LIVER TRANSPLANTATION 15:1508-1513, 2009

ORIGINAL ARTICLE

Endoscopic Variceal Ligation for Primary

Prophylaxis of Esophageal Variceal
Hemorrhage in Pre–Liver Transplant Patients
Eu Jin Lim,1 Paul J. Gow,1 and Peter W. Angus1
1
Department of Gastroenterology and Hepatology, Austin Hospital, Heidelberg, Victoria, Australia

Endoscopic variceal ligation (EVL) is widely used to prevent esophageal variceal bleeding in patients with advanced cirrhosis.
However, the safety and efficacy of EVL in this setting have not been clearly established. This study included 300 adult patients
with cirrhosis on our liver transplant waitlist who underwent upper gastrointestinal endoscopy. Esophageal varices deemed to
be at high risk of bleeding were banded until eradication or transplantation. A retrospective review of patient notes and
endoscopy databases was undertaken, and the number of banding episodes, complications, and patient outcomes were
recorded. Forty-two of 300 patients presented with or had previous variceal bleeding prior to referral and were excluded from
the analysis. Of the remaining 258 patients, 101 underwent a total of 259 banding episodes (2.6 per patient) with a median
follow-up post-banding of 18.4 months per patient (a total of 150 patient years). Failed prophylaxis occurred in 2 patients (2%),
and there were 3 episodes (1.2%) of acute hematemesis from band-induced ulceration. One patient (1%) had mild esophageal
stricturing post-banding without dysphagia. Four of 36 patients (11%) previously found to have moderately sized or larger
varices that were not banded presented with hematemesis due to variceal bleeding and were subsequently banded. None of
the patients that received banding died because of bleeding or failed to receive a transplant as a result of banding
complications. This study shows that in liver transplant candidates, EVL is highly effective in preventing first variceal bleed.
Although banding carries a small risk of band-induced bleeding, this rate is low in comparison with the predicted rate of variceal
bleeding in this population. Liver Transpl 15:1508-1513, 2009. © 2009 AASLD.

Received February 22, 2009; accepted June 23, 2009.

Current guidelines recommend the use of either nonse-
lective beta-blockers or endoscopic variceal ligation
(EVL) to prevent first variceal bleeding in patients at
high risk.1 Nonselective beta-blockers reduce variceal
bleeding2 and overall mortality in patients with cirrho-
sis.3 However, patients with end-stage liver failure often
tolerate therapeutic doses of nonselective beta-blockers
poorly because of the development of symptomatic bra-
dycardia, hypotension, fatigue, or dyspnea. In addition,
about a third of patients will not achieve a reduction in
portal pressures sufficient to prevent variceal bleeding
despite receiving therapeutic doses of beta-blockers.4
Because of these concerns, some units have routinely
employed EVL as primary prophylaxis in preference to
beta-blockers for patients awaiting liver transplanta-
tion who are at high risk of variceal bleeding. EVL has
been shown to be effective in reducing first variceal
bleed5 as well as reducing bleeding-related and all-
cause mortality.6 However, EVL is associated with po-
tential complications such as upper gastrointestinal
bleeding from band-induced ulcers, esophageal perfo-
ration, dysphagia due to scar-induced esophageal stric-
ture, and the development of and bleeding from gastric
varices.
A recent study of patients on a liver transplant wait-
list revealed that 6.5% of 31 patients who underwent
prophylactic variceal banding bled from band-induced
esophageal ulcers, and this resulted in the death of 1
patient.7 However, another similarly sized study of liver
transplant candidates reported no banding-related
bleeding events and complete protection from variceal
bleeding.8 The widely differing results of these 2 studies
may reflect the relatively small patient populations
studied and/or differences in banding experience and

Abbreviations: EVL, endoscopic variceal ligation; MELD, Model for End-Stage Liver Disease; NIEC, North Italian Endoscopic Club.
Address reprint requests to Eu Jin Lim, Department of Gastroenterology and Hepatology, Austin Hospital, 145 Studley Road, Heidelberg, Victoria
3084, Australia. Telephone: 0403 327 841; E-mail: ejlim@rocketmail.com
DOI 10.1002/lt.21857
Published online in Wiley InterScience (www.interscience.wiley.com).

