in the absence of substantial abnormalities in kidney
histology. Pathogenetic mechanisms involve disturbances in circulatory function due to a marked
splanchnic arterial vasodilation, which triggers the activation of vasoconstrictor factors.
An intense systemic inflammatory reaction that is characteristic of advanced cirrhosis may also
be involved. The main triggering factors of AKI-​HRS are bacterial infections, particularly
spontaneous bacterial peritonitis. The diagnosis of AKI-​HRS is a challenge because of a lack of
specific diagnostic tools and mainly involves the differential diagnosis from other forms of AKI,
particularly acute tubular necrosis. The prognosis of patients with AKI-​HRS is poor, with a median
survival of ≤3 months. The ideal treatment for AKI-​HRS is liver transplantation in patients without
contraindications. Medical therapy consists of vasoconstrictor drugs to counteract splanchnic
arterial vasodilation together with volume expansion with albumin. Effective measures to prevent
AKI-​HRS include early identification and treatment of bacterial infections and the administration
of albumin in patients with spontaneous bacterial peritonitis.

Hepatorenal syndrome (HRS) is a disorder character-
ized by marked impairment of kidney function that
occurs in the setting of severe chronic liver disease,
particularly advanced cirrhosis, but it may also occur
during acute liver failure1–3. In HRS, there is a marked
reduction in glomerular filtration rate (GFR) (that is,
the level of kidney function) and an increase in serum
creatinine (a breakdown product of creatine phosphate
in muscles and a marker of kidney function) levels,
which occur in the absence of substantial alterations in
kidney histology; this is in contrast to what occurs
in most instances of kidney function impairment (for
example, acute tubular necrosis (ATN)), in which there
are marked changes in kidney histology. Importantly,
HRS is potentially reversible, which means that serum
creatinine and GFR can return to within the normal
range. Reversal of impaired kidney function may occur
after liver transplantation, after pharmacological ther-
apy using vasoconstrictors, or spontaneously (in a small
number of patients). Additionally, severe liver disease
is frequently associated with impairment of function of
other organs besides the kidney and liver, particularly
the cardiovascular system. The short-​term prognosis of
*e-​mail: pgines@
clinic.cat HRS is usually poor and depends on the reversibility
https://doi.org/10.1038/ of kidney impairment and associated organ failures.
s41572-018-0022-7 In addition, making a diagnosis of HRS is challenging
because there are no specific diagnostic markers.
Moreover, patients with liver disease may develop
kidney impairment owing to causes other than HRS,
particularly volume depletion, ATN, drug toxicity and
glomerulointerstitial diseases. Key terms in HRS are
summarized in Box 1.
Typically, HRS occurs in two different clinical
patterns. In the first pattern, known historically as
type 1 HRS, an abrupt impairment of kidney func-
tion occurs4. These patients meet the modern criteria
of acute kidney injury (AKI) outlined by the Kidney
Diseases Improving Global Outcomes (KDIGO) (Box 2),
and, therefore, this pattern of HRS is now known as
AKI-​HRS. However, the term type 1 HRS is still very
frequently used in clinical practice. For decades, no con-
clusive definition existed for AKI in patients with cir-
rhosis; the most accepted definition was based on serum
creatinine levels >1.5 mg/dl. The definition of AKI in
patients with cirrhosis has now shifted from requiring
an absolute cut-​off of serum creatinine >1.5 mg/dl to
acknowledging the prognostic importance of defin-
ing AKI on the basis of changes in serum creatinine
compared with baseline5,6. As such, AKI in cirrhosis is
now classified into four stages of severity according to
the degree of serum creatinine deviance from baseline
readings (Box 2).

NATURE REvIEwS | DISeASe PRImeRS | Article citation I­D:­ _­# #­# #­# #­# #­# #­# #­# #­# #­# #­# #­# _­ 1
Primer
Author addresses
1
Liver Unit, Hospital Clinic, University of Barcelona,
Barcelona, Catalonia, Spain.
2
Institut d’Investigacions Biomèdiques August Pi i Sunyer
(IDIBAPS), Barcelona, Catalonia, Spain.
3
Centro de Investigación Biomédica en Red de
Enfermedades Hepáticas y Digestivas (CIBEReHD), Madrid,
Spain.
4
Unit of Internal Medicine and Hepatology (UIMH),
Department of Medicine – DIMED, University of Padova,
Padova, Italy.
5
Division of Gastroenterology, Department of Medicine,
University of Toronto, Ontario, Canada.
6
Division of Nephrology and Hypertension, University of
Southern California, Los Angeles, CA, USA.
7
Division of Gastroenterology and Hepatology, Mayo
Clinic, Rochester, MN, USA.
In the second clinical pattern, classically known as
type 2 HRS, there is chronic impairment of kidney
function4,7; these patients therefore fall into the category
of chronic kidney disease (CKD) because of a chronic
reduction of GFR. Owing to this reason, this pattern is
now known as CKD-​HRS5 (Box 1). The pathogenetic
differences between both patterns are not well known.
Despite being labelled CKD, which is not normally
reversible, there is a largely reversible element to kidney
dysfunction in CKD-​HRS.
From a mechanistic perspective, the profound
impairment in kidney function in the absence of sub-
stantial alterations in kidney histology and the potential
reversibility of HRS make it a unique syndrome in med-
icine that has fascinated investigators and clinicians for
decades. Furthermore, the difficulties in the differential
diagnosis and management and the poor prognosis of
HRS make this syndrome a real clinical challenge for
physicians caring for patients with liver diseases.
This Primer is aimed at providing a deep insight into
clinical aspects, diagnosis, prevention and manage­
ment of HRS, focusing on AKI-​HRS owing to recent
advances in this field that may be useful for both inves-
tigators and clinicians dealing with patients with liver
diseases. CKD-​HRS will not be discussed in detail in
this Primer.
Epidemiology
Incidence and mortality
AKI is a well-​recognized complication in patients with
advanced cirrhosis 8,9. The incidence of AKI varies
from 20–50% in patients with cirrhosis admitted into
hospital8. This wide range in the reported incidence of
AKI has been influenced by the definition of AKI used
and the patient population studied. The development
of AKI has been shown to be an independent predic-
tor of in-​hospital mortality and post-​liver transplant
mortality that correlates directly with the severity of
AKI9–17. Survival is dependent on AKI stage and type
of AKI (Boxes 1, 2). In one study including a cohort of
547 patients, 90-day transplant-​free survival for patients
with stage 1A AKI was 84%, stage 1B AKI was 58%,
stage 2 AKI was 48% and stage 3 AKI was 43%, whereas
that of patients without AKI was 89%16. With respect
2 | Article citation ID: (2018) 4:23 www.nature.com/nrdp
to the type of AKI, the worst prognosis is observed
in patients with AKI-​HRS and those with ATN, wher­eas
patients with hypovolaemia-​induced AKI have a
better prognosis10,14,16,18.
Using the new consensus definition of AKI (Box 2),
recent studies have reported incidence rates of AKI
ranging between 27% and 53% in patients with cirrho-
sis who are admitted to hospital for complications of
cirrhosis13,14,16,18. In a prospective observational study
of 547 patients admitted with acute decompensated
cirrhosis (the development of ascites, gastrointestinal
bleeding, hepatic encephalopathy or infection, therefore
representing an acute manifestation of symptoms), 53%
had AKI (68% stage 1, 19% stage 2 and 13% stage 3); of
these patients with AKI, 65% had AKI at the time
of admission, and 35% patients developed AKI during
their hospitalization16.
In a large prospective study of patients with AKI and
cirrhosis (n = 3,458), which incorporated urine output
in addition to serum creatinine into the diagnostic cri-
teria for AKI, the incidence of AKI increased from 58%
to 82%15. When only the serum creatinine criteria of
AKI were applied, 61% of patients who met stage 2 to
stage 3 AKI on the basis of urine output criteria were
misclassified as either no AKI or stage 1 AKI. Patients
who were reclassified on the basis of urine output cri-
teria of AKI had a threefold increased in-​hospital mor-
tality compared with patients with no AKI. Hospital
mortality was greatest for patients who met both urine
output and serum creatinine criteria for AKI. Thus, the
incidence of AKI may be substantially higher than pre-
viously believed if urine output is also included in the
definition of AKI. Nevertheless, these findings require
validation in future studies before they are recom-
mended for use in clinical practice. In the meantime,
criteria based on changes in serum creatinine should
be used; worsening oliguria or development of anuria
could be considered as criteria for AKI but not as man-
dates for the definition of AKI for diagnostic coding
and/or inclusion into trials.
Risk factors for AKI
Few epidemiological studies exist that assess the fre-
quency of the different causes of AKI, particularly HRS
in patients with cirrhosis. Furthermore, some of these
studies were performed before the implementation of
the current AKI definition. In the largest study of this
group, which included 463 consecutive patients with
AKI and cirrhosis who were studied prospectively in
a single institution during a 6-year period, the most
common cause of AKI was bacterial infections (46%),
followed by hypovolaemia (32%), HRS (13%) and dif-
ferent types of parenchymal nephropathy (9%)19. The
cause of AKI influenced the prognosis markedly and was
an independent predictive factor of survival; the worst
prognosis corresponded to HRS and bacterial infections.
However, with the current definition of HRS (Box 1), it is
likely that many patients with AKI attributed to bacterial
infections in this particular study will be reclassified as
having AKI-HRS.
Several recent studies have assessed the frequency
and types of AKI in patients admitted to hospital with

