Hepatology, VOL. 0, NO.

0, 2019 

Terlipressin Is Superior to Noradrenaline

in the Management of Acute Kidney Injury
in Acute on Chronic Liver Failure
Vinod Arora,1 Rakhi Maiwall,1 Vijayaraghavan Rajan,1 Ankur Jindal,1 Saggere Muralikrishna Shasthry,1 Guresh Kumar,2
Priyanka Jain,2 and Shiv Kumar Sarin1

Hepatorenal syndrome (HRS) carries a high short-term mortality in patients with cirrhosis and acute on chronic
liver failure (ACLF). Terlipressin and noradrenaline are routinely used in cirrhosis with HRS and have been found
to be equally effective. There are no data comparing the efficacy of terlipressin with noradrenaline in ACLF pa-
tients with HRS. In an open-label, randomized controlled trial (RCT), consecutive patients with ACLF diagnosed
with HRS acute kidney injury (AKI) were randomized to albumin with infusion of terlipressin (2-12 mg/day;
n = 60) or noradrenaline (0.5-3.0 mg/h; n = 60). Response to treatment, course of AKI, and outcome were studied.
Baseline characteristics, including AKI stage and sepsis-related HRS-AKI, were comparable between groups.
Compared to noradrenaline, terlipressin achieved greater day 4 (26.1% vs. 11.7%; P = 0.03) and day 7 (41.7% vs.
20%; P = 0.01) response. Reversal of HRS was also better with terlipressin (40% vs. 16.7%; P = 0.004), with a sig-
nificant reduction in the requirement of renal replacement therapy (RRT; 56.6% vs. 80%; P = 0.006) and improved
28-day survival (48.3% vs. 20%; P = 0.001). Adverse events limiting use of drugs were higher with terlipressin than
noradrenaline (23.3% vs. 8.3%; P = 0.02), but were reversible. On multivariate analysis, high Model for End-Stage
Liver Disease (MELD; odds ratio [OR], 1.10; confidence interval [CI] = 1.009-1.20; P = 0.03) and noradrenaline
compared to terlipressin (OR, 3.05; CI = 1.27-7.33; P = 0.01) predicted nonresponse to therapy. Use of noradrena-
line compared to terlipressin was also predictive of higher mortality (hazard ratio [HR], 2.08; CI = 1.32-3.30;
P = 0.002). Conclusion: AKI in ACLF carries a high mortality. Infusion of terlipressin gives earlier and higher
response than noradrenaline, with improved survival in ACLF patients with HRS-AKI. (Hepatology
­
2019;0:1-11).

A
cute on chronic liver failure (ACLF) is a dis-
tinct entity where, because of severe acute
hepatic injury, a rapid loss of liver function
develops in a patient with previous chronic liver dis-
ease.(1) These patients have severe hepatic dysfunc-
tion, and outcome is defined by functional hepatic
reserve and extent of extrahepatic organ failures.(2)
Renal failure is a frequent extrahepatic organ fail-
ure, and its presence is an independent prognostic
marker for mortality.(3) The pathophysiological basis
of renal dysfunction in patients with ACLF is differ-
ent compared to those with decompensated cirrhosis
(DC).(4) Systemic inflammation is the hallmark of
ACLF, characterized by a cytokine storm wherein
there is an increase in both pro- and anti-inflam-
matory cytokines, such as interleukin (IL)-6, IL-8,
IL-1β, and IL-10, leading to circulatory dysfunction
and organ failures.(5) These patients therefore have

Abbreviations: ACLF, Acute on chronic liver failure; AKI, acute kidney injury; CI, confidence interval; CVP, central venous pressure; DC,
decompensated cirrhosis; HE, hepatic encephalopathy; HR, hazard ratio; HRS, hepatorenal syndrome; ICA, International Club of Ascites; IL,
interleukin; IQR, interquartile range; IVC, inferior vena cava; ITT, intention-to-treat; MAP, mean arterial pressure; MELD, Model for
End-Stage Liver Disease; OR, odds ratio; PPA, per-protocol analysis; RCT, randomized controlled trial; RRT, renal replacement therapy; SBP,
spontaneous bacterial peritonitis; sCr, serum creatinine.
Received January 18, 2018; accepted August 1, 2018.
Additional Supporting Information may be found at onlinelibrary.wiley.com/doi/10.1002/hep.30208/suppinfo.
© 2018 by the American Association for the Study of Liver Diseases.
View this article online at wileyonlinelibrary.com.
DOI 10.1002/hep.30208
Potential conflict of interest: Nothing to report.

