’Original article

Effects of atorvastatin on portal hemodynamics and

clinical outcomes in patients with cirrhosis with
portal hypertension: a proof-of-concept study
Saptarshi Bishnua, SK.M. Ahammeda, Avik Sarkarb, Jabaranjan Hembramb, Saswata Chatterjeea, Kshaunish Dasc,
Gopal K. Dhalic, Abhijit Chowdhurya and Kausik Dasa

Background and aim Statins can modulate portal microvascular dynamics in patients with cirrhosis. We present data from a
proof-of-concept study aimed at comparing combination of propranolol and atorvastatin versus propranolol alone in reducing
portal pressure in patients with cirrhosis.
Patients and methods In this open-label proof-of-concept study, 23 consecutive patients with cirrhosis were randomized into
group A (incremental dose propranolol, n = 12) or group B (atorvastatin 20 mg daily with propranolol in incremental dose, n = 11).
Hepatic venous pressure gradient (HVPG) was estimated at baseline, and after 30 days, clinical outcomes were evaluated after
1 year.
Results The two groups were matched with respect to etiology of cirrhosis; clinical, biochemical, and endoscopic parameters;
child status; and baseline HVPG. Decreases of wedged hepatic venous pressure, free hepatic venous pressure, and HVPG in
group A and group B after 30 days were 4.67 ± 2.57 versus 6.09 ± 3.56 (P = 0.290), 1.83 ± 2.62 versus 1.27 ± 1.67 (P = 0.546),
and 2.58 ± 1.88 versus 4.81 ± 2.82 mmHg (P = 0.041), respectively. The proportion of HVPG responders in group A and group B
were 50.00 and 90.91%, respectively. The two groups did not, however, differ significantly in terms of clinical outcomes (variceal
bleed, endoscopic variceal ligation sessions, hepatic encephalopathy, requirement of therapeutic paracentesis, spontaneous
bacterial peritonitis, and death).
Conclusion Decrease of HVPG in patients with cirrhosis treated with atorvastatin and propranolol is significantly more than
those treated with only propranolol. Atorvastatin, with its pleiotropic effects, may be useful in portal hypertension in cirrhosis.
Larger data sets are required for ratification. Eur J Gastroenterol Hepatol 00:000–000
Copyright © 2017 Wolters Kluwer Health, Inc. All rights reserved.

Introduction their lipid-lowering effects. They have also been found to

Currently available therapies for portal hypertension, based on
the use of β-adrenergic blockers and organic nitrates, achieve
a favourable portal hemodynamic response in less than 50%
of patients. Moreover, ~ 15% of patients may have contra-
indications and another 15% of patients may not tolerate
β-blockers [1]. This has prompted research of alternative drugs
for long-term treatment of portal hypertension. The ideal drug
in this respect has been postulated to be one that decreases
intrahepatic vascular resistance, maintains or improves hepatic
blood flow, has antifibrotic effects, and is selective to the hepatic
circulation with minimal effects on systemic hemodynamics.
Statins, drugs that inhibit the activity of 3-hydroxyl-
3-methyl coenzyme A reductase, have been long used for
European Journal of Gastroenterology & Hepatology 2017, 00:000–000
Keywords: atorvastatin, cirrhosis, hepatic venous pressure gradient,
portal hypertension, propranolol
Departments of aHepatology, bGI Radiology and cGastroenterology, School of
Digestive and Liver Diseases, Institute of Postgraduate Medical Education and
Research, Kolkata, India
Correspondence to Saptarshi Bishnu, MD, DM, Department of Hepatology,
School of Digestive and Liver Diseases, Institute of Postgraduate Medical
Education and Research, Kolkata 700020, India
Tel: + 91 916 346 5931; fax: + 91 332 223 6383;
e-mail: saptarshibishnu@gmail.com
Received 15 April 2017 Accepted 18 July 2017
have pleiotropic effects, which include improvement in
endothelial function by increasing NO production in
endothelial cells [2]. This last effect is mediated through
activation of protein kinase Akt by the phosphatidylinositol
3-kinase, leading to eNOS phosphorylation at Ser 1177,
with subsequent increased activity [3].
In a study on patients with cirrhosis, Zafra et al. [4]
showed that the use of simvastatin increased hepatic blood
flow by 21%, decreased hepatic sinusoidal resistance by
14%, and significantly increased nitric oxide product
levels in hepatic venous blood but not in peripheral blood.
A randomized controlled trial on the effects of simvastatin
versus placebo in cirrhosis and portal hypertension by
Abraldes and colleagues demonstrated that simvastatin
significantly decreased hepatic venous pressure gradient
(HVPG) (− 8.3%) without deleterious effects on systemic
hemodynamics and increased effective liver perfusion and
function as evidenced from improved hepatic, fractional,
and intrinsic clearance of indocyanine green [4].
Currently, statins are not labeled by the Food and
Drugs Administration for use in portal hypertension;
however, in light of the promising role of statins in portal
hypertension and requirement of more clinical and
experimental data to support this new therapeutic mod-
ality, this present study was designed to investigate the
clinical and hemodynamic effects and 1-year outcomes of