complications of cirrhosis using the new definition of
AKI (Box 2). The results of these studies are summarized
in Table 1, which includes the frequency of the most
common types of AKI: hypovolaemia-​induced AKI,
ATN and AKI-​HRS (Box 1). The most common type of
AKI was hypovolaemia-​induced AKI, which accounted
for approximately one-​third of all AKI cases (27–50%),
whereas the prevalence of ATN ranged between 14% and
35%, and that of AKI-​HRS ranged from 15% to 43%.
Differences among studies are probably secondary to
differences in patient selection and the cause of AKI,
particularly ATN and HRS.
Mechanisms/pathophysiology
The pathophysiology of AKI-​HRS is mainly related
to disturbances in the arterial circulation secondary to
the existence of portal hypertension (an increase in the
blood pressure within the portal venous system), which
is characteristic of advanced cirrhosis. Recent studies
suggest that a systemic inflammatory state, with an
Box 1 | Key glossary terms
Acute kidney injury (AKI)
An acute reduction in glomerular filtration rate (GFR), as indicated by an increase in
serum creatinine over 48 hours.
Chronic kidney disease (CKD)
A chronic reduction in GFR, defined according to the Kidney Disease Improving
Global Outcomes (KDIGO) criteria as a GFR (estimated from the serum creatinine)
<60 ml/min/1.73 m2 for >3 months.
AKI-​hepatorenal syndrome (HRS)
A specific type of AKI seen in patients with advanced cirrhosis, which is now defined
by new criteria set by the KDIGO.
CKD-​HRS
A specific type of CKD that only occurs in cirrhosis characterized by chronic
impairment of kidney function and lack of signs suggestive of intrinsic kidney disease
(for example, haematuria, proteinuria and abnormal kidney ultrasonography).
Type 1 HRS
Old terminology referring to a sudden impairment of kidney function, which is
characteristic of patients with cirrhosis and ascites, defined by a 100% increase
in serum creatinine to a value >2.5 mg/dl (221  µmol/l) within <2 weeks. This term is
still commonly used in clinical practice and was used in all studies on therapy of HRS.
Type 2 HRS
Old terminology referring to chronic impairment of kidney function, which is
characteristic of patients with cirrhosis and ascites, defined by a persistent increase in
serum creatinine >1.5 mg/dl (133  µmol/l). This term has been substituted by CKD-​HRS.
However, it is still commonly used in clinical practice and was used in all studies on
therapy of HRS.
Cirrhosis
A chronic disease of the liver characterized by inflammation and liver fibrosis caused by
the increased synthesis of collagen. Cirrhosis develops after several years of continuous
liver injury and can be associated with a number of aetiological factors, for example,
hepatitis C or hepatitis B virus infection, chronic alcohol consumption and nonalcoholic
fatty liver disease.
Hypovolaemia-​induced AKI
An acute impairment of kidney function that is caused by hypovolaemia associated
with fluid losses within the days before development of AKI (which may be caused by
overdiuresis, diarrhoea or gastrointestinal bleeding).
AKI-​acute tubular necrosis
An acute impairment of kidney function that is caused by necrosis or dysfunction of the
kidney tubules, which is usually associated with the presence of shock or treatment
with nephrotoxic agents.
NATURE REvIEwS | DISeASe PRImeRS | Article citation I­D:­ _­# #­# #­# #­# #­# #­# #­# #­# #­# #­# #­# _­
Primer
increase in the release of inflammatory mediators,
plays a role in the circulatory and kidney dysfunction
observed in HRS.
Circulatory dysfunction
A large body of evidence indicates that impairment in
circulatory function has a key role in the development
of HRS1,9,20–22. In 1988, the ‘arterial vasodilation theory’
was proposed as the explanation for the circulatory
impairment occurring in patients with cirrhosis21. This
impairment is associated with portal hypertension and
cirrhotic cardiomyopathy (a cardiac condition occur-
ring in patients with cirrhosis, which is characterized by
diastolic dysfunction, an impaired systolic response to
physical stress and electrophysiological abnormalities),
which are features of advanced cirrhosis.
The arterial vasodilation theory and portal hyper-
tension. According to the arterial vasodilation theory,
portal hypertension, which is associated with cirrhosis,
causes splanchnic arterial vasodilation, which in turn
triggers systemic circulatory dysfunction. In cirrhosis,
vasodilation is caused by increased production and
activity of various vasodilators, including nitric oxide
(NO), carbon monoxide and endogenous endocanna­
binoids1,9,23–25. The pathogenetic role of these vasodila-
tor factors has been investigated in animal models of
cirrhosis; proof of their role in human disease is based
mainly on the correlation between increased plasma
levels and disease status. Splanchnic arterial vaso­
dilation leads to decreased vascular resistance and, as
a consequence, to a reduction in effective arterial blood
volume and arterial pressure. At earlier stages of liver
cirrhosis (such as compensated, asymptomatic cirrho-
sis), portal hypertension and splanchnic arterial vaso-
dilation are moderate, and, therefore, the decrease in
vascular resistance is also moderate, and the effects on
arterial pressure are counterbalanced by an increase
in cardiac output, which maintains arterial pressure within
the normal range. By contrast, in advanced cirrhosis,
splanchnic arterial vasodilation becomes very severe
and leads to a marked decrease in systemic vascular
resistance. At this point, the increase in cardiac out-
put is no longer able to compensate for the decrease
in systemic vascular resistance, and effective arterial
hypovolaemia and arterial hypotension develop owing
to the disparity between intravascular blood volume
and the dilated arterial circulation9,21–24. Subsequently,
there is a compensatory homeostatic response that leads
to the activation of vasoconstriction factors, such as
the renin–angiotensin–aldosterone system (RAAS), the
sympathetic nervous system (SNS) and, at later stages,
the non-​osmotic secretion of arginine vasopressin,
which is aimed at maintaining arterial pressure within
the normal range. These vasoconstriction systems help
maintain effective arterial blood volume and arterial
pressure near normal limits; however, they also have
detrimental effects on kidney function, which result in
renal sodium retention, impairment of solute-​free water
excretion and, at late stages, marked kidney vasocon-
striction with a subsequent decrease in GFR, causing
HRS9,21–27 (Fig. 1).
3
Primer
There is a large body of evidence showing that
patients with HRS have high levels of plasma renin acti­
vity, which is a marker of the RAAS, and high plasma
noradrenaline concentrations, which is a marker of SNS
hyperactivity9,26,27. In addition, strong evidence for the
role of circulatory dysfunction in the pathophysiology
of HRS is provided by the fact that vasoconstrictor
drugs that act on the splanchnic arterial circulation are
effective in the treatment of HRS28–30.
Effects of cirrhotic cardiomyopathy on circulatory and
kidney dysfunction. As described above, patients with
advanced cirrhosis have increased cardiac output, which
is aimed at compensating for the decrease in effective
arterial blood volume and maintaining arterial pres-
sure. However, evidence exists showing that as cirrhosis
becomes more advanced, cardiac output may decrease
and contribute to the reduction in effective arterial
blood volume31–34. The decrease in cardiac output is
thought to be a consequence of cirrhotic cardiomyo-
pathy33. Diastolic dysfunction may be present in up to
50–60% of patients with cirrhosis; however, a patho-
genetic relationship between diastolic dysfunction and
impairment in circulatory function or development of
HRS has not been demonstrated35. By contrast, there
are data indicating a relationship between a decrease
in cardiac output, the development of kidney under­
perfusion and the occurrence of HRS. For example, data
from a study investigating systemic haemodynamics and
activity of vasoactive systems in patients with cirrhosis
with or without HRS showed that the development of
HRS was associated with a decrease in mean arterial
pressure, together with a reduction in cardiac output
and an increase in plasma renin activity and noradren-
aline concentrations32. In another report, in patients
with cirrhosis and spontaneous bacterial peritonitis,
which is the spontaneous development of infection in
the ascitic fluid, those who developed HRS had a signi­
ficantly lower cardiac output at the time of the diagnosis
Box 2 | A new definition of AKI in cirrhosis
The International Club of Ascites-​Acute Kidney Injury (ICA-​AKI) adapted the Kidney
Disease Improving Global Outcomes (KDIGO) definition of acute kidney injury (AKI)5 to
be useful in the setting of cirrhosis, that is, to define AKI-​hepatorenal syndrome (HRS).
The key feature of the adapted definition is that the time frame for changes in serum
creatinine to become evident is shortened, and the increase in serum creatinine must
be measured against a baseline such that subtle changes can be tracked.
Baseline serum creatinine is defined as a value of serum creatinine obtained in the
previous 3 months. In patients with more than one value obtained within the previous
3 months, the value closest to the admission time to hospital should be used. In patients
without a previous serum creatinine value, the serum creatinine on admission should
be used as baseline.
AKI has four stages:
• Stage 1A: an increase in serum creatinine ≥0.3 mg/dl (26.5  μmol/l) to a value lower
than 1.5 mg/dl (133 µmol/l) from baseline at diagnosis of AKI.
• Stage 1B: an increase in serum creatinine ≥0.3 mg/dl (26.5  μmol/l) to a value
≥1.5 mg/d (133  µmol/l) from baseline at diagnosis of AKI.
• Stage 2: an increase in serum creatinine greater than twofold to threefold from baseline.
• Stage 3: an increase of serum creatinine greater than threefold from baseline or
serum creatinine ≥4.0 mg/dl (353.6  μmol/l) with an acute increase ≥0.3 mg/dl
(26.5 μmol/l) or initiation of renal replacement therapy.
4 | Article citation ID: (2018) 4:23 www.nature.com/nrdp
of spontaneous bacterial peritonitis than those patients
who did not develop HRS31. Moreover, another study
showed that low cardiac output was associated with low
renal blood flow and GFR and high serum creatinine
levels. Interestingly, patients with low cardiac index
(<1.5 l/min/m2) had a higher probability of developing
HRS and lower 3-month and 12-month survivals than
those with higher cardiac index. Therefore, it appears as
though impairment in cardiac function may be a con-
tributing factor to circulatory dysfunction and reduced
kidney perfusion leading to HRS (Fig. 1).
Renal vasoconstriction
The pathophysiological hallmark of HRS is vasocon-
striction of the renal circulation1,9,20,22. The existence of
marked vasoconstriction in the kidneys of patients with
HRS with a characteristic reduction in renal cortical per-
fusion has been demonstrated with a variety of methods,
including renal arteriography, Xe washout technique,
paraaminohippuric acid excretion and duplex Doppler
ultrasonography9,22,36–40.
The mechanism of renal vasoconstriction is prob­
ably multifactorial and may involve changes in systemic
haemodynamics, the activation of vasoconstrictor fac-
tors and the suppression of vasodilator factors acting on
the renal circulation. Assessment of the potential role of
specific vasoactive factors has been problematic owing
to the intrinsic difficulties of performing pathophysio­
logical investigations in humans. Moreover, investigation
of the pathogenetic factors of HRS has been hampered by
the lack of a good animal model of HRS. Angiotensin II
is one of the potential mediators of the renal vaso-
constriction through activation of the RAAS. Several
lines of evidence support a role for this system. First,
cross-​sectional studies, as well as longitudinal studies,
in patients with cirrhosis have demonstrated that the
activity of the RAAS (as estimated by plasma renin
activity) increases from compensated to decompensated
cirrhosis, reaches a maximum in patients with HRS and
correlates inversely with GFR26,32,41. Second, clinical con-
ditions that act as triggers of HRS, such as spontaneous
bacterial peritonitis, are associated with marked activa-
tion of the RAAS that correlates with development of
HRS; interestingly, manoeuvres that prevent this over-
activity of the RAAS, such as albumin administration,
are associated with decreased risk of HRS42. Third, in
patients with HRS that is associated with infection, the
activity of the RAAS predicts the reversibility of HRS;
patients with higher activity of the RAAS have a lower
probability of HRS reversibility than those with
lower activity43. Fourth, resolution of HRS in response
to terlipressin (which is a vasoconstrictor) and albu-
min, which is characterized by a marked increase in
renal blood flow and GFR, is associated with marked
suppression in the activity of the RAAS44.
Another vasoconstrictor system that may play a
role in the pathogenesis of renal vasoconstriction in
HRS is the SNS. Plasma levels of noradrenaline are
markedly increased in patients with HRS compared
with levels in patients with ascites who have normal
renal function. Moreover, noradrenaline levels corre-
late inversely with GFR27. Furthermore, spontaneous

Table 1 | The prevalence and causes of AKI in patients with cirrhosis
Study n AKI prevalence Causes of AKI
(%)
Hypovolaemia ATN HRS
(%) (%) (%)
Fagundes et al., 2013 375 47 35 ND 18
Piano et al., 2013 233 27 36 ND 43
Belcher et al., 2014 110 a
ND 50 35 15
Alegretti et al., 2015 120a ND 33 29 30
Tandon et al., 2017 4,733 36 ND ND ND
Huelin et al., 2017 547 53 27 14 32
All patients with cirrhosis had been hospitalized for complications of the disease. AKI, acute
kidney injury ; ATN, acute tubular necrosis; HRS, hepatorenal syndrome; ND, not determined.
a
Studies included only patients with cirrhosis and AKI.
or therapeutic-​induced resolution of HRS is associ-
ated with a marked suppression of SNS overactivity43,44.
Unfortunately, it has not been possible to assess the
renal effects of blockade of the RAAS and blockade of
the activity of the SNS in patients with HRS because
blockade of these systems in patients with cirrhosis is
associated with marked arterial hypotension owing to
the vasoconstrictor effect of these systems on systemic
arterial circulation45. Other vasoconstrictor factors with
a potential role in kidney vasoconstriction in HRS have
been proposed, specifically endothelin and cysteinyl
leuko­triens46,47. A possible role for endothelin was pro-
posed on the basis of increased plasma levels in patients
with HRS46. However, the administration of a specific
antagonist of endothelin receptors did not improve GFR
in patients with HRS, which suggests that endothelin
does not have a major pathogenetic role48.
An alternative, yet not mutually exclusive possibility
associated with increased activity and/or production of
vasoconstrictor factors that induce renal vasoconstric-
tion in HRS is the existence of the reduced production of
endogenous renal vasodilators, particularly prostaglan-
dins. Prostaglandins, which are lipid mediators prod­
uced by most nucleated human cells, have vasodilator
effects on the kidney circulation and may act by com-
pensating for the enhanced vasoconstrictor effects of the
RAAS and the SNS. Interestingly, renal prostaglandin E2
production is increased in patients with decompensated
cirrhosis41 and may have a protective role. Support for
this model is mainly derived from the fact that treatment
with NSAIDs, which inhibit prostaglandin synthesis,
leads to development of AKI, which resembles HRS, in
many patients with cirrhosis and ascites49.
In healthy individuals, renal autoregulation main-
tains a constant renal blood flow independently of
arterial pressure fluctuations. However, patients with
advanced cirrhosis have a shift to the right of the renal
autoregulation curve, which means that for the same
renal perfusion pressure, renal blood flow is lower than
that of patients with compensated cirrhosis and healthy
subjects. This effect, which is probably related to the
increased activity of the SNS, may be responsible, at least
in part, for the increased risk that patients with advanced
cirrhosis have of developing AKI in general and HRS
in particular50.
NATURE REvIEwS | DISeASe PRImeRS | Article citation I­D:­ _­# #­# #­# #­# #­# #­# #­# #­# #­# #­# #­# _­
Primer
Systemic inflammation
In recent years, evidence has accumulated indicating that
Refs decompensated cirrhosis is associated with persistent
systemic inflammation, which may play an important
part in the progression of cirrhosis and the development
14 of complications, including HRS51,52. Decompensated
18 cirrhosis is associated with increased serum levels of
C-​reactive protein and an increased leukocyte count,
10
which increase in parallel with disease severity, regardless
17
of the presence of infections53,54. Moreover, patients with
13 advanced cirrhosis show increased serum levels of pro-​
inflammatory cytokines, such as IL-6, IL-8 and tumour
16
necrosis factor (TNF), among others, and markers of oxi-
dative stress (such as oxidized albumin)51,52,55. Levels of
all these inflammatory markers increase in parallel with
disease severity and are the highest in patients with acute-​
on-chronic liver failure (ACLF), which is a syndrome that
is characterized by the presence of multiple organ fail-
ures, in which kidney failure is one of the most frequent
organ failures with marked prognostic impact56–58.
Patients with advanced cirrhosis show bacterial
translocation from the gut to mesenteric lymph nodes,
which is associated with increased levels of pro-
inflammatory cytokines59, particularly IL-6 and TNF56,57.
However, the pathogenetic role of each specific cytokine
that is increased in cirrhosis is not known. Overall, it is
hypothesized that pathogen-​associated molecular pat-
terns (PAMPs) (for example, lipopolysaccharide) deriv-
ing from bacterial translocation or bacterial infections
or damage-​associated molecular patterns (DAMPs) (for
example, high-​mobility group protein B1 (HMGB1),
heat shock proteins (HSPs) and hyaluronic acid) that are
released from the injured liver may activate pattern rec-
ognition receptors present in circulating innate immune
cells, leading to the activation of immune cells and the
consequent development of an inflammatory response.
The specific pathogenetic role of DAMPs, however, has
not yet been demonstrated. Inflammatory mediators,
particularly cytokines acting on vascular smooth mus-
cle cells, may lead to further impairment of circulatory
dysfunction already present in patients with cirrhosis,
leading to the development of HRS52 (Figs 1,2).
A potential exacerbating role for bacterial infections. To
date, evidence for the relationship between inflammation
and the development of HRS is derived from clinical
studies. Importantly, bacterial infections are one of the
main triggers of HRS in patients with cirrhosis. Although
spontaneous bacterial peritonitis is the main precipitating
factor of HRS, any type of infection in patients with cir-
rhosis can trigger the development of HRS42,43,60–64. In the
absence of volume replacement by albumin administra-
tion (see below), up to 30% of patients with spontaneous
bacterial peritonitis develop HRS2,42,60,61,65. However, bac-
terial infections other than spontaneous bacterial peri-
tonitis can also trigger HRS, and thus, all patients with
cirrhosis who have a bacterial infection should be closely
monitored for the potential development of HRS43,62–64.
Interestingly, patients with spontaneous bacterial peri-
tonitis who develop HRS show significantly higher
serum levels of IL-6 and TNF than patients with spon-
taneous bacterial peritonitis who do not develop HRS61.
5
Primer

Moreover, the increase in cytokine levels is associated
with an increase in vasodilator mediators, such as NO
and carbon monoxide66,67. These findings suggest that
in the context of the same precipitating factor, such as a
bacterial infection, a higher systemic inflammatory state
leads to development of HRS, and this effect could be
linked to a further impairment of the vasodilatory state.
A prospective study that investigated the relationship
between inflammatory state and AKI in cirrhosis showed
that almost all patients with HRS (78%) had either bacte-
rial infection or a systemic inflammatory response syn-
drome defined as two or more of the following variables:
temperature >38 °C or ≤36 °C; heart rate of ≥90 beats
per minute; respiratory rate of ≥20 breaths per minute,
partial arterial carbon dioxide pressure of ≤32 mmHg
or the use of mechanical ventilation; or a white blood
cell count of ≥12,000/mm3 or ≤4,000/mm3 or a differential
count showing >10% immature forms68. By contrast, only
14% of patients with pre-​renal AKI had a bacterial infec-
tion or a systemic inflammatory response syndrome68.
Interestingly, almost one-​third of patients with HRS have
systemic inflammatory response syndrome without evi-
dence of bacterial infection, indicating that inflammatory
response, independently of the presence of documented

Advanced
cirrhosis

Portal hypertension

Cirrhotic Vasodilator mediators Bacterial

cardiomyopathy (NO, CO and translocation
endocannabinoids) (PAMPs) DAMPs
(HMGB1, HSPs and
↓ Cardiac output Splanchnic arterial vasodilation hyaluronic acid)
Decreased
vascular
resistance Inflammatory
mediators and/or
Effective arterial hypovolaemia vasodilator
factors (IL-6, Activation of
TNF, endotoxin) immune cells
Activation of vasoconstrictor factors Inflammatory response
• RAAS
• SNS
• AVP

Renal vasoconstriction Inflammatory mediators

(cytokines, chemokines
Decreased GFR and NO)

↑ Prostaglandins
Impaired renal Kidney
autoregulation tissue
injury?