© 2009 American Association for the Study of Liver Diseases.


VARICEAL BANDING IN LIVER TRANSPLANT CANDIDATES 1509
technique. The aim of the current study, therefore, was
to evaluate the safety and efficacy of EVL as primary
prophylaxis of variceal bleeding in a large population of
patients awaiting liver transplantation.
PATIENTS AND METHODS
A retrospective audit was undertaken of adult patients
with chronic liver disease on the Victorian liver trans-
plant waitlist from 1998 to 2008 who had undergone
screening upper gastrointestinal endoscopy as part of
their liver transplant workup. Children (age ⬍ 18) and
patients with fulminant hepatic failure were excluded.
Cirrhosis was diagnosed on liver biopsy histology. Data
on the patient’s age, diagnosis, and severity of liver
disease, as determined by the Child-Pugh score, were
collected.
At endoscopy, if esophageal varices were present,
they were graded from I to IV according to Conn’s clas-
sification.9 The presence of red wale markings, cherry
red spots, gastric varices, and portal hypertensive gas-
tropathy was also recorded. Patients with varices
deemed to be at high risk of bleeding [large varices
(grade IV) or medium varices (grade III) with red wale
marking or cherry red spots] underwent prospective
EVL after informed consent was obtained. Each session
was performed by an endoscopist experienced at EVL.
Either the endoscopist or an anesthetist provided seda-
tion. In each session, 1 band was applied to each col-
umn of varices in the distal 5 cm of the esophagus, and
as many bands as possible were applied (the number
varied from 1 to 7). Initially, a single-band ligation de-
vice was used, with or without the aid of an overtube,
but later a multiple-band instrument became standard.
Acid suppression treatment with a proton pump inhib-
itor was used routinely.
Repeat endoscopy with EVL was performed every 4 to
6 weeks until the esophageal varices were eradicated,
the varices became too small to band, or the patient
underwent liver transplantation. Patients with no vari-
ces or small varices underwent annual screening en-
doscopy. The number of banding episodes and the
number of patients who had a variceal bleed despite
EVL (primary prophylaxis failure) were recorded. Com-
plications from EVL, including gastrointestinal bleed-
ing from band-induced ulcers, dysphagia, esophageal
perforation, enlargement of or bleeding from gastric
varices, and death, were noted. The North Italian En-
doscopic Club (NIEC) score, which combines the sever-
ity of liver disease (by Child-Pugh score), the size of
varices, and the presence of red wale markings to esti-
mate the risk of variceal bleeding without prophylaxis,
was calculated for each patient (see Table 1).10 The
ethics committee of the institution approved the study
protocol.
RESULTS
Three hundred patients with chronic liver disease were
placed on the waiting list for liver transplantation be-
tween 1998 and 2008. Forty-two of the 300 patients
LIVER TRANSPLANTATION.DOI 10.1002/lt. Published on behalf of the American Association for the Study of Liver Diseases
15276473, 2009, 11, Downloaded from https://aasldpubs.onlinelibrary.wiley.com/doi/10.1002/lt.21857 by EBMG ACCESS - GHANA, Wiley Online Library on [27/02/2024]. See the Terms and Conditions (https://onlinelibrary.wiley.com/terms-and-conditions) on Wiley Online Library for rules of use; OA articles are governed by the applicable Creative Commons License
TABLE 1. North Italian Endoscopic Club Score
Variable Points to Add
Child-Pugh class A 6.5
B 13.0
C 19.5
Size of varices Small 8.7
Medium 13.0
Large 19.5
Red Wale markings Absent 3.2
Mild 6.4
Moderate 9.6
Severe 12.8
presented with variceal bleeding or had previous
variceal banding done prior to referral to our unit and
were excluded from the analysis. Two hundred and
fifty-eight patients did not have a history of variceal
bleeding and underwent endoscopic surveillance and
primary prophylaxis performed as outlined previously.
There were 186 males and 72 females with a median
age of 50.1 years (age range of 17-69), and they were
followed up for 275 patient years (mean of 12.7 months
per patient). The etiology of cirrhosis in our patient
cohort is listed in Table 2. Nearly half the patients had
hepatitis C, alcohol, or both as the cause for cirrhosis.
Nineteen patients also had hepatocellular carcinomas
(Table 2). Of the 258 patients, 205 (80%) had varices
(Fig. 1). One hundred seventy-seven patients (68.6%)
received a liver transplant, 23 patients (8.9%) were re-
moved from the waiting list, 22 patients (8.5%) died
waiting for a transplant, and 36 patients (14%) re-
mained on the waiting list.
Outcome in Treated Patients
One hundred and one patients with grade III or IV
varices underwent prophylactic EVL with a total of 259
banding episodes (mean of 2.56 episodes per patient).
These patients were followed up post-banding for a total
of 150 patient years with a median follow-up of 18.4
months per patient. Twenty-five of these patients (25%)
were receiving propranolol therapy at the time of refer-
ral, but only 15 (15%) received presumed therapeutic
doses of 20 mg twice a day or greater. Eighty-one of
these patients (80.2%) had Child-Pugh C cirrhosis, and
the mean Child-Pugh score was 11.2 (Table 3). The
mean Model for End-Stage Liver Disease score was 20.
The mean NIEC index was 37.1, with 50 patients
(49.5%) having an NIEC index of more than 40 (Table 4).
Of the patients that underwent variceal band ligation,
35 (34.7%) had grade IV varices, and 66 (65.3%) had
grade III varices. Thirty-seven patients (36.6%) had
mild portal hypertensive gastropathy, 42 (41.6%) had
moderate portal hypertensive gastropathy, and 5 (5%)
had severe portal hypertensive gastropathy.
There were 2 episodes of variceal bleeding (2% of
patients) despite EVL, and this meant an overall bleed-
ing rate from varices of 1.3% per patient year of fol-
low-up post-banding. Both occurred more than 2 weeks
post-EVL, and no band-induced ulcers were seen at
 15276473, 2009, 11, Downloaded from https://aasldpubs.onlinelibrary.wiley.com/doi/10.1002/lt.21857 by EBMG ACCESS - GHANA, Wiley Online Library on [27/02/2024]. See the Terms and Conditions (https://onlinelibrary.wiley.com/terms-and-conditions) on Wiley Online Library for rules of use; OA articles are governed by the applicable Creative Commons License
1510 LIM, GOW, AND ANGUS