0954-691X Copyright © 2017 Wolters Kluwer Health, Inc. All rights reserved. DOI: 10.1097/MEG.0000000000001006 1

Copyright r 2017 Wolters Kluwer Health, Inc. Unauthorized reproduction of this article is prohibited.
2 European Journal of Gastroenterology & Hepatology
administering statins for treatment of portal hypertension
in patients with cirrhosis. The primary objective was to
compare changes in HVPG in patients with cirrhosis and
portal hypertension after 30 days of using atorvastatin and
propranolol versus propranolol only. The secondary
objectives were to compare the rates of variceal bleeding,
hepatic encephalopathy, ascites requiring therapeutic
paracentesis, spontaneous bacterial peritonitis (SBP),
and mortality in patients treated with combination of
atorvastatin and propranolol, as compared with only
propranolol.
Patients and methods
This prospective randomized single-blind, open-labeled,
proof-of-concept study was conducted at a tertiary care
hospital in India. Eligible participants were all patients at
the Hepatology Outpatient Department aged 18–60 years
with cirrhosis of liver of any etiology along with any evi-
dence of portal hypertension [history of variceal bleeding,
ascites, splenomegaly, esophageal varices on upper gas-
trointestinal (GI) endoscopy, or history of endoscopic
variceal ligation (EVL)]. Recruitment of patients was done
from January 2014 to May 2014. Consecutive patients
with chronic liver disease and portal hypertension, ful-
filling the inclusion and exclusion criteria, were considered
eligible for this study.
Inclusion criteria
The following were the inclusion criteria:
(1) Age: 18–60 years.
(2) Cirrhosis (diagnosed clinically, radiologically, or
histopathologically).
(3) Portal hypertension (history of variceal bleed, ascites,
splenomegaly, esophageal varices on upper GI endo-
scopy, or history of having undergone EVL).
Exclusion criteria
The following were the exclusion criteria:
(1) Child-Pugh-Turcott (CPT) class C.
(2) Hepatic encephalopathy grades II–IV.
(3) Hepatocellular carcinoma.
(4) Portal vein thrombosis or cavernomatosis.
(5) Hepatic venous outflow tract obstruction.
(6) Previous portosystemic shunt surgery.
(7) Obstructive airway diseases.
(8) Cardiac conduction abnormalities.
(9) Peripheral vascular disease.
(10) Congestive cardiac failure NYHA class II–IV.
(11) Renal insufficiency (serum creatinine > 2 mg/dl).
(12) Previous episodes of rhabdomyolysis.
(13) Hypersensitivity to HMG-CoA reductase inhibitors.
(14) Previous treatment with HMG-CoA reductase
inhibitor.
(15) Participation in a concurring clinical trial.
(16) Pregnancy or plan to conceive during study period.
Participants fulfilling the eligibility criteria were inter-
viewed and explained the details of the study, both
Copyright r 2017 Wolters Kluwer Health, Inc. Unauthorized reproduction of this article is prohibited.
Month 2017 • Volume 00 • Number 00
verbally and using patient information brochure in
appropriate regional vernacular or English, as applicable.
Participants were allowed up to 7 days to communicate
with the investigator their decision to enter the study,
following which they gave written informed consent to
enter the study as well as to undergo appropriate invasive
procedures (upper GI endoscopy, HVPG estimation). This
study was conducted according to the principles of Good
Clinical Practice as laid down by the Declaration of
Helsinki [5], and after receiving clearance from
Institutional Ethical Committee.
Data collection was done on printed questionnaire
forms at initial interview and assessment, during sub-
sequent outpatient visits, at scheduled appointments for
hemodynamic studies, at end of study period, and during
any emergency admission within the study period. These
data forms were securely stored in the office of the prin-
cipal investigator, and the information was transferred to
Microsoft Excel worksheet (Microsoft Corporation,
Redmond, Washington, USA) on a password-protected
computer. Each patient was allocated a unique study
identity number and was referred in data analysis work-
sheet by their initials to maintain anonymity.
Data collected at initial interview and baseline assess-
ment included the following:
(1) Clinical: age, sex (male/female), etiology of cirrhosis
(alcohol/hepatitis B virus/hepatitis C virus/others),
history of ascites, history of variceal bleed, history of
hepatic encephalopathy, previous treatment with
diuretics, BMI, resting pulse rate, and noninvasive
blood pressure using aneroid sphygmomanometer.
(2) Laboratory: serum values of bilirubin, alanine transa-
minase, aspartate transaminase, alkaline phosphatase,
albumin, globulin, prothrombin time, creatinine,
urinary pregnancy tests for women, and creatinine
phosphokinase.
(3) CPT score and Model For End-Stage Liver Disease
score.
(4) Endoscopic: presence of large/small/absent esophageal
varices, using Olympus GIF Q150L videoendoscope
(Olympus Corporation, Shinjuku, Tokyo, Japan).
(5) Hemodynamic: hepatic hemodynamic measurements
were done following standard procedural details as
enunciated by Bosch et al. [6]. Wedged hepatic venous
pressure (WHVP), free hepatic venous pressure
(FHVP), and HVPG estimations were performed for
each patient at baseline.
After baseline hepatic hemodynamic study, participants
were allocated into either one of two treatment arms [only
propranolol arm (group A), versus combined atorvastatin
and propranolol arm (group B)] using a block randomization
table generated using an online random number generator
(http://www.sealedenvelope.com, Copyright 2001–2015;
Sealed Envelope Ltd, Clerkenwell Close, London, UK). After
randomization for each arm of the study, the patients were
administered either (i) propranolol orally in a dose of 40 mg
daily or (ii) atorvastatin orally in a dose of 20 mg/day and
propranolol orally in a dose of 40 mg daily. For each
allocation group, the dose of propranolol was increased
depending on resting pulse rate at each monthly follow-up
 Atorvastatin in portal hypertension Bishnu et al. www.eurojgh.com 3

