The n e w e ng l a n d j o u r na l of m e dic i n e

Review Article

C. Corey Hardin, M.D., Ph.D., Editor

Acute Kidney Injury in Patients

with Cirrhosis
Mitra K. Nadim, M.D., and Guadalupe Garcia‑Tsao, M.D.​​

A
cute kidney injury (AKI) is a common condition in patients with From the Division of Nephrology and
cirrhosis. It occurs in up to 50% of hospitalized patients with cirrhosis and ­Hypertension, Keck School of Medicine,
University of Southern California, Los
in 58% of such patients in the intensive care unit (ICU).1-5 AKI is associ- Angeles (M.K.N.); and the Section of Di-
ated with high morbidity and mortality and an increased incidence of chronic gestive Diseases, Yale University School
kidney disease (CKD) after liver transplantation.1-3,5,6 Progression to advanced of Medicine, New Haven, and the Sec-
tion of Digestive Diseases, Veterans Af-
stages of AKI (Table 1) portends an even worse prognosis.1,2,5 fairs Connecticut Healthcare System, West
In general, the three main causes of AKI11 are renal hypoperfusion (also re- Haven — both in Connecticut (G.G.-T.).
ferred to as prerenal AKI), which in most cases is due to hypovolemia; intrinsic, Dr. Garcia-Tsao can be contacted at
­[email protected] or at the
structural kidney injury; and postrenal injury due to urinary obstruction. A unique Section of Digestive Diseases, Veterans
cause of AKI due to renal hypoperfusion in patients with cirrhosis is the hepatore- Affairs Connecticut Healthcare System,
nal syndrome (HRS), which is the result of renal vasoconstriction. Hypoperfusion 950 Campbell Ave., West Haven, CT 06516.

from hypovolemia accounts for approximately half the cases of AKI in patients N Engl J Med 2023;388:733-45.
with cirrhosis, intrinsic causes (e.g., acute tubular necrosis) account for approxi- DOI: 10.1056/NEJMra2215289
Copyright © 2023 Massachusetts Medical Society.
mately 30% of cases, and HRS accounts for approximately 15 to 20%, with less
than 1% of cases attributable to postrenal obstruction.3 CME
at NEJM.org

Defini t ion of A K I a nd HR S
Before 2012, criteria defining AKI in cirrhosis were absent, but HRS was defined
as a syndrome that occurred in patients with cirrhosis and portal hypertension and
was characterized by impaired kidney function (defined as a serum creatinine
level of >1.5 mg per deciliter [132.6 μmol per liter]) in the absence of underlying
kidney disease.7 Clinically, HRS was divided into type 1 (HRS-1), characterized by
a rapid reduction in kidney function, with an increase in serum creatinine to a
level exceeding 2.5 mg per deciliter [221.0 μmol per liter] in less than 2 weeks,
and type 2 (HRS-2), a more chronic deterioration in kidney function.
Over the past decade, the definitions of AKI and HRS in patients with cirrhosis
have been revised (Table 1).8-10 The new definition of AKI was harmonized with
the Kidney Disease: Improving Global Outcomes definition as an increase in the
serum creatinine level of 0.3 mg per deciliter (26.5 μmol per liter) or higher
within 48 hours or as an increase in the serum creatinine level that is at least 1.5
times the baseline level and that is known or presumed to have occurred within
the previous 7 days.8-10,12 The new definition of HRS eliminated the threshold
value for the serum creatinine level (1.5 mg per deciliter), allowing for earlier diag-
nosis and treatment in patients with normal serum creatinine levels but reduced
kidney function, such as women, older patients, or those with sarcopenia. The
acute form of HRS, HRS-1, was also renamed HRS-AKI to distinguish it from the
more chronic type, HRS-2, now renamed HRS-CKD.10 Changes in urinary output
should be considered in the definition of AKI, particularly in critically ill patients
with cirrhosis in whom changes in urinary output have been shown to be a sensitive,

n engl j med 388;8  nejm.org  February 23, 2023 733

The New England Journal of Medicine
Downloaded from nejm.org at UNIVERSIDAD NACIONAL AUTONOMA DE MEXICO on February 23, 2023. For personal use only. No other uses without permission.
Copyright © 2023 Massachusetts Medical Society. All rights reserved.
 734
Table 1. Definitions of Acute Kidney Injury (AKI) and the Hepatorenal Syndrome (HRS) in Patients with Cirrhosis.*

Variable or International Ascites Club, Acute Disease Quality Initiative, International Club of Ascites,
Definition Salerno et al.7 (2007) Nadim et al.8 (2012) Angeli et al.9 (2015) Angeli et al.10 (2019)
Baseline serum — — Serum creatinine measured in previous Similar to 2015 definition9
creatinine 3 mo; in patients with >1 value with-
level in previous 3 mo, the value ­closest
to time of hospital admission should
be used; in patients without a previ-
ous serum creatinine value, the value
on admission should be used as
The

baseline
AKI — Increase in serum creatinine ≥0.3 mg/ Increase in serum creatinine ≥0.3 mg/dl Absolute increase in serum creatinine
dl within 48 hr or increase ≥1.5 times within 48 hr or increase ≥1.5 times ≥0.3 mg/dl within 48 hr or≥1.5
baseline level baseline level, which is known or times baseline level or urinary
presumed to have occurred within ­output <0.5 ml/kg/hr in 6 hr
previous 7 days
AKI stage — Stage 1: increase in serum creatinine Similar to 2012 definition8 Similar to 2012 definition8
≥0.3 mg/dl or ≥1.5–2 times baseline
level
Stage 2: increase in serum creatinine
>2–3 times baseline level
Stage 3: increase in serum creatinine
>3 times baseline level or ≥4.0 mg/
n e w e ng l a n d j o u r na l

The New England Journal of Medicine

dl with an acute increase ≥0.5 mg/
of

dl or initiation of renal-replacement
therapy
HRS Cirrhosis with ascites Similar to 2007 definition7 Similar to 2007 definition,7 except for Similar to 2015 definition,9 except

n engl j med 388;8  nejm.org  February 23, 2023

Serum creatinine >1.5 mg/dl with no removal of serum creatinine for addition of urinary output
improvement (decrease ≤1.5 mg/ >1.5 mg/dl and inclusion of AKI <0.5 ml/kg/hr for ≥6 hr as a criterion

Copyright © 2023 Massachusetts Medical Society. All rights reserved.

m e dic i n e

dl) after at least 2 days of diuretic ­diagnosis according to KDIGO se- for AKI; suggestion of HRS-AKI with
withdrawal and volume expansion rum creatinine criteria (i.e., increase FeNa of <0.2% (FeNA <0.1% highly
with albumin (1 g/kg/day, maximum in serum creatinine ≥0.3 mg/dl with- predictive)
of 100 g/day) in 48 hr or ≥1.5 times baseline level)
Absence of shock
No current or recent treatment with
nephrotoxic drugs
Absence of parenchymal kidney disease
as indicated by proteinuria >500 mg/
day, microhematuria >50 red cells/
high-power field, or abnormal renal
findings on ultrasonography