TABLE 2. Etiology of Cirrhosis

Diagnosis Number of Patients Patients with Hepatocellular Carcinoma

Hepatitis C 64 (25.3%) 10
Alcohol 30 (11.8%) 1
Hepatitis C ⫹ alcohol 26 (10.2%) 0
Hepatitis B 27 (10.6%) 4
Nonalcoholic steatohepatitis 13 (5.2%) 2
Cryptogenic cirrhosis 26 (10.2%) 1
Primary sclerosing cholangitis 27 (10.6%) 0
Primary biliary cirrhosis 12 (4.7%) 1
Other 29 (11.4%) 0

at emergent endoscopy. All 3 patients required blood

Figure 1. Patients who underwent endoscopic surveillance
for esophageal varices.
emergent endoscopy. One patient, who had Child-Pugh
B cirrhosis and was not on propranolol, was initially
found to have 4 grade IV varices with red wale markings
(NIEC index of 30) that were banded and was booked to
have repeat endoscopy in 4 weeks but re-presented 3
months later with acute hematemesis. This patient
subsequently underwent 3 further banding sessions
without complications to achieve variceal eradication.
The other patient, who had Child-Pugh C cirrhosis and
was on propranolol (10 mg twice a day), was initially
found to have 3 grade III varices (NIEC index of 35.7)
that were banded and presented at 4 weeks with acute
hematemesis before repeat endoscopy could be done.
This patient subsequently underwent another 2 band-
ing sessions without complications to achieve variceal
eradication. The observed variceal bleeding rate with
EVL intervention, compared to the predicted variceal
bleeding rate without intervention according to the
NIEC, is presented in Table 4.
Complications
Three patients (1.15% of banding episodes) had acute
hematemesis from band-induced ulceration despite
routine proton pump inhibitor use. All 3 patients had
Child-Pugh C cirrhosis with NIEC indices between 42.1
and 46.5. Only 1 of the 3 patients was on propranolol
(at a dose of 20 mg twice a day). Two of the patients had
3 grade III varices with red wale markings, and 1 patient
had 2 grade IV varices with red wale markings. All
bleeding episodes occurred within 2 weeks of the first
banding session, and band-induced ulcers were noted
LIVER TRANSPLANTATION.DOI 10.1002/lt. Published on behalf of the American Association for the Study of Liver Diseases
transfusion, 2 required intensive care unit admission,
and 1 required intubation and inotropes. These pa-
tients subsequently underwent 4 to 6 further banding
sessions without further complications to eventually
eradicate their varices.
One patient (1%) had mild esophageal stricturing
post-banding but no dysphagia. There were no esoph-
ageal perforations. No patient developed gastric varices
post-banding. Eight patients who had gastric varices
underwent EVL of their esophageal varices without
subsequent obvious enlargement of or bleeding from
their gastric varices. None of the patients that received
banding died because of bleeding or failed to receive a
transplant as a result of EVL-associated complications.
Outcome in Untreated Patients
One hundred and fifty-seven patients had endoscopic
surveillance but did not undergo EVL. They had a mean
Child-Pugh score of 10.4 and a mean Model for End-
Stage Liver Disease score of 20. One hundred and four
of these patients had varices: 2 (1.9%) had grade IV
varices, 34 (32.7%) had grade III varices without red
wale markings, and 69 (65.4%) had grade I or II varices.
The 36 patients with grade III or larger varices had a
mean NIEC index of 27.16 (predicted annual bleeding
rate of 22%) and were followed up for 35.6 patient
years. Four of these patients (11%) subsequently pre-
sented with acute hematemesis due to variceal bleeding
and were subsequently banded with a total of 13 band-
ing episodes (mean of 3.25 episodes per patient). None
of these patients had complications relating to EVL.
There were no variceal bleeds among the 68 patients
who had only grade I or II varices or among the 53
patients with no varices. With respect to pharmacolog-
ical therapy, of the 157 patients who were not prophy-
lactically banded, only 20 (12.7%) were on propranolol,
of whom only 11 (7%) were on doses of 20 mg twice a
day or greater.
DISCUSSION
In the absence of prophylaxis, 50% of Child-Pugh C
patients with cirrhosis will suffer a variceal hemorrhage
after 1 year,11 and this is associated with an in-patient
mortality rate of at least 20%.12 Although primary pro-
 15276473, 2009, 11, Downloaded from https://aasldpubs.onlinelibrary.wiley.com/doi/10.1002/lt.21857 by EBMG ACCESS - GHANA, Wiley Online Library on [27/02/2024]. See the Terms and Conditions (https://onlinelibrary.wiley.com/terms-and-conditions) on Wiley Online Library for rules of use; OA articles are governed by the applicable Creative Commons License
VARICEAL BANDING IN LIVER TRANSPLANT CANDIDATES 1511

TABLE 3. Child-Pugh Scores of Candidates for Primary Prophylaxis

Patients Banded Patients Not Banded

Child-Pugh class A 3 (3.0%) 19 (12.1%)
B 17 (16.8%) 41 (26.1%)
C 81 (80.2%) 97 (61.8)
Mean Child-Pugh score 11.2 10.4
Mean MELD Score 20 20

Abbreviation: MELD, Model for End-Stage Liver Disease.

TABLE 4. NIEC Scores in Patients Undergoing Endoscopic Variceal Ligation

Number of Predicted Annual Observed Annual

Risk NIEC Patients Bleeding Risk Bleeding Rate (With
Class Index Banded (Without Intervention) Intervention)
1 ⬍ 20 6 (5.9%) 9.5% 0%
2 20-25 5 (5.0%) 15.8% 0%
3 25.1-30 8 (7.9%) 22.2% 12.5%
4 30.1-35 0 (0%) 32.1% 0%
5 35.1-40 31 (30.7%) 50.0% 2.71%
6 ⬎ 40 50 (49.5%) 63.6% 0%