1
ARORA ET AL.
a higher incidence and progression of acute kidney
injury (AKI), as well as more-frequent structural kid-
ney damage, which is also associated with worse out-
comes.(4,6) Postmortem kidney biopsy data have also
shown the presence of bile cast nephropathy related to
bilirubin toxicity in these patients.(7) Probably because
of similar reasons, hepatorenal syndrome (HRS) in
patients with ACLF has also been shown to respond
poorly to vasoconstrictors.(4,5,8)
Studies done in DC and HRS have shown equal
efficacy of the two commonly used vasoconstrictors,
terlipressin and noradrenaline.(9) However, these stud-
ies were done in the era wherein HRS was diagnosed
based on the absolute value of serum creatinine (sCr).
Considering the numerous fallacies in the measure-
ment of sCr, the International Club of Ascites (ICA)
has recently proposed new diagnostic criteria for AKI
for the diagnosis of hepatorenal syndrome (HRS-AKI)
in patients with cirrhosis.(10) Recent studies have also
shown that using continuous infusion, a significantly
lower dosage of terlipressin is needed, with better
tolerability and fewer adverse events, as compared to
bolus dosing in HRS-AKI, though treatment response
rate was comparable between groups.(11) Considering
a different pathophysiological basis of HRS-AKI in
patients with ACLF and lack of studies comparing
continuous infusion of terlipressin with noradrenaline,
especially in the context of revised diagnostic crite-
ria for HRS-AKI, we initiated this randomized con-
trolled trial (RCT) to compare the efficacy of infusion
of these two vasoconstrictors in the management of
HRS-AKI in patients with ACLF.
Patients and Methods
The study was conducted as a randomized
open-label trial in the Department of Hepatology,
ARTICLE INFORMATION:
From the 1 Department of Hepatology, Institute of Liver and Biliary Sciences, New Delhi, India; 2 Department of Clinical Research
and Biostatistics, Institute of Liver and Biliary Sciences, New Delhi, India.
ADDRESS CORRESPONDENCE AND REPRINT REQUESTS TO:
Shiv Kumar Sarin, M.D., D.M., D.Sc. (Hony.)
Department of Hepatology
Institute of Liver and Biliary Sciences
D-1, Acharya Shree Tulsi Marg
2
Hepatology,  Month 2019
Institute of Liver and Biliary Sciences Hospital
(New Delhi, India) from October 2015 to December
2016. The study included patients with a diagnosis
of ACLF, defined as an acute hepatic insult mani-
festing as jaundice (serum bilirubin ≥5 mg/dL) and
coagulopathy (international normalized ratio [INR]
≥1.5) complicated within 4 weeks by ascites and/or
encephalopathy in a patient with previously diag-
nosed or undiagnosed chronic liver disease or cirrho-
sis and is associated with high 28-day mortality.(12)
All authors had access to the study data and reviewed
and approved the final manuscript. AKI was defined
as per the ICA-AKI criteria.(13) During this period,
340 ACLF patients were screened; 120 who met the
ICA-AKI criteria either at admission or during the
course of hospitalization were randomized to either
terlipressin or noradrenaline. Exclusion criteria were
age <18 years, DC, patients on renal replacement
therapy (RRT), renal or liver transplantation, history
of coronary artery disease, ischemic cardiomyopathy,
ventricular arrhythmia, peripheral vascular disease,
chronic kidney disease, obstructive uropathy, or lack
of informed consent.
This single-center study protocol conformed to
the Declaration of Helsinki and was approved by
the Institutional Ethics Committee vide letter no.
F25/5/80/ILBS/AC/2015/627. The study followed
the Consolidated Standards of Reporting Trials
(CONSORT) guidelines for randomized trials and
was registered at the ClinicalTrials.gov with full pro-
tocol access (identifier: NCT02573727). A written
informed consent was taken at enrollment.
RANDOMIZATION
Block randomization was done with a block size of
12. Allocation concealment was done by the sequen-
tially numbered opaque sealed envelopes technique.
Vasant Kunj, New Delhi
110070, India
E-mail: sksarin@ilbs.in
Tel: +91 11 4630 0000
Hepatology, Vol. 0,  No. 0,  2019
Patients were randomized to receive either terlipressin
or noradrenaline in a 1:1 ratio.
BASELINE EVALUATION
All patients were subjected to detailed history,
examination, routine hematological tests, and kidney
and liver function tests at admission. History regard-
ing use of diuretics, hepatotoxic drugs, and nephro-
toxic drugs was recorded.
Hepatic encephalopathy (HE) was defined as per
West-Haven’s criteria(14) and spontaneous bacterial
peritonitis (SBP) as ascitic fluid absolute neutrophil
count >250 cells/mm3.(15) Acute variceal bleed was
diagnosed as per Asian Pacific Association for the Study
of Liver (APASL) criteria.(16) Severity of liver failure
was assessed by Child-Turcotte-Pugh (CTP) score and
Model for End-Stage Liver Disease (MELD) score.
PRIMARY AND SECONDARY
OBJECTIVES
The primary objective of the study was to compare
reversal of HRS-AKI at day 14, and secondary objec-
tives were to compare early response of noradrenaline
and terlipressin at days 4 and 7 and assess 28-day sur-
vival. We also aimed to determine the predictors of
survival and factors predicting response to terlipressin
or noradrenaline and the adverse events of drugs.
STUDY DEFINITIONS
Baseline creatinine was defined as value of sCr
obtained in the previous 3 months, when available. In
patients with more than one value within the previous
3 months, the value closest to admission time to the
hospital was used. In patients without a previous sCr
value, sCr on admission was used as baseline.
HRS-AKI was defined as ICA-AKI stage ≥II
when other causes of AKI were excluded and the
patient was nonresponsive to volume expansion with
intravenous albumin(10) (Supporting Box S1). Sepsis-
related HRS-AKI was defined as AKI meeting the
criteria for HRS with presence of microbiologically
proven bacterial infection or SBP.
Staging of AKI and response to treatment were
defined as per the ICA-AKI criteria.(10) Response
was assessed every 48 hours.(11) Early response was
defined at days 4 and 7, and reversal of HRS AKI was
defined at day 14.
ARORA ET AL.
Complete and partial response was defined as
return of sCr to a value within 0.3 mg/dL of baseline
and regression of AKI stage with a reduction of sCr to
≥0.3 mg/dL above baseline, respectively. No response
was defined as no regression of AKI.(10,13)
Urine routine microscopy, urine sodium (Na), and
blood urea nitrogen to creatinine ratio were used to
distinguish between prerenal azotemia and renal
causes like acute tubular necrosis.
Standard Management
This included stoppage of diuretics and lactulose
and introduction of standard medical and nutritional
therapy for ACLF as per our hospital protocol.
VOLUME EXPANSION
Plasma expansion was done with albumin for the
initial 2 days (1 g/kg/day dose titrated as per signs
of volume overload, i.e., presence of basal crepitations,
chest x-ray suggestive of volume overload, or ultra-
sound chest showing B-lines; and where central venous
pressure [CVP] was placed, the target was to maintain
CVP ≤15 cm of H20). When CVP was not placed,
the inferior vena cava (IVC) was used for assessment
of volume status with IVC diameter measured 3 cm
from the right atrium greater than 22 mm with col-
lapsibility <50% indicating volume overload. Following
plasma expansion and classification into HRS-AKI,
patients were randomized to receive either terlipres-
sin or noradrenaline. Daily albumin (20-40 g/day)
was given in both groups until the end of reversal of
HRS-AKI or evidence of volume overload (CVP >18
cm of H20 or IVC >22 mm) or requirement of RRT.
Both groups received daily intravenous albumin 20-40
g/day until the end of the study period.
DRUG ADMINISTRATION
Dosage of terlipressin and noradrenaline was
decided as in the studies comparing these two drugs
in management of HRS-AKI.(11,17) Terlipressin was
given as a continuous infusion started at the dosage of
2 mg/24 h,(11,18) and response was assessed every 48
hours and the dosage of terlipressin was modified in
line with studies in which dosage of terlipressin was
doubled every 48 hours in case of nonresponse (<25%
of pretreatment value) to the maximum dosage of 12
mg/24 h.(11,18) Noradrenaline therapy was given as a
3
ARORA ET AL. Hepatology,  Month 2019