Fig. 1 | Factors involved in the pathogenesis of HRS. Circulatory dysfunction is the key mechanism involved in the
pathophysiology of hepatorenal syndrome (HRS). Patients with advanced cirrhosis present a marked splanchnic arterial
vasodilation triggered by portal hypertension. Splanchnic vasodilation leads to a decreased systemic vascular resistance
that, at this stage, cannot be compensated by cardiac output, and therefore, effective arterial hypovolaemia develops.
Consequently , there is activation of endogenous vasoconstrictor systems. The intense activation of these systems leads
to a marked kidney vasoconstriction, inducing a decrease in glomerular filtration rate (GFR) and the development of HRS.
At this advanced stage of cirrhosis, evidence shows that there is a decrease in cardiac output, which may also contribute
to the decrease in effective arterial blood volume. There is growing evidence suggesting that systemic inflammation also
plays an important role in the pathophysiology of HRS. It is hypothesized that pathogen-​associated molecular patterns
(PAMPs) and damage-​associated molecular patterns (DAMPs) deriving from bacterial translocation and from injured
liver, respectively , may activate circulating innate immune cells, leading to the development of a marked inflammatory
response. Inflammatory mediators would lead to further impairment of circulatory dysfunction and, eventually , could also
cause direct kidney tissue damage, but this has not been demonstrated. Finally , local factors, such as impairment in kidney
autoregulation, may also play a role. AVP, arginine vasopressin; CO, carbon monoxide; HMGB1, high-mobility group box 1;
HSPs, heat shock proteins; NO, nitric oxide; RAAS, renin–angiotensin–aldosterone system; SNS, sympathetic nervous
system; TNF, tumour necrosis factor.

6 | Article citation ID: (2018) 4:23 www.nature.com/nrdp

 Primer

Gastrointestinal
tract
Advanced DAMPs PAMPs
cirrhosis (products from injured liver) (from bacterial translocation)

• Phagocytosis TLRs (TLR2, TLR4 and TLR9) • Interferon

• ROS • IL-17
• IL-22
Neutrophil Monocyte Dendritic T cell
cell +
• Cytokines (IL-1, IL-6, IL-8 and TNF)
• Chemokines

• Cell recruitment
• Endothelial activation
• Vasodilation
Blood vessel

Fig. 2 | Inflammatory mediators and pathways that can affect the circulation. The figure summarizes the mechanisms
potentially involved in the systemic inflammatory reaction present in patients with advanced cirrhosis. It is hypothesized
that bacterial products (pathogen-​associated molecular patterns (PAMPs)) derived from bacterial translocation and/or
damage-​associated molecular patterns (DAMPs) (that is, high-mobility group box 1 (HMGB1), heat shock proteins (HSPs),
ATP and hyaluronic acid) from injured liver cells may activate pattern recognition receptors, particularly Toll-​like receptors
(TLRs). Pattern recognition receptors are mainly expressed in monocytes and macrophages, neutrophils and dendritic
cells, are activated upon stimulation and trigger the inflammatory process leading to release of pro-​inflammatory
cytokines and chemokines. Some cytokines have autocrine effects, leading to further monocyte and neutrophil activation,
and when secreted in high amounts, they also have endocrine effects that lead to the production of acute phase protein in
the liver. Antigen-​presenting cells, such as dendritic cells, will also lead to T cell activation. Cytokine production will lead
to activation of endothelial cells, increasing vascular permeability and inducing capillary leakage and vasodilation.
Chemokines will be mainly involved in immune cell recruitment to the site of injury. Among recruited cells, neutrophils will
have a key role in phagocytosis and pathogen killing. ROS, reactive oxygen species; TNF, tumour necrosis factor.

bacterial infection, is frequent in the setting of HRS and
may be involved in its pathophysiology68.
A circulatory effect or direct kidney immunopathology?
Evidence supports the existence of an intense systemic
inflammatory reaction in patients with advanced cirrho-
sis, which is associated with negative clinical outcomes,
including HRS and mortality51,52. However, it is unclear
whether the excessive inflammatory response contrib-
utes to HRS through further impairment of circulatory
function or through direct effects on kidney circulation,
or both, or by effects on other structures.
In recent years, more information has become
available on the role of systemic inflammation and the
development of complications of cirrhosis in the setting
of ACLF (according to the ACLF definition set out in
the CANONIC study)69. ACLF is mainly characterized
by the development of extrahepatic organ failures,
including the kidney56. Several studies show that patients
with ACLF present with a marked systemic inflammatory
reaction compared with patients with decompensated cir-
rhosis without ACLF56–58. Moreover, serum levels of pro-​
inflammatory cytokines have been shown to be associ-
ated with the number of organ failures and mortality.
Specifically, the levels of IL-6 and IL-8 and human oxidized
albumin are elevated in patients with ACLF (as compared
with healthy individuals and patients with cirrhosis with-
out ACLF) and correlate with kidney impairment56,57.
The levels of inflammatory markers in patients with
ACLF are very similar to those described in patients
with sepsis70–72. Therefore, it could be hypothesized that a
confirmed or suspected infection plus a systemic inflam-
matory response syndrome may lead to organ dysfunc-
tion, as is the case in sepsis. The excessive inflammatory
reaction could lead to direct cell and tissue damage by
inflammatory mediators, which is a mechanism known
as immunopathology, causing organ failure (Fig.  1).
To date, however, there are still not sufficient data to sup-
port this hypothesis. Therefore, more investigations are
needed to further clarify the role of inflammation in the
pathophysiology of HRS.
Other precipitating factors
Although less frequent, precipitating factors for the
development of HRS other than bacterial infections are
large-​volume paracentesis (the removal of ascites fluid
from patients with decompensated cirrhosis) and gastro­
intestinal bleeding. Large-​volume paracentesis may lead
to the development of HRS in ~15% of patients if intra-
venous albumin (a treatment and preventive measure
for HRS, which will be described later) is not admin­
istered concomitantly73,74. Indeed, prevention of HRS is
the main indication for the administration of albumin
after large-​volume paracentesis. The development of
HRS after gastrointestinal bleeding is uncommon.
In this context, AKI is usually related to hypovolaemia-​
induced AKI or ATN associated with hypovolemic
shock rather than HRS75. The mechanisms by which
large-​volume paracentesis and gastrointestinal bleeding
cause AKI-​HRS are probably related to an impairment of

NATURE REvIEwS | DISeASe PRImeRS | Article citation I­D:­ _­# #­# #­# #­# #­# #­# #­# #­# #­# #­# #­# _­ 7
Primer
Box 3 | Diagnostic criteria of HRS
Classical diagnostic criteria of hepatorenal syndrome (HRS)a
• Cirrhosis with ascites
• Serum creatinine >1.5 mg/dl (>133 mmol/l)
• No improvement of serum creatinine (decrease to a level of ≤1.5 mg/dl) after at least
2 days with diuretic withdrawal and volume expansion with albumin
• Absence of shock
• No current or recent treatment with nephrotoxic drugs
• Absence of parenchymal kidney disease. Parenchymal kidney disease is indicated
by proteinuria (>500 mg/day), microhaematuria (>50 red blood cells per high-​power
field) and/or abnormal renal ultrasonography
New diagnostic criteria of acute kidney injury (AKI)-HRSa
• Cirrhosis with ascites
• Diagnosis of AKI according to International Club of Ascites-​Acute Kidney Injury
(ICA-​AKI) criteria
• No response after 2 consecutive days of diuretic withdrawal and plasma volume
expansion with albumin
• Absence of shock
• No current or recent use of nephrotoxic drugs (NSAIDs, aminoglycosides or iodinated
contrast media)
• No signs of structural kidney injury. Structural kidney injury is indicated by proteinuria
(>500 mg/day), microhaematuria (>50 red blood cells per high-​power field) and/or
abnormal renal ultrasonography
a
Diagnostic criteria are from definitions in refs4,5.
arterial circulation; however, an inflammatory component
may also play a role76,77.
Diagnosis, screening and prevention
Diagnosis
The first diagnostic criteria of HRS were proposed by
the International Club of Ascites (ICA) in 1996 and
later modified in 2007 (refs4,78). The later modification
allowed the diagnosis of HRS in patients with ongo-
ing bacterial infections in the absence of septic shock78
(Box 3). Overall, these ‘classical’ criteria were based on
the following criteria: first, the presence of decompen-
sated cirrhosis, second, serum creatinine exceeding a
cut-​off value of >1.5 mg/dl); and third, the exclusion
of other specific causes of kidney failure, particularly
hypovolaemia-​induced renal failure and ATN. However,
these criteria had two main problems. The first was that
they did not take into account changes in serum creati-
nine with respect to previous values in the same patient,
which are essential to distinguish between AKI and
CKD. The second was that a serum creatinine >1.5 mg/dl
was used to identify patients who already had markedly
reduced GFR (approximate GFR <30 ml/min/1.73 m2),
therefore identifying too late those who already have
developed kidney disease. To solve these problems,
the hepatology community decided to adapt the crite-
ria proposed by the KDIGO, which are based on small
changes of serum creatinine levels with respect to base-
line values within a short period time79, for the defini-
tion and staging of AKI to patients with cirrhosis (Box 2).
Some factors may influence serum creatinine levels,
particularly nutritional status and sex78.
As mentioned, several studies have recently con-
firmed the usefulness of these modified criteria in the
8 | Article citation ID: (2018) 4:23 www.nature.com/nrdp
assessment and staging of AKI in cirrhosis10,13–16,18,80,81.
Therefore, in the setting of the new definition of AKI in
cirrhosis, the definition of HRS has also been modified.
The only change that has been made to the classic defi-
nition of HRS is the removal of a cut-​off value of serum
creatinine, thus eliminating the major drawback of the
previous definition. Consequently, the acute form of
HRS has been renamed AKI-​HRS5 (Boxes 1,2,3). These
modified criteria will likely lead to earlier diagnosis and
treatment of HRS.
Differential diagnosis of HRS. Besides HRS, other types
of AKI can occur in patients with cirrhosis, such as
hypovolaemia-​induced AKI, nephrotoxicity, ATN or
glomerulonephritis, which should be managed differ-
ently1,9. Moreover, despite the overall poor prognosis of
patients with cirrhosis and AKI, different aetiologies
of AKI are known to be associated with different prog-
noses, with ATN and HRS having the worst survival19.
Therefore, the exclusion of other causes of AKI is a key
step in the diagnosis of HRS.
As mentioned earlier, HRS remains a diagnosis of
exclusion, as to date there are no specific markers or
laboratory tests for its diagnosis. The differential diag-
nosis of the causes of AKI may not always be easy in
daily clinical practice. Differential diagnosis between
HRS and ATN is particularly challenging, considering
that both are conditions that usually appear in criti-
cally ill patients with cirrhosis who present with several
complications of the disease, in whom the detection
of precipitating factors for either HRS or ATN may be
challenging. Moreover, classical urine biomarkers such
as urine sodium, fractional excretion of sodium or urine
osmolality, which are used in the nephrology setting for
the differential diagnosis of AKI, also have limitations
in patients with cirrhosis, as they may be influenced by
diuretic treatment, and urine sodium may be markedly
low owing to increased sodium retention despite being
the cause of AKI4,9,82. Thus, urine sodium concentration
has no place in the diagnosis of HRS.
New urinary biomarkers. A requirement exists for new
objective biomarkers to help in the differential diagnosis
of HRS. In the past decade, several urine biomarkers of
tubular damage have been shown to be potentially use-
ful for the differential diagnosis of AKI in cirrhosis83.
Among them, neutrophil gelatinase-​associated lipocalin
(NGAL) (an iron trafficking protein involved in multiple
processes) has shown promising results. Besides NGAL,
other biomarkers that have been shown to be potentially
useful are the pro-​inflammatory cytokine IL-18, kidney
injury molecule-1 (KIM-1; also known as HAVCR1),
liver-​type fatty acid-​binding protein (L-​FABP; also
known as FABP1) and albumin84–89. Their function in the
renal tubules in the context of ATN is largely unknown.
Studies have shown that urine NGAL (uNGAL) levels
are significantly increased in patients with ATN com-
pared with those in patients with hypovolaemia-​induced
AKI and HRS, implying the usefulness of uNGAL lev-
els in the differential diagnosis of AKI. Patients with
hypovolaemia-​induced AKI show the lowest uNGAL
levels, which are similar to levels in patients without AKI,