Downloaded from nejm.org at UNIVERSIDAD NACIONAL AUTONOMA DE MEXICO on February 23, 2023. For personal use only. No other uses without permission.
 Variable or International Ascites Club, Acute Disease Quality Initiative, International Club of Ascites,
Definition Salerno et al.7 (2007) Nadim et al.8 (2012) Angeli et al.9 (2015) Angeli et al.10 (2019)
HRS type 1 Rapid, progressive renal failure, defined A specific form of AKI — HRS-AKI: absolute increase in serum
by doubling of initial serum creati- creatinine ≥0.3 mg/dl within 48 hr
nine (to a level >2.5 mg/dl) in <2 wk or increase in serum creatinine
>1.5 times baseline level; or urinary
output <0.5 ml/kg/hr for 6 hr
HRS type 2 Serum creatinine increased from 1.5 to A specific form of CKD — HRS-CKD: eGFR <60 ml/min/1.73 m2
2.5 mg/dl, with steady or slowly pro- for ≥3 mo in the absence of other
gressive course; typically associated (structural) causes
with refractory ascites HRS-AKD: eGFR <60 ml/min/1.73 m2
for <3 mo in the absence of other
(structural) causes or <50% increase
in serum creatinine with last out­
patient value within previous 3 mo
as baseline level
Response to Complete response (reversal of HRS): — Full response: return of serum creatinine Similar to 2015 definition9
therapy decrease in serum creatinine to to a level within 0.3 mg/dl of base-
<1.5 mg/dl line level
Partial response: decrease in serum cre- Partial response: regression of AKI
atinine ≥50% of pretreatment level, stage, with reduction of serum
without reaching level of <1.5 mg/dl ­creatinine to ≥0.3 mg/dl above
No response: no decrease in serum ­baseline level
creatinine or decrease to <50% of No response: no regression of AKI
pretreatment level, with final level
>1.5 mg/dl
Relapse: increase in serum creatinine
>1.5 mg/dl after discontinuation
of therapy

* To convert values for creatinine to millimoles per liter, multiply by 88.4. CKD denotes chronic kidney disease, eGFR estimated glomerular filtration rate, FeNa fractional excretion of so-

The New England Journal of Medicine

dium, and KDIGO Kidney Disease: Improving Global Outcomes.

n engl j med 388;8  nejm.org  February 23, 2023

Acute Kidney Injury in Patients with Cirrhosis

Copyright © 2023 Massachusetts Medical Society. All rights reserved.

735

Downloaded from nejm.org at UNIVERSIDAD NACIONAL AUTONOMA DE MEXICO on February 23, 2023. For personal use only. No other uses without permission.
 The n e w e ng l a n d j o u r na l of m e dic i n e

early marker of AKI and are associated with kidney injury that does not respond to volume
worse outcomes (Table 1).5,10,13 One of the major expansion — that is, HRS-AKI. Renal vasocon-
limitations of the definition of HRS is the exclu- striction in patients with cirrhosis is not coun-
sion of underlying parenchymal kidney disease. tered by the release of vasodilatory substances
In an era marked by an increasing incidence (e.g., prostaglandins) because of a decrease in
of nonalcoholic steatohepatitis, the presence of their production and local release of vasocon-
underlying CKD due to diabetes mellitus, hyper- strictors such as endothelin.
tension, or both is also increasing. In patients The pathogenesis of renal vasoconstriction
with preexisting CKD, the prognostic and thera- and decreased renal blood flow in cirrhosis
peutic implications of HRS-AKI have yet to be (“hepatorenal physiology”) represents a continuum
determined, and these implications will proba- of the mechanisms that initially lead to ascites.
bly differ from those in patients with HRS-AKI Thus, patients with cirrhosis who have ascites,
that progresses to HRS-CKD.8,10 particularly those with refractory ascites, are at
the highest risk not only for the development of
AKI but also for its most severe clinical form,
Pathoph ysiol o gy
HRS-AKI (Fig. 1). Although HRS-AKI can occur
Patients with cirrhosis, particularly those with in the absence of a precipitating factor, it is more
ascites, have an increased susceptibility to AKI commonly precipitated by factors that cause a
because of the hemodynamic alterations that decrease in effective arterial blood volume. These
result from portal hypertension (Fig. 1).14,15 The factors include rapid fluid loss (e.g., excessive
initial mechanism in the pathogenesis of portal diuresis or gastrointestinal bleeding), worsening
hypertension is increased intrahepatic resistance vasodilatation induced by drugs (e.g., angiotensin-
due to distortion of the liver architecture (fibro- converting–enzyme inhibitors), and a systemic
sis and nodules) and an increase in intrahepatic inflammatory response (e.g., infection) (Fig. 1).
vascular tone. Subsequent activation of vasodila- Not all cases of AKI in patients with cirrhosis
tors in the splanchnic circulation (the most im- result from renal hypoperfusion; some cases
portant being nitric oxide) leads to progressive may result from structural kidney injury. How-
splanchnic and systemic vasodilatation. Increased ever, renal hypoperfusion may lead to structural
translocation of bacteria and bacterial products kidney injury when prolonged or when coupled
due to intestinal dysbiosis, bacterial overgrowth, with a “second hit,” such as exposure to nephro-
and altered tight-junction proteins contributes toxic medications, resulting in a delay in renal
further to vasodilatation, resulting in a reduc- recovery.
tion in the effective arterial blood volume that
will activate the neurohumoral systems (the re- A sse ssmen t of K idne y F unc t ion
nin–angiotensin–aldosterone, sympathetic, and
arginine–vasopressin systems) leading to sodium Evaluation of kidney function in patients with
and water retention and ascites formation.16,17 In cirrhosis remains critically important and a chal-
advanced stages of cirrhosis, progressive vasodi- lenging problem. The serum creatinine level,
latation leads to more avid retention of sodium which is the most practical and commonly used
and water, resulting in refractory ascites and marker of kidney function and the measure used
dilutional hyponatremia, respectively (Fig. 1). in the Model for End-Stage Liver Disease (MELD)
With progressive vasodilatation, vasoconstric- score to prioritize candidates for liver transplan-
tive systems (mainly renin and angiotensin) are tation, overestimates the GFR in patients with
activated, resulting in renal vasoconstriction and cirrhosis because of a combination of decreased
decreased renal blood flow. In addition, a rela- creatinine production due to liver disease, pro-
tive decrease in cardiac output in this high-out- tein calorie malnutrition, and muscle wasting.
put state of cardiac failure (so-called cirrhotic In addition, in patients with AKI and fluid over-
cardiomyopathy) may further contribute to de- load, an increase in the serum creatinine level
creased renal perfusion.18-20 The reduced renal can lag by several hours to days, despite a de-
blood flow results in a decrease in the glomeru- crease in the GFR.21,22 Exogenous clearance
lar filtration rate (GFR) and a prerenal type of markers such as inulin and iothalamate are not

736 n engl j med 388;8  nejm.org  February 23, 2023

The New England Journal of Medicine

Downloaded from nejm.org at UNIVERSIDAD NACIONAL AUTONOMA DE MEXICO on February 23, 2023. For personal use only. No other uses without permission.
Copyright © 2023 Massachusetts Medical Society. All rights reserved.
 Acute Kidney Injury in Patients with Cirrhosis