Abbreviation: NIEC, North Italian Endoscopic Club.

phylaxis with EVL has been shown to reduce the risk of
a first variceal bleed5 as well as bleeding-related and
all-cause mortality,6 there has been very little study
of its safety and efficacy in patients with high-risk
esophageal varices. The patients in this study had a
mean NIEC index of 37.1 with a predicted variceal
bleeding rate of 50% per year. Nearly half of the pa-
tients had an NIEC index greater than 40, which has
a predicted variceal bleeding rate of 63.6% per year
without intervention. However, only 2 patients failed
primary prophylaxis, and this resulted in a yearly
bleeding rate of less than 2% per year. Thus, the
results of the current study show that primary pro-
phylaxis with EVL produces a major reduction in the
expected variceal bleeding rate in patients with ad-
vanced liver disease listed for liver transplantation.
Although 25% of the patients in our series were re-
ceiving propranolol therapy, often in very low doses,
there was no evidence that this conferred any addi-
tional benefit in comparison with EVL alone. As such,
we routinely perform upper gastrointestinal endos-
copy on all patients with cirrhosis referred to our liver
transplant unit, regardless of beta-blocker status, in
order to screen for high-risk varices and perform pri-
mary prophylaxis as required.
There have been 2 recent studies comparing the effi-
cacy and safety of EVL versus beta-blockade in high-
risk patients awaiting liver transplantation.7,8 Both
showed that EVL was highly effective, reducing the
variceal bleeding rates to zero in patients with a pre-
dicted variceal bleeding rate of approximately 30% or
more after 1 year. However, the conclusions reached
regarding the efficacy of beta-blockade in this patient
LIVER TRANSPLANTATION.DOI 10.1002/lt. Published on behalf of the American Association for the Study of Liver Diseases
group were very different between the 2 studies. Unfor-
tunately, both were interrupted before their recruit-
ment targets were reached. Jutabha et al.8 intended to
enroll 104 patients but halted the study after 62 pa-
tients were recruited (31 for EVL versus 31 for propran-
olol) because of significantly increased bleeding and
mortality in the propranolol group. In contrast, Norb-
erto et al.7 interrupted a study with a similar number of
patients because of a very low bleeding rate in both
groups, which made it very unlikely that a significant
difference between the groups could be demonstrated if
the planned recruitment target of 120 patients was
reached.
Consistent with these previous reports, the findings
of the current study, which is the largest reported ex-
perience using EVL as primary prophylaxis in liver
transplant candidates, show that the risk of dying from
variceal bleeding while a patient is waiting for liver
transplantation can be largely eliminated by a program
of aggressive variceal ligation. We chose EVL as the
cornerstone of primary prophylaxis in this high-risk
population for several reasons. The first was the con-
cern that a substantial proportion of patients (between
30% and 50% of patients in different studies) who re-
ceive apparently therapeutic doses of beta-blockers do
not achieve a significant fall in portal pressure and
therefore remain at high risk of bleeding.4 Furthermore,
without monitoring the hepatic venous pressure gradi-
ent,one cannot determine whether beta-blockade has
been effective in lowering portal pressure. We felt that
this was a particular concern in liver transplantation
candidates with end-stage chronic liver disease who
were at high risk of dying following variceal bleeding. In