whereas patients with HRS typically show intermediate
levels between those of patients with hypovolaemia-​
induced AKI and those of patients with ATN84,85,87–89.
Interestingly, these studies show that uNGAL levels have
a very good accuracy, with an area under the receiver
operating characteristic curve (AUROC; whereby a
value of 1 indicates a perfect predictor and a value of
0.5 or less indicates poor predictive accuracy) ≥0.8 for
the diagnosis of ATN versus other causes of AKI. One
limitation of uNGAL is that it may be increased in the
setting of urinary tract infections because NGAL is also
expressed in leukocytes90. Other urine biomarkers such
as IL-18, albumin, KIM-1 and L-​FABP have also been
reported as having good accuracy in the differential
diagnosis of ATN and HRS, with patients with ATN
showing the highest values and patients with HRS hav-
ing intermediate values in comparison with patients with
hypovolaemia-​induced AKI87,88.
Results of a study that investigated a panel of up to
12 different urinary biomarkers in a series of patients
with cirrhosis and AKI showed that the biomark-
ers with the best performance are uNGAL and IL-18
(AUROC > 0.95), followed by albuminuria (AUROC
0.86)88. By contrast, KIM-1 showed lower accuracy
(AUROC 0.70), which was confirmed in another study87.
Moreover, a meta-​analysis assessing the role of urine
IL-18 and NGAL showed that urinary levels of IL-18
and uNGAL have high diagnostic accuracy in differ-
entiating ATN from other types of AKI91. The AUROC
for the diagnosis of ATN was 0.86 (95% CI 0.68–0.94) for
NGAL and 0.88 (95% CI 0.82–0.93) for IL-18 (ref.91).
Interestingly, this meta-​analysis also showed that NGAL
and IL-18 were useful in predicting short-​term mortality
in patients with cirrhosis91.
In summary, currently, there is a large body of evi-
dence showing that urine biomarkers, particularly
NGAL and IL-18, are useful in the differential diagno-
sis of AKI in cirrhosis and could be used in this setting
to overcome the challenge of differentiating HRS from
ATN in patients with cirrhosis. Prospective studies
should ideally be performed to assess which biomarkers
have the greatest accuracy and the optimum point in
time in which biomarkers should be measured, which
could be either immediately after a diagnosis of AKI or
after the response to albumin administration has been
evaluated (2 days after diagnosis) (Box 3).
Finally, cystatin C is another biomarker of poten-
tial interest in patients with cirrhosis. Two studies have
shown that plasma levels of cystatin C are predictive
of AKI, HRS and mortality in patients hospitalized for
complications of cirrhosis, suggesting that it is a good
biomarker for the early identification of patients who are
at high risk of development of renal events and death92,93.
By contrast, urinary levels of cystatin C are not accurate
in differential diagnosis between ATN and other types
of AKI in cirrhosis88.
Prevention
The first step for the prevention of HRS is the treatment
or prevention of possible precipitating factors potentially
leading to HRS. In patients with cirrhosis who are treated
for ascites by large-​volume paracentesis, intravenous
NATURE REvIEwS | DISeASe PRImeRS | Article citation I­D:­ _­# #­# #­# #­# #­# #­# #­# #­# #­# #­# #­# _­
Primer
albumin should always be administered to improve cir-
culatory function74. Moreover, patients with advanced
cirrhosis should be closely monitored, and antibiotic
therapy should be given very early after diagnosis
of any bacterial infection74.
In patients with cirrhosis who develop spontaneous
bacterial peritonitis, the administration of intravenous
albumin together with antibiotic treatment markedly
reduces the incidence of HRS and mortality42,94. It is
not clear whether patients with cirrhosis with serum
creatinine <1 mg/dl and serum bilirubin (a marker
of liver injury) <4 mg/dl at diagnosis of spontaneous
bacterial peritonitis could also benefit from albumin
because the incidence of HRS in these patients is very
low 95. However, considering the lack of sufficient
information about this particular group of patients,
current guidelines recommend the administration
of albumin to all patients with spontaneous bacterial
peritonitis to prevent the development of HRS74. The
administration of albumin to patients with cirrho-
sis with bacterial infections other than spontaneous
bacterial peritonitis has not shown a clear benefit in
terms of prevention of HRS or survival96,97. Therefore,
albumin treatment is not indicated in these patients
to prevent HRS.
Finally, long-​term treatment with the antibiotic nor-
floxacin as primary prophylaxis of spontaneous bac-
terial peritonitis in patients with ascitic fluid protein
<15 g/l associated with impaired liver and/or kidney
function (bilirubin >3 mg/dl, sodium <130 milliequiv-
alents (mEq)/l, Child–Pugh >10 (a score of cirrhosis
severity) and/or serum creatinine >1.2 mg/dl) is asso-
ciated with a significant decrease in the incidence of
HRS and increased survival74,98. The mechanism under-
lying the beneficial effect of norfloxacin prophylaxis is
probably related to the decrease of bacterial transloca-
tion from the gut and systemic inflammation, which
are known to be related to the impairment of circula-
tory function, thereby predisposing the patient to the
development of HRS59.
Management
When HRS is diagnosed, a specific treatment (see below)
should be started as soon as possible because patients
may rapidly deteriorate. In addition, supportive meas-
ures should be taken. The clinical status of a patient
with HRS, such as fluid balance, arterial pressure and
other vital signs, should be carefully monitored. A con-
comitant bacterial infection should be suspected in all
patients; all patients should undergo blood, urine and
ascites culture to diagnose infection, which can be
treated with broad spectrum antibiotics74. The use of
diuretics should be avoided, and low volume therapeutic
paracentesis with albumin should be performed to con-
trol ascites, if necessary. Although no general consensus
exists, it seems advisable to discontinue β-​blockers (used
to prevent variceal bleeding) in patients during treat-
ment for spontaneous bacterial peritonitis owing to the
potentially negative effect of β-​blockers on circulatory
and kidney functions99.
In 2018, it should be considered that the traditional
classification of HRS into two clinical types is no longer
9
Primer
Table 2 | Clinical trials of the use of vasoconstrictors in the management of HRS
Randomized clinical trial Treatment
response (%)a
Solanki et al., 2003
Neri et al., 2008
Martin-​Llahi et al., 2008
Sanyal et al., 2008
Cavallin et al., 2015 d
Boyer et al., 2016
HRS, hepatorenal syndrome. Complete response. 15-day overall survival. 6-month overall
a b
survival. dTerlipressin plus albumin versus midodrine and octreotide plus albumin.
10 | Article citation ID: (2018) 4:23 www.nature.com/nrdp
3-month overall Refs
survival (%)
Terlipressin Placebo Terlipressin Placebo
and/or and/or
control control
42 0 42b 0b
80 19 54 18
43.5 8.7 27 19
34 13 42.9c 37.5c
55.5 4.8 59 43
19.6 13.1 57.7 54.5
c value of <1.5 mg/dl102,103.
applied and that all published clinical trials are based on
these old criteria. In fact, as mentioned before, in keep-
ing with the new criteria for the diagnosis of AKI, the
new definition of type 1 HRS is ‘AKI-​HRS’ (Boxes 1,2).
Accordingly, there is no longer a cut-​off value of serum
creatinine levels for the diagnosis of AKI-​HRS and to start
pharmacological treatment5. Nonetheless, it is advisable
to use terlipressin in patients with AKI-​HRS stage 1B
or greater because of the lack of information on effi-
cacy and side effects of terlipressin in AKI-​HRS stage 1.
The use of these newer AKI-​HRS criteria in diagno-
sis may mean that patients are diagnosed and treated
earlier. The impact of earlier treatment on the rate of
treatment response in patients with AKI-HRS should
still be investigated; nevertheless, the need to explain
why pharmacological treatment fails in some patients
will remain.
Pharmacological therapy
Vasoconstrictors. Several randomized controlled tri-
als have shown that vasoconstrictors, in particular,
terlipressin, plus volume expansion with albumin
(see below) significantly improved renal function
in patients with HRS28,29,100–104. The rationale behind
the use of vasoconstrictors is to counteract splanch-
nic arterial vasodilation1,44. Albumin further coun-
teracts the reduction in effective circulating volume
and improves cardiac contractility105–107. Three types
of vasoconstrictors are currently available for the treat-
ment of HRS: terlipressin, noradrenaline and the com-
bination of midodrine and octreotide. Terlipressin
is the most investigated vasoconstrictor in this field;
however, it must be emphasized that rando­m­ized clini­
cal trials on terlipressin and albumin are based on
old criteria of type 1 HRS and not AKI-​HRS. Several
pilot studies44,108 as well as six randomized controlled
trials28,29,103,104,109,110 demonstrated that the combination of
terlipressin plus albumin is effective in the treatment
of type 1 HRS (Table 2). The administration of terlipres-
sin by continuous intravenous infusion is associated with
a significantly lower rate of adverse effects than intra­
venous bolus administration owing to the need for lower
doses102. These findings are probably consistent with
the short-​term effect of terlipressin on portal pressure
(3–4 hours). The dose of terlipressin should be increased
in a stepwise manner if serum creatinine does not decrease
by at least 25% with respect to the pretreatment value
after 3 days of treatment.
Treatment with vasoconstrictors plus albumin
should be continued until serum creatinine reaches
a final value <1.5 mg/dl, ideally until it is close to the
baseline value taken before the impairment of kidney
110
function occurred. In patients with no response or
109
partial response (Table 2), the treatment should be dis-
29 continued within 14 days. In two previous randomized
controlled clinical trials performed in Europe in 2015
28
and 2016, 55.5% of patients with type 1 HRS showed a
103
complete response to terlipressin plus albumin, defined
104
as a reduction of serum creatinine by >50% to a final
In comparison with results of previous randomized
trials, a 2016 large randomized double-​blind trial (the
Reverse trial) of terlipressin and albumin compared with
placebo and albumin performed in the US and Canada
showed no significant differences in the response rate
of both treatment arms; 19.6% of patients responded to
treatment in the terlipressin arm, and 13.1% of patients
responded in the placebo arm104. Discrepant findings
between this study and previous studies may be due to
differences in patient populations and the greater use
of alternative therapies, particularly renal replacement
therapy, in the North American study, which prevented
an adequate follow-​up of patients, as discussed else-
where111. However, a pooled analysis of the results of the
Reverse trial104 and a previous placebo-​controlled trial
also performed in the US28 showed a significantly higher
patient response rate of terlipressin and albumin than
placebo and albumin (27% versus 14%; P = 0.004)112.
After the discontinuation of terlipressin, recurrence
of HRS occurs in <20% of patients with type 1 HRS, and
re-​treatment with vasoconstrictors and albumin is usu-
ally effective74. Patients who respond to treatment with
terlipressin plus albumin show a better survival than
nonresponders to treatment; moreover, in two meta-​
analyses of randomized controlled trials, the use of ter-
lipressin was associated with a significant improvement
in short-​term survival30,113. This moderate increase in
survival may be important in patients with HRS who
are on the waiting list for liver transplantation. By con-
trast, patients with type 2 HRS (Box 1) have a high rate of
treatment response to terlipressin and albumin; however,
the recurrence of HRS is quite common114. In addition,
a recent case–control study did not show any difference
in the outcome after liver transplant in patients who
had type 2 HRS that was treated or not treated with
terlipressin115. For these reasons, type 2 HRS cannot be
considered an indication for the use of terlipressin plus
albumin, even in patients who are on the waiting list for
liver transplantation.
The most common adverse effects of treatment
with terlipressin are diarrhoea and abdominal cramps.
Severe adverse effects have also been described, such as
angina, cardiac arrhythmias, intestinal ischaemia, finger,
tongue, scrotum or skin ischaemia and severe hyperten-
sion. Patients with ischaemic heart disease and periph­
e­ral vascular disease probably should not be treated
with terlipressin74.