visit until target heart rate of 55–60/min was attained or pulse n = 12 and group B, n = 11) underwent repeat HVPG
rate decreased by more than 25% from baseline. estimation, and clinical outcomes over a follow-up period
Repeat hepatic hemodynamic study was performed and of 1 year were recorded. The flowchart in Fig. 1 schematically
WHVP, FHVP, and HVPG were estimated for each patient depicts the inclusion, exclusion, and follow-up of patients as
after an interval of 30 days from the start of the allocated described previously.
therapy.
The term HVPG responder was used to define any Baseline demographic, clinical, and endoscopic
patient who, on repeat hemodynamic measurement, had characteristics
either at least 20% decrease of HVPG from baseline, or the
The baseline demographic and clinical characteristics of
repeat HVPG had decreased to less than 12 mmHg.
the patients in groups A and B have been depicted in
Data collected in the interim of 1 year of follow-up and
Table 1. The most common etiology of cirrhosis in group
at end of study included the following:
A was ethanol related [six (50.00%)], followed by crypto-
genic [three (25.00%)] and HBV, NASH, and Wilson’s
(1) Number of episodes of portal hypertension-related
disease [one (8.33%)] whereas in group B, the most
bleeding and details, duration of admission, therapy
common etiology was cryptogenic [six (54.44%)] followed
received, and blood component support required.
by ethanol [four (36.36%)] and autoimmune hepatitis [one
(2) Number of episodes of hepatic encephalopathy and
(9.09%)]. The baseline clinical parameters (pulse rate,
details, and duration of admission.
mean arterial pressure, and BMI), baseline laboratory
(3) Number of therapeutic paracenteses required.
parameters (serum levels of bilirubin, aminotransferases,
(4) Number of episodes of SBP.
albumin, creatinine and international normalized ratio),
(5) Need for alternative therapies (shunt surgery and
endoscopic findings, and median CTP score and Model
transjugular intrahepatic portosystemic shunt).
For End-Stage Liver Disease scores did not differ sig-
(6) Number of deaths. Terminal event for each death.
nificantly in the two groups.
Any participant who was undergoing EVL for esopha-
Baseline portal hemodynamic measurements
geal varices at the time of inclusion in the study continued
to undergo scheduled EVL till obliteration of varices in Before allocation of treatment, WHVP, FHVP, and
addition to allocated pharmacotherapy. Moreover, if any HVPG recorded in groups A and B were 29.83 ± 5.29
participant developed any complication of cirrhosis (variceal versus 25.18 ± 5.02 mmHg (P = 0.421), 12.67 ± 4.08
bleeding, hepatic encephalopathy, SBP, sepsis) during the versus 9.63 ± 2.69 mmHg (P = 0.472), and 17.17 ± 6.12
study period, then that participant was admitted and treated versus 15.55 ± 5.30 mmHg (P = 0.503), respectively.
with established therapy in accordance with institutional
protocols. Repeat portal hemodynamic measurements after
Statistical analysis was performed using student t-test receiving allocated treatment for 1 month
(paired and unpaired) for continuous variables and
After 1 month of receiving allocated treatment, the median
Fischer’s exact test for categorical variables. Significance
daily dose of propranolol at second catheterization for
was established at a two-tailed P value of less than 0.05.
both group A and group B was 80 (40–120) mg (P = NS)
Statistical computation was carried out using GraphPad
and repeat portal hemodynamic estimation revealed that
Prism statistical package 6.07 (GraphPad Software Inc.,
WHVP, FHVP, and HVPG recorded in groups A and B
La Jolla, California, USA) and Microsoft Excel software
were 25.17 ± 4.76 versus 19.09 ± 3.70 mmHg (P = 0.002),
package (Microsoft Corporation).
10.83 ± 2.89 versus 8.36 ± 1.63 mmHg (P = 0.020), and
14.58 ± 5.68 versus 10.73 ± 4.03 mmHg (P = 0.073), respec-
Results tively. The change of HVPG values in each individual patient
in both the groups is depicted as spaghetti plot in Fig. 2.
A total of 33 patients with cirrhosis along with portal
Mean HVPG decrease in groups A and B was sig-
hypertension were included in the study after informed
nificantly different at 2.58 ± 1.88 versus 4.81 ± 2.82 mmHg
consent, and their baseline demographic, anthropometric,
(P = 0.041). The mean percentage decrease of HVPG from
clinical, laboratory, and endoscopic data were recorded.
baseline in group A was 15.96 ± 11.79% as compared with
Three patients were subsequently removed from the
30.48 ± 13.90% in group B (P = 0.0131). Further analysis
study before the baseline HVPG estimation (withdrew
revealed that six (50.00%) patients in group A and nine
consent = 1, started propranolol on personal physician
(81.82%) patients in group B had at least 20% reduction
advice = 2). Baseline HVPG estimation was performed in
of HVPG from before treatment (P = 0.193), whereas in
30 patients who were then randomized using a block
four (33.33%) and seven (63.64%) patients of group A
randomization table to receive propranolol alone (group A,
and B, respectively, the repeat HVPG decreased to less
n = 15) or combination of atorvastatin and propranolol
than 12 mmHg (P = 0.220). Overall, there were six
(group B, n = 15). Of the 15 patients in each treatment arm,
(50.00%) HVPG responders in group A and 10 (90.91%)
the repeat HVPG estimation after receiving 30 days of
in group B (P = 0.069) (Fig. 3).
allocated treatment could not be performed in three patients
in group A (lost to follow-up = 1 and second HVPG not
Clinical outcomes over 1-year follow-up
done owing to technical reasons = 2) and four patients in
group B (refused consent for second HVPG estimation = 1, There were five (41.67%) patients in group A and four
lost to follow-up = 1, and second HVPG not done because (36.36%) patients in group B who experienced one or
of technical reasons = 2). The remaining patients (group A, more episodes of variceal bleeding (P = 1.000), and the