1512 LIM, GOW, AND ANGUS
contrast, the likely efficacy of EVL is immediately ap-
parent at the time of endoscopy, and further sessions of
banding can be performed to ensure rapid eradication
of varices in those with persistent at-risk varices. A
second concern was that patients with advanced liver
failure are less able to tolerate beta-blockade. Indeed, in
the study by Norberto et al.,7 16% of patients in the
beta-blocker arm were forced to cease therapy because
of intolerance, and some were changed to banding for
prophylaxis. A significant number of patients were able
to tolerate only what many would consider to be sub-
therapeutic doses. Similarly in our study, of patients
who were given beta-blockers, 40% were able to tolerate
only 20 mg or less of propranolol per day.
The very low rate of variceal bleeding in our high-risk
patient group supports the use of EVL as primary pro-
phylaxis in this population. In fact, bleeding rates in
our study and the previous 2 randomized studies in
liver transplant candidates were lower than those in
many other studies of EVL prophylaxis. A recent meta-
analysis of studies comparing EVL with beta-blockade
for the primary prophylaxis of esophageal variceal hem-
orrhage reported that 10% of patients in the EVL arm
developed esophageal variceal bleeding.13 The results
of EVL in our patients may reflect a more aggressive
approach to banding and the fact that it was performed
by a small group of gastroenterologists who are highly
experienced in this procedure.
The current study confirmed that EVL is associated
with a small but significant risk of serious and poten-
tially life-threatening bleeding from band-induced ul-
ceration. The 3 patients who bled from band-induced
ulcers had Child-Pugh C cirrhosis with NIEC indices
of more than 40, which indicated that they were at
very high risk from variceal bleeding in the absence of
prophylaxis. All 3 bled soon after their first banding
session, and this implied that this may be the time of
greatest risk, perhaps because the variceal wall and
esophageal mucosa are most fragile at this time.
These episodes resulted in significant morbidity, with
all 3 patients having severe bleeding with hemody-
namic compromise necessitating blood transfusion.
Two patients required intensive care unit admission,
1 of whom was intubated and required inotrope sup-
port.
These findings are consistent with those of Norberto
et al.,7 who reported 2 episodes of band-induced ulcer
bleeding, 1 of which was fatal. As a result, these inves-
tigators suggested that beta-blockade should be used
in preference to EVL as primary prophylaxis in patients
with advanced cirrhosis. In contrast, Jutabha et al.8
reported no episodes of EVL ulcer bleeding in those
patients randomized to EVL prophylaxis, and they
found that EVL had greater efficacy than beta-blockers
in preventing variceal bleeding. However, this study
appears to be an outlier because ulcer bleeding rates in
our study and that of Norberto et al. are consistent with
the recent meta-analysis of EVL for the prevention of
first variceal bleeding, which found that across 6 stud-
ies, the mean band-induced ulcer-bleeding rate was
3.3%, and there was a 1.9% bleeding-related mortality
LIVER TRANSPLANTATION.DOI 10.1002/lt. Published on behalf of the American Association for the Study of Liver Diseases
15276473, 2009, 11, Downloaded from https://aasldpubs.onlinelibrary.wiley.com/doi/10.1002/lt.21857 by EBMG ACCESS - GHANA, Wiley Online Library on [27/02/2024]. See the Terms and Conditions (https://onlinelibrary.wiley.com/terms-and-conditions) on Wiley Online Library for rules of use; OA articles are governed by the applicable Creative Commons License