A combination of two vasoconstrictors, midodrine
(an α1-agonist drug) and octreotide (a somatostatin ana-
logue), together with albumin infusion showed efficacy
in treating type 1 HRS in a small series of patients 116,117.
Midodrine and octreotide can be administered in
increasing doses if there is no treatment response
(a reduction in serum creatinine of at least 25% compared
with baseline) by day 3 of treatment. Octreotide is usu-
ally administered subcutaneously but can also be given
intravenously117. In a pilot study, the combination of
midodrine, octreotide and albumin restored renal func-
tion in ~40% of patients with type 1 HRS116. However,
in a randomized controlled trial, the combination of
terlipressin and albumin was significantly more effec-
tive than the combination of midodrine, octreotide and
albumin (improvement of renal function in 70% versus
29%; P = 0.01)103.
The administration of noradrenaline plus albumin
has been investigated as a treatment of HRS. In two
small randomized trials in patients with type 1 HRS,
noradrenaline showed an efficacy similar to that of
terlipressin100,101. Using noradrenaline to treat HRS is
tempting because it is a cheaper drug than terlipres-
sin; however, it should be administered using a central
venous line and under continuous monitoring, and thus,
it cannot be used outside intensive care units. Therefore,
further studies are needed to determine the usefulness
of this treatment.
Albumin. Volume expansion by albumin infusion
appears to be crucial for the effectiveness of the treat-
ment of HRS. Only one study was carried out in which
terlipressin was used alone for the treatment of HRS; in
this study, the efficacy of terlipressin was much lower
than when it was used together with albumin108. One
possible explanation is that a fall in cardiac output has
been shown to be a crucial event in the pathophysio­
logy of HRS31,34 and that cardiac output could be fur-
ther reduced by the effect of terlipressin118. By contrast,
albumin infusion is capable of maintaining or increas-
ing cardiac output even in the most advanced phases
of liver disease106. The dose of albumin administered
is determined by the level of central venous pressure.
However, evidence is accumulating that the increase in
systemic vascular resistance and cardiac output owing
to albumin is also related to properties of the molecule
other than plasma volume expansion105,107. The mech-
anisms and role of these actions fall beyond the scope
of this Primer; nevertheless, it is important to highlight
that while exerting these non-​oncotic actions, albumin
molecules go through post-​translational changes that
increase the antioxidative and anti-​inflammatory prop-
erties of albumin55,119. Thus, either the pretreatment
concentration of endogenous effective albumin or
the dose of exogenous effective albumin administered
could be crucial for the patient response to the treatment
of HRS.
Antibiotics. Patients with AKI-​HRS have a high chance
of having a bacterial infection; however, antibiotics are
administered only if the patient has a demonstrated
infection or a high suspicion of infection. Prophylactic
NATURE REvIEwS | DISeASe PRImeRS | Article citation I­D:­ _­# #­# #­# #­# #­# #­# #­# #­# #­# #­# #­# _­
Primer
antibiotics are not recommended unless a bacterial
infection is confirmed by blood, urine or ascites culture.
Nonresponse. In randomized controlled studies, the
average proportion of patients who are nonrespond-
ers to terlipressin is 54%, with wide variations among
studies (from 20.0% to 80.4%) (Table 2). This variation
may be related, among other factors, to differences in
the patient population and the use of alternative ther-
apies28,29,103,104,109,110. The following should be considered
as factors that may contribute to the lack of treatment
response observed in some patients: the pretreatment
level of serum creatinine, the type of HRS (AKI-​HRS or
CKD-​HRS) and the pathophysiology of HRS.
The pretreatment level of serum creatinine is of
importance, as it has been shown that the higher the
baseline serum creatinine, the lower the rate of response
to therapy120. This is related to the fact that higher pre-
treatment serum creatinine requires a greater increase in
arterial pressure to achieve a treatment response121.
In terms of the type of AKI-​HRS, it should be hypoth-
esized that there may be some degree of parenchymal
damage in patients with HRS. This possibility arises
from two lines of evidence. First, renal biopsy in patients
with cirrhosis and renal dysfunction shows that, even in
absence of substantial proteinuria, there may be tubule-​
interstitial injury, glomerular injury and vasculitis122.
Second, studies that have applied new biomarkers of
tubular damage in the differential diagnosis of AKI (in
particular, between HRS-​AKI and ATN)81–89 have shown
moderately increased values of several biomarkers in
patients with HRS-​AKI compared with in patients with
hypovolaemia AKI. Thus, the presence of some degree of
parenchymal damage may theoretically limit the efficacy
of the treatment in some patients with HRS. However,
this has not yet been convincingly demonstrated.
In regard to pathophysiology, it has been previously
recognized that HRS involves not only macrovascular
dysfunction but also microvascular dysfunction, systemic
inflammation and direct tubular damage, particularly in
the setting of ACLF. In patients with ACLF, the number of
organ failures, which is associated with the degree of sys-
temic inflammation, affects the response of AKI-​HRS to
pharmacological treatment123,124. In addition, non-​renal
organ failure may directly impair the response of patients
with HRS to terlipressin plus albumin. For example,
adrenal dysfunction, which is quite common in patients
with cirrhosis or ACLF125, can induce more pronounced
systemic haemodynamic instability, therefore reducing
the effect of vasoconstrictors.
Non-​pharmacological therapy
A substantial proportion of patients with cirrhosis and
AKI-​HRS do not respond to vasoconstrictor therapy.
With the exception of liver transplantation, all other
non-pharmaceutical therapies that have been tried in the
management of AKI-HRS have proved to be applicable to
only a very few patients, with questionable results. Some
of these therapies were tried before pharmaco­logical
therapies became available, and certainly not all the
studies report results in patients who are nonresponders
to treatment.
11
Primer
Molecular adsorbent recirculating system. The molec-
ular adsorbent recirculating system is a form of albu-
min dialysis whereby albumin recirculates repeatedly
to adsorb various bacterial products and cytokines that
are thought to be responsible for maintaining the vaso­
dilatory state of advanced cirrhosis. A small study in
2000 showed reductions in serum creatinine and mor-
tality when a molecular adsorbent recirculating system
was added to standard medical therapy for HRS and
haemodiafiltration126. A more recent 2010 randomized
controlled trial enrolled 189 patients with ACLF, 50%
of whom had type 1 HRS, and confirmed the reduc-
tion in serum creatinine when a molecular adsorbent
recirculating system was compared with standard med-
ical treatment. However, there was no effect on 28-day
survival, irrespective of whether the entire study popu­
lation or the subgroup of patients with type 1 HRS
was evaluated127. The reduction in serum creatinine was
related to the molecular adsorbent recirculating system
dialysis working as a very efficient system for remov-
ing creatinine from the serum without affecting renal
blood flow or GFR128. The most recent clinical guide-
lines from the European Association for the Study of
the Liver (EASL) indicate that a molecular adsorbent
recirculating system is not recommended for manage-
ment of HRS in clinical practice, but their potential
beneficial effects should be investigated further 129.
The question of a molecular adsorbent recirculating
system removing inflammatory mediators in patients
with AKI-​HRS has not been specifically studied. This
may not be possible, as the filters used in molecular
adsorbent recirculating system therapy can filter out
only molecules of certain size, and this may be a limit­
ing factor in the removal of inflammatory mediators in
patients with AKI-​HRS.
Transjugular intrahepatic portosystemic shunt. Portal
hypertension may be treated with the use of a trans-
jugular intrahepatic portosystemic shunt, a stent that is
placed via a transjugular approach within the liver, that
connects the portal vein with one of the hepatic veins.
This procedure has limited use as a treatment for type 1
HRS, however, as most of these patients also have severe
liver dysfunction. Although lowering portal pressure
makes physiological sense in patients with AKI-​HRS,
the relative liver ischaemia that immediately follows
transjugular intrahepatic portosystemic shunt insertion
can precipitate liver failure in patients with marginal
liver reserve. However, in carefully selected patients
with type 1 HRS and relatively preserved liver function,
such as patients with alcoholic liver disease who are now
abstinent or patients with viral hepatitis that has been
eradicated, the use of a transjugular intrahepatic porto-
systemic shunt following pharmacotherapy can lead to
an improved outcome117.
Renal replacement therapy. Renal replacement ther-
apy (that is, haemodialysis) has no role in the manage-
ment of AKI-​HRS in cirrhosis as a stand-​alone therapy.
A 2015 retrospective study showed that the routine use
of renal replacement therapy in patients with cir-
rhosis and type 1 HRS who were nonresponders to
12 | Article citation ID: (2018) 4:23 www.nature.com/nrdp
vasoconstrictor therapy (that is, midodrine, noradren-
aline or vasopressin) did not provide any significant
improvement in either 30-day or 180-day survival but
led to significantly longer stays in hospital130. In patients
who have a reversible cause of liver failure, or those who
are listed for liver transplant, renal replacement therapy
can be initiated until either the recovery of liver function
or liver transplantation6.
Liver transplantation. Liver transplantation is the
definitive treatment for AKI-​HRS, as it removes liver
dysfunction and eliminates portal hypertension, which
are the two pivotal pathogenetic mechanisms respon-
sible for the development of type 1 HRS. However, the
recovery of renal function after liver transplantation is
not universal131,132, despite what is reported in the older
literature36,133. The mechanisms behind this are unclear,
but prolonged renal ischaemia in patients with persis-
tent AKI-​HRS that does not respond to pharmacother-
apy may induce tubular damage, which prevents full
recovery of renal function after liver transplantation122.
The various biomarkers becoming available, which indi-
cate renal tubular damage, will be able to better define
the likelihood of renal recovery after liver transplanta-
tion (see above). Patients with non-​pharmacotherapy-
responsive AKI-​HRS who are on the liver transplant
waitlist should receive renal replacement therapy if indi-
cated. A 2015 retrospective study enrolling a homogen­
eous group of patients with type 1 HRS showed that
a pre-transplant dialysis period of >14 days was pre-
dictive of non-​recovery of renal function after trans-
plant131. Thus, timely liver transplant is imperative for
renal recovery. Renal recovery seems to be independent
of pharmacotherapy131,134, the need for pre-transplant
dialysis130,131 or whether the donor liver is from a living
donor or cadaveric135,136.
There is much debate as to whether patients with
type 1 HRS should receive liver transplant alone or a
simultaneous liver and kidney transplant. A retrospec-
tive study using the United Network for Organ Sharing
database found that there was no difference in recipi-
ent mortality or graft survival between liver transplant
alone and simultaneous liver and kidney transplant for
patients with AKI-​HRS137 when a dialysis period of
>8 weeks was used as an indication for simultaneous
liver and kidney transplant138. Therefore, the long-term
renal benefits of simultaneous liver and kidney trans-
plant for patients with type 1 HRS are still unproven.
In 2018, the policy for simultaneous liver and kidney
transplant is evolving. The most recent recommenda-
tions for simultaneous liver and kidney transplant in
the United States include: first, for patients with AKI, an
estimated GFR of ≤25 ml/min/1.73 m2 for 6 weeks or a
period of kidney dialysis of ≥6 weeks; second, the pres-
ence of stage ≥3B CKD, which is equivalent to a GFR
of <44 ml/min/1.73 m2 for >90 days; and third, the comor-
bid presence of metabolic diseases139 (Fig. 3). European
guidelines suggest that patients with end-​stage liver dis-
ease who have a GFR of <30 ml/min/1.73 m2 or type 1
HRS requiring renal replacement therapy of >8–12 weeks
and patients with renal biopsy samples revealing
>30% fibrosis and glomerulosclerosis should receive

simultaneous liver and kidney transplant74. Canadian
guidelines are similar to the European guidelines140.
In patients who reverse their type 1 HRS with liver
transplant, both renal function and survival are excel-
lent, with survival at >90% at 6 months to 1 year131,141,
which is comparable to patients without any form of AKI
who receive a liver transplant11.
Quality of life
The development of HRS is also associated with poor
survival. Patients with traditional type 2 HRS have a
median survival of ~6 months, and patients with type 1
HRS in the absence of pharmacological treatment or
liver transplantation have a median survival of only
~1  month7. No studies on quality of life (QOL) in
patients with HRS have been reported. Therefore, this
section deals with QOL in patients in the pre-​HRS phase
of the disease. Patients with advanced liver disease may
have complications such as ascites, jaundice, hepatic
encephalopathy and variceal bleeding. Ascites is asso-
ciated with impairment of QOL and reduced survival.
The median survival of patients with ascites is 2 years,
and ~40% of patients die within 1 year142. Thus, liver
transplantation should be considered in all patients who
develop ascites. Other variables associated with poor
prognosis include high serum creatinine, hyponatrae-
mia, low mean arterial pressure and decreased urinary
excretion of sodium143. Additionally, patients with ascites
have a risk of developing spontaneous bacterial peritoni-
tis. Following an episode of spontaneous bacterial peri-
tonitis, the probability of survival at 1 year is ~30–50%
Assess renal function by serum creatinine measurement
Normal renal AKI CKD
function or CKD • GFR <25 ml/min/1.73 m2 • Stage 3b with GFR
stage 1, stage 2 (US) 41–44 ml/min/1.73 m2
or stage 3a • GFR <30 ml/min/1.73 m2 for >90 days (US)
(Europe and Canada) • >30% fibrosis or
or needing dialysis glomerulosclerosis
on renal biopsy
(Europe)
Dialysis time Dialysis time
<6 weeks (US) >6 weeks (US)
<8-12 weeks (Europe) >8-12 weeks (Europe)
<12 weeks (Canada) >12 weeks (Canada)
Liver transplant alone Simultaneous liver and kidney transplant
After liver transplant, consider early kidney transplant if:
• Non-recovery of renal function
• Dialysis >3 months
• Simultaneous kidney transplant deferred owing to high
risk for kidney transplantation at time of liver transplantation
Fig. 3 | Algorithm for liver transplant alone versus simultaneous liver and kidney
transplant in HRS. A suggested algorithm for decision making regarding liver
transplantation alone versus simultaneous liver and kidney transplantation in patients
with hepatorenal syndrome is provided. The algorithm varies according to the regional
guidelines that are used. AKI, acute kidney injury ; CKD, chronic kidney disease; GFR ,
glomerular filtration rate; HRS, hepatorenal syndrome. Data were originally presented
in this manner in ref.156.
NATURE REvIEwS | DISeASe PRImeRS | Article citation I­D:­ _­# #­# #­# #­# #­# #­# #­# #­# #­# #­# #­# _­
Primer
and is ~25–30% at 2 years. Moreover, in the absence of
antibiotic prophylaxis, the risk of recurrent spontaneous
bacterial peritonitis is ~70% at 1 year144.
QOL in patients with ascites is related to abdominal
distension and lower extremity oedema, as well as other
changes related to advanced liver disease such as hypo­
natraemia145. Therefore, QOL in patients with ascites
might be best measured by combining scores of ascites,
lower extremity oedema and QOL related to advanced
liver disease. Generic instruments like the Chronic Liver
Disease Questionnaire may be used to capture liver-​
related QOL. Ascites-​related QOL is best determined
by the Ascites-​Q instrument, though the Functional
Assessment of Chronic Illness Therapy-​Ascites Index
(FACIT-​AI) and Ascites Symptom Inventory-7 (ASI-7)
are also used. Lower limb oedema-​related QOL may be
measured using lymphedema instruments developed in
patients with cancer but has not been adequately tested
in patients with cirrhosis.
Outlook
In recent years, much progress has been made in the
management of HRS, which is mainly due to the use of
vasoconstrictors, specifically terlipressin146. With terli-
pressin treatment, HRS has changed from a condition
characterized by the relentless progression of kidney
failure leading rapidly to death unless timely liver trans-
plantation could be performed to a condition in which
reversibility of kidney failure may be achieved with
pharmacological therapy, and short-​term survival has
markedly improved in responders to therapy111. This
improvement in management has been possible thanks
to an improved knowledge of the pathophysiology and
a better awareness of the condition in patients with
advanced liver disease.
Future challenges
Despite this moderate optimism, a number of important
challenges still remain in order to improve patient out-
comes and renal function in HRS. The main challenges
are discussed below.
Improve early diagnosis. Previously, most patients
with type 1 HRS were diagnosed late, when GFR
is very low and serum creatinine has increased substan-
tially28,102–104,123,147,148. This is largely due to the require-
ment of a cut-​off level of 2.5 mg/dl as diagnostic criteria
but also to a delay in the diagnostic process owing to lack
of awareness of the condition. Early diagnosis is impor-
tant because the response to terlipressin and albumin
is inversely related to serum creatinine levels29,112,120.
The use of the new AKI-​HRS criteria, which no longer
require a cut-​off level of 2.5 mg/dl, may help improve
early diagnosis. However, a high degree of attention
to changes in kidney function, by the periodic assess-
ment of serum creatinine levels, is crucial if HRS is to be
diagnosed at an early stage.
Improved tools for differential diagnosis. A critical issue
in the diagnosis of HRS is to differentiate it from other
forms of AKI, particularly ATN. Because currently there
are no objective and specific methods, the differential
13
Primer