INTESTINAL LUMEN

Bacterial
overgrowth
Cirrhosis with ascites

Disruption of
tight junctions

Translocation Increased gut

of bacteria and permeability
bacterial products

CAPILLARY

Strategies to prevent Treatment strategies

all forms of AKI for HRS-AKI

Antibiotic prophylaxis after

Bleeding Bacterial ↑ Intrahepatic Liver transplantation
gastrointestinal bleeding
and SBP translocation resistance
Transjugular intrahepatic
portosystemic shunt

SBP, sepsis ↑ Proinflammatory Splanchnic and

Vasoconstrictors
cytokines systemic vasodilatation

Albumin with LVP and SBP; Infection, sepsis,

avoid vasodilators LVP, vasodilators
↑ Cardiac output
Cautious use of diuretics
Cirrhotic
(<500 g of weight loss per day) Bleeding, ↓ Effective arterial cardiomyopathy Albumin
and lactulose (adjust to yield volume depletion blood volume
2–3 bowel movements per day); (diuretics, diarrhea)
prevent variceal hemorrhage ↓ Cardiac output

Avoid nephrotoxic drugs; Activation of neurohumoral systems Renal-replacement

cautious use of IV contrast RAAS, sympathetic nervous system, therapy
arginine–vasopressin system
Kidney transplantation

Sodium ↑ Sodium and Renal

retention water retention vasoconstriction

Ascites Refractory ascites ↓ Renal

±hyponatremia blood flow

HRS-AKI

Figure 1. Pathophysiology of the Hepatorenal Syndrome and Acute Kidney Injury (HRS-AKI) in Patients with Cirrhosis.
A sine qua non for the development of HRS-AKI is the presence of ascites, which is often tense ascites and is often associated with hy-
ponatremia, a low mean arterial pressure, and oliguria. Factors that can precipitate AKI in a patient with cirrhosis (even without ascites)
or HRS-AKI are indicated by red arrows. Strategies to prevent all forms of AKI (including HRS-AKI) in patients with cirrhosis are shown.
IV denotes intravenous, LVP large-volume paracentesis, RAAS renin–angiotensin–aldosterone system, and SBP spontaneous bacterial
peritonitis.

n engl j med 388;8 nejm.org February 23, 2023 737

The New England Journal of Medicine
Downloaded from nejm.org at UNIVERSIDAD NACIONAL AUTONOMA DE MEXICO on February 23, 2023. For personal use only. No other uses without permission.
Copyright © 2023 Massachusetts Medical Society. All rights reserved.
 The n e w e ng l a n d j o u r na l of m e dic i n e

readily available and are confounded by changes clinically indicated (Fig. 2). Microscopical ex-
in the volume of distribution due to ascites and amination of urinary sediment is important to
extracellular volume expansion. Measurements rule out intrinsic causes of AKI.
of creatinine clearance with the use of timed Measurements of fractional excretion of so-
urinary collection (typically over 24 hours) are dium and urea have been used by clinicians to
subject to inaccuracy because of errors in uri- confirm hypovolemia and HRS-AKI and to rule
nary collection (i.e., incomplete collection or out acute tubular necrosis. However, patients
overcollection) or increased tubular secretion of with cirrhosis and ascites already have avid so-
creatinine as the GFR declines. dium retention, and fractional excretion of
Estimated GFR (eGFR) equations based on sodium of less than 1.0% is common in such
serum creatinine, cystatin C, or both are a sim- patients, even in the absence of AKI.28 Differen-
ple method of determining kidney function in tiating between HRS-AKI and acute tubular ne-
the general population of persons with stable crosis is therefore often challenging. In general,
serum creatinine levels. In patients with cirrho- cutoff values of less than 0.1% for fractional
sis, however, eGFR equations tend to overesti- excretion of sodium, less than 21% for frac-
mate the true GFR by 10 to 20 ml per minute per tional excretion of urea, and less than 44 mg per
1.73 m2 of body-surface area, especially in pa- deciliter for urinary albumin may help to con-
tients with a GFR of less than 40 ml per minute firm HRS-AKI and rule out acute tubular necro-
per 1.73 m2, ascites, or both and should be used sis.28-30 These cutoff values should be interpreted
with caution.23-25 The accuracy of eGFR measure- cautiously, however, and always in the context of
ments is particularly important in patients with the patient’s clinical presentation, since they are
cirrhosis because eGFR is one of the factors used not very sensitive or specific and have not been
to determine candidacy for simultaneous liver correlated with histologic findings. Urinary neu-
and kidney transplantation. The removal of race trophil gelatinase–associated lipocalin, a marker
from all eGFR equations was recently recom- of tubular injury, has been shown to differenti-
mended as an important step in efforts to ate between HRS-AKI and acute tubular necro-
eliminate disparities in the care of patients with sis. However, its use in clinical practice has been
kidney disease,26 and this removal has been ad- limited because of lack of standardization, un-
opted within the transplantation community; certainty regarding the cutoff value, and the
however, the effect of this change in eGFR equa- unavailability of the biomarker in many coun-
tions on candidacy for simultaneous liver and tries.28,31-34 Kidney biopsy should be considered
kidney transplantation remains to be adequately only if the results might change management
studied.27 (e.g., treatment of glomerular diseases), since
biopsy is an invasive procedure and may be as-
sociated with bleeding complications.
Di agnos t ic Wor k up
a nd M a nagemen t of A K I Volume expansion is central in reversing AKI
due to volume depletion, and the response to
Once AKI is diagnosed, it is important to dis- volume expansion will help to determine the
continue any medications that could have a role cause of AKI (Fig. 2). The type of resuscitation
in precipitating or worsening AKI — specifical- fluid (crystalloids vs. albumin) and the amount
ly, diuretics, vasodilators, nonselective beta- should be individualized according to the cause
blockers, nonsteroidal antiinflammatory drugs of AKI and the patient’s volume status (Fig. 2).
— and to rule out infection, particularly sponta- Volume resuscitation is recommended for a trial
neous bacterial peritonitis, which is frequently a period of 24 to 48 hours in patients who are
precipitant of AKI and HRS-AKI (Fig. 2). At the clinically hypovolemic or euvolemic. However, it
same time, a diagnostic workup for AKI is es- is important to exercise caution in administering
sential for treating early AKI and for preventing fluids in patients with AKI in order to avoid
progression, which is associated with increased fluid overload and pulmonary edema. Assess-
mortality.1,2,5 The cause of AKI is determined on ment of intravascular volume and volume re-
the basis of the patient’s history and physical sponsiveness is desirable but has been challeng-
examination, urinary tests, the response to di- ing, since many of the hemodynamic tools have
uretic withdrawal, and a volume challenge when not been studied in patients with cirrhosis and