continuous intravenous infusion starting at 0.5 mg/h clinical and biochemical indices are expressed either
with doubling of dose after every 4 hours designed to as mean ± SD or median (interquartile range; IQR).
achieve an increase in mean arterial pressure (MAP) Frequencies (percentages) were used to study qualita-
of at least 10 mm Hg or an increase in 4-hour urine tive variables. To compare the difference of means and
output >200 mL.(19,20) When one of these goals was medians, the independent Student t test and Mann-
not achieved, noradrenaline dose was increased every Whitney U test, respectively, were used. Pearson’s
4 hours in steps of 0.5 mg/h, up to a maximum dose χ2 tests were used for comparisons between rates of
of 3 mg/h. complications, mortality, and sex. Survival curves were
described by means of the Kaplan-Meier method.
We performed uni- and multivariate Cox and binary
DIALYSIS PROTOCOL
logistic regression analysis to determine the predictors
Dialysis was done for patients with failure of stan- of mortality and treatment response, respectively. A P
dard medical management with complications such value <0.05 was considered statistically significant.
as presence of features of volume overload nonre-
sponsive to intravenous furesmide, metabolic acidosis,
refractory hyperkalemia, and symptoms of uremia, like
pericardial rub or recurrent vomiting.(3,8) Slow low-­ Results
efficiency dialysis (SLED) was the predominant mode Of 340 ACLF patients screened, 181 (53.2%) had
of dialysis whereas continuous RRT was considered AKI at admission or during the course of stay in the
for hemodynamically unstable patients. hospital, and 120 met the inclusion criteria (Fig. 1). On
per-protocol analysis (PPA), 51 patients were included
ADVERSE EVENTS in the terlipressin and 55 in the noradrenaline arm.
Of these 340 patients, prevalence of AKI at admission
Adverse events were defined as “any untoward was in 26.2% (89 of 340), and the remaining patients
medical occurrence in a patient or clinical investiga- developed AKI during their stay in the hospital.
tion subject administered a pharmaceutical product, Baseline demographic characteristics of both
which does not necessarily have to have a causal rela- patient groups were comparable (Table 1). Alcohol
tionship with this treatment.” remained the most common acute and chronic insult,
Serious adverse events were defined on the basis of followed by reactivation of hepatitis B. AKI stage
previous studies(11) (Supporting Box S2). II and III were equally distributed between study
In patients with nonserious adverse events, dose groups. Bacterial-infection–associated HRS was
reduction and monitoring for adverse events were noted in 58 (48.3%) of 120 patients. Of these, pneu-
done. In patients who had serious adverse events, res- monia was the most common infection, observed
cue therapy was considered. The latter included injec- in 31 (25.33%), followed by SBP in 19 (15.8%;
tion of octreotide given as 100-200 μg/8 hourly and Table 1). None of the patients underwent transplan-
midodrine at a dose of 2.5-12.5 mg/8 h to maintain tation during the study period. Because the patients
MAP >75 mm Hg. had AKI at time of presentation, they were steroid
ineligible.
STATISTICAL ANALYSIS
PRIMARY ENDPOINT
This study was a single-center, prospective RCT.
All authors had access to the study data and reviewed Median dose of intravenous albumin in both ter-
and approved the final manuscript. Assuming a lipressin and noradrenaline groups was 60 g/day (60
response rate in terlipressin as 56% and in noradren- [IQR, 60-80] vs. 60 [IQR, 60-75] g/day; P = 0.81).
aline as 43%, and a 10% noninferiority margin, with This dose of 60 g/day is also related to the body
an alpha error of 5% and power of 80%, we needed to weight of our patients.
enroll 57 cases in each arm.(21,22) Mean dose of terlipressin and noradrenaline was
Statistical analyses were performed using SPSS soft- 2.02 ± 0.70 mg/day and 1.11 ± 0.40 mg/h, respec-
ware (version 20.0; IBM Corp., Armonk, NY). Patient tively. MAP achieved was comparable in both arms