rate.13 It should be noted that this meta-analysis re-
viewed trials that included less sick patients (eg, only
14% of patients in Lo et al.’s study14 who underwent
EVL had Child-Pugh C cirrhosis versus 80.2% of our
patients, and only 9.1% of the patients in Lui et al.’s
study15 had grade III/IV varices in their EVL arm ver-
sus 100% of our patients).
In deciding which prophylactic approach should be
preferred in liver transplant candidates, we need to
weigh the major reduction in the risk of esophageal
variceal bleeding achieved with EVL against the small
risk of EVL-induced ulcer bleeding and the ongoing
uncertainty regarding the relative efficacy and tolera-
bility of beta-blockade in this population. The current
study shows that in this population, EVL is highly ef-
fective in preventing first variceal bleed. Prophylactic
EVL does carry a small risk of band-induced bleeding,
which is associated with significant morbidity. How-
ever, the rate of bleeding from band-induced ulcers is
very low in comparison with the predicted rate of
variceal bleeding.
REFERENCES
1. Garcia-Tsao G, Sanyal AJ, Grace ND, Carey WD, for the
Practice Guidelines Committee of the American Associa-
tion for the Study of Liver Diseases and the Practice Pa-
rameters Committee of the American College of Gastroen-
terology. Prevention and management of gastroesophageal
varices and variceal hemorrhage in cirrhosis. Am J Gas-
troenterol 2007;102:2086-2102.
2. D’Amico G, Pagliaro L, Bosch J. Pharmacological treat-
ment of portal hypertension: an evidence-based approach.
Semin Liver Dis 1999;19:475-505.
3. Chen W, Nikolova D, Frederiksen SL, Gluud C. Beta-
blockers reduce mortality in cirrhotic patients with oe-
sophageal varices who have never bled (Cochrane review).
J Hepatol 2004;40(suppl 1):67
4. Garcia-Tsao G, Grace ND, Groszmann RJ, Conn HO, Ber-
mann MM, Patrick MJ, et al. Short-term effects of propran-
olol on portal venous pressure. Hepatology 1986;6:101-
106.
5. Schepke M, Kleber G, Nurnberg D, Willert J, Koch L,
Veltzke-Schlieker W, et al. Ligation versus propranolol for
the primary prophylaxis of variceal bleeding in cirrhosis.
Hepatology 2004;40:65-72.
6. Imperiale, TF, Chalasani, N. A meta-analysis of endo-
scopic variceal ligation for primary prophylaxis of esoph-
ageal variceal bleeding. Hepatology 2001;33:802.
7. Norberto L, Polese L, Cillo U, Grigoletto F, Burroughs AK,
Neri D, et al. A randomized study comparing ligation with
propranolol for primary prophylaxis of variceal bleeding in
candidates for liver transplantation. Liver Transpl 2007;
13:1272-1278.
8. Jutabha R, Jensen DM, Martin P, Savides T, Han SH,
Gornbein J. Randomized study comparing banding and
propranolol to prevent initial variceal hemorrhage in cir-
rhotics with high-risk esophageal varices. Gastroenterol-
ogy 2005;128:870-881.
9. Conn HO. Ammonia tolerance in the diagnosis of esoph-
ageal varices: a comparison of endoscopic, radiologic
and biochemical techniques. J Lab Clin Med 1967;70:
442-451.
10. North Italian Endoscopic Club for the Study and Treat-
ment of Esophageal Varices. Prediction of the first variceal
hemorrhage in patients with cirrhosis of the liver and
 15276473, 2009, 11, Downloaded from https://aasldpubs.onlinelibrary.wiley.com/doi/10.1002/lt.21857 by EBMG ACCESS - GHANA, Wiley Online Library on [27/02/2024]. See the Terms and Conditions (https://onlinelibrary.wiley.com/terms-and-conditions) on Wiley Online Library for rules of use; OA articles are governed by the applicable Creative Commons License
VARICEAL BANDING IN LIVER TRANSPLANT CANDIDATES 1513