738 n engl j med 388;8  nejm.org  February 23, 2023

The New England Journal of Medicine

Downloaded from nejm.org at UNIVERSIDAD NACIONAL AUTONOMA DE MEXICO on February 23, 2023. For personal use only. No other uses without permission.
Copyright © 2023 Massachusetts Medical Society. All rights reserved.
 Acute Kidney Injury in Patients with Cirrhosis

may be misleading because of increased intraab- dose can be increased, maintained, or discontin-
dominal pressures due to ascites.13 In a small, ued, according to the response (changes in the
single-center, retrospective study that assessed serum creatinine level) (Fig. 3). In a large ran-
volume status with the use of point-of-care ultra- domized, controlled trial showing the efficacy of
sonography in patients with a diagnosis of terlipressin in patients with HRS-AKI, which led
HRS-AKI who were considered to have adequate to its approval in the United States, the agent
volume repletion, 21% of the patients had intra- was associated with an increased incidence of
vascular fluid overload, and 28% continued to respiratory failure due to pulmonary edema.43
have intravascular volume depletion.35 A combi- Therefore, it is essential to withhold terlipressin
nation of assessments, including a careful history and albumin if there is clinical evidence of intra-
taking and physical examination, point-of-care vascular volume overload (i.e., anasarca, jugular
ultrasonography, and static and dynamic mea- venous distention, hypoxemia, pulmonary con-
surements, when available, should be used to gestion on a chest radiography, or elevated right
evaluate volume status and fluid responsiveness. ventricular systolic pressure on an echocardiog-
There is insufficient evidence supporting the raphy).
use of paracentesis to improve intraabdominal Small single-center studies have shown that
pressure (and theoretically, kidney function) in the responses to norepinephrine, an alternative
patients with cirrhosis and AKI. Partial ascites vasopressor, are similar to those to terlipressin.
removal is recommended for comfort in patients However, norepinephrine requires continuous
with tense ascites and should be accompanied by infusion in an ICU.39,40 The combination of octreo-
intravenous administration of albumin to pre- tide and midodrine has weak vasoconstrictive ac-
vent circulatory dysfunction, a vasodilatory state tivity, and a randomized, controlled trial showed
that follows large-volume paracentesis (removal that the combined treatment was inferior to
of >5 liters) and can lead to worsening kidney terlipressin in reversing HRS-AKI.44 Therefore,
function.36-38 octreotide and midodrine should be used only
temporarily, for 24 to 48 hours, and only if terli-
pressin is unavailable or contraindicated.
T r e atmen t of HR S-A K I
In patients with cirrhosis, HRS-AKI should be Transjugular Intrahepatic Portosystemic
treated only after other causes of AKI have been Shunt
investigated and excluded (Fig. 2). An important therapeutic option in patients with
portal hypertension is the placement of a trans­
Pharmacologic Therapy jugular intrahepatic portosystemic shunt (TIPS)
The mainstay of pharmacologic management of that may improve kidney function by redistribut-
HRS-AKI is the use of vasoconstrictors com- ing blood volume and reducing portal pres-
bined with intravenous albumin (Fig. 3).39,40 As sure.45-48 A meta-analysis of nine small studies
proof of concept, changes in the serum creati- suggested that treatment with TIPS led to a sig-
nine level correlate inversely with changes in nificant improvement in kidney function, with a
mean arterial pressure induced by vasoconstric- pooled response of 93% among patients with
tors.41 Vasoconstrictors have not been shown to HRS-AKI.49 However, there is currently insuffi-
improve survival, so their use should be consid- cient evidence to recommend TIPS for HRS-AKI.
ered a bridge to transplantation rather than a
cure for HRS-AKI. Renal-Replacement Therapy
Terlipressin, a vasopressin analogue, is the Renal-replacement therapy (also known as kid-
most common vasoconstrictor used worldwide, ney-replacement therapy) has been viewed as a
and both U.S.37 and European38 guidelines rec- bridge to liver transplantation, but renal-replace-
ommend it as a first-line agent for HRS-AKI. ment therapy for patients with HRS-AKI who are
Terlipressin can be administered intravenously not candidates for liver transplantation has been
as a bolus or as a continuous infusion, with controversial because of high mortality.50 How-
similar efficacy. However, the cumulative daily ever, because mortality associated with renal-
dose and the incidence of adverse events are replacement therapy has been shown to be simi-
lower with a continuous infusion.42 The initial lar among patients with HRS-AKI and those

n engl j med 388;8  nejm.org  February 23, 2023 739

The New England Journal of Medicine
Downloaded from nejm.org at UNIVERSIDAD NACIONAL AUTONOMA DE MEXICO on February 23, 2023. For personal use only. No other uses without permission.
Copyright © 2023 Massachusetts Medical Society. All rights reserved.
 The n e w e ng l a n d j o u r na l of m e dic i n e

Presentation with cirrhosis and AKI

Conduct history and physical examination: consider risk factors that may suggest cause of AKI
(e.g., recent shock, fever, recent initiation of diuretics, diarrhea)
Review medications list for active or recent nephrotoxin or exposure to IV contrast
Discontinue diuretics, NSAIDs, ACEI or ARB, NSBB
Perform diagnostic paracentesis to rule out SBP
Perform urinary tests: microscopical assessment of urinary sediment, FeNa, FeUrea,
NGAL (if available)

Renal Hypoperfusion (most common) Intrinsic or Parenchymal Kidney Injury Urinary Obstruction (least common)
Hypovolemia (e.g., from diuretics, ATN (most common; e.g., from sepsis, Urethral obstruction, ureteral obstruction
diarrhea, gastrointestinal bleeding) IV contrast, medications, bile cast
Hepatorenal syndrome (HRS-AKI) nephropathy)
AIN: from antibiotics, NSAIDs, PPI
Glomerular diseases: e.g., IgA, MPGN,
cryoglobulinemia

Assess volume status (physical exam- ATN: discontinue nephrotoxic agents Determine postvoid residual volume
ination, POCUS, chest radiography, AIN: discontinue offending agent Renal ultrasonography
echocardiography) Consider kidney biopsy
Glomerular disease: specific treatment

Intravascular volume Equivocal volume

depletion status

Crystalloid solutions 25% albumin 1g/kg

(≤100 g/day in divided
doses)

Measure serum creatinine level

after 24 hr

Unchanged or increasing serum

Serum creatinine level decreases creatinine level

Continue monitoring Reassess volume status

for renal recovery Volume expansion with 25%
albumin (1g/kg in divided doses)
if no volume overload

Measure serum creatinine level

after an additional 24 hr

Unchanged or increasing serum

Serum creatinine level decreases creatinine level

Continue monitoring Reassess causes of intrinsic AKI

for renal recovery Repeat urinary tests

ATN HRS-AKI

Administer diuretics if there is Administer vasoconstrictor

evidence of volume overload plus albumin
Renal-replacement therapy in
select patients

740 n engl j med 388;8  nejm.org  February 23, 2023

The New England Journal of Medicine

Downloaded from nejm.org at UNIVERSIDAD NACIONAL AUTONOMA DE MEXICO on February 23, 2023. For personal use only. No other uses without permission.
Copyright © 2023 Massachusetts Medical Society. All rights reserved.
 Acute Kidney Injury in Patients with Cirrhosis