4
Hepatology, Vol. 0,  No. 0,  2019
FIG. 1. CONSORT chart.
(82.48 ± 3.96 vs. 81.70 ± 4.74 mm Hg; P = 0.33). On
intention-to-treat (ITT) analysis, the primary end-
point of reversal of HRS at day 14 was noted in 24
(40%) of 60 patients in the terlipressin group com-
pared to 10 (16.7%) of 60 in the noradrenaline group
(P = 0.004; Fig. 2). Similarly, on PPA, reversal of HRS
was noted in 22 of 51 (43.13%) patients in the terli-
pressin arm compared to 9 of 55 (16.3%) in the nor-
adrenaline arm (P = 0.002).
SECONDARY ENDPOINTS
Early complete response at day 4 was noted in 11
patients (18.3%) in the terlipressin arm, but none in
the noradrenaline arm. At day 7, complete response
was noted in 21 (35%) versus 5 (8.3%) patients
ARORA ET AL.
(P < 0.001) in the terlipressin compared to the nor-
adrenaline group (Table 2). Overall, higher early and
significant response at day 7 (partial and complete)
was noted in the terlipressin group 25 (41.7%) ver-
sus the noradrenaline group 12 (20%), respectively
(P = 0.01; Fig. 2).
CHANGE IN URINE FLOW
RATE OVER PERIOD OF 7 DAYS
WITH TERLIPRESSIN AND
NORADRENALINE
Compared to noradrenaline, terlipressin achieved
a 2-fold increase in urine/h from baseline. However,
both drugs caused a significant increase in urine flow
in patients with AKI-HRS from baseline. In the
5
ARORA ET AL. Hepatology,  Month 2019

TABLE 1. Baseline Characteristics of the Study Groups

Terlipressin Noradrenaline
Variable (n = 60) (n = 60) P Value
Demographic details
Age (years) 40.26 ± 6.25 38.80 ± 6.95 0.24
Sex distribution (% males) 58/60 (96) 55/60 (92) 0.44
Etiology (%)
Alcohol 44/60 (73) 43/60 (71.7) 0.83
Reactivation of hepatitis B 5/60 (8.3) 7/60 (11.7) 0.54
Drug-induced liver injury 4/60 (6.7) 5/60 (8.3) 1.00
HEV infection 4/60 (6.7) 2/60 (3.3) 0.68
Wilson’s flare 1/60 (1.7) 0 —
Autoimmune hepatitis flare 2/60 (3.3) 3/60 (5) 1.00
Stage of AKI (%)
II 31/60 (51.6) 32/60 (53.3) 0.85
III 29/60 (48.3) 28/60 (46.66)
Distribution of infection (%)
SBP 10/60 (16.6) 9/60 (15) 0.78
Pneumonia 14/60 (20) 17/60 (28.3) 0.53
Urinary tract infection 1/60 (1.6) 4/60 (6.6) 0.36
Cellulitis 1/60 (1.6) 2/60 (3.3) 1.00
Clinical/biochemical parameters
Hemoglobin (g/dL) 9.51 ± 2.31 9.03 ± 2.33 0.26
TLC (per cmm) 16.10 (8, 29.6) 15.6 (6, 39) 0.46
Platelets (per cmm) 161 (30.7, 262) 129 (24, 379) 0.98
Total bilirubin (mg/dL) 22.15 ± 9.71 22.96 ± 8.44 0.63
Aspartate transaminases (IU/mL) 130 (80, 726) 116 (37, 224) 0.10
Alanine transaminase (IU/mL) 54 (20, 710) 55 (31, 308) 0.68
Alb (g/dL) 2.13 ± 0.53 2.14 ± 0.43 0.91
Blood urea (mg/dL) 80 (9, 205) 68 (8, 160) 0.17
Serum creatinine (mg/dL) (at admission) 1.71 (0.01, 4.71) 1.69 (0.04, 5.82) 0.71
Serum creatinine (at randomization) 1.79 (1.09, 5.30) 2.02 (1.02-5.10) 0.21
Peak creatinine (mg/dL) 2.68 ± 1.22 2.988 ± 1.090 0.14
Sodium (Na [Meq/L]) 130.34 ± 7.14 129.87 ± 8.15 0.74
Potassium (K [Meq/L]) 4.20 ± 1.03 3.98 ± 1.00 0.24
INR 2.80 ± 0.64 2.79 ± 0.726 0.94
MAP (mm Hg) at baseline 68.08 ± 4.62 67.85 ± 4.20 0.76
Urine sodium (Meq/L) 16.48 ± 3.81 16.24 ± 4.24 0.74
CTP 11.01 ± 1.06 11.07 ± 1.10 0.77
MELD 33.27 ± 4.98 33.75 ± 5.00 0.60
Baseline urine flow rate (mL/h) 24.34 ± 2.02 23.64 ± 2.68 0.11

Abbreviations: cmm, cubic millimeter; TLC, total leucocyte count.

terlipressin arm, urine flow rate increased from 24.34
± 2.02 to 47.21 ± 2.48 mL/h (P < 0.001) and in the
noradrenaline group from 23.68 ± 2.68 to 35.46 ±
2.97 mL/h (P < 0.001).
REQUIREMENT OF DIALYSIS
A total of 82 of 120 (68.3%) patients required
RRT at a median of 4 {2, 21} days. The predominant
modality used was SLED, in 82.9% (68 of 82) of
patients and continuous RRT in the remaining 17.07%
(14 of 82). Indications of dialysis included severe met-
abolic acidosis (51.2%), hyperkalemia (36.6%), symp-
toms of uremia (31.7%), and volume overload (30.5%).
A small proportion of the latter group of patients
received a single dose of furosemide to look for diuretic
responsiveness; however, none of the patients showed a
response.