Assessment of AKI stage according to modified ICA-AKI criteria

Stage 1A AKI Stage 1B, stage 2 or stage 3 AKI

• Close monitoring • Remove risk factors (nephrotoxic drugs,

• Remove risk factors (nephrotoxic drugs, vasodilators, NSAIDs and diuretics)
vasodilators, NSAIDs and diuretics) • Volume expansion with
• Treatment of infections, if present albumin (1 g/kg) for 2 days
• Plasma volume expansion, if fluid losses

Resolutiona Stable Progression Resolutiona Stable or progression

Criteria of AKI-HRS

No Yes

Futher treatment of AKI decided Renal replacement therapy, Vasoconstrictors, (terlipressin)

on a case-by-case basis if required and albuminb

Fig. 4 | An algorithm for diagnosis and management of AKI in cirrhosis. A newly proposed algorithm for the diagnosis
and management of acute kidney injury (AKI) in cirrhosis is shown. AKI-​HRS, AKI-​hepatorenal syndrome; ICA-​AKI,
International Club of Ascites-​AKI. aReturn of serum creatinine levels to <0.3 mg/dl from baseline. bIntravenous
noradrenaline can be used if terlipressin is not available. The combination of midodrine, octreotide and albumin is much
less effective than terlipressin or noradrenaline plus albumin. Modified with permission from ref.129, Elsevier.

diagnosis is usually made on the basis of clinical findings
and certain urine parameters, the value of which has not
yet been convincingly demonstrated82,83. In 2016, it was
shown that some kidney biomarkers, particularly NGAL
and IL-18, are of value in the differential diagnosis
because they are markedly increased in ATN compared
with in HRS83. These biomarkers will likely be incor-
porated into a diagnosis and management algorithm of
AKI in cirrhosis, which was recently proposed in new
European guidelines for the management of patients
with decompensated cirrhosis129 (Fig. 4).
Nonresponse to pharmacological therapy. With the
currently available treatments, only ~50% of patients
respond to therapy. The reason why such a high pro-
portion of patients do not respond to pharmacological
therapy is currently unknown. There are a number of
possible causes of the lack of response. First, a delay in
the initiation of treatment owing to delayed diagnosis, as
mentioned above, could cause this nonresponse. Second,
the misdiagnosis of HRS may also play a part, as it is
possible that a proportion of patients do not respond to
therapy because they do not have HRS but other causes
of AKI, particularly ATN or some kind of intrinsic
nephropathy. The use of kidney biomarkers, particularly
NGAL or IL-18, may help prevent or reduce misdiagno-
sis. Third, associated severe cardiomyopathy may lead
to nonresponse. Cirrhotic cardiomyopathy is common
in patients with advanced cirrhosis and is characterized
by a poor increase in cardiac output in demanding situ­
ations33,149. Low cardiac output has been demonstrated
in some patients with type 1 HRS31,32,34. Moreover, it has
been shown that terlipressin may have negative effects
on cardiac function150. Therefore, the possibility exists
that in some patients with HRS, terlipressin admin-
istration may result in a further impairment of an
already depressed cardiac function, resulting in a lack
of improvement of effective arterial blood volume and
no effect on kidney function. This hypothesis deserves
future investigation.
Finally, patients with HRS frequently have associated
ACLF111; as mentioned before, a recent study showed
that response to terlipressin in HRS is dependent on
severity of ACLF. Patients with ACLF grade 3 (the most
severe form) have a lower rate of response to terlipressin
than those with ACLF grade 2 or ACLF grade 1 (ref.124).
This suggests that patients with severe ACLF have
causes of AKI other than HRS or that the multiorgan
failure present in more severe cases of ACLF precludes
an effective haemodynamic response to terlipressin.
The presence of severe systemic inflammation may
be one of the factors responsible for the reduced effect
of terlipressin.
Prevention. Some effective methods of prevention have
been evaluated. The first and more effective method is
the administration of albumin to patients with cirrho-
sis and spontaneous bacterial peritonitis42,65. This pre-
ventive treatment not only prevents the development
of HRS, which is very common in these patients, but
also improves survival. This method of prevention is
recommended by international clinical guidelines74,129.
By contrast, albumin is not effective in the prevention
of HRS that is associated with infections other than
spontaneous bacterial peritonitis96,97. A second pre-
ventive treatment is the administration of norfloxacin
to patients with advanced cirrhosis and ascites, which
decreases the risk of occurrence of HRS independently

14 | Article citation ID: (2018) 4:23 www.nature.com/nrdp

 Primer

from its effect in the prevention of spontaneous bacte-
rial peritonitis recurrence98. However, this method has
the potential drawback of increasing the risk of infec-
tions caused by multidrug-​resistant organisms. Long-​
term treatment with rifaximin has also been reported
to prevent the occurrence of HRS, but the information
is limited and requires confirmation in larger prospec-
tive randomized controlled studies151. Finally, two other
methods have been evaluated: treatment with granulo-
cyte colony-​stimulating factor (G-​CSF) in patients with
ACLF and pentoxifylline. G-​CSF has shown efficacy in a
randomized study with a low sample size, and its efficacy
requires validation in other studies152. Reasons for the
efficacy of G-​CSF are not known but may be related at
least in part to the effect of this drug in increasing leuko-
cyte number in the peripheral blood and thus decreasing
infections that may trigger HRS. With respect to pentoxi­
fylline, studies have shown contradictory findings153–155.
In summary, methods of prevention of HRS are urgently
needed and should be investigated.
Published online xx xx xxxx

1. Ginès, P. & Schrier, R. W. Renal failure in cirrhosis.
N. Engl. J. Med. 361, 1279–1290 (2009).
2. Durand, F., Graupera, I., Ginès, P., Olson, J. C. &
Nadim, M. K. Pathogenesis of hepatorenal syndrome:
implications for therapy. Am. J. Kidney Dis. 67,
318–328 (2016).
3. Wong, F. The evolving concept of acute kidney injury
in patients with cirrhosis. Nat. Rev. Gastroenterol.
Hepatol. 12, 711–719 (2015).
4. Arroyo, V. et al. Definition and diagnostic criteria
of refractory ascites and hepatorenal syndrome in
cirrhosis. Hepatology 23, 164–176 (1996).
This is the first consensus published on the
definition and diagnostic criteria of HRS in
cirrhosis.
5. Angeli, P. et al. Diagnosis and management of
acute kidney injury in patients with cirrhosis: revised
consensus recommendations of the international club
of ascites. J. Hepatol. 62, 968–974 (2015).
6. Nadim, M. K. et al. Hepatorenal syndrome: the 8th
international consensus conference of the acute
dialysis quality initiative (ADQI) group. Crit. Care 16,
R23 (2012).
7. Alessandria, C. et al. MELD score and clinical type
predict prognosis in hepatorenal syndrome: relevance
to liver transplantation. Hepatology 41, 1282–1289
(2005).
8. Garcia-​Tsao, G., Parikh, C. R. & Viola, A. Acute kidney
injury in cirrhosis. Hepatology 48, 2064–2077 (2008).
9. Ginès, P., Cárdenas, A., Solà, E., Schrier, R. in Schrier’s
Diseases of the Kidney (eds Schrier, R. W. et al)
1965–1996 (Lippincott Williams & Wilkins, 2012).
10. Belcher, J. M. et al. Association of AKI with mortality
and complications in hospitalized patients with
cirrhosis. Hepatology 57, 753–762 (2012).
11. Nadim, M. K. et al. Impact of the etiology of
acute kidney injury on outcomes following liver
transplantation: acute tubular necrosis versus
hepatorenal syndrome. Liver Transplant. 18,
539–548 (2012).
12. Hilmi, I. A. et al. Acute kidney injury following
orthotopic liver transplantation: incidence, risk
factors, and effects on patient and graft outcomes.
Br. J. Anaesth. 114, 919–926 (2015).
13. Tandon, P. et al. Relevance of new definitions to
incidence and prognosis of acute kidney injury in
hospitalized patients with cirrhosis: a retrospective
population-​based cohort study. PLOS One 11,
e0160394 (2016).
14. Fagundes, C. et al. A modified acute kidney injury
classification for diagnosis and risk stratification of
impairment of kidney function in cirrhosis. J. Hepatol.
59, 474–481 (2013).
15. Amathieu, R. et al. Significance of oliguria in critically
ill patients with chronic liver disease. Hepatology 66,
1592–1600 (2017).
16. Huelin, P. et al. Validation of a staging system for
acute kidney injury in patients with cirrhosis and
association with acute-​on-chronic liver failure.
Clin. Gastroenterol. Hepatol. 15, 438–445 (2017).
This prospective study reports the frequency and
aetiology of AKI episodes in a large series of
patients hospitalized for complications of cirrhosis.
This study also proposes a modification of AKI
staging specifically for cirrhosis.
17. Allegretti, A. S. et al. Prognosis of acute kidney injury
and hepatorenal syndrome in patients with cirrhosis:
a prospective cohort study. Int. J. Nephrol. 2015, 1–9
(2015).
18. Piano, S. et al. Evaluation of the acute kidney
injury network criteria in hospitalized patients with
cirrhosis and ascites. J. Hepatol. 59, 482–489
(2013).
19. Martín–Llahí, M. et al. Prognostic importance of the
cause of renal failure in patients with cirrhosis.
Gastroenterology 140, 488–496 (2011).
20. Fagundes, C. & Ginès, P. Hepatorenal syndrome:
a severe, but treatable, cause of kidney failure in
cirrhosis. Am. J. Kidney Dis. 59, 874–885 (2012).
21. Schrier, R. W. et al. Peripheral arterial vasodilation
hypothesis: a proposal for the initiation of renal
sodium and water retention in cirrhosis. Hepatology
8, 1151–1157 (1988).
This is a seminal paper describing the
haemodynamic abnormalities in patients with
cirrhosis and proposing a theory on the
mechanisms leading to renal sodium retention
and development of ascites and kidney function
abnormalities in cirrhosis.
22. Epstein, M. et al. Renal failure in the patient with
cirrhosis. Am. J. Med. 49, 175–185 (1970).
23. Martin, P.-Y., Ginès, P. & Schrier, R. W. Nitric oxide
as a mediator of hemodynamic abnormalities and
sodium and water retention in cirrhosis. N. Engl.
J. Med. 339, 533–541 (1998).
24. Iwakiri, Y. & Groszmann, R. J. The hyperdynamic
circulation of chronic liver diseases: from the
patient to the molecule. Hepatology 43, S121–S131
(2006).
25. Ros, J. et al. Endogenous cannabinoids: A new system
involved in the homeostasis of arterial pressure in
experimental cirrhosis in the rat. Gastroenterology
122, 85–93 (2002).
26. Bernardi, M., Trevisani, F., Gasbarrini, A. &
Gasbarrini, G. Hepatorenal disorders: role of the
renin-​angiotensin-aldosterone system. Semin. Liver Dis.
14, 23–34 (1994).
27. Henriksen, J. H., Møller, S., Ring-​Larsen, H. &
Christensen, N. J. The sympathetic nervous
system in liver disease. J. Hepatol. 29, 328–341
(1998).
28. Sanyal, A. J. et al. A randomized, prospective,
double-​blind, placebo-​controlled trial of terlipressin
for type 1 hepatorenal syndrome. Gastroenterology
134, 1360–1368 (2008).
29. Martín–Llahí, M. et al. Terlipressin and albumin
versus albumin in patients with cirrhosis and
hepatorenal syndrome: a randomized study.
Gastroenterology 134, 1352–1359 (2008).
30. Gluud, L. L., Christensen, K., Christensen, E. &
Krag, A. Systematic review of randomized trials on
vasoconstrictor drugs for hepatorenal syndrome.
Hepatology 51, 576–584 (2009).
31. Ruiz-​del-Arbol, L. Systemic, renal, and hepatic
hemodynamic derangement in cirrhotic patients with
spontaneous bacterial peritonitis. Hepatology 38,
1210–1218 (2003).
32. Ruiz-​del-Arbol, L. et al. Circulatory function and
hepatorenal syndrome in cirrhosis. Hepatology 42,
439–447 (2005).
33. Wong, F. Cirrhotic cardiomyopathy. Hepatol. Int. 3,
294–304 (2008).
This is a comprehensive review of the aetiology,
diagnosis and clinical consequences of cirrhotic
cardiomyopathy.
34. Krag, A., Bendtsen, F., Henriksen, J. H. & Moller, S.
Low cardiac output predicts development of
hepatorenal syndrome and survival in patients with
cirrhosis and ascites. Gut 59, 105–110 (2009).
35. Nazar, A. et al. LEFT ventricular function assessed by
echocardiography in cirrhosis: relationship to systemic
hemodynamics and renal dysfunction. J. Hepatol. 58,
51–57 (2013).
36. Gonwa, T. A. et al. Impact of pretransplant renal
function on survival after liver transplantation.
Transplantation 59, 361–365 (1995).
37. Schroeder, E. T., Shear, L., Sancetta, S. M. &
Gabuzda, G. J. Renal failure in patients with cirrhosis
of the liver: 3. Evaluation of intrarenal blood flow by
para-​aminohippurate extraction and response to
angiotensin. Am. J. Med. 43, 887–896 (1967).
38. Kew, M. C. et al. Renal and intrarenal blood-​flow in
cirrhosis of the liver. Lancet 2, 504–510 (1971).
39. Sacerdoti, D., Merlo, A., Merkel, C., Zuin, R. &
Gatta, A. Redistribution of renal blood flow in patients
with liver cirrhosis: the role of renal PGE2. J. Hepatol.
2, 253–261 (1986).
40. Maroto, A. et al. Diagnosis of functional kidney failure
of cirrhosis with Doppler sonography: prognostic
value of resistive index. Hepatology 20, 839–844
(1994).
41. Arroyo, V., Ginés, P., Rimola, A. & Gaya, J. Renal
function abnormalities, prostaglandins, and effects of
nonsteroidal anti-​inflammatory drugs in cirrhosis with
ascites: an overview with emphasis on pathogenesis.
Am. J. Med. 81, 104–122 (1986).
42. Sort, P. et al. Effect of intravenous albumin on renal
impairment and mortality in patients with cirrhosis
and spontaneous bacterial peritonitis. N. Engl. J. Med.
341, 403–409 (1999).
This fundamental study demonstrates the efficacy
of albumin administration in preventing HRS and
reducing mortality in patients with cirrhosis and
spontaneous bacterial peritonitis.
43. Barreto, R. et al. Type-1 hepatorenal syndrome
associated with infections in cirrhosis: Natural history,
outcome of kidney function, and survival. Hepatology
59, 1505–1513 (2014).
44. Uriz, J. et al. Terlipressin plus albumin infusion:
an effective and safe therapy of hepatorenal syndrome.
J. Hepatol. 33, 43–48 (2000).
45. Schroeder, E. T., Anderson, G. H., Goldman, S. H. &
Streeten, D. H. Effect of blockade of angiotensin II on
blood pressure, renin and aldosterone in cirrhosis.
Kidney Int. 9, 511–519 (1976).
46. Moore, K. et al. Plasma endothelin immunoreactivity
in liver disease and the hepatorenal syndrome.
N. Engl. J. Med. 327, 1774–1778 (1992).
47. Moore, K. P. et al. Increased production of cysteinyl
leukotrienes in hepatorenal syndrome. J. Hepatol. 11,
263–271 (1990).
48. Wong, F., Moore, K., Dingemanse, J. & Jalan, R.
Lack of renal improvement with nonselective endothelin
antagonism with tezosentan in type 2 hepatorenal
syndrome. Hepatology 47, 160–168 (2008).
49. Elia, C. et al. Severe acute kidney injury associated
with non-​steroidal anti-​inflammatory drugs in
cirrhosis: a case-​control study. J. Hepatol. 63,
593–600 (2015).
50. Stadlbauer, V. P. et al. Relationship between activation
of the sympathetic nervous system and renal blood
flow autoregulation in cirrhosis. Gastroenterology
134, 111–119 (2008).
51. Albillos, A., Lario, M. & Álvarez-​Mon, M. Cirrhosis-​
associated immune dysfunction: distinctive features
and clinical relevance. J. Hepatol. 61, 1385–1396
(2014).
52. Bernardi, M., Moreau, R., Angeli, P., Schnabl, B. &
Arroyo, V. Mechanisms of decompensation and organ
failure in cirrhosis: from peripheral arterial vasodilation
to systemic inflammation hypothesis. J. Hepatol. 63,
1272–1284 (2015).
53. Cazzaniga, M. et al. The systemic inflammatory
response syndrome in cirrhotic patients: relationship
with their in-​hospital outcome. J. Hepatol. 51,
475–482 (2009).
54. Cervoni, J. P. et al. C-​reactive protein predicts short-​
term mortality in patients with cirrhosis. J. Hepatol.
56, 1299–1304 (2012).