Figure 2 (facing page). Workup and Management of AKI dysfunction before liver transplantation, since
in a Patient with Cirrhosis. recovery of kidney function is less likely in these
AKI is defined by a serum creatinine level that exceeds patients than in those with a shorter duration of
the baseline level by 0.3 mg per deciliter (26.5 mmol kidney dysfunction. However, predicting the se-
per liter) within 48 hours or a serum creatinine level verity and duration of kidney dysfunction that
that is at least 1.5 times the baseline level (an increase
make recovery unlikely remains challenging.6,58
that is known or presumed to have occurred within the
previous 7 days) or urinary output of less than 0.5 ml In 2017, the Organ Procurement and Transplan-
per kilogram of body weight per hour over a period of tation Network set forth criteria for simultane-
6 hours. Urinary tests are important in differentiating ous liver and kidney transplantation on the basis
between HRS-AKI and various causes of AKI that are of previous national guidelines. These criteria
due to intrinsic or parenchymal kidney injury. Findings
included factors such as prolonged duration of
suggestive of HRS-AKI include the following: normal
urinary sediment, fractional excretion of sodium (FeNa) AKI and dialysis and the presence of CKD. The
of less than 0.1%, and fractional excretion of urea (FeUrea) organization also implemented a safety-net pol-
of less than 21%. A urinary neutrophil gelatinase–asso- icy that gave priority to patients with persistent,
ciated lipocalin (NGAL) level greater than 220 to 244 μg severe kidney dysfunction after liver transplanta-
per gram of creatinine is suggestive of acute tubular ne-
tion to receive a kidney transplant within the
crosis (ATN). Volume overload is suggested by the pres-
ence of anasarca, jugular venous distention, a chest film first year.59,60 Kidney biopsy may assist in estab-
showing pulmonary congestion, or an elevated right ven- lishing the diagnosis and determining the re-
tricular systolic pressure. Diuretics may need to be initi- versibility of kidney dysfunction and the need for
ated or continued if there is evidence of volume overload. simultaneous liver and kidney transplantation.61
ACEI denotes angiotensin-converting–enzyme inhibitor,
Predictors of reversibility of AKI after transplan-
AIN acute interstitial nephritis, ARB angiotensin-recep-
tor blocker, MPGN membranoproliferative glomerulo- tation, such as biomarkers, are needed to guide
nephritis, NSAIDs nonsteroidal antiinflammatory drugs, the allocation of kidney grafts.62,63
NSBB nonselective beta-blocker, POCUS point-of-care
ultrasonography, and PPI proton-pump inhibitor.
Pr e v en t ion of A K I
The development of AKI is a common, yet severe
with cirrhosis and acute tubular necrosis,51 a complication in patients with cirrhosis. There-
trial of renal-replacement therapy in selected fore, it is imperative to recognize and manage
patients with HRS-AKI could be considered.13,37 events that may result in AKI, particularly in
There is no consensus on when renal-replace- patients with ascites (Fig. 1).64 Volume depletion
ment therapy should be initiated in patients with should be prevented through the careful use of
cirrhosis and AKI. Although several random- diuretics (with a goal of <1 lb [0.45 kg] of body-
ized, controlled trials have not shown a benefit weight loss per day), careful use of lactulose
of early initiation of renal-replacement therapy (with dose adjustment to yield two or three
in the general population of critically ill patients formed bowel movements a day), and prevention
in the ICU, patients with cirrhosis were excluded of variceal hemorrhage.65 Intravenous albumin
or were largely underrepresented in those stud- infusions at a dose of 4 to 6 g per liter of ascites
ies.52-55 Therefore, in patients with cirrhosis, the removed have been shown to ameliorate circula-
decision about when to initiate renal-replace- tory dysfunction and prevent AKI after large-
ment therapy should be individualized on the volume paracentesis (removal of >5 liters).66 Al-
basis of life-threatening indications that are re- bumin infusions have also been shown to reduce
fractory to medical treatment (e.g., hyperkale- the incidence of AKI and to decrease mortality
mia, acidosis, or fluid overload), uremic compli- among patients with spontaneous bacterial peri-
cations, the trajectory of kidney function, or the tonitis.67 However, in a study involving patients
overall prognosis.13,37,56,57 with infections other than spontaneous bacterial
peritonitis, intravenous albumin did not prevent
Liver Transplantation AKI and was actually associated with an in-
Liver transplantation is the treatment of choice creased incidence of pulmonary edema.68 The
in patients with HRS-AKI. Simultaneous liver long-term use of intravenous albumin infusions
and kidney transplantation is a potential thera- (weekly or every 2 weeks) in the outpatient set-
peutic option for patients with prolonged kidney ting has led to controversial results. An open-

n engl j med 388;8  nejm.org  February 23, 2023 741

The New England Journal of Medicine
Downloaded from nejm.org at UNIVERSIDAD NACIONAL AUTONOMA DE MEXICO on February 23, 2023. For personal use only. No other uses without permission.
Copyright © 2023 Massachusetts Medical Society. All rights reserved.
 The n e w e ng l a n d j o u r na l of m e dic i n e

Presentation with suspected HRS-AKI

Administer vasoconstrictor plus albumin (20–25 g/day)

Monitor for side effects before each bolus administration
or every 6 hr (if administered by infusion) and discontinue
if side effects present
Assess volume status daily

Terlipressin Terlipressin contraindicated or unavailable

FDA label
Norepinephrine
Administer 1 mg intravenously every 6 hr
as IV bolus
EASL guidelines
Administer 1 mg intravenously every 4–6 hr as
IV bolus or by continuous IV infusion at initial Administer 0.5–3.0 mg/hr (continuous IV infusion)
dose of 2 mg/day Titrate to achieve an increase in MAP >10 mm Hg

Measure serum creatinine level

on day 3 of treatment
(per EASL guidelines) or on day 4
of treatment (per FDA label)

Serum creatinine level ≥baseline level Serum creatinine level decreases

Per FDA label, if serum creatinine

level decreases by <30% from baseline,
increase dose of terlipressin to 2 mg
intravenously every 6 hr; per EASL
guidelines, if serum creatinine level
decreases by <25% from baseline,
increase dose in a stepwise manner to a
maximum of 12 mg/day

Continue vasoconstrictor
Development of complications
Ischemic complications
Respiratory failure
Volume overload HRS-AKI reversal
(two consecutive measurements
≥2 hr apart of serum creatinine of
≤1.5 mg/dl [per FDA] or serum
creatinine level <1.5 mg/dl [per EASL])

Discontinue vasoconstrictor Discontinue vasoconstrictor

Evaluate for renal-replacement therapy Maximum duration 14 days

label, randomized, controlled trial showed that AKI) and death,69 whereas a randomized, placebo-
this approach was associated with a reduced controlled trial showed no significant reductions
incidence of complications of ascites (including in the incidence of AKI or death with long-term

742 n engl j med 388;8  nejm.org  February 23, 2023

The New England Journal of Medicine

Downloaded from nejm.org at UNIVERSIDAD NACIONAL AUTONOMA DE MEXICO on February 23, 2023. For personal use only. No other uses without permission.
Copyright © 2023 Massachusetts Medical Society. All rights reserved.
 Acute Kidney Injury in Patients with Cirrhosis