6
Hepatology, Vol. 0,  No. 0,  2019
FIG. 2. Rate of response (complete and partial) in patients who
were randomized to terlipressin and noradrenaline arms.
Dialysis was required in 34 patients (56.66%) in
the terlipressin arm compared to 48 (80%) in the nor-
adrenaline arm (P = 0.006). Time to dialysis was not
different in terlipressin (6.63 ± 3.29 days) and nor-
adrenaline (5.65 ± 4.06 days; P = 0.15), and, similarly,
median numbers of sessions were comparable (4(1,10);
3(1,11); P = 0.34).
Of the patients who underwent dialysis, 14- and
28-day mortality in the terlipressin and noradrena-
line arms were 35% (21 of 34) and 65% (39 of 48;
P = 0.05) and 91.2% (31 of 34) and 100% (48 of 48;
P = 0.07), respectively.
SURVIVAL ANALYSIS
Of the 120 ACLF patients, 79 (66.7%) died.
Overall survival was significantly better (29 of 60;
48.3%) for patients in the terlipressin as compared to
the noradrenaline group (12 of 60; 20%; P = 0.001) on
ITT analysis. Probability of short-term survival at 14
and 28 days was also significantly higher in the ter-
lipressin group (Fig. 3A). On PPA, terlipressin had a
survival advantage over noradrenaline, given that 25 of
51 (49.01%) patients survived in the terlipressin arm
compared to 9 of 55 (18.1%) in the noradrenaline arm
(P < 0.001; Fig. 3B).
TABLE 2. Secondary Outcomes (Early Response at Days 4 and 7)
Variable
Response day 4 (partial response; complete response)
Response day 7 (partial response; complete response)
ARORA ET AL.
PREDICTORS OF TREATMENT NO
RESPONSE
On univariate analysis, noradrenaline compared to
terlipressin (odds ratio [OR], 3.33; confidence interval
[CI], 1.42-7.82; P = 0.006), HE (OR, 2.69; CI = 1.09-
6.62; P = 0.03), and high MELD score (OR, 1.13;
CI = 1.03-1.23; P = 0.005) were predictors of nonre-
sponse. However, using multivariate logistic regression,
high MELD (OR, 1.10; CI = 1.009-1.20; P = 0.03)
and use of noradrenaline compared to terlipressin
(OR, 3.05; CI = 1.27-7.33; p = 0.01) were independent
­predictors of nonresponse in HRS-AKI (Table 3).
PREDICTORS OF MORTALITY
On univariate analysis, MELD (hazard ratio [HR],
1.08; CI = 1.03-1.14; P = 0.001), HE (HR, 2.06;
CI = 1.32-3.21; P = 0.001), pneumonia (HR, 1.56;
CI = 1.92-14.68; P = 0.001), and use of noradrenaline
compared to terlipressin (HR, 2.32; CI = 1.47-3.66;
P < 0.001) were significant predictors of 28-day mor-
tality. However, on multivariate Cox regression, high
MELD score (HR, 1.08; CI = 1.02-1.13; P = 0.003),
HE (HR, 1.88; CI = 1.20-2.94; P = 0.002), and use
of noradrenaline compared to terlipressin (HR, 2.08;
CI = 1.32-3.30; P = 0.002) were independent predic-
tors of high 28-day mortality (Table 3).
RESPONSE TO TERLIPRESSIN
AND NORADRENALINE IN SEPSIS
AND NON-SEPSIS-RELATED AKI
Sepsis was observed in 48.3% (58 of 120) of the
included patients with HRS-AKI. Patients with
­sepsis-related HRS-AKI had a lower response rate to
therapy (13 of 58; 22.41%) compared to non-­sepsis-
related AKI (21 of 62; 33.87%), though the differ-
ence was not significant (P = 0.17). Survival was also
comparable in sepsis and non-sepsis-related HRS-
AKI (Supporting Fig. S1). On subgroup analysis in
Terlipressin Noradrenaline
(N = 60) (N = 60) P Value
16/60 (26.7%) 7/60 (11.7%) 0.03
(5;11) (7;0)
25/60 (41.7%) 12/60 (20%) 0.01
(4;21) (7:5)
7
ARORA ET AL. Hepatology,  Month 2019

TABLE 3. Table Depicting Predictor of Nonresponse and

Mortality Using Uni- and Multivariate Analysis
Predictor of Nonresponse

Variable OR 95% CI P Value

Univariate analysis
Noradrenaline compared to terlipressin 3.33 1.42-7.82 0.006
High MELD 1.13 1.03-1.23 0.005
HE 2.69 1.09-6.62 0.03
Multivariate analysis
Noradrenaline compared to terlipressin 3.05 1.27-7.33 0.01
High MELD 1.10 1.009-1.200 0.03
Predictor of Mortality
Variable HR 95% CI P Value
Univariate analysis
MELD 1.08 1.03-1.14 0.001
HE 2.06 1.32-3.21 0.001
Pneumonia 1.56 1.92-14.68 0.001
Noradrenaline compared to terlipressin 2.32 1.47-3.66 <0.001
Multivariate analysis
HE 1.88 1.20-2.94 0.002
Noradrenaline compared to terlipressin 2.08 1.32-3.30 0.002
MELD 1.08 1.02-1.13 0.003

compared to 5 (8.3%) in the noradrenaline group

(P = 0.02). Adverse events requiring discontinuation
of the drug were reported in 9 of 60 (15%) patients in
the terlipressin arm compared to 5 of 60 (8.3%) in the
noradrenaline arm (P = 0.39). Adverse events related
and unrelated to drug are shown in Table 4.

FIG. 3. (A) Kaplan-Meier curve comparing patient survival

in terlipressin (solid) group compared to noradrenaline group
(dotted; P = 0.001; ITT analysis). (B) Kaplan-Meier curve
comparing survival in the terlipressin (solid) group compared to
the noradrenaline group (dotted; P < 0.001; PPA).
sepsis-related AKI, there was a significantly better
response to terlipressin than noradrenaline (32.6%
vs. 6.2%; P = 0.03). In non-sepsis-related HRS-AKI,
reversal of HRS was also more frequent in the terli-
pressin group (50% vs. 28.6%), though the difference
did not reach statistical significance (P = 0.08).
ADVERSE EVENTS OF
TERLIPRESSIN AND
NORADRENALINE INFUSION
Adverse events related to the drug were reported
in 14 (23.3%) patients in the terlipressin group as
Discussion
This study compares two vasoconstrictors, namely
terlipressin and noradrenaline, in patients with ACLF
with HRS-AKI. Results show that HRS-AKI is com-
mon and is observed in around one third of patients
of ACLF at presentation, and its incidence increases
over time. Infusion of terlipressin showed earlier and
greater response than noradrenaline in these patients
and resulted in improved survival.
Presence of AKI in the presence of liver disease
is known to have a poor outcome and carries a poor
posttransplant outcome.(23) Mean MELD score in
each of our study groups was higher than 20, indicat-
ing severe hepatic dysfunction in these patients with
cirrhosis with liver failure. It is therefore very import-
ant if we can reverse AKI in this subset of patients by
careful management and make a proportion of them
transplant eligible. Mean age of presentation of ACLF