NATURE REvIEwS | DISeASe PRImeRS | Article citation I­D:­ _­# #­# #­# #­# #­# #­# #­# #­# #­# #­# #­# _­ 15
Primer
55. Jalan, R. et al. Alterations in the functional capacity
of albumin in patients with decompensated cirrhosis is
associated with increased mortality. Hepatology 50,
555–564 (2009).
56. Clària, J. et al. Systemic inflammation in
decompensated cirrhosis: characterization and role
in acute-​on-chronic liver failure. Hepatology 64,
1249–1264 (2016).
57. Solé, C. et al. Characterization of inflammatory
response in acute-​on-chronic liver failure and
relationship with prognosis. Sci. Rep. 6 (2016).
58. Moreau, R. et al. Acute-​on-chronic liver failure is a
distinct syndrome that develops in patients with acute
decompensation of cirrhosis. Gastroenterology 144,
1426–1437 (2013).
59. Wiest, R., Lawson, M. & Geuking, M. Pathological
bacterial translocation in liver cirrhosis. J. Hepatol.
60, 197–209 (2014).
60. Follo, A. et al. Renal impairment after spontaneous
bacterial peritonitis in cirrhosis: incidence, clinical
course, predictive factors and prognosis. Hepatology
20, 1495–1501 (1994).
This is the first study to provide a detailed
description of the association between renal
function impairment and spontaneous bacterial
peritonitis in cirrhosis.
61. Navasa, M. et al. Tumor necrosis factor and
interleukin-6 in spontaneous bacterial peritonitis
in cirrhosis: relationship with the development
of renal impairment and mortality. Hepatology 27,
1227–1232 (1998).
62. Terra, C. et al. Renal failure in patients with cirrhosis
and sepsis unrelated to spontaneous bacterial
peritonitis: value of MELD score. Gastroenterology
129, 1944–1953 (2005).
63. Fasolato, S. et al. Renal failure and bacterial infections
in patients with cirrhosis: Epidemiology and clinical
features. Hepatology 45, 223–229 (2006).
64. Pereira, G. et al. Renal failure and hyponatremia in
patients with cirrhosis and skin and soft tissue infection.
A retrospective study. J. Hepatol. 56, 1040–1046
(2012).
65. Terg, R. et al. Serum creatinine and bilirubin predict
renal failure and mortality in patients with spontaneous
bacterial peritonitis: a retrospective study. Liver Int.
29, 415–419 (2009).
66. De Las Heras, D. Increased carbon monoxide
production in patients with cirrhosis with and without
spontaneous bacterial peritonitis. Hepatology 38,
452–459 (2003).
67. Such, J. et al. Nitric oxide in ascitic fluid is an
independent predictor of the development of renal
impairment in patients with cirrhosis and spontaneous
bacterial peritonitis. Eur. J. Gastroenterol. Hepatol.
16, 571–577 (2004).
68. Thabut, D. et al. Model for end-​stage liver disease
score and systemic inflammatory response are major
prognostic factors in patients with cirrhosis and acute
functional renal failure. Hepatology 46, 1872–1882
(2007).
69. Arroyo, V. et al. Acute-​on-chronic liver failure in
cirrhosis. Nat. Rev. Dis. Primers 2, 16041 (2016).
70. Angus, D. C. & van der Poll, T. Severe sepsis and septic
shock. N. Engl. J. Med. 369, 840–851 (2013).
71. Sriskandan, S. & Altmann, D. M. The immunology
of sepsis. J. Pathol. 214, 211–223 (2008).
72. Wasmuth, H. E. et al. Patients with acute on chronic
liver failure display ‘sepsis-​like’ immune paralysis.
J. Hepatol. 42, 195–201 (2005).
73. Ginès, P. et al. Randomized comparative study
of therapeutic paracentesis with and without
intravenous albumin in cirrhosis. Gastroenterology
94, 1493–1502 (1988).
74. European Association for the Study of the Liver.
EASL clinical practice guidelines on the management
of ascites, spontaneous bacterial peritonitis, and
hepatorenal syndrome in cirrhosis. J. Hepatol. 53,
397–417 (2010).
75. Cárdenas, A. Renal failure after upper gastrointestinal
bleeding in cirrhosis: Incidence, clinical course,
predictive factors, and short-​term prognosis.
Hepatology 34, 671–676 (2001).
76. Bernardi, M., Caraceni, P., Navickis, R. J. &
Wilkes, M. M. Albumin infusion in patients undergoing
large-​volume paracentesis: a meta-​analysis of
randomized trials. Hepatology 55, 1172–1181 (2012).
This study proves the beneficial effects of albumin
treatment in patients with cirrhosis undergoing
large-​volume paracentesis.
77. Carl, D. E. et al. Post-​paracentesis circulatory
derangements are related to monocyte activation.
Liver Int. 34, 1001–1007 (2014).
16 | Article citation ID: (2018) 4:23 www.nature.com/nrdp
78. Salerno, F., Gerbes, A., Gines, P., Wong, F. & Arroyo, V.
Diagnosis, prevention and treatment of hepatorenal
syndrome in cirrhosis. Postgrad. Med. J. 84, 662–670
(2008).
79. Kidney Disease: Improving Global Outcomes
(KDIGO) CKD-​MBD Update Work Group. KDIGO
2017 Clinical Practice Guideline Update for the
Diagnosis, Evaluation, Prevention, and Treatment of
Chronic Kidney Disease–Mineral and Bone Disorder
(CKD-​MBD). Kidney Int. Suppl. 7, issue 1 (2017).
80. Tsien, C. D., Rabie, R. & Wong, F. Acute kidney injury in
decompensated cirrhosis. Gut 62, 131–137 (2012).
81. Wong, F. et al. New consensus definition of acute
kidney injury accurately predicts 30-day mortality in
patients with cirrhosis and infection. Gastroenterology
145, e16–e17 (2013).
82. Schrier, R. W., Shchekochikhin, D. & Ginès, P. Renal
failure in cirrhosis: prerenal azotemia, hepatorenal
syndrome and acute tubular necrosis. Nephrol. Dial.
Transplant. 27, 2625–2628 (2012).
83. Francoz, C., Nadim, M. K. & Durand, F. Kidney
biomarkers in cirrhosis. J. Hepatol. 65, 809–824
(2016).
84. Fagundes, C. et al. Urinary neutrophil gelatinase-​
associated lipocalin as biomarker in the differential
diagnosis of impairment of kidney function in cirrhosis.
J. Hepatol. 57, 267–273 (2012).
85. Verna, E. C. et al. Urinary neutrophil gelatinase-​
associated lipocalin predicts mortality and identifies
acute kidney injury in cirrhosis. Dig. Dis. Sci. 57,
2362–2370 (2012).
86. Belcher, J. M. et al. Urinary biomarkers and
progression of AKI in patients with cirrhosis. Clin. J. Am.
Soc. Nephrol. 9, 1857–1867 (2014).
87. Belcher, J. M. et al. Kidney biomarkers and differential
diagnosis of patients with cirrhosis and acute kidney
injury. Hepatology 60, 622–632 (2014).
88. Ariza, X. et al. Analysis of a urinary biomarker
panel for clinical outcomes assessment in cirrhosis.
PLOS One 10, e0128145 (2015).
89. Barreto, R. et al. Urinary neutrophil gelatinase-​
associated lipocalin predicts kidney outcome and
death in patients with cirrhosis and bacterial infections.
J. Hepatol. 61, 35–42 (2014).
90. Decavele, A.-S. C., Dhondt, L., De Buyzere, M. L. &
Delanghe, J. R. Increased urinary neutrophil gelatinase
associated lipocalin in urinary tract infections and
leukocyturia. Clin. Chem. Lab. Med. 49, 999–1003
(2011).
91. Puthumana, J. et al. Urine interleukin 18 and lipocalin
2 are biomarkers of acute tubular necrosis in patients
with cirrhosis: a systematic review and meta-​analysis.
Clin. Gastroenterol. Hepatol. 15, 1003–1013.e3
(2017).
92. Markwardt, D. et al. Plasma cystatin C is a predictor
of renal dysfunction, acute-​on-chronic liver failure, and
mortality in patients with acutely decompensated liver
cirrhosis. Hepatology 66, 1232–1241 (2017).
93. Maiwall, R. et al. Cystatin C predicts acute kidney
injury and mortality in cirrhotics: a prospective cohort
study. Liver Int. 38, 654–664 (2018).
94. Salerno, F., Navickis, R. J. & Wilkes, M. M. Albumin
infusion improves outcomes of patients with
spontaneous bacterial peritonitis: a meta-​analysis of
randomized trials. Clin. Gastroenterol. Hepatol. 11,
123–130.e1 (2013).
95. Poca, M. et al. Role of albumin treatment in
patients with spontaneous bacterial peritonitis.
Clin. Gastroenterol. Hepatol. 10, 309–315 (2012).
96. Guevara, M. et al. Albumin for bacterial infections
other than spontaneous bacterial peritonitis in
cirrhosis. A randomized, controlled study. J. Hepatol.
57, 759–765 (2012).
97. Thévenot, T. et al. Effect of albumin in cirrhotic
patients with infection other than spontaneous
bacterial peritonitis. A randomized trial. J. Hepatol.
62, 822–830 (2015).
98. Fernández, J. et al. Primary prophylaxis of spontaneous
bacterial peritonitis delays hepatorenal syndrome and
improves survival in cirrhosis. Gastroenterology 133,
818–824 (2007).
99. Mandorfer, M. et al. Nonselective β blockers increase
risk for hepatorenal syndrome and death in patients
with cirrhosis and spontaneous bacterial peritonitis.
Gastroenterology 146, 1680–1690.e1 (2014).
100. Singh, V. et al. Noradrenaline versus terlipressin in the
treatment of hepatorenal syndrome: a randomized
study. J. Hepatol. 56, 1293–1298 (2012).
101. Sharma, P., Kumar, A., Shrama, B. C. & Sarin, S. K.
An open label, pilot, randomized controlled trial of
noradrenaline versus terlipressin in the treatment
of type 1 hepatorenal syndrome and predictors of
response. Am. J. Gastroenterol. 103, 1689–1697
(2008).
102. Cavallin, M. et al. Terlipressin given by continuous
intravenous infusion versus intravenous boluses
in the treatment of hepatorenal syndrome:
A randomized controlled study. Hepatology 63,
983–992 (2016).
This study proves that continuous intravenous
infusion of terlipressin is safer than bolus
administration in patients with cirrhosis and HRS.
103. Cavallin, M. et al. Terlipressin plus albumin versus
midodrine and octreotide plus albumin in the
treatment of hepatorenal syndrome: A randomized
trial. Hepatology 62, 567–574 (2015).
104. Boyer, T. D. et al. Terlipressin plus albumin is more
effective than albumin alone in improving renal
function in patients with cirrhosis and hepatorenal
syndrome type 1. Gastroenterology 150,
1579–1589.e2 (2016).
105. Fernández, J. et al. A randomized unblinded pilot
study comparing albumin versus hydroxyethyl starch
in spontaneous bacterial peritonitis. Hepatology 42,
627–634 (2005).
106. Brinch, K., Møller, S., Bendtsen, F., Becker, U. &
Henriksen, J. H. Plasma volume expansion by albumin
in cirrhosis. Relation to blood volume distribution,
arterial compliance and severity of disease. J. Hepatol.
39, 24–31 (2003).
107. Bortoluzzi, A. et al. Positive cardiac inotropic effect of
albumin infusion in rodents with cirrhosis and ascites:
molecular mechanisms. Hepatology 57, 266–276
(2013).
108. Ortega, R. Terlipressin therapy with and without
albumin for patients with hepatorenal syndrome:
Results of a prospective, nonrandomized study.
Hepatology 36, 941–948 (2002).
109. Neri, S. et al. Terlipressin and albumin in patients with
cirrhosis and type i hepatorenal syndrome. Dig. Dis. Sci.
53, 830–835 (2008).
110. Solanki, P. et al. Beneficial effects of terlipressin in
hepatorenal syndrome: a prospective, randomized
placebo-​controlled clinical trial. J. Gastroenterol.
Hepatol. 18, 152–156 (2003).
111. Ginès, P. Management of hepatorenal syndrome in
the era of acute-​on-chronic liver failure: terlipressin
and beyond. Gastroenterology 150, 1525–1527
(2016).
112. Sanyal, A. J. et al. Reversal of hepatorenal syndrome
type 1 with terlipressin plus albumin versus placebo
plus albumin in a pooled analysis of the OT-0401
and REVERSE randomised clinical studies.
Aliment. Pharmacol. Ther. 45, 1390–1402 (2017).
113. Facciorusso, A. et al. Comparative efficacy of
pharmacological strategies for management of type 1
hepatorenal syndrome: a systematic review and
network meta-​analysis. Lancet Gastroenterol. Hepatol.
2, 94–102 (2017).
114. Alessandria, C. et al. Noradrenalin versus terlipressin
in patients with hepatorenal syndrome: a prospective,
randomized, unblinded, pilot study. J. Hepatol. 47,
499–505 (2007).
115. Rodriguez, E. et al. Treatment of type 2 hepatorenal
syndrome in patients awaiting transplantation: effects
on kidney function and transplantation outcomes.
Liver Transplant. 21, 1347–1354 (2015).
116. Angeli, P. et al. Reversal of type 1 hepatorenal
syndrome with the administration of midodrine and
octreotide. Hepatology 29, 1690–1697 (1999).
117. Wong, F., Pantea, L. & Sniderman, K. Midodrine,
octreotide, albumin, and TIPS in selected patients
with cirrhosis and type 1 hepatorenal syndrome.
Hepatology 40, 55–64 (2004).
118. Narahara, Y. et al. Effects of terlipressin on systemic,
hepatic and renal hemodynamics in patients with
cirrhosis. J. Gastroenterol. Hepatol. 24, 1791–1797
(2009).
119. Domenicali, M. et al. Posttranscriptional changes of
serum albumin: clinical and prognostic significance in
hospitalized patients with cirrhosis. Hepatology 60,
1851–1860 (2014).
120. Boyer, T. D. et al. Predictors of response to terlipressin
plus albumin in hepatorenal syndrome (HRS) type 1:
relationship of serum creatinine to hemodynamics.
J. Hepatol. 55, 315–321 (2011).
121. Velez, J. C. Q. & Nietert, P. J. Therapeutic response to
vasoconstrictors in hepatorenal syndrome parallels
increase in mean arterial pressure: a pooled analysis
of clinical trials. Am. J. Kidney Dis. 58, 928–938
(2011).
122. Trawalé, J.-M. et al. The spectrum of renal lesions in
patients with cirrhosis: a clinicopathological study.
Liver Int. 30, 725–732 (2010).