Figure 3 (facing page). Management Algorithm lactam antibiotics and proton-pump inhibitors)
for Suspected HRS-AKI. cause allergic interstitial injury. Thus, kidney
The Food and Drug Administration (FDA) label and the function should be closely monitored in patients
European Association for the Study of the Liver (EASL)38 with cirrhosis and ascites while they are receiv-
guidelines recommend terlipressin as first-line therapy ing these medications. Data on the use of intra-
for HRS-AKI. Side effects include myocardial infarction,
venous contrast material in these patients are
peripheral or mesenteric ischemia, and pulmonary edema.
Albumin infusion should be decreased, and cautious use lacking. Cautious use of intravenous contrast
of terlipressin is recommended in patients with evidence material is recommended, especially in patients
of intravascular volume overload. Terlipressin should with an eGFR of less than 45 ml per minute per
be withheld if oxygen saturation (SpO2) is less than 90%. 1.73 m2.72
Norepinephrine use is limited to the intensive care unit
(ICU) and requires placement of a central catheter. It
may be considered as an alternative if terlipressin is C onclusions
contraindicated or unavailable. Side effects of norepi-
nephrine include ischemic events and cardiac dysrhyth- Advances over the past decade in the classifica-
mias. The combination of midodrine (7.5 to 15 mg orally tion, pathophysiological understanding, diagno-
three times a day) and octreotide (100 to 200 μg subcu-
sis, and management of AKI in patients with
taneously three times a day), given over a period of 24
to 48 hours, may be considered if terlipressin is unavail- cirrhosis will allow for earlier diagnosis and
able or contraindicated and a transfer to the ICU for nor- treatment of HRS-AKI. Since previous studies of
epinephrine infusion is not an option. Midodrine may the treatment of HRS-AKI used the old criterion
cause bradyarrhythmias. The baseline serum creatinine of a serum creatinine level exceeding 2.5 mg per
level is the level determined just before the initiation of
deciliter, it is possible that with the new defini-
vasoconstrictor therapy. MAP denotes mean arterial
pressure. tion, reversal of HRS-AKI with terlipressin and
albumin may occur more frequently and with
lower doses or for a shorter duration than re-
ported in recent randomized trials. However, the
albumin infusions.70 In a randomized, controlled extent to which the presence of systemic inflam-
trial involving hospitalized patients, an intrave- mation or underlying kidney parenchymal dam-
nous albumin infusion targeted at maintaining age limits the efficacy of treatment in patients
serum albumin levels at approximately 3 g per with HRS-AKI remains unknown. Further devel-
deciliter did not improve outcomes (one of opment of urinary biomarkers and their inclu-
which was the development of AKI) and was as- sion in the diagnostic or treatment algorithm
sociated with an increased incidence of pulmo- could potentially improve the differential diag-
nary edema.71 Therefore, long-term use of albu- nosis of AKI, guide vasoconstrictor therapy for
min in the outpatient or inpatient setting is HRS-AKI, and assist in predicting the reversibil-
currently not recommended. ity of AKI after liver transplantation. Additional
Various medications such as nonsteroidal investigations are needed to determine the
antiinflammatory drugs and renin–angiotensin– amount of albumin necessary for the prevention
aldosterone system blockers have direct nephro- and treatment of AKI and HRS-AKI and to sup-
toxic effects by impairing intrarenal blood flow. port wider use of point-of-care ultrasonography
Some agents (e.g., radiocontrast dye, aminogly- to guide fluid repletion.
cosides, vancomycin, and amphotericin B) have Disclosure forms provided by the authors are available with
direct renal tubule toxicity, and some (e.g., beta- the full text of this article at NEJM.org.

References
1. Tandon P, James MT, Abraldes JG, and complications in hospitalized patients with cirrhosis — a US population-based
Karvellas CJ, Ye F, Pannu N. Relevance of with cirrhosis. Hepatology 2013;​57:​753- study. J Hepatol 2020;​73:​1092-9.
new definitions to incidence and progno- 62. 5. Amathieu R, Al-Khafaji A, Sileanu
sis of acute kidney injury in hospitalized 3. Garcia-Tsao G, Parikh CR, Viola A. FE, et al. Significance of oliguria in crit-
patients with cirrhosis: a retrospective Acute kidney injury in cirrhosis. Hepatol- ically ill patients with chronic liver dis-
population-based cohort Study. PLoS One ogy 2008;​48:​2064-77. ease. Hepatology 2017;​66:​1592-600.
2016;​11(8):​e0160394. 4. Desai AP, Knapp SM, Orman ES, et al. 6. Nadim MK, Genyk YS, Tokin C, et al.
2. Belcher JM, Garcia-Tsao G, Sanyal AJ, Changing epidemiology and outcomes of Impact of the etiology of acute kidney in-
et al. Association of AKI with mortality acute kidney injury in hospitalized patients jury on outcomes following liver trans-

n engl j med 388;8  nejm.org  February 23, 2023 743

The New England Journal of Medicine
Downloaded from nejm.org at UNIVERSIDAD NACIONAL AUTONOMA DE MEXICO on February 23, 2023. For personal use only. No other uses without permission.
Copyright © 2023 Massachusetts Medical Society. All rights reserved.
 The n e w e ng l a n d j o u r na l of m e dic i n e