8
Hepatology, Vol. 0,  No. 0,  2019 ARORA ET AL.

TABLE 4. Adverse Events Related/Unrelated to the Randomized Treatment Group and Serious Event to Cause
Discontinuation of Treatment
Variable Terlipressin Noradrenaline P Value

Diarrhea (more than 4 times/day) 5/60 (8.3%) (2 patients* had more than 6 stools/day) — —
Abdominal pain 1/60* (1.6%) — —
† †
Atrial fibrillation 2/60* (3.3%) — —
Cyanosis 4/60† (6.6%) (3*† patients had persistent cyanosis on decreasing the dose of — —
terlipressin)
Chest pain 2/60 (3.3%) (1patient*† had progressive symptoms) — —
†
Ventricular premature complex (VPCs) 3/60*† (5%) —
Hypertension (MAP >130 mm Hg) 2/60*† (3.3%) —
‡
Pruritus 1/60 (1.6%) —
Asthenia‡ — 1/60 (1.6%)
Dry skin‡ 1/60 (1.6%)
‡
Dyspepsia 2/60 (3.3%) 1/60 (1.6%) 1.00
Hiccups‡ — 1/60 (1.6%)
Cheilitis‡ 1/60 (1.6%) —
Petechiae‡ — 1/60 (1.6%)
Toothache‡ 1/60 (1.6%) —

*Adverse event requiring discontinuation of drug.

†
Serious adverse event (grade 3 or 4).
‡
Adverse effect unrelated to the drug.