123. Rodríguez, E. et al. Terlipressin and albumin for
type-1 hepatorenal syndrome associated with sepsis.
J. Hepatol. 60, 955–961 (2014).
124. Piano, S. et al. Impact of acute-​on-chronic liver failure
on response to treatment with terlipressin and
albumin in patients with type 1 hepatorenal syndrome.
J. Hepatol. 66, S572 (2017).
125. Acevedo, J. et al. Relative adrenal insufficiency in
decompensated cirrhosis: relationship to short-​term
risk of severe sepsis, hepatorenal syndrome, and
death. Hepatology 58, 1757–1765 (2013).
126. Mitzner, S. et al. Improvement of hepatorenal
syndrome with extracorporeal albumin dialysis mars:
results of a prospective, randomized, controlled clinical
trial. Liver Transplant. 6, 277–286 (2000).
127. Bañnares, R. et al. Extracorporeal liver support
with the molecular adsorbent recirculating system
(MARS) in patients with acute-​on-chronic liver failure
(AOCLF). The RELIEF trial. J. Hepatol. 52, S459–S460
(2010).
128. Wong, F., Raina, N. & Richardson, R. Molecular
adsorbent recirculating system is ineffective in the
management of type 1 hepatorenal syndrome in
patients with cirrhosis with ascites who have failed
vasoconstrictor treatment. Gut 59, 381–386 (2009).
129. European Association for the Study of the Liver. EASL
Clinical Practice Guidelines for the management of
patients with decompensated cirrhosis. J. Hepatol.
69, 406–460 (2018).
130. Zhang, Z., Maddukuri, G., Jaipaul, N. & Cai, C. X.
Role of renal replacement therapy in patients with
type 1 hepatorenal syndrome receiving combination
treatment of vasoconstrictor plus albumin. J. Crit.
Care 30, 969–974 (2015).
131. Wong, F., Leung, W., Al Beshir, M., Marquez, M. &
Renner, E. L. Outcomes of patients with cirrhosis
and hepatorenal syndrome type 1 treated with liver
transplantation. Liver Transplant. 21, 300–307 (2015).
132. Marik, P. E., Wood, K. & Starzl, T. E. The course of type 1
hepato-​renal syndrome post liver transplantation.
Nephrol. Dial. Transplant. 21, 478–482 (2005).
133. Iwatsuki, S. et al. Recovery from hepatorenal syndrome
after orthotopic liver transplantation. N. Engl. J. Med.
289, 1155–1159 (1973).
134. Boyer, T. D. et al. Impact of liver transplantation on the
survival of patients treated for hepatorenal syndrome
type 1. Liver Transplant. 17, 1328–1332 (2011).
135. Chok, K. S. H. et al. Outcomes of living donor liver
transplantation for patients with preoperative
type 1 hepatorenal syndrome and acute hepatic
decompensation. Liver Transplant. 18, 779–785
(2012).
136. Goldaracena, N. et al. Living versus deceased donor
liver transplantation provides comparable recovery of
renal function in patients with hepatorenal syndrome:
a matched case-​control study. Am. J. Transplant. 14,
2788–2795 (2014).
NATURE REvIEwS | DISeASe PRImeRS | Article citation I­D:­ _­# #­# #­# #­# #­# #­# #­# #­# #­# #­# #­# _­
137. Martin, E. F. et al. Recipient survival and graft survival
are not diminished by simultaneous liver-​kidney
transplantation: an analysis of the united network
for organ sharing database. Liver Transplant. 18,
914–929 (2012).
138. Eason, J. D. et al. Proceedings of consensus conference
on simultaneous liver kidney transplantation (SLK).
Am. J. Transplant. 8, 2243–2251 (2008).
139. Formica, R. N. Simultaneous liver kidney
transplantation. Curr. Opin. Nephrol. Hypertens. 25,
577–582 (2016).
140. Cantarovich, M. et al. Canadian forum on combined
organ transplantation. Transplantation 100,
1339–1348 (2016).
141. Sethi, A., Estrella, M. M., Ugarte, R. & Atta, M.
Kidney function and mortality post-​liver transplant
in the Model for End-​Stage Liver Disease era.
Int. J. Nephrol. Renovasc. Dis. 4, 139–144 (2011).
142. Ginés, P. et al. Compensated cirrhosis: natural history
and prognostic factors. Hepatology 7, 122–128
(1987).
143. Llach, J. et al. Prognostic value of arterial pressure,
endogenous vasoactive systems, and renal function in
cirrhotic patients admitted to the hospital for the
treatment of ascites. Gastroenterology 94, 482–487
(1988).
This seminal study demonstrates the prognostic value
of variables related to systemic haemodynamics and
renal function in patients with cirrhosis and ascites.
144. Titó, L. et al. Recurrence of spontaneous bacterial
peritonitis in cirrhosis: frequency and predictive
factors. Hepatology 8, 27–31 (1988).
145. Solà, E. et al. Factors related to quality of life in
patients with cirrhosis and ascites: relevance of serum
sodium concentration and leg edema. J. Hepatol. 57,
1199–1206 (2012).
146. Ginès, P. Terlipressin for hepatorenal syndrome: ready
for prime time. Lancet Gastroenterol. Hepatol. 2,
75–76 (2017).
147. Nazar, A. et al. Predictors of response to therapy with
terlipressin and albumin in patients with cirrhosis
and type 1 hepatorenal syndrome. Hepatology 51,
219–226 (2009).
148. Salerno, F. et al. Diagnosis, treatment and survival
of patients with hepatorenal syndrome: a survey
on daily medical practice. J. Hepatol. 55, 1241–1248
(2011).
149. Liu, H., Jayakumar, S., Traboulsi, M. & Lee, S. S.
Cirrhotic cardiomyopathy: implications for liver
transplantation. Liver Transplant. 23, 826–835
(2017).
150. Krag, A., Bendtsen, F., Mortensen, C., Henriksen, J. H.
& Møller, S. Effects of a single terlipressin administration
on cardiac function and perfusion in cirrhosis. Eur. J.
Gastroenterol. Hepatol. 22, 1085–1092 (2010).
151. Dong, T., Aronsohn, A., Gautham Reddy, K. & Te, H. S.
Rifaximin decreases the incidence and severity of
Primer
acute kidney injury and hepatorenal syndrome in
cirrhosis. Dig. Dis. Sci. 61, 3621–3626 (2016).
152. Garg, V. et al. Granulocyte colony–stimulating factor
mobilizes cd34+ cells and improves survival of
patients with acute-​on-chronic liver failure.
Gastroenterology 142, 505–512.e1 (2012).
153. Lebrec, D. et al. Pentoxifylline does not decrease
short-​term mortality but does reduce complications in
patients with advanced cirrhosis. Gastroenterology
138, 1755–1762.e2 (2010).
154. Mathurin, P. et al. Prednisolone with versus without
pentoxifylline and survival of patients with severe
alcoholic hepatitis. JAMA 310, 1033 (2013).
155. Park, S. H. et al. Pentoxifylline versus corticosteroid
to treat severe alcoholic hepatitis: a randomised,
non-inferiority, open trial. J. Hepatol. 61, 792–798
(2014).
156. Pham, P.-T. T., Lunsford, K. E., Bunnapradist, S. &
Danovitch, G. M. Simultaneous liver–kidney
transplantation or liver transplantation alone for
patients in need of liver transplantation with renal
dysfunction. Curr. Opin. Organ. Transplant. 21,
194–200 (2016).
Acknowledgements
Some of the work cited has been funded from public grants
from Instituto de Salud Carlos III through the Plan Estatal de
Investigación Científica y Técnica y de Innovación 2013–2016
(project reference PI 16/00043). This grant was co-​funded by
the European Regional Development Fund (FEDER), Agencia
de Gestió d’Ajuts Universitaris I de Recerca (AGAUR) 2014/
SGR 1281 and AGAUR 2017/SGR 1281 and the European
Union (EU)-funded HORIZON 2020 project number 731875
(Acronym: LIVERHOPE). P.G. is a recipient of an ICREA
Academia award. This review article is dedicated to the memory
of Juan Rodés (1938–2017), who was a mentor and a friend.
Author contributions
Introduction (P.G.); Epidemiology (M.K.N.); Mechanisms/
pathophysiology (E.S.); Diagnosis, screening and prevention
(E.S.); Management (P.A. and F.W.); Quality of life (P.K.);
Outlook (P.G.); Overview of the Primer (P.G.).
Competing interests
P.G. declares that he is a member of advisory boards for
Ferring Pharmaceuticals, Intercept Pharmaceuticals, Martin
Pharmaceuticals and Novartis. He has received research
funds from Grifols and Sequana Medical AG. P.A. declares
that he is a member of the advisory board for Sequana
Medical AG and that he has no financial disclosures. F.W.
declares grant support from and consultancy for Mallinckrodt
and consultancy for Conatus and Ferring Pharmaceuticals.
E.S., M.K.N. and P.K. declare no competing interests.
Publisher’s note
Springer Nature remains neutral with regard to jurisdictional
claims in published maps and institutional affiliations.
17