plantation: acute tubular necrosis versus et al. Fluid accumulation, recognition and Huggins JT. Point-of-care echocardiogra-
hepatorenal syndrome. Liver Transpl 2012;​ staging of acute kidney injury in critically- phy unveils misclassification of acute kid-
18:​539-48. ill patients. Crit Care 2010;​14(3):​R82. ney injury as hepatorenal syndrome. Am J
7. Salerno F, Gerbes A, Ginès P, Wong F, 22. Liu KD, Thompson BT, Ancukiewicz M, Nephrol 2019;​50:​204-11.
Arroyo V. Diagnosis, prevention and treat- et al. Acute kidney injury in patients with 36. Ruiz-del-Arbol L, Monescillo A, Ji-
ment of hepatorenal syndrome in cirrho- acute lung injury: impact of fluid accumu- menéz W, Garcia-Plaza A, Arroyo V, Rodés J.
sis. Gut 2007;​56:​1310-8. lation on classification of acute kidney Paracentesis-induced circulatory dysfunc-
8. Nadim MK, Kellum JA, Davenport A, injury and associated outcomes. Crit Care tion: mechanism and effect on hepatic
et al. Hepatorenal syndrome: the 8th In- Med 2011;​39:​2665-71. hemodynamics in cirrhosis. Gastroenter-
ternational Consensus Conference of the 23. Singapura P, Ma T-W, Sarmast N, et al. ology 1997;​113:​579-86.
Acute Dialysis Quality Initiative (ADQI) Estimating glomerular filtration rate in 37. Biggins SW, Angeli P, Garcia-Tsao G,
Group. Crit Care 2012;​16(1):​R23. cirrhosis using creatinine-based and cys- et al. Diagnosis, evaluation, and manage-
9. Angeli P, Ginès P, Wong F, et al. Diag- tatin C-based equations: systematic review ment of ascites, spontaneous bacterial
nosis and management of acute kidney and meta-analysis. Liver Transpl 2021;​27:​ peritonitis and hepatorenal syndrome:
injury in patients with cirrhosis: revised 1538-52. 2021 practice guidance by the American
consensus recommendations of the Inter- 24. Asrani SK, Jennings LW, Trotter JF, Association for the Study of Liver Diseas-
national Club of Ascites. J Hepatol 2015;​ et al. A model for glomerular filtration es. Hepatology 2021;​74:​1014-48.
62:​968-74. rate assessment in liver disease (GRAIL) 38. European Association for the Study of
10. Angeli P, Garcia-Tsao G, Nadim MK, in the presence of renal dysfunction. Hep- the Liver. EASL clinical practice guide-
Parikh CR. News in pathophysiology, defi- atology 2019;​69:​1219-30. lines for the management of patients with
nition and classification of hepatorenal 25. Asrani SK, Jennings LW, Kim WR, decompensated cirrhosis. J Hepatol 2018;​
syndrome: a step beyond the International et al. MELD-GRAIL-Na: glomerular filtra- 69:​406-60.
Club of Ascites (ICA) consensus document. tion rate and mortality on liver-transplant 39. Pitre T, Kiflen M, Helmeczi W, et al.
J Hepatol 2019;​71:​811-22. waiting list. Hepatology 2020;​71:​1766-74. The comparative effectiveness of vasoac-
11. Ronco C, Bellomo R, Kellum JA. Acute 26. Delgado C, Baweja M, Crews DC, et al. tive treatments for hepatorenal syndrome:
kidney injury. Lancet 2019;​394:​1949-64. A unifying approach for GFR estimation: a systematic review and network meta-
12. Kidney Disease: Improving Global recommendations of the NKF-ASN Task analysis. Crit Care Med 2022;​ 50:​
1419-
Outcomes (KDIGO). KDIGO clinical prac- Force on Reassessing the Inclusion of 29.
tice guideline for acute kidney injury. Kid- Race in Diagnosing Kidney Disease. Am J 40. Mohamed MMG, Rauf A, Adam A,
ney Int Suppl 2012;​2:​1-138. Kidney Dis 2022;​79(2):​268-88. Kheiri B, Lacasse A, El-Halawany H. Terli-
13. Nadim MK, Durand F, Kellum JA, et al. 27. Panchal S, Serper M, Bittermann T, pressin effect on hepatorenal syndrome:
Management of the critically ill patient Asrani SK, Goldberg DS, Mahmud N. Im- updated meta-analysis of randomized con-
with cirrhosis: a multidisciplinary per- pact of race-adjusted glomerular filtra- trolled trials. JGH Open 2021;​5:​896-901.
spective. J Hepatol 2016;​64:​717-35. tion rate estimation on eligibility for si- 41. Velez JC, Nietert PJ. Therapeutic re-
14. Francoz C, Durand F, Kahn JA, Genyk multaneous liver-kidney transplantation. sponse to vasoconstrictors in hepatorenal
YS, Nadim MK. Hepatorenal syndrome. Liver Transpl 2022;​28:​959-68. syndrome parallels increase in mean arte-
Clin J Am Soc Nephrol 2019;​14:​774-81. 28. Belcher JM, Sanyal AJ, Peixoto AJ, rial pressure: a pooled analysis of clinical
15. Schrier RW, Arroyo V, Bernardi M, Ep- et al. Kidney biomarkers and differential trials. Am J Kidney Dis 2011;​58:​928-38.
stein M, Henriksen JH, Rodés J. Peripheral diagnosis of patients with cirrhosis and 42. Cavallin M, Piano S, Romano A, et al.
arterial vasodilation hypothesis: a propos- acute kidney injury. Hepatology 2014;​60:​ Terlipressin given by continuous intrave-
al for the initiation of renal sodium and 622-32. nous infusion versus intravenous boluses
water retention in cirrhosis. Hepatology 29. Alsaad AA, Wadei HM. Fractional ex- in the treatment of hepatorenal syndrome:
1988;​8:​1151-7. cretion of sodium in hepatorenal syn- a randomized controlled study. Hepatol-
16. Traykova D, Schneider B, Chojkier M, drome: clinical and pathological correla- ogy 2016;​63:​983-92.
Buck M. Blood microbiome quantity and tion. World J Hepatol 2016;​8:​1497-501. 43. Wong F, Pappas SC, Curry MP, et al.
the hyperdynamic circulation in decom- 30. Patidar KR, Kang L, Bajaj JS, Carl D, Terlipressin plus albumin for the treat-
pensated cirrhotic patients. PLoS One Sanyal AJ. Fractional excretion of urea: ment of type 1 hepatorenal syndrome.
2017;​12(2):​e0169310. a simple tool for the differential diagno- N Engl J Med 2021;​384:​818-28.
17. Wiest R, Das S, Cadelina G, Garcia- sis of acute kidney injury in cirrhosis. 44. Cavallin M, Kamath PS, Merli M, et al.
Tsao G, Milstien S, Groszmann RJ. Bacte- Hepatology 2018;​68:​224-33. Terlipressin plus albumin versus mido-
rial translocation in cirrhotic rats stimu- 31. Huelin P, Solà E, Elia C, et al. Neutro- drine and octreotide plus albumin in the
lates eNOS-derived NO production and phil gelatinase-associated lipocalin for treatment of hepatorenal syndrome: a ran-
impairs mesenteric vascular contractility. assessment of acute kidney injury in cir- domized trial. Hepatology 2015;​62:​567-74.
J Clin Invest 1999;​104:​1223-33. rhosis: a prospective study. Hepatology 45. Charilaou P, Devani K, Petrosyan R,
18. Turco L, Garcia-Tsao G, Magnani I, 2019;​70:​319-33. Reddy C, Pyrsopoulos N. Inpatient mor-
et al. Cardiopulmonary hemodynamics 32. Asrani SK, Shankar N, da Graca B, tality benefit with transjugular intrahe-
and C-reactive protein as prognostic indi- Nadim MK, Cardenas A. Role of novel patic portosystemic shunt for hospital-
cators in compensated and decompensat- kidney biomarkers in patients with cir- ized hepatorenal syndrome patients. Dig
ed cirrhosis. J Hepatol 2018;​68:​949-58. rhosis and after liver transplantation. Dis Sci 2020;​65:​3378-88.
19. Danielsen KV, Wiese S, Busk T, et al. Liver Transpl 2022;​28:​466-82. 46. Ponzo P, Campion D, Rizzo M, et al.
Cardiovascular mapping in cirrhosis from 33. Allegretti AS, Parada XV, Endres P, Transjugular intrahepatic porto-systemic
the compensated stage to hepatorenal et al. Urinary NGAL as a diagnostic and shunt in cirrhotic patients with hepatore-
syndrome: a magnetic resonance study. prognostic marker for acute kidney injury nal syndrome — chronic kidney disease:
Am J Gastroenterol 2022;​117:​1269-78. in cirrhosis: a prospective study. Clin impact on renal function. Dig Liver Dis
20. Ruiz-del-Arbol L, Monescillo A, Aro- Transl Gastroenterol 2021;​12(5):​e00359. 2022;​54:​1101-8.
cena C, et al. Circulatory function and 34. Francoz C, Nadim MK, Durand F. 47. Boike JR, Thornburg BG, Asrani SK,
hepatorenal syndrome in cirrhosis. Hepa- Kidney biomarkers in cirrhosis. J Hepatol et al. North American practice-based rec-
tology 2005;​42:​439-47. 2016;​65:​809-24. ommendations for transjugular intrahe-
21. Macedo E, Bouchard J, Soroko SH, 35. Velez JCQ, Petkovich B, Karakala N, patic portosystemic shunts in portal hy-