in our patient cohort was 41 years, that is, a consid-
erably young population who may benefit from liver
transplant, if kidney functions could be normalized
or else may require simultaneous liver-kidney trans-
plantation, which could be difficult in a resource-con-
strained country like ours.
Our results show that prevalence of AKI at time
of admission in ACLF patients was 26.3%, similar to
reports by Jindal et al.(8) of 22.1% and Garg et al.(24) of
35%. Alcohol was the commonest cause for acute insult
leading to precipitation of ACLF in our study. This
reflects the change in profile of acute insult of ACLF, as
alcohol, denoting the change in the trend from reactiva-
tion of hepatitis B as the common insult to ethanol as
the most common insult in the Indian population.(8,25)
Our study reported a response rate of 40% in rever-
sal of HRS-AKI in the terlipressin arm compared to
the noradrenaline arm, which showed a low response
rate of 16.7% (P = 0.004). Interestingly, these differ-
ences between response rates in the two groups were
distinctly evident as early as day 4 (26.7% vs. 11.7%;
P = 0.03). These observations have a great relevance
in overall management of patients with ACLF. The
results of our study are contrary to the studies done in
the past, including a recent meta-analysis comparing
the response rate of noradrenaline and terlipressin in
patients with DC, which has reported a comparable
response with both the drugs. This is possibly because
we included only patients with ACLF, which is a dis-
tinct group of patients who usually have more-severe
hemodynamic alterations and AKI than those with
DC.(4) Moreover, we used a revised consensus defini-
tion for AKI-HRS, not relying on the absolute value
of sCr, and compared noradrenaline to a continuous
infusion of terlipressin in the present study.
In patients with ACLF, our results demonstrate
superiority of terlipressin over noradrenaline in these
subsets of patients. The revised criteria for HRS-AKI
and administration of terlipressin as continuous infu-
sion could have also contributed to its superior effi-
cacy over noradrenaline.(4,26) Terlipressin as a drug has
several additional attributes, which could contribute to
its additional beneficial effects over noradrenaline. It
reduces portal pressure(27) and inhibits inducible nitric
oxide synthase,(28) which may contribute to its greater
response as compared to noradrenaline. However, in
our study, we did not measure portal pressure and
inducible nitric oxide synthease levels. These param-
eters could be measured in future studies using vaso-
active drugs.
Our results suggest that patients with ACLF who
are sick have a reduced response to terlipressin. These
results are supported by a previous study by Jindal
et al.,(8) who had observed, in a nonrandomized study,
that response to terlipressin is observed in only around
one third of patients with AKI in ACLF.
9
ARORA ET AL.
There was a significant reduction in the require-
ment of RRT in the terlipressin infusion group
compared to noradrenaline infusion (56.6% vs. 80%;
P = 0.006). This is a very important observation of our
study. If, by giving terlipressin, we could prevent pro-
gression to AKIN III requiring RRT or cause reversal
of AKI from AKI III in patients with HRS-AKI, it
would immensely improve the overall management
and outcome of ACLF. This benefit was observed in
our study wherein use of terlipressin over noradrena-
line was an independent predictor of 28-day mortality.
This highlights not only the prognostic relevance, but
also the impact of management of AKI on mortality
in patients with ACLF. Furthermore, it is possible that
the beneficial effects of terlipressin could have been
more pronounced. These could have been masked by
ready availability of RRT in advanced centers. If dial-
ysis was used less frequently, terlipressin would have
been used more, with possibly better response rates.
There was no difference between incidence of sep-
sis- and non-sepsis-related AKI. We observed a better
response with terlipressin as compared to noradrena-
line in sepsis-related AKI (32.6% vs. 6.2%), which was
significant (P = 0.03). A previous study by Rodríguez
et al.(18) showed that early treatment of type 1 HRS
associated with sepsis with terlipressin and albumin is
effective and safe. This could be related to an increase
in the tissue blood flow and tissue microcirculation
and increase in renal flow leading to an increase in
urine output in patients with sepsis.(29)
Similarly, a 28-day survival advantage was observed
with terlipressin compared to noradrenaline (48.3%
vs. 20%; P = 0.001). Our results are in contrast to a
previous meta-analysis,(30) which showed comparable
response between noradrenaline and terlipressin, which
may be attributed to a sicker cohort of patients in our
study. The terlipressin arm had lower mortality as
compared to the noradrenaline group (51.7% vs. 80%;
P = 0.002). Previous studies have also shown a survival
advantage with terlipressin when used as a vasocon-
strictor in critically ill patients with cirrhosis.(31) In
this study, terlipressin was shown to be an effective
vasopressor compared to noradrenaline and offered
a survival advantage over noradrenaline. In another
study, Gluud et al.(32) had reported a 20% reduction in
short-term mortality by use of terlipressin.
Presence of AKI in ACLF carries a high mor-
tality rate, and 2 of the 3 patients in our study suc-
cumbed to this. Previous studies have also reported a
10
Hepatology,  Month 2019
high mortality rate varying from 30% to 77% in the
presence of renal failure and presence of single or
more than one organ failure in ACLF as defined by
APASL-ACLF and the CANONIC study.(1,2,26,28)
High MELD score and use of noradrenaline com-
pared to terlipressin were predictors of nonresponse to
therapy and mortality, as shown in the previous studies.(8)
Severity of liver disease predicts response to therapy, as
shown in previous studies,(8,33) that the more severe the
liver disease, the more the number of extrahepatic organ
failures and higher MELD score, and the lower the
response to either terlipressin or noradrenaline.
Adverse events were more frequent in the terlipres-
sin group compared to the noradrenaline group, but
were reversible after stoppage of the drug. This was
not entirely unexpected. Similar results were noted in
studies by Choudhury et al.(32) and Cavallin et al.(11)
Concerns regarding the adverse events of terlipres-
sin are present, but survival advantage and preserving
kidneys favor usage of terlipressin over noradrenaline.
Incidence of adverse events was 14 of 60 (23.3%) in our
study with continuous infusion of terlipressin, and drug
discontinuation was required in 9 of 60 (15%) patients,
which was less compared to that reported by Cavallin
et al. with a bolus dose of terlipressin (32.4%).(11)
Considering the sick group of patients (mean MELD
>20) and survival advantage of terlipressin over nor-
adrenaline, we recommend using a continuous infu-
sion of terlipressin in HRS-AKI in ACLF patients.
However, it would be worthwhile to initiate further
studies using a lower dosage of terlipressin along
with another vasoconstrictor, such as noradrenaline or
midodrine, to see whether equal or higher efficacy in
reversal of AKI could be achieved with reduced adverse