Copyright r 2017 Wolters Kluwer Health, Inc. Unauthorized reproduction of this article is prohibited.
4 European Journal of Gastroenterology & Hepatology Month 2017 • Volume 00 • Number 00

Fig. 1. Flowchart depicting inclusion, exclusion, and follow-up of patients during the course of study. HVPG, hepatic venous pressure gradient.

Table 1. Baseline characteristics of two groups

Group A (propranolol only) (n = 12) Group B (propranolol + atorvastatin) (n = 11) P value
Age (mean ± SD) (years) 46.67 ± 7.10 44 ± 12.73 0.549
Males [n (%)] 12 (100.00) 9 (81.82) 0.217
Etiology of cirrhosis [n (%)]
Hepatitis B virus related 1 (8.33) 0 (0.00) –
Hepatitis C virus related 0 (0.00) 0 (0.00) –
Ethanol related 6 (50.00) 4 (36.36) –
Nonalcoholic steatohepatitis related 1 (8.33) 0 (0.00) –
Autoimmune hepatitis related 0 (0.00) 1 (9.09) –
Wilson’s disease related 1 (8.33) 0 (0.00) –
Cryptogenic 3 (25.00) 6 (54.55) –
Presence of ascites in past [n (%)] 5 (41.67) 7 (63.64) 0.414
Presence of ascites at inclusion [n (%)] 6 (50.00) 5 (45.45) 1.000
Use of diuretics in past [n (%)] 4 (33.33) 7 (63.64) 0.220
Use of diuretics at inclusion [n (%)] 5 (41.67) 6 (54.55) 0.684
History of variceal bleed [n (%)] 5 (41.67) 6 (54.55) 0.684
Pulse rate (mean ± SD) (beats/min) 80.83 ± 13.06 76.55 ± 12.64 0.433
Mean arterial pressure (mean ± SD) (mmHg) 92.39 ± 15.27 94.21 ± 9.61 0.734
BMI (mean ± SD) (kg/m2) 22.70 ± 2.64 19.71 ± 3.01 0.020
Laboratory parameters (mean ± SD)
Bilirubin (mg/dl) 1.90 ± 1.82 1.48 ± 0.74 0.477
ALT (IU/ml) 47.58 ± 24.74 35.27 ± 10.03 0.134
AST (IU/ml) 80.08 ± 54.25 49.45 ± 17.63 0.086
ALP (IU/ml) 304.75 ± 183.77 181.45 ± 73.87 0.049
Serum albumin (g/dl) 3.38 ± 0.57 3.73 ± 0.73 0.213
International normalized ratio 1.30 ± 0.16 1.20 ± 0.13 0.097
Serum creatinine (mg/dl) 0.99 ± 0.11 0.97 ± 0.29 0.868
Child-Pugh-Turcot score [median (range)] 6.5 (5–8) 6 (5–9) 0.552
MELD score [median (range)] 11 (8–16) 11 (7–14) 0.538
Esophageal varices [n (%)]
Large 9 (75.00) 9 (81.82) 1.000
Small 3 (25.00) 2 (18.18) –
None 0 (0.00) 0 (0.00) –

ALP, alkaline phosphatase; ALT, alanine transaminase; AST, aspartate transaminase; MELD, Model For End-Stage Liver Disease.

median number of EVL sessions required being four (0–8)
and three (0–6), respectively. Four (33.33%) patients in
group A and two (18.18%) patients in group B experi-
enced one or more episodes of hepatic encephalopathy
(P = 0.640). Four (33.33%) patients in group A and three
(27.27%) patients in group B experienced one or more
episodes of hepatic encephalopathy (P = 1.000). Only
one (8.33%) patient in group A developed SBP during
the study period and one patient also from group A died
because of sepsis during the study period. The clinical
outcomes are summarized in Table 2.
Discussion
This prospective randomized single-blind study demon-
strated that combination of atorvastatin and propranolol
when administered for 1 month resulted in a significant
decrease of HVPG in patients with cirrhosis (Child A/B)

Copyright r 2017 Wolters Kluwer Health, Inc. Unauthorized reproduction of this article is prohibited.
 Atorvastatin in portal hypertension Bishnu et al. www.eurojgh.com 5

Fig. 2. Spaghetti plot showing change of hepatic venous pressure gradient (HVPG) in two groups before and after receiving allocated pharmacotherapy.

Fig. 3. Comparison of hepatic venous pressure gradient (HVPG) responders in two groups.