esophageal varices: a prospective multicenter study.
N Engl J Med 1988;319:983-989.
11. Pagliaro L, D’Amico G, Pasta L, Politi F, Vizzini G, Traina
M, et al. Portal hypertension in cirrhosis: natural history.
In: Bosch J, Groszmann RJ. Portal Hypertension: Patho-
physiology and Treatment. Oxford, United Kingdom:
Blackwell Scientific; 1994:72-92.
12. Carbonell N, Pauwels A, Serfaty L, Fourdan O, Levy VG,
Poupon R. Improved survival after variceal bleeding in
patients with cirrhosis over the past two decades. Hepa-
tology 2004;40:652-659.
13. Khuroo MS, Khuroo NS, Farahat KLC, Khuroo YS, Sofi AA,
Dahab ST. Meta-analysis: endoscopic variceal ligation for
primary prophylaxis of oesophageal variceal bleeding. Al-
iment Pharmacol Ther 2005;21:347-361.
14. Lo GH, Chen WC, Chen MH, Lin CP, Lo CC, Hsu PI, et al.
Endoscopic ligation vs. nadolol in the prevention of first
variceal bleeding in patients with cirrhosis. Gastrointest
Endosc 2004;59:333-338.
15. Lui HF, Stanley AJ, Forrest EH, Jalan R, Hislop WS, Mills
PR, et al. Primary prophylaxis of variceal hemorrhage: a
randomized controlled trial comparing band ligation, pro-
pranolol, and isosorbide mononitrate. Gastroenterology
2002;123:735-744.

LIVER TRANSPLANTATION.DOI 10.1002/lt. Published on behalf of the American Association for the Study of Liver Diseases