events. The limitation of our study remains that it was
an open-label study. However, very stringent patient
allocation was done, and strict inclusion and exclusion
criteria were followed to avoid any bias.
To summarize, terlipressin used as an infusion
is superior to noradrenaline in the management of
HRS-AKI in patients with ACLF, and it should be
considered as the first-line therapy in these patients.
Even though adverse events were more frequent with
terlipressin, these were mild and reversible in the
majority of cases with reduction in dose or drug dis-
continuation. Severity of liver disease is an important
determinant of response to therapy, and use of terlip-
ressin is an independent predictor of better response
and clinical outcome in these patients.
Hepatology, Vol. 0,  No. 0,  2019
REFERENCE
1) Moreau R, Jalan R, Gines P, Pavesi M, Angeli P, Cordoba J,
et al. CANONIC Study Investigators of the EASL–CLIF
Consortium. Acute-on-chronic liver failure is a distinct syn-
drome that develops in patients with acute decompensation of
cirrhosis. Gastroenterology 2013;144:1426-1437.
2) Sarin SK, Choudhury A. Acute-on-chronic liver failure: termi-
nology, mechanisms and management. Nat Rev Gastroenterol
Hepatol 2016;13:131-149.
3) Maiwall R, Sarin SK, Moreau R. Acute kidney injury in acute on
chronic liver failure. Hepatol Int 2016;10:245-257.
4) Maiwall R, Kumar S, Sarin SK, Kumar G, Rastogi A, Bihari
C, et al. AKI in patients with acute on chronic liver failure is
different from acute decompensation of cirrhosis. Hepatol Int
2015;9:627-639.
5) Arroyo V, Jalan R. Acute-on-chronic liver failure: defini-
tion, diagnosis, and clinical characteristics. Semin Liver Dis
2016;36:109-116.
6) Davenport A, Sheikh MF, Lamb E, Agarwal B, Jalan R. Acute
kidney injury in acute-on-chronic liver failure: where does hepa-
torenal syndrome fit? Kidney Int 2017;92:1058-1070.
7) Nayak SL, Kumar M, Bihari C, Rastogi A. Bile cast nephropa-
thy in patients with acute kidney injury due to hepatorenal syn-
drome: a postmortem kidney biopsy study. J Clin Transl Hepatol
2017;5:92-100.
8) Jindal A, Bhadoria AS, Maiwall R, Sarin SK. Evaluation of acute
kidney injury and its response to terlipressin in patients with
acute-on-chronic liver failure. Liver Int 2016;36:59-67.
9) Mattos ÂZ, Mattos AA, Ribeiro RA. Terlipressin versus nor-
adrenaline in the treatment of hepatorenal syndrome: systematic
review with meta-analysis and full economic evaluation. Eur J
Gastroenterol Hepatol 2016;28:345-351.
10) Angeli P, Gines P, Wong F, Bernardi M, Boyer TD, Gerbes A,
et al. International Club of Ascites. Diagnosis and management
of acute kidney injury in patients with cirrhosis: revised consen-
sus recommendations of the International Club of Ascites. Gut
2015;64:531-537.
11) Cavallin M, Piano S, Angeli P, Romano A, Fasolato S, Frigo AC,
et al. Terlipressin given by continuous intravenous infusion versus
intravenous boluses in the treatment of hepatorenal syndrome: a
randomized controlled study. Hepatology 2016;63:983-992.
12) Sarin SK, Kedarisetty CK; APASL ACLF Working Party.
Acute-on-chronic liver failure: consensus recommendations of
the Asian Pacific Association for the Study of the Liver (APASL)
2014. Hepatol Int 2014;8:453-471.
13) Wong F. Acute kidney injury in liver cirrhosis: new definition
and application. Clin Mol Hepatol 2016;22:415-422.
14) Riordan SM, Williams R. Treatment of hepatic encephalopathy.
N Engl J Med 1997;337:473-479.
15) Wiest R, Krag A, Gerbes A. Spontaneous bacterial peritonitis:
recent guidelines and beyond. Gut 2012;61:297-310.
16) Sarin SK, Kumar A, Angus PW, Baijal SS, Baik SK, Bayraktar
Y, et al. Diagnosis and management of acute variceal bleeding:
Asian Pacific Association for Study of the Liver recommenda-
tions. Hepatol Int 2011;5:607-624.
17) Sharma P, Kumar A, Sarin SK. An open label, pilot, randomized
controlled trial of noradrenaline versus terlipressin in the treat-
ment of type 1 hepatorenal syndrome and predictors of response.
Am J Gastroenterol 2008;103:1689-1697.
18) Rodríguez E, Elia C, Solà E, Barreto R, Graupera I, Andrealli
A, et al. Terlipressin and albumin for type-1 hepatorenal syn-
drome associated with sepsis. J Hepatol 2014;60:955-961.
ARORA ET AL.
19) Duvoux C, Zanditenas D, Hézode C, Chauvat A, Monin JL,
Thoraval F, et al. Effects of noradrenalin and albumin in patients
with type I hepatorenal syndrome: a pilot study. Hepatology
2002;36:374-380.
20) Velez JC, Nietert PJ. Therapeutic response to vasoconstrictors
in hepatorenal syndrome parallels increase in mean arterial
pressure: a pooled analysis of clinical trials. Am J Kidney Dis
2011;58:928-936.
21) Angeli P, Tonon M, Pilutti C, Morando F, Piano S. Sepsis-
induced acute kidney injury in patients with cirrhosis. Hepatol
Int 2016;10:115-123.
22) Singh V, Ghosh S, Singh B, Kumar P, Sharma N, Bhalla A,
et al. Noradrenaline vs. terlipressin in the treatment of
hepatorenalsyndrome: a randomized study. J Hepatol
2012;56:1293-1298.
23) Wiesen P, Massion PB, Joris J, Detry O, Damas P. Incidence and
risk factors for early renal dysfunction after liver transplantation.
World J Transplant 2016;6:220-232.
24) Garg H, Kumar A, Garg V, Sharma P, Sharma BC, Sarin SK.
Clinical profile and predictors of mortality in patients of acute-
on-chronic liver failure. Dig Liver Dis 2012;44:166-171.
25) Shalimar, Saraswat V, Singh SP, Duseja A, Shukla A, Acharya
SK, et al. Acute-on-chronic liver failure in India: The Indian
National Association for Study of the Liver consortium experi-
ence. J Gastroenterol Hepatol 2016;31:1742-1749.
26) Bajaj JS, O’Leary JG, Reddy KR, Wong F, Biggins SW, Patton
H, et al. Survival in infection-related acute–on–chronic liver
failure is defined by extrahepatic organ failures. Hepatology
2014;60:250-256.
27) Massicotte L, Perrault MA, Denault AY, Klinck JR, Beaulieu
D, Karakiewicz P, et al. Effects of phlebotomy and phenyleph-
rine infusion on portal venous pressure and systemic hemody-
namics during liver transplantation. Transplantation 2010;89:
920-927.
28) Moreau R, Barrière E, Tazi KA, Lardeux B, Dargère D, Lebrec
D, et al. Terlipressin inhibits in vivo aortic iNOS expression
induced by lipopolysaccharide in rats with biliary cirrhosis.
Hepatology 2002;36:1070-1078.
29) Xiao X, Zhang J, Wang Y, Zhou J, Zhu Y, Li T, et al. Effects
of terlipressin on patients with sepsis via improving tissue blood
flow. J Surg Res 2016;200:274-282.
30) Junior AP, Farias AQ , D’ Albuquerque LA, Carrilho FJ,
Malbouisson LM. Terlipressin versus norepinephrine in the
treatment of hepatorenal syndrome: a systematic review and me-
ta-analysis. PLoS One 2014;9:e107466.
31) Choudhury A, Kedarisetty CK, Vashishtha C, Saini D, Kumar
S, Sarin SK, et al. A randomized trial comparing terlipressin and
noradrenaline in patients with cirrhosis and septic shock. Liver
Int 2017;37:552-561.
32) Gluud LL, Kjaer MS, Christensen E. Terlipressin for hepatore-
nal syndrome. Cochrane Database Syst Rev 2012:CD005162.
33) Romano TG, Schmidtbauer I, Silva FM de Q , Pompilio CE,
D’Albuquerque LAC, Macedo E. Role of MELD score and
serum creatinine as prognostic tools for the development of
acute kidney injury after liver transplantation. PLoS One
2013;8:e64089.
Supporting Information
Additional Supporting Information may be found at
onlinelibrary.wiley.com/doi/10.1002/hep.30208/suppinfo. 
11