Table 2. Clinical outcomes over 1-year follow-up

Group A (propranolol only) (n = 12) Group B (propranolol + atorvastatin) (n = 11) P value
Patients having variceal bleed [n (%)] 5 (41.67) 4 (36.36) 1.000
Number of variceal bleed episodes [median (range)] 0 (0–3) 0 (0–1) –
Number of EVL sessions required [median (range)] 4 (0–8) 3 (0–6) –
Patients having one or more episodes of HE [n (%)] 4 (33.33) 2 (18.18) 0.64
Patients requiring therapeutic paracentesis [n (%)] 4 (33.33) 3 (27.27) 1.000
Patients having SBP [n (%)] 1 (8.33) 0 (0.00) 1.000
Deaths [n (%)] 1 (8.33) 0 (0.00) 1.000

EVL, endoscopic variceal ligation; HE, hepatic encephalopathy; SBP, spontaneous bacterial peritonitis.

as compared with only propranolol (4.81 ± 2.82 vs.
2.58 ± 1.88 mmHg). This decrease is more than the HVPG
decrease observed in an earlier study by Abraldes et al. [7],
where administration of simvastatin use for 30 days
was shown to decrease the HVPG from 18.5 ± 7.2 to
17.1 ± 4.6 mmHg.
Although moderate, the magnitude of the HVPG-
lowering effect of atorvastatin in this study has potential
clinical relevance. Portal hypertension-related complica-
tions and mortality have been shown to be reduced sig-
nificantly when HVPG is decreased by at least 20% of
baseline values [8,9]. In the earlier study by Abraldes and
colleagues, 32% of patients treated with simvastatin had
an HVPG reduction of 20% or greater to less than
12 mmHg. In this study, nine of 11 (81.82%) patients
treated with atorvastatin and propranolol had at least
20% reduction of HVPG from baseline [7]. In the same
treatment arm, repeat HVPG decreased to less than
12 mmHg in seven of 11 (63.64%) patients. The total
number of HVPG responders in the combined atorvastatin
and propranolol arm of our study was 10 (90.91%) of 11.
Similar studies on a larger number of patients will enable
us to corroborate this figure.
In earlier studies, use of β-blockers alone has shown to
result in HVPG decrease of 10–15% from baseline
[10–12]. In consonance with this, the mean HVPG decline