744 n engl j med 388;8  nejm.org  February 23, 2023

The New England Journal of Medicine

Downloaded from nejm.org at UNIVERSIDAD NACIONAL AUTONOMA DE MEXICO on February 23, 2023. For personal use only. No other uses without permission.
Copyright © 2023 Massachusetts Medical Society. All rights reserved.
 Acute Kidney Injury in Patients with Cirrhosis

pertension. Clin Gastroenterol Hepatol mechanical fluid removal in critical ill- 65. de Franchis R, Bosch J, Garcia-Tsao G,
2022;​20(8):​1636.e36-1662.e36. ness. Br J Anaesth 2014;​113:​764-71. Reiberger T, Ripoll C. Baveno VII — re-
48. Allegretti AS, Ortiz G, Cui J, et al. 57. Bouchard J, Mehta RL. Timing of kid- newing consensus in portal hypertension.
Changes in kidney function after trans­ ney support therapy in acute kidney injury: J Hepatol 2022;​76:​959-74.
jugular intrahepatic portosystemic shunts what are we waiting for? Am J Kidney Dis 66. Bernardi M, Caraceni P, Navickis RJ,
versus large-volume paracentesis in cir- 2022;​79:​417-26. Wilkes MM. Albumin infusion in patients
rhosis: a matched cohort analysis. Am J 58. Levitsky J, Baker T, Ahya SN, et al. undergoing large-volume paracentesis:
Kidney Dis 2016;​68:​381-91. Outcomes and native renal recovery fol- a meta-analysis of randomized trials. Hep-
49. Song T, Rössle M, He F, Liu F, Guo X, lowing simultaneous liver-kidney trans- atology 2012;​55:​1172-81.
Qi X. Transjugular intrahepatic portosys- plantation. Am J Transplant 2012;​12:​2949- 67. Sort P, Navasa M, Arroyo V, et al. Ef-
temic shunt for hepatorenal syndrome: 57. fect of intravenous albumin on renal im-
a systematic review and meta-analysis. 59. Nadim MK, Sung RS, Davis CL, et al. pairment and mortality in patients with
Dig Liver Dis 2018;​50:​323-30. Simultaneous liver-kidney transplantation cirrhosis and spontaneous bacterial peri-
50. Martín-Llahí M, Guevara M, Torre A, summit: current state and future direc- tonitis. N Engl J Med 1999;​341:​403-9.
et al. Prognostic importance of the cause tions. Am J Transplant 2012;​12:​2901-8. 68. Thévenot T, Bureau C, Oberti F, et al.
of renal failure in patients with cirrho- 60. Formica RN, Aeder M, Boyle G, et al. Effect of albumin in cirrhotic patients
sis. Gastroenterology 2011;​140(2):​488.e4- Simultaneous liver-kidney allocation pol- with infection other than spontaneous
496.e4. icy: a proposal to optimize appropriate bacterial peritonitis: a randomized trial.
51. Allegretti AS, Parada XV, Eneanya ND, utilization of scarce resources. Am J Trans- J Hepatol 2015;​62:​822-30.
et al. Prognosis of patients with cirrhosis plant 2016;​16:​758-66. 69. Caraceni P, Riggio O, Angeli P, et al.
and aki who initiate RRT. Clin J Am Soc 61. Pichler RH, Huskey J, Kowalewska J, Long-term albumin administration in de-
Nephrol 2018;​13:​16-25. et al. Kidney biopsies may help predict compensated cirrhosis (ANSWER): an
52. Zarbock A, Kellum JA, Schmidt C, et al. renal function after liver transplantation. open-label randomised trial. Lancet 2018;​
Effect of early vs delayed initiation of re- Transplantation 2016;​100:​2122-8. 391:​2417-29.
nal replacement therapy on mortality in 62. Levitsky J, Asrani SK, Klintmalm G, 70. Solà E, Solé C, Simón-Talero M, et al.
critically ill patients with acute kidney et al. Discovery and validation of a bio- Midodrine and albumin for prevention of
injury: the ELAIN randomized clinical marker model (PRESERVE) predictive of complications in patients with cirrhosis
trial. JAMA 2016;​315:​2190-9. renal outcomes after liver transplanta- awaiting liver transplantation: a random-
53. Gaudry S, Hajage D, Schortgen F, et al. tion. Hepatology 2020;​71:​1775-86. ized placebo-controlled trial. J Hepatol
Initiation strategies for renal-replacement 63. Levitsky J, Asrani SK, Abecassis M, 2018;​69:​1250-9.
therapy in the intensive care unit. N Engl Ruiz R, Jennings LW, Klintmalm G. Ex- 71. China L, Freemantle N, Forrest E, et al.
J Med 2016;​375:​122-33. ternal validation of a pretransplant bio- A randomized trial of albumin infusions
54. Barbar SD, Clere-Jehl R, Bourredjem A, marker model (REVERSE) predictive of in hospitalized patients with cirrhosis.
et al. Timing of renal-replacement therapy renal recovery after liver transplantation. N Engl J Med 2021;​384:​808-17.
in patients with acute kidney injury and Hepatology 2019;​70:​1349-59. 72. Davenport MS, Perazella MA, Yee J,
sepsis. N Engl J Med 2018;​379:​1431-42. 64. O’Leary JGO, Levitsky J, Wong F, et al. Use of intravenous iodinated con-
55. Wald R, Bagshaw SM. Timing of ini- Nadim MK, Charlton M, Kim WR. Pro- trast media in patients with kidney dis-
tiation of renal-replacement therapy in tecting the kidney in liver transplant can- ease: consensus statements from the
acute kidney injury. N Engl J Med 2020;​ didates: practice-based recommendations American College of Radiology and the
383:​1797-8. from the American Society of Transplan- National Kidney Foundation. Kidney Med
56. Rosner MH, Ostermann M, Murugan tation Liver and Intestine Community of 2020;​2:​85-93.
R, et al. Indications and management of Practice. Am J Transplant 2016;​16:​2516-31. Copyright © 2023 Massachusetts Medical Society.

n engl j med 388;8  nejm.org  February 23, 2023 745

The New England Journal of Medicine
Downloaded from nejm.org at UNIVERSIDAD NACIONAL AUTONOMA DE MEXICO on February 23, 2023. For personal use only. No other uses without permission.
Copyright © 2023 Massachusetts Medical Society. All rights reserved.