Copyright r 2017 Wolters Kluwer Health, Inc. Unauthorized reproduction of this article is prohibited.
6 European Journal of Gastroenterology & Hepatology
from baseline on use of propranolol alone was found to be
16% in this present study. In contrast, combination of
statin with propranolol was found to cause 11.0 ± 14.3%
decrease from baseline HVPG in the study by Abraldes
et al. [7], whereas in this present study, a decrease of
30.48 ± 13.90% from baseline HVPG was observed with
the use of atorvastatin with propranolol.
Considering the findings that both the absolute
(4.81 ± 2.82 vs. 2.58 ± 1.88 mmHg) and relative (16 vs.
30%) decrease of HVPG is less with use of only propra-
nolol, it might be stated that atorvastatin has an additive
role to propranolol in decreasing HVPG. The additive
effect of atorvastatin may enable to augment the expected
10–15% HVPG decrease with propranolol alone to more
than 20% and thus, markedly increase the number of
patients who are effectively protected from the compli-
cations of portal hypertension.
The quanta of HVPG decrease by use of atorvastatin alone
cannot be estimated from this proof-of-concept study, which
was not designed for a head-to-head trial between atorvastatin
and propranolol. An atorvastatin-only arm in the study design
was deemed unethical as it would have entailed certain
patients to not receive propranolol, the current recommended
therapy, in the background of variceal bleeding or presence of
large varices [13].
In this present study, the decrease of HVPG in both the
treatment arms was mainly because of decrease of WHVP,
accompanied by a smaller decrease of FHVP. This con-
trasts with the findings of Abraldes et al. [7] where
decrease of HVPG was mainly because of increase of
FHVP rather than decrease of WHVP. Decrease of WHVP
in this study probably reflects a corresponding vasodila-
tation and decrease of pressure at the sinusoidal level
owing to sinusoidal vasodilation and NO release on use of
atorvastatin. However, definite proof of that would
require simultaneous measurements of portal blood flow,
NO, and eNOS levels in portal circulation, which was
beyond the scope of this study.
Previous studies had not been designed to provide
follow-up clinical data on the incidence of complications
of cirrhosis while on statins. In this present study, the
number of patients having one or more episodes of variceal
bleeds or hepatic encephalopathy did not differ sig-
nificantly between the two study arms. Larger numbers of
study participants are undoubtedly required to con-
clusively answer whether addition of statins to propranolol
lessens the incidence of complications of cirrhosis.
This study was designed as a proof-of-concept study to
study the effects of statins on portal hemodynamics and
hence is limited in its scope. A major limitation of this
present study is the small number of study participants in
each treatment arm. Moreover, the choice of atorvastatin
among all available statins was arbitrary, as was the dose
of atorvastatin used.
This proof-of-concept study has shown in a single-
blind, open-labelled, randomized trial that addition of
atorvastatin to propranolol leads to greater reduction of
portal pressure in patients with cirrhosis and portal
hypertension, most likely through a reduction in hepatic
vascular resistance. Larger studies are required to establish
Copyright r 2017 Wolters Kluwer Health, Inc. Unauthorized reproduction of this article is prohibited.
Month 2017 • Volume 00 • Number 00
whether the HVPG decrease seen in a short time frame
translates, over longer follow-up, into clinical benefit in
terms of preventing the complications of cirrhosis.
Acknowledgements
Saptarshi Bishnu, Abhijit Chowdhury, Gopal K Dhali,
Kshaunish Das were involved in planning of the study;
Saptarshi Bishnu, Avik Sarkar, Saswata Chatterjee,
Jabaranjan Hembram were involved in conducting the
study and performing invasive procedures; Saptarshi
Bishnu, Kausik Das, Kshaunish Das, Saswata Chatterjee
performed interpretation of data, Saptarshi Bishnu, Abhijit
Chowdhury, Gopal K Dhali, Kausik Das were involved in
preparation of manuscript. All the authors have approved
the final draft submitted.
Conflicts of interest
There are no conflicts of interest.
References
1 de Franchis R. Evolving consensus in portal hypertension report of the
Baveno IV consensus workshop on methodology of diagnosis and
therapy in portal hypertension. J Hepatol 2005; 43:167–176.
2 Kalinowski L, Dobrucki L, Brovkovych V, Malinski T. Increased nitric
oxide bioavailability in endothelial cells contributes to the pleiotropic
effect of cerivastatin. Circulation 2002; 105:933–938.
3 Kureishi Y, Luo Z, Shiojima I, Bialik A, Fulton D, Walsh K, et al. The
HMG-CoA reductase inhibitor simvastatin activates the protein kinase
Akt and promotes angiogenesis in normocholesterolemic animals. Nat
Med 2000; 6:1004–1010.
4 Zafra C, Abraldes J, Turnes J, Berzigotti A, Ferbandez M, Bosch J, et al.
Simvastatin enhances hepatic nitric oxide production and decreases the
hepatic vascular tone in patients with cirrhosis. Gastroenterology 2004;
126:749–755.
5 World Medical Association. World Medical Association Declaration of
Helsinki: ethical principles for medical research involving human sub-
jects. JAMA 2013; 310:2191–2194.
6 Bosch J, Garcia-Pagan JC, Berzigotti A, Abraldes JG. Measurement of
portal pressure and its role in the management of chronic liver disease.
Semin Liver Dis 2006; 26:348–362.
7 Abraldes J, Albilos A, Banares R, Turnes J, Gonzales R, Bosch J, et al.
Simvastatin lowers portal pressure in patients with cirrhosis and portal 3.
Hypertension: a randomized controlled trial. Gastroenterology 2009;
136:1651–1658.
8 Abraldes JG, Tarantino I, Turnes J, Garcia-Pagan JC, Rodés J,
Bosch J. Hemodynamic response to pharmacological treatment of
portal hypertension and longterm prognosis of cirrhosis. Hepatology
2003; 37:902–908.
9 D’Amico G, Garcia-Pagan J, Luca A, Bosch J. Hepatic vein pressure
gradient reduction and prevention of variceal bleeding in cirrhosis:
a systematic review. Gastroenterology 2006; 131:1611–1624.
10 Thalheimer U, Bosch J, Burroughs A. How to prevent varices from
bleeding: shades of grey—the case for nonselective beta blockers.
Gastroenterology 2007; 133:2029–2036.
11 Bosch J, Mastai R, Kravetz D, Bruix J, Gaya J, Rigau J, Rodes J. Effects
of propranolol on azygos venous blood flow and hepatic and systemic
hemodynamics in cirrhosis. Hepatology 1984; 4:1200–1205.
12 Banares R, Moitinho E, Matilla A, García-Pagán JC, Lampreave JL,
Piera C, et al. Randomized comparison of long-term carvedilol and
propranolol administration in the treatment of portal hypertension in
cirrhosis. Hepatology 2002; 36:1367–1373.
13 de Francis R. On behalf of the Baveno VI Faculty. Expanding consensus
in portal hypertension report of the Baveno VI Consensus Workshop:
stratifying risk and individualizing care for portal hypertension. J Hepatol
2015; 63:743–752.